Prohibition of Abuse of Law: A New General Principle of EU Law? 9781472565570, 9781841139388

The Court of Justice has been alluding to ‘abuse and abusive practices’ for more than thirty years, but for a long time

190 21 4MB

English Pages [626] Year 2011

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Prohibition of Abuse of Law: A New General Principle of EU Law?
 9781472565570, 9781841139388

Citation preview

Preface Is there a general principle of European Union law, according to which ‘abuse of law’ is prohibited? The contributions to this volume attempt to provide an answer to this question. They are based on papers delivered at a symposium held at the University of Oxford in October 2008 under the coordinating auspices of the Centre for Business Taxation and the Institute of European and Comparative Law at Oxford. Whilst the Introduction to this volume sets out the research pathway that led to this event – as well as giving an overview of the following chapters – it was clear to us from the outset that this academic gathering should have three key dimensions. First, reflecting the somewhat fragmented research context in this field, the ethos of the event was to be more of a workshop than a conference in the traditional sense. Second, we wanted it to be truly intra-disciplinary, bringing together EU lawyers, tax lawyers and comparative lawyers, with different interests, different perspectives and different areas of expertise. Third, and as a consequence of the preceding characteristics, the event was structured with a view to ensuring a productive discussion and encouraging exchange of ideas between these lawyers from diverse research backgrounds. Over the course of two intense days almost forty papers were presented by invited speakers from across Europe and further afield, each bringing their own specialised knowledge in different areas of EU, comparative and tax law, and all demonstrating remarkable commitment and patience. The event, and this resultant compilation, presented challenge and reward in equal measure. The topicality of our subject became all the more evident after the conference, as the Court of Justice was increasingly called to address questions of abuse. Even as this book went to press, the Court issued three decisions referring to ‘abuse of law’, ‘abuse of rights’, and the ‘principle of prohibiting abusive practices’ within a single day (Case C-303/08 Bozkurt, C-103/09 Weald Leasing, C-277/09 RBS Deutschland – all on 22 December 2010), and it reiterated that this principle applies across various spheres of EU law (Weald Leasing). We hope that the contributions to this volume will prove beneficial in the future development of the emerging general principle of abuse of law. Since this book deals with such a highly dynamic area, the authors had to contend with a long and arduous editorial process. Whilst it has not been possible to include the most recent developments in all of the chapters, in the majority of them the law is stated on the basis of the information available to the authors in mid 2010. In addition, all chapters were amended in order to take into account the changes necessitated by the Treaty of Lisbon that came into force in December 2009. Finally, in a project as long and testing as this one was, there are always many people and institutions to thank. We are particularly grateful to Richard Hart for taking on the risk of publishing a book compilation dealing with such a new area of law. His editorial team, notably Mel Hamill and Tom Adams, were exceptionally efficient and helpful throughout. The Oxford Centre for Business Taxation provided generous financial support and, together with the Institute of European and Comparative Law, assisted with organising the symposium, for which Miguel Poiares Maduro was a superlative

vi Preface chairman. Judith Freedman and Stephen Weatherill were constant presences throughout this whole process, since the initial idea for the symposium started to emerge, and we are grateful for their continuous support. Lindsay Loomer, our copy editor, worked tirelessly on all the chapters, liaising directly with both the authors and the publisher. Philipp Eichenhofer and Ioannis Papathanasiou provided invaluable assistance, namely in bringing the chapters into line with the terminology and the numbering of the Treaty of Lisbon. Our final words, however, are of profound gratitude towards the speakers at the symposium and the authors of the contributions to this volume, who remained patient and faithful to this project throughout the period that it took for this compilation to come to light. It was the outstanding academic standard of their work, and their belief in what we were trying to achieve with this project, which kept us going. Rita de la Feria Stefan Vogenauer Oxford, January 2011

Foreword Several recent high profile cases at the Court of Justice of the EU have made the principle of abuse of rights and/or abuse of law one of the most discussed and relevant current issues of EU law. This is a timely and important book, discussing whether there is an emerging general principle of abuse of law in EU law and its possible implications. Though its profile has been raised by recent cases, the concept of abuse of law has long been present in EU law. That is even more so if one takes into account the many instances where the concept is not referred to as an autonomous principle but guides a constructive interpretation of the applicable rules so as to exclude from their scope circumstances of abuse. This is natural: any legal system needs to protect itself from different forms of attacks which the legislator has not explicitly foreseen. It is a safety valve protecting the integrity of the system. Nevertheless, although one can find it in any legal system, a concept of either abuse of law or abuse of rights remains controversial. This is for two main reasons: first, because of the inherent difficulty in its practical application and the possible negative consequences for legal certainty; and, second, because it is normatively conceived as a restriction to liberty. As Rousseau put it, ‘obedience to the law which one imposes on oneself is freedom’. From this, it is deduced that anything that it is not prohibited by the law is allowed by it. For some such prohibitions need to be clear and precise so as to protect the presumption of freedom. On the other hand, the law is not whatever one can do with its words. To allow that would also promote legal uncertainty but, moreover, it is also challengeable in normative terms. In fact, the prevention of abuses can also be defended by reference to the democratic character of the legislative process. It can be conceived as limiting opportunistic behaviour that, through interpretative manipulation of the existent rules, would derive from the legislation effects which were neither intended nor deliberated in the democratic process. Underlying the debate on the concept of abuse of law is therefore a tension between two sources of legal uncertainty and two conflicting normative concerns. The discussion of the concept of abuse of law in the EU law context has to take into account this background but is made more complex by three factors: First, an important dimension of the EU project is mobility, including the idea that individuals can choose between legal systems. It therefore becomes difficult to distinguish between an evasion of national law (to be considered abusive) and a choice of law that is intended or protected by EU law itself. Second, the EU legal system is frequently ‘incomplete’ and fragmented, with plenty of ambiguous legal terms and legislative gaps. Potentially, reliance on a concept abuse of law may solve some of those problems but it may also contribute to legislative inertia and, in the long run, exacerbate these problems further. Third, the EU legal system operates in the context of different legal cultures. These have developed different concepts and instruments to prevent abuses and protect the integrity of the legal system and these may not always be easy to coordinate and accommodate. The contributions to this book compilation explore these themes whilst, at the same time, testing the application of the principle of abuse of law in specific contexts of EU

viii Foreword law. From these contributions, several questions emerge as to the extent to which the concept is used in the case law of the Court of Justice, how it is used and whether it should be used at all. Is there a uniform approach to abuse in different EU law areas and ought there to be one? What is the right approach: abuse of rights, abuse of law or a constructive interpretation of the legal rules which takes into account the concept of abuse? Should the determination of abuse depend on objective or subjective criteria? What are the different implications of the concept of abuse for legal certainty? What is the right balance between the role to be played by Member States and EU law in the prevention of abuses? And what is the right balance between the role to be played by courts and the legislator in developing such a principle? These and many other questions are addressed in the different contributions to this book. It is a unique and long awaited contribution for a topic that is in urgent need of the kind of serious and in depth analysis that is undertaken in the discussions that follow. Miguel Poiares Maduro Florence, January 2011

Contributors John Armour is Lovells Professor of Law and Finance at the University of Oxford. He is a member of the Oxford-Man Institute for Quantitative Finance and a Research Associate of the European Corporate Governance Institute. Anthony Arnull is Barber Professor of Jurisprudence at the University of Birmingham, where he was Director of the Institute of European Law from 2000 to 2006. He is Consultant Editor of the European Law Review. Catherine Barnard is Professor in European Union Law and Employment Law at the University of Cambridge and a Fellow of Trinity College, Cambridge. She specialises in EU law, employment law and discrimination law. Adrian Briggs is Professor of Private International Law at the University of Oxford, Fellow of St Edmund Hall, Oxford, and a Barrister in private practice. Cathryn Costello is Fellow and Tutor in EU and Public Law at Worcester College, Oxford. She is a member of the Faculty of Law at the University of Oxford. Gilles Cuniberti is Professor of Private International Law and Comparative Law at the University of Luxembourg. Rita de la Feria is Senior Research Fellow at the Oxford University Centre for Business Taxation. She is a faculty member of the Saïd Business School and Associate of the Faculty of Law at the University of Oxford. Michael Dougan is Professor of European Law, and Jean Monnet Chair in EU Law, at the University of Liverpool. Dimitrios Doukas is Lecturer in European Law at the School of Law, Queen’s University Belfast. Ana Paula Dourado is Professor of Tax Law and International and European Tax Law at the Faculty of Law, University of Lisbon and Vice-President of the Institute for Economic, Fiscal and Tax Law of the same faculty. Horst Eidenmüller is Professor of Private Law, German, European and International Company Law at Munich University (Research Professorship) and Visiting Professor at the University of Oxford. Paul Farmer is a Barrister and Partner equivalent in Dorsey & Whitney’s London office, where his practice covers tax litigation and EU regulatory matters. He is Visiting Professor at King’s College London and an International Research Fellow at the Oxford University Centre for Business Taxation.

Contributors  xiii Jonathan Faull is Director-General of Justice, Freedom and Security at the European Commission. He is Professor of Law at the Free University of Brussels (VUB) and the College of Europe. He writes here in a personal capacity. Judith Freedman is KPMG Professor of Taxation Law at the University of Oxford. She is a Fellow of Worcester College, Oxford and Director of Legal Research at the Oxford University Centre for Business Taxation. Julian Ghosh is a QC at Pump Court Tax Chambers, whose practice covers all areas of taxation. He is Visiting Professor at the International Tax Centre, University of Leiden; Lector at Trinity College, Cambridge; and Deputy Special Commissioner. James Gordley is WR Irby Chair in Law at Tulane University Law School. Panos Koutrakos is Professor of European Union Law and Jean Monnet Professor of European Law at the University of Bristol. Michael Lang is Head of the Institute for Austrian and International Tax Law at WU Vienna University of Economics and Business Administration. He is Director of the LLM Program in International Tax Law at this university. Richard Lyal is Legal Adviser to the Commission of the European Communities. Timothy Lyons QC practises at Tax Chambers, Lincoln’s Inn, London. He is also a member of the Irish Bar and a visiting professor at the London School of Economics. Miguel Poiares Maduro is Professor in the Department of European Law at the Robert Schuman Centre for Advanced Studies, European University Institute, Florence. He served as Advocate-General at the European Court of Justice in Luxembourg from 2003 to 2009. Axel Metzger is Professor of Civil Law, Intellectual Property, IT Law and Private International Law at the Law Faculty of Leibniz University, Hannover, Germany. Pasquale Pistone is EURYI–ESF Professor of European and International Tax Law at the WU Vienna and Professor of Tax Law at the University of Salerno. Conor Quigley QC is a Barrister specialising in EU law at Serle Court, Lincoln’s Inn. He is a Research Fellow at the Oxford Institute of European and Comparative Law and an Honorary Research Fellow at Lady Margaret Hall, Oxford. Wolf-Georg Ringe is Lecturer in Law at the University of Oxford. He is a Fellow of Christ Church College, Oxford, and Deputy Director of the Oxford Institute of European and Comparative Law. Pierre Schammo is Lecturer in Law at the University of Manchester, School of Law. Before joining the University of Manchester, he was Research Fellow in European Financial and Corporate Law at the British Institute of International and Comparative Law.

xiv Contributors Edwin Simpson is Barclays Bank Lecturer in the Law of Taxation at the University of Oxford, Official Student and Tutor in Law at Christ Church, Oxford, and a Barrister in private practice. Jukka Snell is Professor of European Law and the Deputy Head of the School of Law at Swansea University. Karsten Engsig Sørensen is Professor and Head of Department at the Department of Business Law at the Aarhus School of Business, Aarhus University. He is also Director of the Centre for International Business Law (CIBL). Eleanor Spaventa is Reader in Law and Director of the Durham European Law Institute at Durham University. Takis Tridimas is the Sir John Lubbock Professor of Banking Law at Queen Mary, University of London. He is also the Nancy A Patterson Distinguished Scholar and Professor at the School of Law, Pennsylvania State University and a Barrister in Matrix Chambers. Frans Vanistendael is Professor Emeritus of the Faculty of Law of the KU Leuven (Belgium) and Director of the European Tax College. John Vella is Senior Research Fellow at the Oxford University Centre for Business Taxation and a member of the Faculty of Law at the University of Oxford. Stefan Vogenauer is Professor of Comparative Law at the University of Oxford and Fellow of Brasenose College, Oxford. He is Director of the Oxford Institute of European and Comparative Law. Stephen Weatherill is Jacques Delors Professor of European Law at the University of Oxford. He is a Fellow of Somerville College, Oxford. Dennis Weber is Professor of European Corporate Tax Law and Director of the Amsterdam Centre for Tax Law (ACTL) of the University of Amsterdam (UvA). Simon Whittaker is Professor of European Comparative Law at the University of Oxford and Fellow and Tutor in Law at St John’s College, Oxford. Katja S Ziegler is Reader in European and Comparative Law at the University of Oxford and Fellow of St Hilda’s College, Oxford.

1 Prohibition of Abuse of (European) Law: The Creation of a New General Principle of EU Law through Tax: A Response Paul Farmer

T

he recently published paper by Rita de la Feria1 provides a comprehensive survey of the development of the case law on abuse of law and of the many issues meriting discussion. It offers an excellent introduction to this volume, and for the most part there is little I can usefully add. I confine myself to a few thoughts which were prompted by Dr de la Feria’s paper. I.  Is abuse of law a general principle of law?

One way to approach this question is to ask whether, and to what extent, the abuse of law concept performs the functions of a general principle of law. It seems clear that it performs some functions, but not all. General principles, such as equality, proportionality, legal certainty and protection of legitimate expectations, fundamental rights and effectiveness, have an interpretative function, but also serve as a criterion for the review of the legality of EU measures. As part of their interpretative function, general principles serve, indirectly, as a means of review of national measures: general principles may assist the Court of Justice or national courts in giving an interpretation of a provision of substantive EU law from which it can be concluded that a national rule is incompatible with that provision. It is important to note, however, that the incompatibility is not with the general principle, but with the substantive provision of European law. All of the abovementioned recognised general principles, however, also serve as a criterion of review of the legality of EU measures. Thus, such a measure may be held to be unlawful because it offends against eg the principle of legal certainty or the principle of proportionality. The concept of abuse of law, like general principles, has an interpretative function. According to the abuse of law concept, provisions of European primary or secondary law are to be interpreted as permitting or requiring the denial of a right where, although the formal conditions laid down are met, the exercise of the right would be abusive. 1  R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395.

4  Paul Farmer Alternatively, in more recent cases on the four freedoms, a restriction may be considered justified by the imperative interest in preventing abuse. Either way, the concept of abuse serves as a principle governing the interpretation of the substantive provision. It does not have an independent life of its own. As yet there has been no case where abuse of law has served as an independent criterion for review of EU measures; and it is not immediately obvious how it could do so. In short, therefore, the abuse of law concept seems to perform some of the functions of recognised general principles but not all of them. It can certainly be described as a general principle of interpretation but does not have all the facets of other recognised general principles. II. Shift from pre-emption to justification

In Emsland-Stärke2 and value-added tax (VAT) cases such as Halifax3 the abuse of law concept operated at the level of determining whether the abusive conduct led to the forfeiture of the rights otherwise conferred by the relevant EU law provision. Similarly, in a number of the earlier cases on the EC Treaty freedoms, the question of abuse arose more often than not in the context of whether the Treaty freedom invoked was applicable at all. However, in some of the earlier cases and certainly in more recent cases the Court has tended to deal with the issue of abuse at the level of justification.4 In ICI5 the Court accepted that a fiscal restriction could in principle be justified by the imperative interest in preventing tax avoidance. At the time of the ICI judgment it was not immediately obvious that there was a direct connection with the concept of abuse of law. In more recent cases, however, such as Cadbury Schweppes6 and Thin Cap7, the concept of preventing abuse seems increasingly to have been folded into the imperative interest defence. It seems clear that the range of situations in which the imperative interest defence of preventing avoidance or abuse may be relied upon goes beyond situations involving what might be termed ‘classic’ abuse, ie cases where the relevant Treaty or other provision is not engaged at all or the rights thereunder are considered forfeited. This may be illustrated by the Thin Cap case. The question whether there was abuse in the classic sense did not arise at all. It was clear that the claimant companies were genuinely established in the United Kingdom. It was clear therefore that Article 43 EC [now Article 49 TFEU] was engaged on the facts of the case. The issue, however, was whether the subsidiaries were so highly geared with debt finance that the restriction imposed by the UK legislation, namely the  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569.  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 4   For a more extensive survey of the cases, see V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law—Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205. 5  Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I-4695. 6  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 7  Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107. 2 3

Prohibition of Abuse of (European) Law: A Response  5 refusal of an interest deduction, could be justified by the need to prevent tax avoidance or abuse. In Cadbury Schweppes the Court was asked to consider whether UK anti-avoidance rules imposing a charge on a UK company in respect of the undistributed profits of its Irish subsidiary on account of the low rate of tax in Ireland were compatible with Article 43 EC. There the Court dealt with abuse initially in the classic sense and ruled that the fact that the subsidiary was established in Ireland for the purposes of benefiting from more favourable tax legislation did not in itself suffice to constitute an abuse of the freedom of establishment. The Court then went on to consider whether the legislation was justified because it was necessary to prevent wholly artificial arrangements. In doing so, the Court drew not only on the previous case law on tax avoidance, but also referred to the principles on abuse articulated in Emsland-Stärke and Halifax. It concluded that a Member State was entitled to impose restrictions only where the taxpayer did not genuinely establish an economic activity but incorporated a purely artificial subsidiary solely with the intent of circumventing UK tax rules. The point about Cadbury Schweppes is that the UK legislation was capable of covering not only abuse in the classic sense, ie cases where there was no genuine establishment, but also other cases where there was genuine establishment but the establishment was tax motivated. The Court was focusing on the legislation rather than the facts of the particular case. In effect what the Court was saying was that the restriction imposed by the UK rules was justified by the requirement to prevent abuse and was proportionate to the extent that it was directed at cases which did not involve any genuine establishment. In other words, the rule was justified insofar as it was targeted at cases of classic abuse. The exact overlap between the two concepts in Cadbury Schweppes is attributable to the fact that, where it applied, the domestic rule negated the establishment for tax purposes. Its application to cases of genuine establishment was necessarily disproportionate. The same was not true of Thin Cap. Pitching the discussion of abuse at the level of justification has the advantage that the Court can introduce the proportionality principle into the equation. Generally, the questions put to the Court are formulated in terms of the legality of the rule (described in abstract), the facts of the case merely providing the background. Fiscal restrictions generally have some underlying justification—the problem is that the legislation is sometimes over-engineered to make life simpler for the tax authorities. The Court’s approach avoids the need to make any distinction between classic cases of abuse where the EC Treaty is not engaged in the first place and cases where, as in Thin Cap, the Treaty is engaged but a restriction may be justified in the interests of preventing abuse. But does it matter that shifting the discussion to the level of justification blurs any such distinction? Are there circumstances in which such a distinction might be meaningful? An area yet to be fully explored is its possible relevance to the consequences of the unlawfulness of domestic provisions. Where legislation is unlawful because it is disproportionate to the interest in preventing abuse and cannot, under the Marleasing8 principle, be read so as to be consistent with the TFEU, it falls to be disapplied. The question arises whether the requirement to disapply the domestic provisions would also arise where, on the facts of the case, there was abuse in the classic sense. The answer must  Case C-106/89 Marleasing SA v La Comercial Internacional de Alimentacion SA [1990] ECR I-4135.

8

6  Paul Farmer surely be ‘No’ if, under the classic concept of abuse, the Treaty provision is not engaged in the first place. It is perhaps less obvious if the Treaty provision is engaged (because there is genuine establishment) but the restriction is justified in the particular circumstances. Where legislation cannot be given a consistent interpretation, does it fall to be disapplied erga omnes or only in cases where its application is disproportionate? For example, if in Thin Cap a restriction were considered justified in the particular circumstances because there was no commercial rationale for the group’s high gearing, then should the taxpayer fail regardless of whether the legislation can be interpreted as including a commercial rationale test? There is certainly a case for saying that the restriction should be disapplied only for those so circumstanced as to render its application disproportionate. On that view, it matters not at what stage of the analysis abuse is dealt with. III. Can Cadbury Schweppes be reconciled with Centros?

A final question which merits discussion in this short response is whether Cadbury Schweppes can be reconciled with the Centros9 decision. Cadbury Schweppes stands for the proposition that a restriction on the freedom of a corporate shareholder resident in the United Kingdom to set up a company in Ireland (consisting, in effect, in the piercing of the corporate veil of the Irish company for tax purposes) may be justified by the need to prevent abuse of the United Kingdom’s tax rules (only) in circumstances where the company is not genuinely established in Ireland (in terms, in particular, of premises, staff and equipment). At first sight it might be thought that Centros is dealing with another issue entirely, namely, the right of a company established in the United Kingdom to operate exclusively through an establishment in Denmark, the home State of the company’s shareholders/directors. Understandably, viewed in those terms, the Court in Centros held that the company had such a right. On closer analysis, however, Cadbury Schweppes and Centros did raise a similar issue. The real issue in Centros—not addressed by the Court—was whether a restriction imposed by Denmark on the ability of Danish-resident founders/shareholders to set up a company in the United Kingdom could be justified in the interests of preventing abuse of Danish company law in circumstances where the company was not genuinely established in the United Kingdom in terms of premises, staff and equipment. True enough, Cadbury Schweppes was concerned with tax rules; Centros was concerned with company rules. Cadbury Schweppes concerned a parent company setting up a subsidiary in another Member State. Centros concerned individuals setting up a company in another Member State. But are those points a material distinction? It is far from clear that the restriction on Danish-resident shareholders in establishing a company in the UK would not be justified by the interest in preventing circumvention of Danish company law in circumstances where there was evidence that the company was formed solely for that purpose and there was no genuine establishment in the UK.

 Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459.

9

2 What is a General Principle of EU Law? ANTHONY ARNULL* I. INTRODUCTION

T

his chapter begins by considering the notion of general principles of law and the special meaning that term has acquired in European Union parlance. It looks at the authority for the application of general principles in the law of the Union and the sources on which the Court of Justice draws in identifying such principles. There is then a brief discussion of their evolution from principles of Community law to principles of Union law. This is followed by an examination of the effect general principles of law may produce. The chapter concludes by examining whether there is, or should be, a general principle in the strict sense of the term prohibiting abuse of law in the Union. II.  THE NOTION OF GENERAL PRINCIPLES

The term ‘general principles’ is deceptively anodyne. At its broadest, it may be used to describe legal standards that are ‘derived from legal rules by way of inductive generalisation’.1 European Union law undoubtedly contains such standards, but there is nothing distinctive about it in this respect, for the same could be said of all developed legal systems. The term could also be used to refer to legal principles or doctrines which are autonomous to the Union’s own legal order. Examples might include primacy, direct effect, national procedural autonomy and institutional balance. The Court of Justice considers these principles inherent in the Treaties. The former are potentially applicable to any situation falling within the scope of the latter. The content of such principles is of considerable interest, but again they are not qualitatively distinct as principles from autonomous national concepts. However, there is another set of principles which are recognised by the Court of Justice, but whose origins lie outside the Union legal order. Examples include legal certainty, proportionality, non-discrimination or equality and respect for fundamental rights. EU lawyers normally use the term ‘general principles’ as a term of art to refer to principles of this nature. Such principles are used mainly to fill gaps and resolve ambiguity in the written law, but they have a significance which is more far-reaching. Their capacity to legitimise Union law and the scope for judicial activism inherent in identifying and   *  I am grateful to those who took part in the discussion at the conference in Oxford on 3 and 4 October 2008. I have tried to take account of some of the points made there in what follows. 1  See A Metzger, ‘Abuse of Law in EU Private Law: A (Re-)Construction from Fragments’, ch 16 below.

8  Anthony Arnull defining them gives the term general principles of law in this technical sense a special resonance.2 III.  AUTHORITY FOR THE APPLICATION OF GENERAL PRINCIPLES

As is well known, the Statute of the International Court of Justice3 requires that Court to apply ‘the general principles of law recognised by civilised nations’. There has never been so clear an injunction in the Union legal order. However, Article 31 of the now-defunct European Coal and Steel Community (ECSC) Treaty required the Court of Justice to ‘ensure that in the interpretation and application of this Treaty, and of rules laid down for the implementation thereof, the law is observed’. That wording was substantially reproduced in what is now Article 19(1) TEU and is generally accepted to embrace something more than the European Treaties themselves. In addition, the grounds on which acts may be declared invalid by the Court of Justice include ‘infringement of the Treaties or of any rule of law relating to their application’.4 It is evident that this formula envisages recourse to rules found outside the Treaties themselves. The second paragraph of Article 340 TFEU is even clearer, requiring the Union to make good damage caused by its institutions or its servants ‘in accordance with the general principles common to the laws of the Member States’.5 So there has always been some authority in the Treaties for the application of general principles of law. Moreover, as Matthias Herdegen observes:6 ‘The Member States, having entrusted the Court with adjudicating certain categories of cases, must be deemed to have endowed the judges with a sufficient body of law to enable them to guarantee effective judicial protection.’ IV. SOURCES OF GENERAL PRINCIPLES

In formulating general principles of Union law, the Court of Justice draws on two main sources: (a) the national laws and constitutional traditions of the Member States; and (b) international treaties signed by the Member States. It was in cases brought under the ECSC Treaty that the Court of Justice first began to turn to the national laws of the Member States to fill gaps in the written law. A wellknown example is Algera,7 a staff case concerning the right of an institution to revoke acts conferring rights on individuals. The Court of Justice observed: 2   Cf AG Trstenjak in Case C-101/08 Audiolux SA and Others v Groupe Bruxelles Lambert SA (GBL) and Others, Bertelsmann AG and Others, Judgment of 15 October 2009, nyr, paras [66–73]. See generally T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press 2006); U Bernitz and J Nergelius (eds), General Principles of European Community Law (The Hague, Kluwer 2000); J Usher, General Principles of EC Law (London, Longman 1998); A Arnull, The General Principles of EEC Law and the Individual (Leicester, Leicester University Press 1990). 3   Art 38(1)(c). 4  See Art 263 TFEU. A similar formula appeared in Art 33 ECSC. 5   The corresponding provision of the ECSC Treaty, Art 40, did not contain any reference to general principles. 6   M Herdegen, ‘The Origins and Development of the General Principles of Community Law’ in Bernitz and Nergelius (n 2 above) 3, 5. There is even authority in national law for recourse to general principles of law: see T Koopmans, ‘General Principles of Law in European and National Systems of Law: A Comparative View’ in Bernitz and Nergelius (n 2 above) 25, 26–31, and Herdegen, ibid, 6, both referring to the principes généraux du droit employed by the French Conseil d’Etat. 7   Joined Cases 7/56 and 3-7/57 Dineke Algera et al v Common Assembly of the European Coal and Steel Community [1957–58] ECR 39, 55.

What is a General Principle of EU Law?  9 The possibility of withdrawing such measures is a problem of administrative law, which is familiar in the case-law and learned writing of all the countries of the Community, but for the solution of which the Treaty does not contain any rules. Unless the Court is to deny justice it is therefore obliged to solve the problem by reference to the rules acknowledged by the legislation, the learned writing and the case-law of the member countries.

The duty of courts to avoid a denial of justice, mentioned in the above quotation, is reflected in the French Civil Code8 and has been described as ‘in itself a general principle of law’.9 It should not be thought that, when the Court of Justice turns to the national laws of the Member States for guidance on the solution to a problem of Union law, it is searching for a consensus or the lowest common denominator. As a Judge of the Court of Justice once explained extrajudicially:10 There is complete agreement that when the Court interprets or supplements Community law on a comparative law basis it is not obliged to take the minimum which the national solutions have in common, or their arithmetic mean or the solution produced by a majority of the legal systems as the basis of its decision. The Court has to weigh up and evaluate the particular problem and search for the ‘best’ and ‘most appropriate’ solution. The best possible solution is the one which meets the specific objectives and basic principles of the Community . . . in the most satisfactory way.

In AM & S v Commission,11 those remarks were cited by Advocate-General Slynn, who went on to explain that: ‘Such a course is followed not to import national laws as such into Community law, but to use it as a means of discovering an unwritten principle of Community law.’ Once a principle inspired by national law is incorporated into Union law, it acquires a life of its own independent of its domestic forebears. Thus, in the CILFIT case,12 the Court of Justice emphasised that legal concepts did not necessarily have the same meaning in Union law as in the national laws of the Member States. In AM & S, the Court of Justice held that the confidentiality of written communications between lawyer and client was in principle protected in proceedings under the competition rules of the EC Treaty. There was no express provision on the subject in Community law, but such protection was generally recognised in the legal systems of the Member States, even if its scope varied. It was also in an ECSC case that the Court of Justice first recognised that international treaties signed by the Member States might constitute a source of general principles. Nold v Commission13 was the third of the triumvirate of early cases in which, under pressure from the German courts, the Court of Justice recognised a general principle of respect     8  See Art 4, which provides: ‘Le juge qui refusera de juger, sous prétexte du silence, de l’obscurité ou de l’insuffisance de la loi, pourra être poursuivi comme coupable de déni de justice.’ The Legifrance website (www. legifrance.gouv.fr/home.jsp) offers the following translation into English: ‘A judge who refuses to give judgment on the pretext of legislation being silent, obscure or insufficient, may be prosecuted for being guilty of a denial of justice.’ That provision, which appears in the Preliminary Title of the Code on the publication, effects and application of statutes in general, was cited by AG Lagrange in Case 8/55 Fédération Charbonnière de Belgique v High Authority [1954–56] ECR 245, 277–78.     9  See Herdegen (n 6 above) 3, 5. 10   H Kutscher, ‘Methods of Interpretation as Seen by a Judge at the Court of Justice’ in Judicial and Academic Conference (Luxembourg, Court of Justice of the European Communities 1976) I‑29. See also AG Lagrange in Case 14/61 Hoogovens v High Authority [1962] ECR 253, 283–84. 11   Case 155/79 AM & S Europe Limited v Commission [1982] ECR 1575, 1649. 12   Case 283/81 Srl CILFIT and Lanificio di Gavardo SpA v Ministry of Health [1982] ECR 3415. 13   Case 4/73 J Nold, Kohlen- und Baustoffgroßhandlung v Commission [1974] ECR 491.

10  Anthony Arnull for fundamental rights. The Court of Justice had acknowledged in an earlier case14 that the protection of such rights was ‘inspired by the constitutional traditions common to the Member States’.15 In Nold, the Court of Justice added that ‘international treaties for the protection of human rights’ signed by the Member States ‘can supply guidelines which should be followed within the framework of Community law’. The applicant in Nold sought to rely on (inter alia) the European Convention on Human Rights,16 which was later described as having ‘special significance’ in this context.17 The breadth of the discretion enjoyed by the Court of Justice in this process will immediately be apparent.18 Although the Advocate-General will sometimes conduct a more or less detailed analysis (often based on a note compiled by the Court’s Research and Documentation Service) of the relevant rules of national or international law, such an analysis rarely appears in the judgment itself.19 This practice may be justified by reasons of practicality or comity, but it may also provoke scepticism about how conscientiously the Court of Justice has actually examined national and international law and expose it to criticism that it is, in reality, pursuing an agenda of its own. V.  FROM COMMUNITY TO UNION LAW

At Maastricht, a reference to the case law of the Court of Justice on the protection of fundamental rights was included in the Treaty on European Union, Article F(2) [later 6(2)] requiring the Union to ‘respect fundamental rights, as guaranteed by the [European Convention on Human Rights] and as they result from the constitutional traditions common to the Member States, as general principles of Community law.’ (Emphasis added.) The drafting of that provision—requiring the Union to respect principles of Community law rather than declaring them to be principles of Union law—was curious. The expression ‘general principles of Community law’ was presumably used because it appeared in the case law of the Court of Justice. Even before the entry into force of the Treaty of Lisbon, however, there were strong grounds for maintaining that all general principles, not just that of respect for fundamental rights, applied wherever the Court had jurisdiction under the Treaties.20 Under what is now Article 2 TEU, the principles on which the Union is founded include that of the rule of law. The general principles of Union law operate typically (though not always exclusively) as principles of administrative law designed to uphold the rule of law: they encompass ‘all essential features of a Rechtsstaat, including the protection of fundamental human rights’.21 Moreover, Article 46 of the pre-Lisbon TEU said that the   Case 11/70 Internationale Handelsgesellschaft v Einfuhr- und Vorratsstelle Getreide [1970] ECR 1125.   Internationale Handelsgesellschaft (n 14 above), para [4]. 16   Cf Case 44/79 Hauer v Land Rheinland-Pfalz [1979] ECR 3727, paras [17–19]; Case 222/84 Johnston v Chief Constable of the Royal Ulster Constabulary [1986] ECR 1651, para [18]. 17  See Case C-260/89 Elliniki Radiophonia Tiléorassi AE (ERT) and Panellinia Omospondia Syllogon Prossopikou v Dimotiki Etairia Pliroforissis and Sotirios Kouvelas and Nicolaos Avdellas et al [1991] ECR I-2925, para [41]; Case C-299/95 Kremzow v Austrian State [1997] ECR I-2629, para [14]; Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat v Council [2008] ECR I-6351, para [283]. 18  See T Hartley, The Foundations of European Community Law, 6th edn (Oxford, Oxford University Press 2007) 133. 19  See eg, the Mangold case (n 42 below). 20  See Tridimas (n 2 above) 36. 21   Herdegen (n 6 above) 4. Cf AG Trstenjak in Audiolux (n 2 above), para [120]. 14 15

What is a General Principle of EU Law?  11 provisions of (inter alia) the EC Treaty concerning the powers of the Court of Justice, and the exercise of those powers, applied to certain provisions of the TEU. They included those of Title VI, the third pillar, over which the Court of Justice exercised some jurisdiction. Article 220 EC [now 19(1) TEU], the provision imposing on the Court a duty to ensure observance of the law, was clearly one which concerned the exercise of its powers. The Court therefore had an obligation to observe that duty when exercising its jurisdiction over third pillar matters. The case law seemed consistent with that line of reasoning. In two actions for damages arising out of an act based partly on the third pillar, Gestoras Pro Amnistía22 and Segi,23 the Court said that the institutions and the Member States when implementing the law of the Union were ‘subject to review of the conformity of their acts with the treaties and the general principles of law’. In Advocaten voor de Wereld,24 a reference under Article 35 of the preLisbon TEU on the validity of the framework decision on the European arrest warrant,25 the Court said that the general principles of law included ‘the principle of the legality of criminal offences and penalties and the principle of equality and non-discrimination’. It proceeded to test the validity of the contested framework decision against those principles (concluding that no factor capable of affecting its validity had emerged). The Union-wide scope of the general principles of law was underlined by the Treaty of Lisbon, which entered into force on 1 December 2009. Article 1 TEU, as amended, provides that ‘[t]he Union shall replace and succeed the European Community.’ The EC Treaty was accordingly renamed the Treaty on the Functioning of the European Union.26 The disappearance of the Community as a legal entity distinct from the Union is reflected in the new Article 6(3) [ex 6(2)] TEU, which refers to ‘general principles of the Union’s law’. The implication that such principles apply in all fields coming within the Union’s jurisdiction was confirmed in Kücükdevici v Swedex, where the Court of Justice used the expression ‘general principle of European Union law’.27 The Kücükdevici case is discussed in more detail below. 22   Case C-354/04 P Gestoras Pro Amnistía, Juan Mari Olano Olano and Julen Zelarain Errasti v Council of the European Union [2007] ECR I-1579, para [51]. 23   Case C-355/04 P Segi, Araitz Zubimendi Izaga and Aritza Galarraga v Council of the European Union [2007] ECR I-1657, para [51]. 24   Case C-303/05 Advocaten voor de Wereld VZW v Leden van de Ministerraad [2007] ECR I-3633. 25   Framework Decision 2002/584/JHA of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States [2002] OJ L190/1. 26   The TEU and the TFEU ‘have the same legal value’: Art 1 TEU. 27   Case C-555/07 Seda Kücükdeveci v Swedex GmbH & Co. KG, judgment of 19 January 2010, nyr, paras [21], [27] and [50]. However, the Court’s usage since the entry into force of the Treaty of Lisbon has not been consistent. In some cases it has continued to refer to general principles of Community law: see eg, Case C-399/06 P Faraj Hassan v Council of the European Union and European Commission and Case C-403/06 P Chafiq Ayadi v Council of the European Union [2009] ECR I-11393; Case C-376/08 Serrantoni Srl, Consorzio stabile edili Scrl v Comune di Milano, judgment of 23 December 2009, nyr, para [33]; Case C-403/09 PPU Jasna Detiˇcek v Maurizio Sgueglia, judgment of 23 December 2009, nyr, para [34]. The General Court (formerly the Court of First Instance) also continued to use that term: see eg, Case T-27/09 Stella Kunststofftechnik v OHIM [2009] ECR II-4481. In other cases the Court of Justice spoke of general principles of law tout court (see eg, Case C-45/08 Spector Photo Group NV, Chris Van Raemdonck v Commissie voor het Bank-, Financieen Assurantiewezen, judgment of 23 December 2009, nyr, para [40]) or a general principle ‘inherent in the Community legal order’ (see eg, Case C-284/05 European Commission v Republic of Finland [2009] ECR I-11705; Case C-387/05 European Commission v Italian Republic, judgment of 15 December 2009, nyr, para [57]; Case C-239/06 European Commission v Italian Republic, judgment of 15 December 2009, nyr, para [57]). It is submitted that these variations in terminology are transitional only and should not be taken to imply that general principles apply only in areas that once fell within the scope of the first (or Community) pillar of the Union.

12  Anthony Arnull

VI.  WHAT EFFECT MAY GENERAL PRINCIPLES OF THE UNION’S LAW PRODUCE?

A.  Relationship with Treaty Provisions General principles belong to the primary law of the Union and are described by the Court of Justice as having ‘constitutional status’.28 Although general principles cannot be used to challenge the validity of provisions contained in the Treaties, they may affect the way such provisions are interpreted. Indeed, they may sometimes lead the Court of Justice to construe the Treaties contra legem. An example is Foto-Frost,29 where the principle of legal certainty was cited by the Court in concluding that national courts lacked jurisdiction to declare Community acts invalid, notwithstanding the apparently contrary terms of Article 234 EC [now 267 TFEU]. Other examples are Les Verts30 and Chernobyl,31 where the Court’s conclusion that the European Parliament could both defend and bring annulment proceedings under the EC Treaty, notwithstanding the then absence of any reference to that institution in the relevant provision,32 was based on the rule of law. B.  Effect on the Institutions and the Member States General principles are binding on the institutions of the Union. They may therefore affect the validity and interpretation of Union acts and may (as Article 340 TFEU makes clear) be invoked in actions for damages against the Union. General principles also bind the Member States when applying Union law. They may be invoked in infringement actions brought by the European Commission and are capable of affecting the interpretation (and indirectly the validity) of provisions of national law. This may both limit and enlarge the discretion of national courts. The principle of consistent interpretation laid down in the case law of the Court of Justice requires national courts, where possible, to interpret rules of national law in a way which gives effect to the objectives of overlapping rules of Union law.33 However, that duty is qualified by the general principles of Union law, particularly the principles of legal certainty and non-retroactivity. It cannot therefore have the effect of determining or aggravating a person’s criminal liability.34

28  See Audiolux and Others (n 2 above), para [63]; Case C-174/08 NCC Construction Danmark A/S v Skatteministeriet [2009] ECR I-10567. Cf Kadi (n 17 above), paras [307-8]; Tridimas (n 2 above) 50–51. 29   Case 314/85 Foto-Frost v Hauptzollamt Lübeck-Ost [1987] ECR 4199, para [15]. 30   Case 294/83 Parti écologiste ‘Les Verts’ v European Parliament [1986] ECR 1339. 31   Case 70/88 European Parliament v Council of the European Communities [1990] ECR I-2041. 32   At Maastricht, the EC Treaty was amended to give effect to the Court’s decisions. See now Art 263 TFEU. 33   The duty of consistent interpretation does not require national courts to interpret domestic provisions in a manner which is contra legem. For a useful summary of the case law, see Joined Cases C-378/07 and C-380/07 Angelidaki and Others v Nomarkhiaki Aftodiikisi Rethimnis and Another, judgment of 23 April 2009, paras [197-201]. 34  See eg, Case 80/86 Kolpinghuis Nijmegen [1987] ECR 3969, paras [13–14]; Joined Cases C-387/02, C-391/02 and C-403/02 Berlusconi and Others [2005] ECR I-3565, para [78].

What is a General Principle of EU Law?  13 C.  Direct Effect Can general principles of Union law have direct effect, that is, confer rights on individuals that national courts are bound to protect? That issue has normally arisen in the context of written provisions of Union law. The test for direct effect in the case of such provisions is essentially whether they are unconditional and sufficiently precise to be suitable for judicial application.35 The general principles of Union law are regularly applied by the Court of Justice (and the General Court—formerly the Court of First Instance) and sometimes produce a decisive effect on the outcome of a case. It is therefore hard to deny that at least some of them are, or have become, sufficiently precise to be applied by a court.36 Examples would include legal certainty,37 legitimate expectations,38 proportionality39 and fundamental rights.40 Indeed, the Court of Justice sometimes expressly leaves the application of general principles to the national courts.41 However, they do not normally constitute an independent source of rights for individuals but rather help to clarify the scope of rights conferred on them by written provisions of Union law. i. The Mangold Case The question whether general principles could confer rights directly on individuals was raised in Mangold,42 which concerned the interpretation of Directive 2000/78 on equal treatment in employment and occupation.43 That Directive was based on Article 13 EC [now 19 TFEU]. Although it did not itself prohibit any form of discrimination, Article 13 empowered the Council, within certain limits, to ‘take appropriate action to combat discrimination based on sex, racial or ethnic origin, religion or belief, disability, age or sexual orientation’. Directive 2000/78 laid down a general framework for combating discrimination on some of those grounds, including age. In Mangold, the Court was asked for guidance on the effect of the Directive in the context of proceedings against a lawyer brought by one of his employees, who claimed he had been discriminated against because of his age. It seemed impossible for the employee to rely on the Directive for two reasons. First, the dispute was a horizontal one; secondly, 35  See eg, Case 43/75 Defrenne v SABENA [1976] ECR 455 (Treaty article); Case 8/81 Becker [1982] ECR 53 (Directive). 36   Ronald Dworkin, who is sometimes cited in this context (see n 39 of the Opinion of AG Sharpston in Bartsch (n 60 below); Tridimas (n 2 above) 2), recognises that principles are regularly applied by courts: see R Dworkin, Taking Rights Seriously (London, Duckworth 1977) 23–30. 37  See eg, Case 314/85 Foto-Frost v Hauptzollamt Lübeck-Ost [1987] ECR 4199; Defrenne (n 35 above). 38  See eg, Case 120/86 Mulder v Minister van Landbouw en Visserij [1988] ECR 2321; Case C-189/89 Spagl [1990] ECR I-4539. 39  See eg, Case 261/81 Rau v De Smedt [1982] ECR 3961. 40  See eg, Joined Cases 46/87 and 227/88 Hoechst v Commission [1989] ECR 2859; Case 374/87 Orkem v Commission [1989] ECR 3283; Case C-94/00 Roquette Frères v Commission [2002] ECR I-9011; Case C-404/92 P X v Commission [1994] ECR I-4737; Case 222/84 Johnston v Chief Constable of the Royal Ulster Constabulary [1986] ECR 1651; Case 36/75 Rutili v Minister for the Interior [1975] ECR 1219; Case C-60/00 Carpenter v Secretary of State for the Home Department [2002] ECR I-6279. 41  See eg, Case 145/88 Torfaen Borough Council v B & Q plc [1989] ECR 3851; Case C-368/95 Familiapress v Bauer Verlag [1997] ECR I-3689. 42   Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9981. 43   Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation [2000] OJ L303/16.

14  Anthony Arnull the deadline for implementation of the Directive in the Member State concerned44 had not expired at the material time. It is well established that directives cannot produce direct effect until after the deadline for implementation has passed and that even then they may not produce horizontal direct effect. Having found that the provisions of national law at issue in the main action could not be justified under Directive 2000/78, the Court sought to overcome both those obstacles. The source of the principle of equal treatment in the field of employment and occupation, it said, was not the Directive itself but ‘various international instruments’ and ‘the constitutional traditions common to the Member States’.45 The principle of nondiscrimination on grounds of age therefore had to be considered a general principle of Community law. Observance of that principle could not be conditional on the expiry of the deadline for implementing the Directive. On the contrary, it was ‘the responsibility of the national court to guarantee the full effectiveness of the general principle of nondiscrimination in respect of age, setting aside any provision of national law which may conflict with Community law’.46 ii.  Responses to the Mangold Case The Court of Justice did not say in Mangold that it was abandoning its case law on the direct effect of directives and the decision proved highly controversial. Its application in the German courts led to a complaint to the Bundesverfassungsgericht.47 It attracted criticism from academic writers48 as well as from Advocates-General in other cases. In Chacón Navas,49 Advocate-General Geelhoed argued that ‘the Court must respect the choices made by the Community legislature in the rules on the application of Article 13 EC . . . and must not stretch them . . . There is even less room, in my view, for widening the scope of Article 13 EC by relying on the general principle of equality.’50 He advocated ‘a more restrained interpretation and application of Directive 2000/78 than adopted by the Court in the Mangold case’.51 In Palacios de la Villa,52 the Court of Justice was asked for a preliminary ruling on the interpretation of Article 13 EC [now Article 19 TFEU] and Directive 2000/78. AdvocateGeneral Mazák pointed out53 that, although various international instruments and 44  Germany, which had, in accordance with Art 18 of the Directive, asked for an additional period to implement its provisions on age and disability discrimination. 45   Mangold (n 42 above), para [74]. 46   Mangold (n 42 above), para [78]. See also Case C-246/06 Josefa Velasco Navarro v Fondo de Garantía Salarial (Fogasa) [2008] ECR I-105, a vertical case where the fourth chamber of the Court held, without an Opinion from its Advocate-General, that, in the period between the entry into force of a directive and the expiry of the deadline for its transposition, national rules falling within its scope had to be applied in accordance with the general principles and fundamental rights recognised in the Community legal order. Cf Case C-442/00 Rodríguez Caballero [2002] ECR I-11915. 47  See R Herzog and L Gerken, ‘Stop the European Court of Justice’ (Centre for European Policy, published in EUobserver, 10 September 2008 (in English) and Frankfurter Allgemeine Zeitung, 8 September 2008 (in German)). See judgment of 6 July 2010. 48  See eg, A Dashwood, ‘From Van Duyn to Mangold via Marshall: Reducing Direct Effect to Absurdity?’ (2006–07) 9 Cambridge Yearbook of European Legal Studies 81; Editorial Comments (2006) 43 Common Market Law Review 1; A Arnull, ‘Out With The Old . . . ’ (2006) 31 European Law Review 1. 49   Case C-13/05 Sonia Chacón Navas v Eurest Colectividades SA [2006] ECR I-6467. 50   AG Geelhoed in Chacón Navas (n 49 above), para [53]. 51   AG Geelhoed in Chacón Navas (n 49 above), para [56]. 52   Case C-411/05 Félix Palacios de la Villa v Cortefiel Servicios SA [2007] ECR I-8531. 53  AG Mazák in Palacios de la Villa (n 52 above), para [88].

What is a General Principle of EU Law?  15 national constitutions enshrined the general principle of equal treatment, few referred specifically to non-discrimination on grounds of age. While it was correct to state that prohibitions of discrimination on specific grounds constituted particular expressions of the general principle of equality,54 ‘to infer . . . from the general principle of equality, the existence of a prohibition of discrimination on a specific ground is quite different and far from compelling’.55 He did not think that either Article 13 EC or Directive 2000/78 necessarily reflected a pre-existing prohibition of all the forms of discrimination they mentioned. He added: ‘[I]f the reasoning in Mangold were followed to its logical conclusion, not only prohibition on grounds of age, but all specific prohibitions of the types of discrimination referred to in . . . Directive 2000/78 would have to be regarded as general principles of Community law.’56 He concluded that the limits to the capacity of directives to produce direct effect should not be ‘undermined by recourse to a general principle’. This would ‘raise serious concerns in relation to legal certainty’ and ‘call into question the distribution of competences between the Community and the Member States, and the attribution of powers under the [EC] Treaty in general’.57 iii.  Case Law Following Mangold The Court of Justice itself was initially rather coy about mentioning Mangold or the general principle of equality. In Chacón Navas, the Court held that Directive 2000/78 did not protect an employee dismissed solely on account of sickness. The Directive was not to be extended beyond the grounds specifically laid down. There was no suggestion that the general principle of equality in itself had been infringed or that it could offer a remedy over and above the Directive. In Palacios de la Villa, the Court found that there had been no infringement of Directive 2000/78 and that there was no need to rule on the interpretation of Article 13 EC. It made no reference to the general principle of equality.58 In Lindorfer v Council,59 the oral procedure was reopened (after Advocate-General Jacobs had given an Opinion) so that observations could be presented to the Court on (inter alia) the implications of the Mangold case, which had in the meantime been decided. A second Advocate-General’s Opinion (this time by Advocate-General Sharpston) was then delivered, but Mangold was not mentioned in the Court’s judgment.

54   The Court has frequently made that claim in relation to specific prohibitions of discrimination: see eg, Joined Cases 201 and 202/85 Klensch v Secrétaire d’Etat [1986] ECR 3477; Case C-13/94 P v S and Cornwall County Council [1996] ECR I-2143. 55  AG Mazák in Palacios de la Villa (n 52 above), para [94]. 56  AG Mazák in Palacios de la Villa (n 52 above), para [96]. Cf Case C-249/96 Grant v South-West Trains [1998] ECR I-621; and Joined Cases C-122/99 P and C-125/99 P D and Sweden v Council [2001] ECR I-4319, the facts of which occurred before Dir 2000/78 was adopted, where the Court refused to treat discrimination on grounds of sexual orientation as prohibited under Community law. In Case C-267/06 Tadao Maruko v Versorgungsanstalt der deutschen Bühnen [2008] ECR I-1757, a reference on the interpretation of Dir 2000/78, AG Ruiz-Jarabo Colomer none the less claimed that the fundamental nature of the principle of non-discrimination on grounds of sexual orientation gave it ‘a different status from the one attributed to the prohibition of discrimination on grounds of age in [Mangold]’ (n 82 of his Opinion). 57  AG Mazák in Palacios de la Villa (n 52 above), para [138]. 58  See L Waddington, ‘Case C-411/05, Félix Palacios de la Villa v Cortefi el Servicios SA, Judgment of the Court (Grand Chamber) of 16 October 2007’ (2008) 45 Common Market Law Review 895; C Canazza, ‘Arrêt Palacios: la Cour tempère l’interdiction des discriminations fondées sur l’âge’ [2008] Journal de Droit Européen n ° 147, 79. 59   Case C-227/04 P Maria-Luise Lindorfer v Council of the European Union [2007] ECR I-6767.

16  Anthony Arnull The Court was unable to keep its head buried in the sand in Bartsch,60 where the Bundesarbeitsgericht asked the Court a series of questions on the effect of the general principle prohibiting age discrimination. Following the advice of Advocate-General Sharpston,61 the Court held that the Community law prohibition of age discrimination did not have to be applied in cases containing no link with Community law. Such a link could not simply be provided by Article 13 EC or by Directive 2000/78 where the deadline for its implementation had not expired. Whereas the rules at issue in Mangold had been enacted by the State to give effect to an earlier Directive,62 those at issue in Bartsch63 had not been introduced to implement Community law. The latter case therefore fell outside its scope. Bartsch seemed to mark a retreat from Mangold, for the Court is not usually as squeamish about the scope of Union law. A striking contrast is provided by Forcheri,64 where the Court held on the basis of slender evidence that access to vocational training fell within the scope of the EEC Treaty as it then stood for the purposes of its Article 7 [later 12 EC, now 18 TFEU], prohibiting discrimination on grounds of nationality. If Bartsch was a retreat, however, it was not a wholesale one. The implication that the outcome would have been different if the deadline for the transposition of Directive 2008/78 had expired was confirmed in Kücükdevici v Swedex, where that condition was satisfied. In those circumstances, the Grand Chamber of the Court held that the principle of non-discrimination on grounds of age, as given expression in Directive 2000/78, required inconsistent provisions of national law to be disapplied.65 Mangold suggests that national provisions may also be brought within the scope of Union law for these purposes where they give effect to other Union measures, at least where those measures have some substantive link with acts based on Article 19 TFEU. Where do Mangold and Kücükdevici leave the direct effect of general principles of Union law? The possibility that some such principles might have direct effect was acknowledged in Palacios de la Villa by Advocate-General Mazák, who referred to ‘fully fledged norms like fundamental rights or the highly developed body of Community principles of sound administration and procedure’.66 In Bartsch, Advocate-General Sharpston described it as ‘trite Community law that general principles of law are capable of being invoked vertically against the State’.67 She added that ‘one should be slow to exclude the possibility that a general principle of Community law may, in appropriate circumstances, be applied 60   Case C-427/06 Birgit Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH [2008] ECR I-7245. 61   AG Sharpston in Bartsch (n 60 above), in particular paras [69–75]. 62   Council Directive 1999/70/EC of 28 June 1999 concerning the framework agreement on fixed-term work concluded by ETUC, UNICE and CEEP [1999] OJ L175/43. That Directive was based on Art 139(2) EC [now 155(2) TFEU], the English text of which used the term ‘decision’. This was interpreted broadly as embracing any of the binding acts referred to in Art 249 EC [now 288 TFEU]: see C Barnard, EC Employment Law, 3rd edn (Oxford, Oxford University Press 2006) 92–93. 63  Guidelines relating to retirement pensions adopted by the employer of the claimant’s deceased husband. 64   Case 152/82 Sandro Forcheri and his wife Marisa Forcheri, née Marino, v Belgian State and asbl Institut Supérieur de Sciences Humaines Appliquées - Ecole Ouvrière Supérieure [1983] ECR 2323. See also Case 293/83 Gravier v City of Liège [1985] ECR 593; Case 186/87 Cowan v Trésor Public [1989] ECR 195; Case C-85/96 Martínez Sala v Freistaat Bayern [1998] ECR I-2691; Case C-281/98 Angonese v Cassa di Risparmio di Bolzano SpA [2000] ECR I-4139. 65  See Case C-555/07 (n 27 above). Presumably the same would be true of all cases falling within the scope of measures based on Art 19 TFEU. These do not have to take the form of directives: Art 19(1) refers to ‘appropriate action’. 66  AG Mazák in Palacios de la Villa (n 52 above), para [134]. 67   AG Sharpston in Bartsch (n 60 above), para [79].

What is a General Principle of EU Law?  17 horizontally’.68 However, she resisted describing general principles as directly effective on the basis that ‘a general principle of Community law is applied to a set of legislative rules and affects the interpretation to be given to them . . . the general principle does not, as such, act as a substitute for an existing legislative text.’69 The question raised in Audiolux70 was whether Community law recognized a general principle of equality of shareholders which protected a company’s minority shareholders. Advocate-General Trstenjak concluded that no such general principle was known to Community law. Because of the possibility that the Court might take a different view, she went on to consider whether any such principle might bind one group of shareholders in its relations with other shareholders. She pointed out that Community legislation on company law laying down a principle of equal treatment was concerned with the conduct of the company itself rather than that of its shareholders. This, combined with the fact that general principles were not binding in relations between individuals, suggested that any general principle of equal treatment of shareholders could apply only between a company and its shareholders and not between one group of shareholders and another. According to Advocate-General Trstenjak, Mangold did not provide support for a possibility of relying directly on general principles in relations between private individ­uals, as the Court did not clarify in that case whether the prohibition of discrimination on grounds of age also has a horizontal effect. Irrespective of the fact that the main pro­ceed­ings in that case concerned a matter of civil law, in the reference for a preliminary ruling the Court essentially had to decide whether Community law precluded a provi­sion of domestic law under which employers could conclude fixed-term contracts of employment without restrictions with workers who have reached the age of 52. It related primarily to the compatibility of national law with the requirements of Community law.71

Advocate-General Trstenjak’s interpretation of Mangold was hard to reconcile with the terms of the Court’s judgment in that case72 and suggested that she too had reservations about its effect. Perhaps chastened by the hostility that judgment had encountered, the fourth chamber of the Court accepted the advice of its Advocate-General in Audiolux that there was no general principle of equality of shareholders and did not consider the effect of Mangold. However, the Court’s later decision in Kücükdevici reiterated that the general principle of Union law prohibiting discrimination on grounds of age, as given expression in Directive 2000/78, produced horizontal direct effect, at least where its effect was merely to exclude the application of an inconsistent provision of national law (rather than to substitute for that provision a new substantive rule).73 In its reasoning in Kücükdevici, the Court appeared to attach significance to the fact that discrimination based on age is prohibited by Article 21(1) of the Charter of Fundamental Rights. This seems inappropriate. Article 6(1) TEU now gives the Charter ‘the same legal value as the Treaties’, but it is not addressed to individuals and should not therefore

  AG Sharpston in Bartsch (n 60 above), para [85]. See also Tridimas (n 2 above) 47-50.   AG Sharpston in Bartsch (n 60 above), fn 72. 70  See n 2 above; I Corbisier, ‘Arrêt “Audiolux”: inexistence d’un principe général de droit communautaire protégeant les actionnaires minoritaires en cas de cession d’une participation de contrôle’ [2010] Journal de Droit Européen n ° 165, 9. 71  See n 2 above, para [122] of the Opinion (footnote omitted). 72  See in particular paras [77–8]. 73  See in particular paras [44–56]. 68 69

18  Anthony Arnull be treated as relevant to the obligations imposed on them by Union law. 74 Be that as it may, the case confirms that the general principle of non-discrimination may affect the obligations as well as the rights of individuals and therefore has implications outside the realm of public law. That principle is particularly susceptible to infringement by private as well as public parties. However, in both Mangold and Kücükdevici, it was possible to identify the practical implications of the principle. By contrast, in Audiolux the Court observed that a principle of equality of shareholders ‘presupposes legislative choices, based on a weighing of the interests at issue and the fixing in advance of precise and detailed rules . . . and cannot be inferred from the general principle of equal treatment.’75 It is evident that, even in the case of established general principles, the requirements of legal certainty would prevent them from having direct effect where there was doubt about their practical implications in the circumstances of a particular case. VII.  IS THERE A GENERAL PRINCIPLE PROHIBITING ABUSE OF THE LAW?

A.  Preliminary Remarks The importance of general principles of law in the legal order of the Union is profound. They enable the Court of Justice to reinforce the Union’s legitimacy by binding it to the values which underpin national law and the international obligations of the Member States. Sometimes the result is to limit the effect of Union law. The Court has held that the general principle of respect for fundamental rights may justify national restrictions on the obligations imposed by Union law, even under fundamental freedoms guaranteed by the TFEU, such as the free movement of goods76 and the freedom to provide services.77 Does the class of general principles include one which prohibits abuse of the law? In attempting to answer that question, a number of preliminary points should be made. First, although a principle prohibiting abuse is not universally recognised in the national laws of the Member States,78 this would not prevent it from being baptised by the Court of Justice as a general principle of Union law if there were sufficient support for such a development in the national systems and it were considered conducive to the proper functioning of the Union’s legal order. Secondly, the Court of Justice has long been ready to examine claims of abuse by private parties of their rights under Union law. The Court will not entertain references made by national courts where there is no real dispute

74   For the addressees of the Charter, see Art 51(1). Cf AG Bot in Kücükdevici (n 27 above), para [90], where he raises the question of the effect on individuals of directives which give effect to rights now enshrined in the Charter. 75  Para [62]. 76  See Case C-112/00 Eugen Schmidberger, Internationale Transporte und Planzüge v Republik Österreich [2003] ECR I-5659. 77  See Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-9609. 78  See AG Tesauro in Case C-367/96 Kefalas and Others v Greek State and OAE [1998] ECR I-2843, para [22]; LN Brown, ‘Is There a General Principle of Abuse of Rights in European Community Law?’ in D Curtin and T Heukels (eds), Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers, Vol II (Dordrecht, Martinus Nijhoff 1994) 511, 513–15; R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 437.

What is a General Principle of EU Law?  19 between the parties to the main action.79 It has been recognised since van Binsbergen80 that an individual’s right to freedom of movement may be restricted where his or her purpose in exercising it is to evade national rules which are justified in the public interest.81 The Court of Justice also has regard in interpreting Union acts to the need to prevent abuse of rights.82 A particular form of abuse by undertakings is expressly prohibited by Article 102 TFEU. Thirdly, a principle which varies in its effect from one area to another may still be classified as general, for the same is true of several such principles that have become well-established.83 On the other hand, Audiolux shows that the threshold that must be attained before a general principle is recognized is a high one, even where the suggested principle may be considered a specific expression of one whose existence is not in doubt. In that case, the referring court referred to several provisions of Community legislation that might have provided support for the existence of general principle of equality of shareholders. The Court observed: ‘the mere fact that secondary Community legislation lays down certain provisions relating to the protection of minority shareholders is not sufficient in itself to establish the existence of a general principle of Community law, in particular if the scope of those provisions is limited to rights which are well defined and certain.’84 Two of the key measures cited by the referring court were ‘essentially limited to regulating very specific company-law situations by imposing on companies certain obligations for the protection of all shareholders. They do not therefore possess the general, comprehensive character which is otherwise naturally inherent in general principles of law.’85 The same was true of a third act, the preamble to which did not state, ‘either expressly or by implication’,86 that the rules it laid down derived from a general principle of Community law. Nor could the suggested principle be considered a specific application in the company law field of the general principle of equal treatment, for the latter principle could not be regarded as imposing particular obligations on shareholders or determining when any such obligations might arise. This would require balancing of the various interests involved and ‘specific expression, in accordance with the principle of legal certainty, so that those concerned could ascertain unequivocally what their rights and obligations were and take steps accordingly . . .’87

79  See eg, Case 104/79 Foglia v Novello [1980] ECR 745; Case C-83/91 Wienand Meilicke v ADV/ORGA F A Meyer AG [1992] ECR I-4871. 80  Case 33/74 Johannes Henricus Maria van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299. Brown (n 78 above) 521, uses the term détournement de procédure to describe this type of situation. 81   The technique used by the Court varies. Sometimes it treats the question of abuse as relevant to whether the situation is covered by the respective provision of Union law (see eg, Case 229/83 Leclerc v Au Blé Vert [1985] ECR 1), sometimes as relevant to whether a restriction on a right conferred by Union law may be justified (see eg, Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459). Whether the former or the latter technique is adopted does not seem to make much practical difference: see V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205, 208–15 and 226. 82  See eg, Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 83  See Arnull (n 2 above). 84   Audiolux (n 2 above), para [34]. 85   Audiolux (n 2 above), para [42]. 86   Audiolux (n 2 above), para [48]. 87   Audiolux (n 2 above), para [58].

20  Anthony Arnull B.  A General Principle? i.  The Pronouncements of the Court of Justice It is now clear that Union law seeks to prevent the rights it confers from being abused. In Halifax,88 a value-added tax case decided by the Grand Chamber, the Court of Justice said that ‘according to settled case-law, Community law cannot be relied on for abusive or fraudulent ends’.89 It continued: The application of Community legislation cannot be extended to cover abusive practices by economic operators, that is to say transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law.90

The Court added: ‘That principle of prohibiting abusive practices also applies to the sphere of VAT.’91 In Kofoed,92 a case involving a national decision to tax an exchange of shares, the Court went further, declaring that a provision of a relevant directive reflected ‘the general Community law principle that abuse of rights is prohibited [French: le principe général du droit communautaire selon lequel l’abus de droit est prohibé]. Individuals must not improperly or fraudulently take advantage of provisions of Community law.’ However, it may be doubted whether those pronouncements were intended to mark the birth of a new general principle of Union law in the strict sense. The judgment in Halifax was consistent with the Opinion of Advocate-General Poiares Maduro, who said that the ‘notion of abuse operates as a principle governing the interpretation of Community law’,93 although he acknowledged that it was a strong principle which could lead to a departure from the literal meaning of a provision. Halifax was among the cases cited in Kofoed as authority for the existence of a principle the Court there described as general. But Kofoed was only a decision of the first chamber and it seems improbable that such a formation was intending to give birth to a new general principle. It seems more likely that it simply intended to apply the principle (general in the broader sense) recognised by the Grand Chamber in Halifax. The language of the first chamber in Kofoed was noticeably more equivocal than that of the Grand Chamber in Mangold, which made it absolutely clear that a new general principle was being inaugurated and what difference that principle made to the outcome of the case. The absence in Kofoed of any analysis of the type conducted by the fourth chamber in Audiolux hardly seems consistent with an intention to launch a principle of constitutional status. More fundamentally, the prohibition of abuse does not protect the rights enjoyed by private parties under Union law but deprives them of advantages it would otherwise

88   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 89   Halifax (n 88 above), para [68]. 90   Halifax (n 88 above), para [69]. Cf the definition of abuse of rights given by AG Trstenjak in fn 53 of her Opinion in Case C-520/06 Stringer and Others v Her Majesty’s Revenue and Customs [2009] ECR I-179. 91   Halifax (n 88 above), para [70]. See also AG Poiares Maduro in Halifax (n 88 above), para [64]. 92   Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795, para [38]. 93   AG Poiares Maduro in Halifax (n 88 above), para [69]. See also AG Tesauro in Kefalas (n 78 above), para [25].

What is a General Principle of EU Law?  21 confer on them. It is typically (albeit not invariably)94 invoked by the administration.95 Unlike established general principles, the prohibition of abuse does not therefore operate as a principle of administrative law.96 It does not contribute to that aspect of the rule of law which requires the exercise of public power to be subject to review by the courts.97 ii.  Abuse by the Institutions and the Member States It is true that the law of the Union contains safeguards against abuse of the law by the institutions. The grounds on which their acts may be declared void under Article 263 TFEU include misuse of powers (détournement de pouvoir). In R v Secretary of State, ex parte British American Tobacco,98 the Court of Justice said that a measure will be vitiated by misuse of powers if it is adopted ‘with the exclusive or main purpose of achieving an end other than that stated or evading a procedure specifically prescribed by the [EC] Treaty for dealing with the circumstances of the case.’ In addition, as we have seen, under the second paragraph of Article 340 TFEU, the Union is required to make good any damage caused by its institutions or by its servants in the performance of their duties. The Court of Justice has acted to deter abuse by Member States in a variety of ways. The infringement procedure under Article 258 TFEU allows it to declare Member States in breach of their Treaty obligations. Under the second paragraph of Article 260 TFEU, the Court may impose financial sanctions on States which fail to comply with adverse judgments against them under Article 258. The Court has used the preliminary rulings procedure to enlist the support of the national courts to ensure compliance by Member States with Union law. It was under that procedure that the Court developed the principle of State liability,99 which reinforces the doctrines of primacy and direct effect by preventing Member States from evading the financial consequences of breaches of Union law for which they are responsible. In a reference from the House of Lords, the Court famously required a British domestic rule, which prevented interim relief from being awarded against the Crown, to be set aside in a case where an Act of Parliament was said to be incompatible with Community law (an argument which was ultimately upheld).100 In other cases referred by national courts, the Court showed increasing willingness to permit traders to recover charges imposed by the State in breach of Community law but which may have been passed on to third parties.101     94  See eg, Case C-441/93 Panagis Pafitis et al v Trapeza Kentrikis Ellados AE et al [1996] ECR I-1347; Case C-341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet, Svenska Byggnadsarbetareförbundets avdelning 1, Byggettan and Svenska Elektrikerförbundet [2007] ECR I-11767.     95   This is sometimes called ‘inverse vertical direct effect’: see Kolpinghuis Nijmegen (n 34 above) 3974.     96   Cf AG Tesauro in Kefalas (n 78 above), para [19].     97  See F Jacobs, The Sovereignty of Law: The European Way (Cambridge, Cambridge University Press 2007) 62.     98   Case C-491/01 R v Secretary of State, ex p British American Tobacco [2002] ECR I-11453, para [189]. See also Joined Cases C-133/93, C-300/93 and C-362/93 Antonio Crispoltoni v Fattoria Autonoma Tabacchi and Giuseppe Natale and Antonio Pontillo v Donatab Srl [1994] ECR I-4863, para [27].     99   Itself an expression of the general principle reflected in Art 340 TFEU and ‘familiar to the legal systems of the Member States that an unlawful act or omission gives rise to an obligation to make good the damage caused’: Joined Cases C-46/93 and C-48/93 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and R v Secretary of State for Transport, ex p Factortame Ltd et al [1996] ECR I-1029, para [29]. 100   Case C-213/89 R v Secretary of State for Transport, ex p Factortame Ltd et al [1990] ECR I-2433. 101  See Case 68/79 Hans Just I/S v Danish Ministry for Fiscal Affairs [1980] ECR 501; Case 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595; Joined Cases C-192/95 to C-218/95 Société Comateb et al v Directeur Général des Douanes et Droits Indirects [1997] ECR I-165.

22  Anthony Arnull Furthermore, Article 54 of the Union’s Charter of Fundamental Rights, which is headed ‘Prohibition of abuse of rights’, provides: Nothing in this Charter shall be interpreted as implying any right to engage in any activity or to perform any act aimed at the destruction of any of the rights and freedoms recognised in this Charter or at their limitation to a greater extent than is provided for herein.

The Charter is addressed to the institutions and bodies of the Union, as well as to the Member States when they are implementing Union law.102 However, the law relating to what may loosely be called abuse by the institutions and the Member States has different normative bases: the need to provide effective judicial remedies; primacy and direct effect; national procedural autonomy. Article 54 of the Charter merely corresponds to Article 17 of the European Convention on Human Rights (which is addressed to ‘any State, group or person’). It is presumably for these reasons that recent literature on prohibiting abuse103 does not deal with the position of the institutions and the Member States. The consequence is that the prohibition of abuse in Union law does not possess ‘the general, comprehensive character which is otherwise naturally inherent in general principles of [Community] law.’104 It is confined to alleged abuses by private parties. iii.  Ramifications of a General Principle Prohibiting Abuse In substantive terms, the prohibition of abuse, if allowed to develop too strongly, could undermine the foundations of the Internal Market. Conscious of this danger, AdvocateGeneral Geelhoed maintained in Akrich105 that Member States could not rely on the need to prevent abuse to restrict ‘a matter inherent in the exercise of a freedom guaranteed by the EC Treaty’. He went on to point out that ‘the dividing line between abuse and use for a purpose not contemplated by the legislature is hard to define’.106 The Court of Justice also seems sensitive to this danger for, in its case law on freedom of movement, it has treated claims that rights are being abused with a degree of scepticism.107 It is a danger which might be exacerbated if the prohibition of abuse were elevated to the status of a general principle of Union law.  See Art 51(1).  See de la Feria (n 78 above); KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 425, 458. 104   Audiolux (n 2 above), para [42]. 105   AG Geelhoed in Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607, para [100]. 106   AG Geelhoed in Akrich (n 105 above), para [173]. Cf Centros (n 81 above), paras [18] and [26–27]; Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155, paras [98] and [137–38]; Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995, para [38]. 107  See eg, Akrich (n 105 above); Case C-413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I-13187; Case C-200/02 Kunqian Catherine Zhu, Man Lavette Chen v Secretary of State for the Home Department [2004] ECR I-9925; Case C-147/03 Commission v Austria [2005] ECR I-5969; Centros (n 81 above); Laval (n 94 above). The restrained use made by the Court of the notion of abuse was applauded by AG Poiares Maduro in Case C-210/06 Cartesio Oktató és Szolgáltató Bt [2008] ECR I-9641, para [29]. (The Court did not use the term ‘abuse’ in its judgment in that case, which concerned the right of establishment of companies.) Cf Art 35 (headed ‘Abuse of rights’) of Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States [2004] OJ L229/35. For a general discussion, see K Ziegler, ‘“Abuse of Law” in the Context of the Free Movement of Workers’, ch 21 below. 102 103

What is a General Principle of EU Law?  23 In Emsland-Stärke, the Court elaborated a test, comprising an objective and a subjective element, for establishing when abuse had taken place. However, it left the application of that test to the national courts.108 This creates obvious risks of lack of uniformity and confusion with national approaches to abuse of law. There is also a risk that national courts acting in bad faith will themselves abuse the margin of discretion they enjoy under the Emsland-Stärke test. This is particularly true of the subjective element of that test, ‘consisting in the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it’.109 It makes little difference in this respect whether, strictly speaking, that element of the test is genuinely ‘subjective’ or not.110 The result is that a true general principle prohibiting abuse of the law would be more dangerous for the integrity of Union law than a maxim of interpretation to similar effect, the application of which could more easily be supervised by the Court under the preliminary rulings procedure. VIII. Conclusion

It is therefore submitted that the prohibition of abuse is too uncertain in its application and potentially damaging to the proper functioning of the Union to be accorded the constitutional status of a general principle of Union law. It should instead be considered a maxim of interpretation which can help to determine the scope of Union law rights and the circumstances in which restrictions on those rights may be justified.111 It should not be used as the basis for contra legem interpretation, though this can sometimes be hard to distinguish from legitimate teleological interpretation. In applying such a maxim, the objective of the right at stake will play a central role.112 Here the Union legislature has a role to play in communicating its objectives to those affected by its acts, as well as to the Court of Justice, and helping them to strike the right balance between competing considerations, such as protecting legal certainty and legitimate expectations on the one hand and preventing tax evasion on the other.

  Emsland-Stärke (n 82 above), paras [52–54].   Emsland-Stärke (n 82 above), para [53]. See J Snell, ‘The Notion of and a General Test for Abuse of Rights: Some Normative Reflections’, ch 15 below. A general principle prohibiting abuse of the law might have strengthened the hand of the first instance judge in Case C-206/01 Arsenal Football Club plc v Matthew Reed [2002] ECR I-10273. See A Arnull, ‘Casenotes: Arsenal Football Club’ (2003) 40 Common Market Law Review 753. 110  See Edwards and Farmer (n 81 above) 221–24. 111  See Edwards and Farmer (n 81 above) 205. 112  See eg, Emsland-Stärke (n 82 above), para [52]; Centros (n 81 above), paras [25–27]; Halifax (n 88 above), para [74]; Cadbury Schweppes (n 106 above), para [64]. 108 109

3 What is a General Principle of EU Law? A Response Karsten Engsig Sørensen

I

n the following chapter, I will concentrate on two of the conclusions drawn by Professor Anthony Arnull at the end of his chapter. The first is the conclusion that the principle prohibiting abuse is not a general principle of EU law, but is instead a principle of construction. In my opinion it can be argued that it is a general principle, and I will present some of the arguments in favour of this. Secondly, I will address the assumption that a general principle on abuse would be more likely to undermine the foundation of the Internal Market than a principle of construction. In my opinion, in most cases it makes little difference what kind of principle it is, and I cannot see why a general principle is more likely to have an undermining effect. Finally, I will address the issue of whether the principle prohibiting abuse is a general principle in the sense that it applies to several or to most areas of EU law. It is the judgment in Kofoed1 which raises doubts as to how generally the principle applies, and it is therefore this judgment which needs closer examination. I. A general principle of EU law prohibiting abuse

The question whether the principle prohibiting abuse is a general principle or something else has attracted much comment from Advocates-General and legal scholars.2 These arguments will not all be repeated here as I will instead highlight the arguments which I consider most clearly indicate that it is a general principle of EU law. Professor Arnull relies mainly on the judgment in Halifax3 for the view that the principle prohibiting abuse operates as a principle of interpretation. While it is very clear in his Opinion that Advocate-General Poiares Maduro takes the view that it is a principle of interpretation, in its judgment the ECJ did not make that clear. In Halifax the ECJ referred to a principle but not to a principle of interpretation. Neither in that case nor in other recent cases does the Court indicate that the principle is a principle of interpretation  Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795.  For references to the relevant literature and opinions see A Arnull, ‘What is a General Principle of EU Law?’, ch 2 above; R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395; and KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423. 3  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 1 2

26  Karsten Engsig Sørensen or that abusive practice is to be countered by interpretation. This is in contrast to other principles of construction, such as the principle of consistent interpretation, where the Court normally stresses that it implements the principles through interpretation. In several cases the Court has held that the issue of abuse should not be addressed when deciding whether a situation is covered by a provision.4 If the principle prohibiting abuse were a principle of interpretation, then that would mean that the question of abuse would be a question of the scope of EU rights. However, this approach seems to have been dismissed in the cases referred to. If, on the other hand, it is accepted that the principle prohibiting abuse is a general principle, this means that the question of abuse can arise after the scope of an EU right has been decided, as a question as to whether a transaction falling within the scope of a right should nevertheless be excluded. Therefore, as a general principle it may operate as an exception which is only considered after the scope of the right has been decided. Another way to approach the problem is to examine whether the principle prohibiting abuse has effects which are normally attributed to a general principle. A general principle may have the functions of adding to interpretation, acting as a basis for a review of EU acts and national law, and filling in legislative gaps.5 On the other hand, a principle of interpretation will only have the first function, so that if it can be established that the principle prohibiting abuse has the latter two functions, this would be a clear indication that it is a general principle. If the principle prohibiting abuse is only a principle of construction, then it would not be possible to use it to reach a contra legem interpretation.6 As a consequence, in cases where neither EU law nor national law has a proviso excluding abusive practices, there would be a problem in interpreting the rule as allowing the setting aside of abusive practices where the conditions for relying on an EU right are formally fulfilled.7 However, there were several cases where the ECJ indicated that abusive practices should be prohibited even though there is not a proviso excluding abuse in the European rules8 or national 4  See Case C-413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I-13187, para [32]; and Case C-23/93 TV10 v Commissariaat voor de Media [1994] ECR I-4795, para [15]. Along the same lines, see Case C-223/03 University of Huddersfield Higher Education Corporation v Commissioners of Customs & Excise [2006] ECR I-1751, para [51]. 5  See further, T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press 2006) 29–35; A Arnull, The European Court of Justice, 2nd edn (Oxford, Oxford University Press 2006) 336; J Nergelius in U Bernitz and J Nergelius (eds), General Principles of Community Law (Utrecht, Kluwer Law International 2000) 226–27. 6  This at least has been established by the Court in the cases on the extent of the duty of consistent interpretation; see Case C-212/04 Konstantinos Adeneler et al v Ellinikos Organismos Galaktos [2006] ECR I-6057, para [110]. It has been argued that if a general principle is used for the interpretation of European law, the same limitation applies; see Tridimas (n 5 above) 30. In his chapter (n 2 above), Professor Arnull disagrees, as he suggests that a general principle may lead the Court to construe the European Treaties contra legem. The question whether a general principle allows contra legem interpretation probably rests on how one draws the line between interpretation and gap-filling. One could argue that a solution which does not find support in the wording of a provision cannot be reached by interpretation but requires gap-filling. Whichever is the case, the conclusion is that a general principle allows the construction of EU law contra legem, whereas a principle of construction cannot have that effect. 7  This also seems to be implicitly acknowledged by the ECJ in Kofoed (n 1 above). The ECJ acknowledged that abuse could not be prevented just by interpreting the provisions implementing the Directive which gave rights but did not address abuse. Therefore it was necessary to rely on a national principle preventing abuse, see Kofoed (n 1 above), paras [40–46]. 8  In Case C-456/04 Agip Patroli SpA v Capitaneria di porto di Siracusa et al [2006] ECR I-3395, para [23], the ECJ pointed out that abuse can be found to exist where the conditions laid down by the Community rule are fulfilled technically but where the aim of the provision is frustrated. Here the ECJ seems to reach a result

What is a General Principle of EU Law? A Response  27 law.9 In none of these cases has the Court considered whether this would amount to a contra legem interpretation. This indicates that the Court does not find that this is an issue of interpretation but rather that it is a question of applying a general principle to set aside (or modify) EU law by ensuring that EU law is not used for abusive practices. Since a general principle can go beyond what can be achieved by interpretation, this indicates that we are dealing with a general principle. That the principle prohibiting abuse works more like a filler of legislative gaps than as an aid to interpretation also seems to be indicated by the form the principle has taken. The principle has created a new rule that makes it possible to exclude abusive practices. The principle has been developed by the Court to be a two-prong test supplemented with a number of procedural requirements, such as the requirement that the evaluation of abuse should be made on a case-by-case basis. This principle goes hand in hand with different kinds of EU rights which do not in themselves include any proviso excluding abusive practices. There can be a debate about the extent to which the principle prohibiting abuse can act as a ground for the annulment of EU measures. The principle creates an exception to an EU right. It excludes abusive practices even when the EU right does not hint at such exclusion. Therefore the EU right is at least modified, but this is hardly the same as an annulment. However, there could be an annulment if the EU law would hinder the principle prohibiting abuse taking effect. Such cases are difficult to imagine and so far we have seen none. The principle may lead to national law being set aside. In some cases it can be debated whether national law is set aside because it breaches the principle, or because national law is in breach of the EU right which is supplemented by the principle prohibiting abuse.10 However, if national law were to hinder the principle prohibiting abuse from taking effect, then the national law would be set aside by the principle. This could be because national law explicitly allows some conduct that should be prohibited or because national principles prevent the EU principle from having full effect. Again this is not likely, but on the other hand, it is not unthinkable. Therefore to summarise, I find that the most likely reading of the case law is that the principle prohibiting abuse is a general principle of EU law and I find that the abuse principle can have the effects normally attributed to a general principle, going beyond a mere principle of construction.

which runs against the wording of the Community rules, in a situation where there was no rule indicating that abuse should be excluded.     9  There is no reference to an abuse doctrine in either national or EU law in Case C-8/92 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas [1993] ECR I-779; Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569; and in several of the VAT cases, including Case C-32/03 I/S Fini H v Skatteministeriet [2005] ECR I-1599 and Halifax (n 3 above). 10  This can be illustrated by looking at what happened in Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. In this case the ECJ found that the controlled foreign companies (CFC) rules infringed Art 43 EC [now Article 49 TFEU] as they restricted freedom also in situations falling outside the exemptions made for abuse under the principle prohibiting abuse. Thus it can be said that in the end it was Art 43 EC that the national law infringed. Others see it as an example that national law was in breach of the principle prohibiting abuse and thus a case demonstrating that the principle prohibiting abuse acts as an overriding rule of law; see de la Feria (n 2 above) 438.

28  Karsten Engsig Sørensen

II.  Does it matter what kind of principle it is?

As mentioned above, a general principle and a principle of construction may have different effects in some situations. However, in most situations they will have similar effects11 and this is one of the main reasons why it is difficult to decide what kind of principle we are dealing with. Where there is abuse, the similarities would include the following: principles will have the effect of prohibiting abusive practices. ·· Both Both principles will have the effect that the EU institutions and the Member States, including the national courts, will have a duty to apply the principle to prohibit abusive practices. Both principles are themselves the basis for denying a person or company the · opportunity to rely on an EU right. Thus national abuse rules are not necessary, 12

but to the extent that there are such national rules, both principles have the effect of allowing them to function. Both principles have the effect that individuals may be barred from relying on a right. If the principle is a principle of construction, then the issue of horizontal effect does not arise, but if it is a general principle, it could. However, since the principle prohibiting abuse means that an individual is barred from exercising a right but is not subject to an obligation, it is not clear that the principle requires horizontal effect. In Kofoed the ECJ seems to assume that the application of the principle against an individual is equivalent to imposing an obligation.13 However, there is agreement that general principles may have horizontal effect and therefore it should be possible for the principle prohibiting abuse to be applied as a general principle to prevent individuals undertaking abusive practices.

·

Given that, whether as a general principle or as a principle of interpretation, the principle has so many similar effects that I cannot see why a general principle would be more likely to undermine the Internal Market than a principle of interpretation. If it is a principle of construction, it is a very strong one which may result in what appears to be contra legem interpretation. Furthermore, there is nothing to indicate that as a principle of construction it would be more limited in scope than it would be as a general principle; see the discussion below. It seems that Professor Arnull is of the opinion that, as a general principle, it may do more harm in the hands of the national courts. However, if it is a principle of construction, it should be applied by the national courts in all cases where they apply EU law or interpret national law implementing EU law. A correct application of EU law will require the courts to make sure that EU rights are not relied on for abusive practices. It may even be argued that national courts would be more willing to apply a principle of construction on their own than a general principle. First, national courts often use principles of construction, such as the duty to interpret national law consistently with EU law, whereas they may be more reluctant to apply general principles on their own. Secondly, whereas it is clear that a principle of construction can be relied on by any party to a case, and also in a case 11  This has been noted by many, including AG Poiares Maduro in Halifax (n 3 above), see n 62; de la Feria (n 2 above) 438; and D Anderson, ‘Abuse of Rights’ (2006) 4 Judicial Review 348, 353. 12  This may be different if a directive authorises the Member States to adopt rules to counter abuse; see below. 13   Kofoed (n 1 above), para [42].

What is a General Principle of EU Law? A Response  29 between two individuals, the reliance on a general principle can raise the issue of whether it has horizontal effect and can thus prove more complicated.14 Therefore the risk of undermining the Internal Market will not depend on whether a principle prohibiting abuse is a principle of construction or a general principle. It will depend on how broadly the Court interprets the concept of abuse. So far, it seems that the Court is very much aware of the fact that a broad application may undermine the effectiveness and aim of EU law, so it has given the principle a very narrow scope. In most cases it has limited the principle to situations of ‘wholly artificial arrangements’, and has imposed very strict procedural requirements, especially by requiring case-by-case examination. Therefore I do not see any risk at present that the principle will undermine the application of EU law. On the contrary, it seems just as likely that the Court of Justice will end up restricting the principle too much, thereby creating a principle without substance.15 III. In which areas of EU law does the principle apply?

An important question to ask here is to what extent is the principle prohibiting abuse general in the sense that it applies to different areas of EU law? In recent years the principle has been applied mainly to tax cases but there are many cases outside tax law. Most cases have focused on abuse in commercial cases, and even though it is probably easier to establish abuse in such cases, the Court has at least held the door open for the application of the principle in non-commercial cases (for instance, marriages of convenience).16 Furthermore, the principle has been applied to many different types of secondary legislation and to association agreements entered into by the EU.17 Thus the case law seems to indicate that the principle applies broadly in EU law, although slightly different approaches have been used in the different areas. However, in Kofoed, the ECJ seems to have reached a result which limits the application of the principle prohibiting abuse in a situation where a tax directive applies. The case concerned the distribution of profits which were closely connected to a cross-border exchange of shares. The Danish tax authorities claimed that the distribution should be considered a taxable cash payment made by the acquiring company in exchange for shares. The transaction was covered by the Merger Tax Directive18 and the ECJ found it 14  As mentioned above, the Court seems to assume that the application of the principle to deny an individual a Community right would require horizontal effect. 15  I have argued this in more detail in Sørensen (n 2 above) 423. However the later judgments in Cadbury Schweppes (n 10 above) and Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107 have given some scope for setting aside abuse practices. 16  See for instance, Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607, para [57]; and Case C-127/08 Blaise Baheten Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I-6241, para [75]. However, as explained below, the consequence of Art 35 (of Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States [2004] OJ L158/77) will be that the general principle prohibiting abuse no longer applies to the free movement of persons covered by this Directive. 17  The cases concerning association agreements are Case C-36/96 Faik Günaydin, Hatice Günaydin, Günes Günaydin and Seda Günaydin v Freistaat Bayern [1997] ECR I-5143; Case C-285/95 Suat Kol v Land Berlin [1997] ECR I-3069; and Case C-16/05 R, Veli Tum and Mehmet Dari v Secretary of State for the Home Department [2007] ECR I-7415. 18  Council Directive 90/434/EEC of 23 July 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States (Merger Tax Directive) [1990] OJ L225/1.

30  Karsten Engsig Sørensen was not possible to interpret ‘cash payments’ as including the dividend paid. However, since the ECJ could see from the reference that the real concern was that there was an abuse of right, it chose to explain how the Directive should be used in such cases of abuse. The Court noted that Article 11(1)(a) of the Directive allowed the Member States to refuse the benefits of the Directive in cases of tax evasion or tax avoidance. This reflects the general principle that abuse of rights is prohibited.19 The ECJ agreed with the AdvocateGeneral that there was evidence pointing at abuse in the case pending before the Danish court. However, instead of just directing the national court to apply the general principle prohibiting abuse of rights, it directed the Danish court to find authority for prohibiting abuse in Danish law. Thus it seems that the ECJ did not find that the national court could rely on the Community principle prohibiting abuse of rights, which indicates that the general principle does not apply in this area. The judgment itself does not make explicitly clear why the ECJ found that the principle prohibiting abuse did not apply in this case. The ECJ concentrated on considering whether Article 11(1)(a) can apply without specific transposition. Under the principle of legal certainty, a directive is not directly applicable to individuals and thus cannot be relied upon by Member States against individuals.20 By this the Court explains why Article 11(1)(a) cannot be authority for prohibiting abuse when it has not been transposed, but the Court does not explain why the general principle prohibiting abuse of rights does not apply. However, an explanation can be found in the Opinion of Advocate-General Kokott, at paragraph [67] (footnotes omitted): The competent authorities would also be precluded from relying directly, against an individual, on any existing general principle of Community law prohibiting the misuse of law. Namely, in cases falling within the scope of Directive 90/434, such a principle has been given specific effect in Article 11(1)(a) of the Directive and has been expressed in a concrete manner. If it were to be permitted, in addition, to have recourse to a general principle which in terms of content is much less clear and precise, there would be a danger, thus, that the harmonisation objective of Directive 90/434 would be undermined and the legal certainty surrounding the restructuring of companies which it seeks to achieve would be jeopardised. Moreover, such an approach would undermine the prohibition, already mentioned, on directly applying untransposed provisions of directives to the detriment of individuals.

Thus Advocate-General Kokott suggests that the general principle does not apply in this area since the principle prohibiting abuse had been given specific effect in Article 11(1) (a). The application of the general principle has ceased to ensure that the Directive works as intended. The interpretation of the Advocate-General has the advantage of promoting legal certainty, as taxpayers could otherwise be confused if they assumed that the Directive deals with the issue of abuse. In a footnote, the Advocate-General refers to the unclear situation created by Mangold,21 which has also been raised by Professor Arnull in his chapter. It is clear that the interpretation given by Advocate-General Kokott would ensure that the general principle prohibiting abuse does not undermine the restriction on the direct effect of directives, and would thus avoid the criticism made of the result reached in the Mangold judgment.   Kofoed (n 1 above), para [38].   Kofoed (n 1 above), para [42]. AG Kokott has repeated this point in Case C-352/08 Modehuis A Zwijnenburg BV Judgment of 16 July 2009, nyr, para [62]. 21  Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9981. 19 20

What is a General Principle of EU Law? A Response  31  Even though the Court does not address the question as directly as the AdvocateGeneral, the reason why the ECJ only considered Article 11(1)(a) and not the general principle can be seen as an acceptance of the argument presented by the AdvocateGeneral. At first sight it seems odd that a general principle can be suppressed by secondary legislation. However, as long as the principle is only made obsolete where the secondary legislation has set up an alternative system that has the capacity to prevent abusive practices, this seems to be acceptable. It is the nature of a general principle that it acts as an interpretive tool and gap-filler, but that if there are no gaps, the principle does not apply. Furthermore, as mentioned by the Advocate-General, there are other general principles, such as the principle of effectiveness and the principle of legal certainty, which may override the principle prohibiting abuse if there is specific regulation of abuse in the secondary legislation. The consequence of accepting the suggested interpretation will be that the general principle prohibiting abuse of rights will not apply in situations where special rules have been adopted to address the issue of abuse. For instance, the general principle will no longer apply to the free movement of persons since, just as with Article 11(1)(a) of the Merger Tax Directive, Article 35 of Directive 2004/38/EC allows the Member States to take measures against abuse in the area. This also explains why in the Metock case the ECJ merely refers to the fact that Article 35 allows Member States to take measures, instead of restating the general principle prohibiting abuse. Since there are many areas where there are specific abuse provisions in secondary legislation, the reach of the general principle may be restricted in many such cases.22 However, we also know that it is not all special regulations of abuse that have the effect that the general principle no longer applies. For instance, it is clear that the Court finds that the general principle applies in the common system of value-added tax (VAT), even though Article 395 of Directive 2006/112/EC on the common system of VAT provides a procedure whereby the Council can authorise Member States to introduce measures to prevent tax evasion or avoidance.23 This is evidently not enough to rule out the application of the general principle prohibiting abuse, and the reason may be that this provision does not address the issue of abuse in a way which the legislator intended to be exhaustive.24 Therefore it seems that the general principle prohibiting abuse does apply to most, if not all, areas of law, but it will cease to apply once the problem of abuse has been 22  For instance, it could mean that the area of anti-dumping or even broader areas of customs law are outside the scope of the general principle prohibiting abuse, since there are several specific abuse measures in this area. 23  This provision replaces Art 27 of the Sixth VAT Directive (77/388/EEC) which was very similarly worded. AG Poiares Maduro addressed the question whether this provision prevented the application of the principle prohibiting abuse in Halifax (n 3 above), para [76]. He reached the conclusion that this was not the case. The consequence of the fact that the general principle applies in this area is that the national authority can rely on EU law to prohibit abuse, and does not have to look for a national authority, as was the case in Kofoed (n 1 above). The question whether a national authority for prohibiting abuse is required in the area of VAT has been much debated in Swedish law; see K Norberg and F Petersen, ‘Förfarandemissbruk inom mervärdesskatteområdet’ [2008] Skattenytt 2. The authors criticise the decision of a Swedish court that a national authority was required. For the reasons given above, I think the authors are right in their criticism. 24  Interestingly, one of AG Poiares Maduro’s arguments for reading Article 27 of the Sixth VAT Directive as not precluding the application of the general principle was that the area of VAT ‘would become a peculiar legal domain where virtually any opportunistic behaviour by taxable persons relying on the literal meaning of its provisions to improperly gain tax advantages against the tax authorities would have to be tolerated unless the Member States had previously adopted legislative measures forbidding such behaviour’. This thinking also indicates that the general principle should cease to be applied to an area of law if the legislator seems to have dealt exhaustively with the problem of abuse in secondary legislation.

32  Karsten Engsig Sørensen regulated through specific rules in a way which is intended to be exhaustive. This clearly means that there are several areas of law where the principle no longer applies, but it does not affect its status as a general principle.

4 The Abuse of Rights in the Civil Law Tradition James Gordley I. Introduction

T

his chapter examines whether the doctrine of ‘abuse of rights’, as found in the civil law tradition, can be of help in understanding the problems of EU law that other authors will address in this volume. I will discuss the civil law before the doctrine was formulated in the late nineteenth and early twentieth century; then the controversy that surrounded its formulation; and, finally, how the doctrine has been applied since the controversy died down. I will conclude that there is little to be learned by studying this doctrine that is of use for EU law. Perhaps the EU should have a doctrine of ‘abuse of rights’. But to be useful, it would have to be a different doctrine, similar only in name to the doctrine I am going to discuss. While we ought to learn from the past, sometimes the best lesson is that the past cannot guide us. II. Historical Roots

Continental jurists who discuss the historical roots of the doctrine of abuse of rights mention two sets of Roman texts, and the interpretations that those texts received. One set concerns what came to be called aemulatio. The second concerns the so-called exceptio doli generalis. Aemulatio is the label that became fastened to a few Roman texts which said that a person was not allowed to use his land in a way that otherwise would be permissible if his purpose was to bother his neighbour. A landowner could divert a torrent of water from his own property and so harm his neighbour if he did so to avoid harm to himself but not if his purpose was to inflict harm on the neighbour.1 The owner of lower premises who made smoke that bothered his upstairs neighbour could not do so if his sole purpose was to bother him.2 The phrase aemulatio was taken from a text that dealt with a quite different problem. It said that a person can construct a new building without permission from the Emperor as long as the purpose is not to outshine another city (ad aemulationem alterius civitatis).3  Digest (‘Dig’) 39.3.2.9.  Dig 47.10.44. 3  Dig 50.10.3.pr. On this text, see D Johnston, ‘Munificence and Municipia: Bequests to Towns in Classical Roman Law’ (1985) 75 Journal of Roman Studies 105, 123–24. 1 2

34  James Gordley Later jurists concluded that an owner could not do something on his land for the sole purpose of bothering his neighbours. That was the view of the seventeenth-century jurist Jean Domat,4 from whom the drafters of the French Civil Code borrowed a great deal, although the Code itself contained no provisions about aemulatio or anything similar. Nevertheless, in the nineteenth century, some French commentators thought that relief should be given when a landowner’s sole purpose was to harm those nearby. They cited Jean Domat and the Roman texts. Charles Demolombe, who disagreed, admitted that he was in the minority.5 Nineteenth-century French courts gave relief in such cases. In one decision that became famous later on, the Court of Appeal at Colmar held that a house owner could not build a false chimney on his roof for the sole purpose of blocking his neighbour. The court said that although: [I]n principle the right to property is in some way absolute, authorizing the owner to use and abuse an object, nevertheless the exercise of this right, like that of any other, must have as its limit the satisfaction of a serious and legitimate interest, and the principles of morality and equity are opposed to a justice that would sanction an action inspired by malevolence and performed under the dominion of an evil passion, not justified by any personal utility and bringing a grave harm to another.6

Although the case, like the Roman texts, concerned relations among landowners, the language of the court suggested a larger principle: one could not exercise any right solely to harm another person. III.  the Controversy

In Germany, the Roman texts remained in force in some regions, and were the basis of university study until the German Civil Code came into force in 1900. For reasons we will discuss later on, some jurists, such as Carl Georg von Wächter, believed that these texts had no general application, while others, such as Bernhard Windscheid, thought that they stood for the general principle that one could not use a right simply to harm someone else. His view passed into Section 226 of the German Civil Code which provides: ‘the use of a right is not permitted when it can only have the purpose of causing harm to another’. Later on, proponents of the doctrine of abuse of rights in France and in Germany were to claim that their doctrine explained aemulatio, the French cases just cited, and Section 226 of the German Civil Code. At a later point we will consider whether that doctrine enshrines some larger principle that might be of help in understanding EU law. Here we need only note that if the principle is only that one cannot exercise a right for the sole purpose of harming another, that principle, though it may be perfectly correct, will not help with EU law. In the cases that have come before the European Court of Justice it may matter greatly that a person acted for the sole purpose of escaping taxation or regulation. But they are not cases in which his purpose was to harm someone else. Aemulatio, and its French and German versions, would not be helpful unless it does stand for a larger principle.

  J Domat, Les Loix civiles dans leur ordre naturel (Paris, 1713) book 2, title 8, s 3.  C Demolombe, Cours de Code Napoléon, 3rd edn, vol 11 (Paris, Durand & Hachette 1882) § 66. 6  Colmar, 2 May 1855, Recueil Dalloz 1856.II.9, 10. 4 5

The Abuse of Rights in the Civil Law Tradition  35 The other set of Roman texts concerned the so-called exceptio doli generalis. An exceptio was like what we call a defence. Some have described the relief that German courts give for abuse of rights as similar to that which a defendant once obtained with the help of this exceptio.7 It is true that under Roman law, a party would have used the exceptio in cases where, today in Germany, he would seek relief for abuse of rights. But it does not follow that Roman law is helpful in suggesting when such relief should be given. The exceptio doli generalis authorised a judge to give relief when the plaintiff’s action was inequitable. But the distinguishing feature of this exceptio was that it committed that issue to the judge’s discretion without suggesting what sort of inequity would warrant relief. The phrase exceptio doli generalis was coined, not by the Romans, but by later jurists who were trying to understand their texts. The text in question was part of a formula or set of instructions given the judge or iudex telling him when to find for the plaintiff. The formula said that the plaintiff should prevail only if ‘in this matter [he] neither did nor is doing dolus malus’. That language meant to later commentators, and, perhaps, to the Romans, that the plaintiff loses, not only if he ‘did’ commit fraud in the past, but also if he ‘is doing’ fraud now by the very bringing of his action. The latter case was the one the commentators referred to as the exceptio doli generalis. The plaintiff would have committed fraud in the past, for example, by telling a lie. But when he was said to do so by the very bringing of the action, the term ‘fraud’ was presumably being used more broadly to mean that by merely bringing it, he was acting inequitably. The Roman sources do not indicate when he would be acting inequitably. The point of the exceptio was to leave that matter for the iudex to consider. Until late classical times, the iudex was a layman appointed to serve ad hoc in a particular lawsuit. He decided the case before him guided by the formula approved by an official called the praetor. No records were kept of what the iudex decided. His decisions did not count in the building of Roman law. Consequently, all we know of the so-called exceptio doli generalis is that the iudex could find for the defendant when he thought it would be inequitable to find for the plaintiff. On the eve of codification, the German jurists understood this exceptio in a similar way: as a discretionary power vested in the judge to refuse to allow a claim he deemed to be inequitable. They argued over whether the judge should have such a broad and undefined power. Ferdinand Regelsberger and Heinrich Dernburg thought that he should. According to Regelsberger the exceptio gave the judge the power to declare ‘that a legal proceeding is not permissible which is in contradiction with the principles of good faith’ (Treu und Glauben). He acknowledged that one might fear that ‘so general a legal principle . . . gives unusual free play to judicial discretion’. But, he claimed, only by doing so could one keep the law ‘in accord with the living legal consciousness of the people’.8 Dernburg described the principle in an equally indefinite way: it allowed relief when the plaintiff’s ‘right to bring the complaint contradicts true right in a given case, and therefore has only the form of law but not the essence’. Relief should be given, he said, because the exceptio is the ‘organ of true right’.9 Alois Brinz, however, thought that the exceptio had no place in modern law. It required the judge to say that the defendant was ‘legally 7   F Ranieri, ‘Bonne foi et exercice du droit dans la tradition du civil law’ (1998) 4 Révue internationale du droit comparé 1055, 1057–59; GH Roth in Münchener Kommentar zum Bürgerlichen Gesetzbuch, 5th edn (Munich, CH Beck 2007) § 242, paras [184], [212]. 8   F Regelsberger, Pandekten 1 (Leipzig, Duncker & Humblot 1893). 9  H Dernburg, System des Römischen Rechts, 8th edn, vol 1 (Berlin, HW Müller 1911) § 118.

36  James Gordley bound and yet free’.10 Contemporary jurists look back on the exceptio doli generalis as an ancestor of the doctrine of abuse of rights. It is true that a judge would grant the exceptio if he thought that a plaintiff has misused his right to bring a claim. Dernburg actually used that phrase—misuse of formal right—although he added that the exceptio should be granted in other cases as well ‘without regard to the intention of the plaintiff’.11 But the exceptio was not a doctrine, as abuse of rights purports to be. The point of the exceptio was to prevent inequities when there was no formal doctrinal basis for relief. That is why examination of the exceptio doli generalis cannot give any definite guidance for EU law. The exceptio allowed a judge to do what was fair without definite guidance. For guidance, then, we would need an account of the abuse of rights that is more definite than the exceptio doli generalis, and less narrow than the principle that one cannot use a right solely to harm another person, as in aemulatio and its later incarnations. In the late nineteenth and early twentieth century, French jurists tried to give such an account. Their doctrine of abuse of rights was later borrowed in Germany. In France, the doctrine became the focus of a bitter debate over the nature of legal rights. To understand the doctrine, we will have to take a close look at that debate. In the end, I am afraid, we will still not know whether EU law should adopt a doctrine of abuse of rights. We will know merely that it should not adopt the French doctrine. The French proponents of this doctrine claimed to be leading an attack on the school of legal thought that had dominated the nineteenth century. The members of this school, they said, had regarded legal rights as ‘absolute’ and so misunderstood the nature of legal rights. The exercise of a right should be limited by the purpose that led the law to recognise such a right. A person abused a right when he exercised it in a way that did not serve this purpose. René Demogue explained, ‘the expression . . . abuse of rights’ indicates ‘a problem of the limits of every right . . . Does it not have limits of a kind which are teleological or social? Is it not necessary to understand, as included in every provision of law, [the qualification that] the right hereby recognized can only be exercised for motives in accordance with good social order?’12 According to Louis Josserand, ‘to abuse [a right] is to proceed, intentionally or unintentionally, against the purpose of the institution of which one has misunderstood the finality and the function’.13 They claimed that French courts had already retreated from the old view that rights are absolute in decisions like that of the Court of Appeal of Colmar described earlier,14 a decision they considered to be path-breaking.15 If rights were absolute, a property owner could not have been prevented from building a false chimney to block his neighbour’s window. Therefore, beginning with that decision, the courts had accepted a new conception in which rights have limits that are ‘teleological and social’ and correspond to their ‘finality and function’. Tacitly, the courts had accepted a doctrine of abuse of rights.

 A Brinz, Lehrbuch der Pandekten, 2nd edn, vol 1 (Erlangen, A Deichert 1873) § 108.  Dernburg (n 9 above) § 118.  R Demogue, Traité des obligations en général, vol 4 (Paris, Rousseau 1924) 679. 13  L Josserand, De l’esprit des droits et de leur relativité: théorie dite de l’abus des droits, 2nd edn (Paris, Dalloz 1939) 245. 14  Colmar (n 6 above). They claimed that a similar principle was at stake in cases of liability for vexatious litigation: Demogue (n 12 above) 635–41; Josserand (n 13 above) 39–62. Josserand found analogies in cases involving mortgages, family authority, and corporate and administrative law: Josserand (n 13 above) 33–38 (mortgages); 63–83 (family authority); 155–91 (corporate law); 192–97 (administrative law). 15  A di Robilant, Abuse of Rights: The Continental Drug and the Common Law (Manuscript, 2007) 66. 10 11 12

The Abuse of Rights in the Civil Law Tradition  37 Critics of the doctrine, such as Marcel Planiol and Georges Ripert, claimed that the courts had done no such thing. In these cases, they had merely prevented a person from exercising a right for the sole purpose of harming another. That was nothing new. It was an application of a principle that was born of the Roman aemulatio, accepted into old French law, and endorsed by the nineteenth-century jurists themselves.16 No one had ever believed that rights were ‘absolute’.17 Consequently ‘the only novelty’ in the new doctrine was the expression ‘abuse of rights’,18 an expression which they regarded as misleading and dangerous. As we have seen, Planiol and Ripert were correct that that there was nothing new about giving relief when a person used his land solely to harm his neighbour. The decision by the court of Colmar was neither path-breaking nor unprecedented, as Antonio Gambaro has pointed out.19 When Charles Demolombe criticised relief in such cases, he admitted that his was the minority view. But it was misleading for them to say that no one had ever believed that rights were ‘absolute’, and to claim that, therefore, the proponents of the new doctrine were not innovating. The jurists who had dominated the nineteenth century in France, Germany, and elsewhere had defined rights in terms of the will of the right holder without regard to the purposes which the right served. The right to property was defined as the entitlement of the owner to dispose of what he owned according to his will.20 It was a ‘despotic’ power,21 ‘a total legal subjection of a thing’,22 a ‘right . . . of complete domination over a physical object’.23 Contractual rights were created by the will of the parties.24 The innovation, as I have described elsewhere,25 was not to use the concept of will to define property or contract. Jurists had done so for centuries. The innovation was to define contract and tort solely in terms of the will, without any limit to what the parties willed, and without regard to the purposes served by institutions such as property and contract. Jurists such as Demogue and Josserand broke with this tradition when they claimed that rights must be understood ‘teleologically’ by looking to their ‘finality and function’. For that reason, the truth of Ripert’s and Planiol’s claim that no one had ever believed that rights were ‘absolute’ depends upon what one means by ‘absolute’. The nineteenth century jurists did believe that, conceptually, property entailed an unlimited right to use 16  M Planiol and G Ripert, Traité pratique de droit civil français, vol 6 (Paris, LGDJ 1930) 573; G Ripert, La règle morale dans les obligations civiles, 3rd edn (Paris, LGDJ 1935) 90. 17  Planiol and Ripert (n 16 above) 573. 18  Planiol and Ripert (n 16 above) 573. 19  A Gambaro, ‘Abuse of Rights in the Civil Law Tradition’ (1995) 3 European Review of Private Law 561, 566. 20  C Aubry and C Rau, Cours de droit civil français d’après la méthode de Zachariae, 4th edn, vol 2 (Paris, Cosse, Marchal & Billard 1869) § 190; F Laurent, Principes de droit civil français, 3rd edn, vol 6 (Paris, Bruylant 1875) § 101; B Windscheid and T Kipp, Lehrbuch des Pandektenrechts, 9th edn, vol 3 (Frankfurt, Rütten & Loenig 1906) §167. 21  Demolombe (n 5 above) 9: § 543. 22   GF Puchta and A Rudorff, Cursus der Institutionen, 3rd edn, vol 2 (Leipzig, Breitkopf & Härtel 1851) § 231. 23  L Arndts von Arnesberg, Lehrbuch der Pandekten, 14th edn (Stuttgart, Cotta 1889) §130. 24  Demolombe (n 5 above) 24: §12; L La­rom­bière, Théorie et pratique des obligations, vol 1 (Paris, Durand 1857) § 41; Laurent (n 20 above) 15: §§ 424–27; GF Puchta, Pan­dekten (Leipzig, Barth 1844) §§ 49, 54; FC von Savigny, System des heutigen römischen Rechts, vol 3 (Berlin, Veith & Comp 1840) § 134; B Wind­scheid and T Kipp, Lehrbuch, vol 1 (Frankfurt, Rütten & Loenig, 1891) § 69. 25  J Gordley, The Philosophical Origins of Modern Contract Doctrine (Oxford, Clarendon Press 1991) 161–213.

38  James Gordley what one owned as one willed. In that sense, the right was absolute. Nevertheless, it did not follow that absolutely anything an owner did counted as a valid exercise of his will. Nor did it follow that his rights were absolute in the sense of being sacrosanct or inviolable. While the right to property was unlimited conceptually, the law could still impose limits on its exercise. One can subscribe to a ‘will theory’ of property or contract and not believe that anything the right holder does will count as a valid exercise of his will. One might think, for example, that to allow a person to sell himself into slavery would be inconsistent with the reasons why the will should be respected. Some will theorists, such as François Laurent in France and Bernhard Windscheid in Germany, thought it was inconsistent with the reasons for respecting the will to allow an owner to use his property for the sole purpose of harming another. To respect the will was to respect a person’s right to order his own life. That right, Laurent said, is ‘necessary for [one’s] physical, intellectual and moral life’.26 Others disagreed, such as the French jurist Demolombe and the German jurist Carl Georg von Wächter. They thought that any enquiry into an owner’s motives violated his ‘absolute’ right to do ‘as he wills’,27 ‘to exercise his right as far as it extends . . . ’.28 Demolombe thought that French courts should not be giving relief, and Wächter read the Roman texts on aemulatio narrowly to concern only the cultivation of fields. All of them agreed that property was to be defined in terms of the owner’s will, and that his will should be respected. They disagreed over what that respect entailed. Moreover, one can believe that, conceptually, property or contract entail an unlimited right for the owner to do as he wills, or the contracting parties to agree on what they will, and still believe that the law can impose its own limits on the exercise of this right. A good illustration is the way the will theorists explained the limits the law placed on a property owner’s right to conduct activities that made noise or smoke or otherwise bothered his neighbours when he was not acting for the sole purpose of bothering them. French jurists, such as Demolombe and Laurent, and German jurists, such as Windscheid, agreed that these limits conflicted with the unlimited right of an owner to use his property as he chose. Nevertheless, it was necessary for the law to impose these limits. As Demolombe said, if all property owners could ‘invoke their absolute right, it is clear that none would have one in reality’. What would be the result? ‘It would be war! It would be anarchy!’29 Laurent explained that because the owner could use his property as he chose ‘[a]ccording to the rigor of the law, each proprietor would be able to object if one of his neighbors released on his property smoke or exhalations of any kind, because he has a right to the purity of air for his person and his goods’.30 If that were so, he admitted, the existence of towns would be impossible.31 German jurists typically defined property as an unlimited right and then described these limits as a ‘statutory abridgement of property’ (gesetzliche Beschränkung des Eigenthums).32 According to Windscheid, the abridgement was needed 26  Laurent (n 20 above) 20: § 411; Windscheid and Kipp (n 24 above) 1: § 121. The same position was taken by KA von Vangerow, Lehrbuch der Pandekten, 7th edn (Marburg, Elwert 1863) § 297. 27  Demolombe (n 5 above) 11: § 66; 12: § 19. 28  CG von Wächter, Pandekten (Leipzig, Breitkopf & Härtel 1880) § 33. 29  Demolombe (n 5 above) 9: § 543. 30  Laurent (n 20 above) 6: § 144. 31  Laurent (n 20 above) 6: § 144. In a later volume of his work, he finally decided that ‘[t]he Code was wrong to say that the owner has the right to enjoy and to dispose of his thing in the most absolute manner . . . ’ Laurent (n 20 above) 20: §417. Nevertheless, he did not suggest any other way that property could be defined. 32  Puchta and Rudorff (n 22 above) 2: § 231; KA von Vangerow, Leitfaden für Pandekten-Vorlesungen (Marburg, Elwert 1847) § 297 Anm II; Windscheid and Kipp (n 24 above) 1: § 169.

The Abuse of Rights in the Civil Law Tradition  39 because the ‘reckless realization of the consequences of the concept of property is not possible without serious disadvantages’.33 For these jurists, then, property is an unlimited right to use what one owns as one chooses because that is what the concept of property means. In employing that concept, the law cannot mean anything else, or it would be employing a different concept. But the ‘consequences of the concept of property’ may have ‘serious disadvantages’—even ‘anarchy’ or the end of ‘the existence of towns’. Therefore, the law sets limits to the exercise of the right to property which it recognises. The nineteenth-century jurists explained the matter that way because their approach was formalistic or conceptualistic as distinguished from teleological or functional. Property rights were not defined as those which an owner should have for the institution of property to serve the purposes which led the law to recognise that institution. Property was defined without regard to these purposes, and the law then needed to impose limitations on the exercise of property rights to ensure that its purposes were accomplished. The right of the owner was unlimited conceptually, and in that sense absolute, but it was not absolute in the sense that the law could not impose limits. The innovation of the proponents of the doctrine of abuse of rights was that they rejected this older formalistic or conceptualistic approach in favour of one in which rights are understood ‘teleologically’ in terms of their ‘finality and function’. Moreover, their adversaries such as Planiol and Ripert did not defend the old formalism. They themselves agreed rights were not ‘unlimited’34 or ‘absolute’.35 They did not try to explain institutions such as property as entailing, by definition, an unlimited right of the owner. They ignored the fact that the nineteenth-century jurists had defined property that way. Ripert built a theory of the inherent limitations of rights that was sufficiently close to the doctrine of abuse of rights that he admitted that the doctrine ‘seems in appearance like the one we are defending’.36 No one, then, was trying to revive the old formalism. The issue was what to do without it: to adopt the doctrine of abuse or rights, or something else. Planiol and Ripert rejected the doctrine because it seemed dangerous. It was too extensive and unlimited. It suggested, not merely that a court should consider the purposes for which rights are recognised, but that it should ‘remedy every wrong’,37 that ‘the judge can ask of each man an account of the motives of his acts’.38 One reason that critics such as Planiol and Ripert thought it dangerous to give a judge such a power was that they thought that it would be used to sacrifice individual rights to judicial conceptions of social needs. One source of their fear was that the proponents of the doctrine themselves did not distinguish clearly between interpreting rights according to purposes and interpreting them to promote social purposes, which, they believed, meant promoting a society that was less individualistic and more socially concerned. When Demogue said that there must be ‘teleological or social’ limits to rights, the two words meant much the same thing to him. For him and for Josserand, the error of the will theorists was not simply to misunderstand the nature of rights but to allow individual   Windscheid and Kipp (n 24 above) 1: § 169.  Planiol and Ripert (n 16 above) 574.  Ripert (n 16 above) 91. 36  Ripert (n 16 above) 103. 37  Planiol and Ripert (n 16 above) 574. 38  Ripert (n 16 above) 103. 33 34 35

40  James Gordley choice to trump social concerns. Demogue favoured the doctrine of abuse of rights because it was ‘in harmony with the reaction which is taking place against individualistic ideas. Today, one wants the individual to act in the general interest’.39 Josserand described the ‘dogma of the absolutism of rights’ not merely as formalistic but as ‘egotism in its juridical form’ which is ‘no less dangerous or less sterile than in any other form’.40 The doctrine of ‘abuse of rights constitutes a living and moving theory of great suppleness, an instrument of progress, a method of adapting law to social needs’.41 Their critics responded by claiming that their doctrine would subvert individual rights. It threatened ‘individualistic doctrines’, which, according to Planiol, ‘have been an effective agent in the struggle against despotism’.42 According to Ripert, Josserand’s theory ‘tends to destroy the idea of right. Now this idea, far from being anti-social, is indispensible to the preservation of a society threatened by statism and communism’.43 Thus, as Anna di Robilant has noted: In the skirmish over abuse of rights, the highest peak of tension was reached when the two factions contended over opposite visions of society. The question of abuse of rights was critical to the debate over individualism and solidarism, urging jurists of various tendencies to confront it.44

It would be a mistake, then, to see the doctrine of abuse of rights merely as a critique of the formalism of the nineteenth century jurists. Moreover, as a critique of formalism, it was flawed for reasons that its critics were eager to point out. In a famous passage, echoed by Ripert,45 Planiol argued: ‘To speak of an abuse of rights is . . . to fall into a logomachie; for, if I use my right, my act is licit, and when it is illicit, that is when I go beyond my right and act without right.’46 The argument was persuasive even to some, like Léon Duguit, who sympathised politically with Demogue and Josserand, and wished for a ‘socialisation’ of private law. The doctrine was a ‘contradiction’. ‘To say that . . . one who abuses his right does not enjoy any legal protection is to say, quite simply, that he is doing a thing that he does not have the right to do’.47 In answering this argument, Josserand reintroduced the formalism he had been trying to escape. The word droit in French means both ‘right’ and ‘law’. In the one sense, as Josserand noted, it means ‘a determinate prerogative’, such as ‘property, servitudes, paternal power’. In the other sense, it means ‘the ensemble of social rules’, such as ‘the civil law’, for which he coined the unfortunate word juricité. ‘The abusive act,’ he said, ‘is, quite simply, one which is done by virtue of a right whose limits have been respected, [droit in the one sense] but is nevertheless contrary to law envisaged in its juricité, that is to say, as an obligatory body of social rules [droit in the other sense].’48 That answer makes sense, however, only if the ‘limits’ of a right do not correspond to what Josserand called ‘finality and function’. If they did, the scope of the right,  Demogue (n 12 above) 679.   Josserand (n 13 above) 233. 41   Josserand (n 13 above) 247. 42  Planiol and Ripert (n 16 above) 574. 43  Ripert (n 16 above) 103. 44   di Robilant (n 15 above) 90. 45  Ripert (n 16 above) 91. 46  Planiol and Ripert (n 16 above) 574. 47  L Duguit, Les transformations générales du droit privé depuis le Code Napoléon, 2nd edn (Paris, Alcan 1920) 200 (citing Planiol). 48   Josserand (n 13 above) 245. 39 40

The Abuse of Rights in the Civil Law Tradition  41 interpreted in terms of its finality and function, would exclude the possibility of abuse. But if its scope has been defined without regard to the purposes which the recognition of the right serves, then we have returned to formalism. As we have seen, the will theorists defined property without regard to its purpose, and so they had to imagine that the law first granted too broad a right, and then had to curtail the right it had just granted. Josserand had to imagine that the law set the ‘limits’ of a right too broadly to accord with its ‘finality and function’ so that the right holder would have something to abuse. The abuse of rights doctrine was then needed to curtail the right the law had just granted. The doctrine may have appealed to its proponents, less because it was a sound critique of formalism than because they hoped it would be ‘an instrument of progress’ toward a less individualistic and more socially concerned society. IV. How the Doctrine has been applied since the controversy died down

Despite its weaknesses as a critique of formalism, the doctrine passed into Germany in the 1930s. Hans-Peter Haferkamp49 has traced it to a book by Wolfgang Siebert published in 1934.50 As Haferkamp notes, Siebert borrowed the doctrine from the French, where, Siebert himself explained, ‘the doctrine of the limits to the use of rights’ is ‘best developed’.51 Siebert formulated the doctrine in the same way as the French: whether rights are abused is to be determined by ‘an immediate consideration of the end of the rights’.52 He called that formulation an Innentheorie. He discussed Planiol’s objection that the doctrine is a logomachie, and adopted Josserand’s answer to Planiol.53 Like the French, Siebert claimed that the doctrine had been implicitly accepted in German law. For this purpose, the German equivalent of aemulatio was unsuitable. As we have seen, it had been enacted as Section 226 of the German Civil Code which provided that ‘the use of a right is not permitted when it can only have the purpose of causing harm to another’. Demogue and Josserand had claimed that their French cases of false chimneys and the like stood for a larger principle. In Germany, the principle was limited and clear. Siebert acknowledged that it did not go far enough.54 So he turned to cases of so-called Verwirkung, decided by the German courts, and pigeonholed by commentators under the requirement of Section 242 that the parties act in good faith. In these cases, a person who failed to exercise a right for a sufficiently long time was said to lose it. The delay had made the exercise of the right unfair to the other party. Siebert called them cases of ‘disloyal delay’.55 It might seem that they, too, could be explained by narrower principles: that a person can lose rights by sleeping on them or by leading others to believe he will not exercise them. Siebert claimed that the correct explanation was that such a party has made an ‘impermissible use of right’, an ‘abuse of right’ (Rechtsmissbrauch). 49  H-P Haferkamp, Die heutige Rechtsmissbrauchslehre: Ergebnis nationalsozialistischen Rechtsdenkens? (Berlin, Spitz 1995); H-P Haferkamp in M Schmoeckel, J Rückert and R Zimmermann (eds), Historischkritischer Kommentar zum BGB, Band II, Schuldrecht: Allgemeiner Teil §§ 241–432 (Tübingen, Mohr Siebeck 2007) § 242, paras [74–76]. 50   W Siebert, Verwirkung und Unzulässigkeit der Rechtsausübung (Marburg, Elwert 1934). 51  Siebert (n 50 above) 68. 52  Siebert (n 50 above) 155. 53  Siebert (n 50 above) 73–74. 54  Siebert (n 50 above) 226. 55  Siebert (n 50 above) 2.

42  James Gordley Siebert published his book in 1934. His theory was an immediate success. Haferkamp cites seventeen jurists who adopted it by 1942.56 Presumably, the reason was not that German academics felt an urgent need to explain Verwirkung and were dissatisfied with a narrower explanation. The reason, Haferkamp concluded,57 was that these jurists thought it would promote a society that was less individualistic and more in line with national socialist principles, just as its French proponents thought it would promote one that was less individualistic and more socially concerned. Siebert’s own allegiance is clear. He explained that ‘[t]he penetration of formal rights with legal-ethical and with social-ethical content not only accords with German legal thought but appears necessary to avoid a legal order that is foreign to the Volk and to the development of a unified Weltanschauung for the law’.58 In support, he cited Adolf Hitler addressing the 1933 Juristentag: ‘The total state will suffer no distinction between law and morality.’59 Siebert acknowledged that the doctrine could lessen the certainty of the law. But, he said, ‘legal certainty matters less to the interest of the Volk than rightness (Gerechtigkeit)’.60 Moreover, ‘[l]egal uncertainty will vanish more and more, the stronger and surer the German national-socialist concept of law takes hold’.61 Haferkamp believes that the doctrine of abuse of rights in Germany is still stamped by its Nazi origins. He points out that ‘the principal aspects of Siebert’s doctrine remain’. For example, abuse of rights is still explained in the way Siebert explained his Innentheorie.62 It is true that the doctrine has remained. It had become entrenched in the German case law as courts used it as a peg on which to hang decisions in many more contexts than Verwirkung. But the reason that so many jurists still speak of the ‘abuse of law’ is that no one thinks today that taking account of the purpose of a right commits one to building a national socialist society. It is the same in France. As Anna di Robilant observed, the doctrine disappointed those who thought it would help forge a new social order.63 Consequently, it no longer seems threatening. V. Conclusion

We conclude that the way the doctrine came to be formulated in France and Germany does not suggest that it can be useful for understanding EU law. It began as a critique of formalism, and it spread, not because it was a good critique, but because its proponents thought it could advance their political and social vision. Today, when jurists speak of ‘abuse of rights’ they do not have in mind any specific alternative to formalism, nor are they seeking to promote one sort of society rather than another. They seem to have little in mind beyond the principle that laws recognising rights, like other laws, should not be applied in a manner inconsistent with the purpose for which the law is made. But that principle is as old as Aristotle. It is how he described ‘equity’ or epikeia. If the principle is correct, the European Court of Justice, like any other court, should follow it whatever  Haferkamp (n 49 above) Historisch-kritischer Kommentar to § 242, fn 594.  Haferkamp (n 49 above) Rechtsmissbrauchslehre. 58  Siebert (n 50 above) 155–56. 59   Völkischer Beobachter, 5 October 1933, cited in Siebert (n 50 above) 156, fn 8. 60  Siebert (n 50 above) 155. 61  Siebert (n 50 above) 154. 62  Haferkamp (n 49 above) Historisch-kritischer Kommentar to § 242, para [82]. 63   di Robilant (n 15 above) 2. 56 57

The Abuse of Rights in the Civil Law Tradition  43 law the Court is interpreting. But such a principle would underlie all legal doctrine, not a specific doctrine of ‘abuse of rights’. For the doctrine of abuse of rights to be helpful, the doctrine must enshrine some narrower principle. Nothing about the way the doctrine was originally formulated suggests what that narrower principle might be. Although the controversy that attended its formulation has died down, courts continue to speak of abuse of rights. The French do so largely in the cases like aemulatio which we have discussed. The Germans do so in many other situations as well. According to Filippo Ranieri, the courts of other continental jurisdictions reach similar results in similar situations even though, unlike the Germans, they do not use the expression ‘abuse of rights’. Ranieri concludes that abuse of rights is ‘a legal concept immanent in Europeanwide Judikatur and Praxis—at least in continental law’.64 The final possibility we will examine is that if we look at this Praxis, we can find a more specific principle which could be of use. According to Ranieri, the case law enshrines ‘[t]he principle that a misuse of right and an exercise of right that is contrary to good faith is prohibited is a principle immanent in all modern European legal orders’.65 One wonders if that principle is any more specific than the general one just discussed: that laws which recognise rights, like all laws, should be applied in a manner consistent with their purpose. When one looks at the case law, it is hard to see a more specific principle underlying the doctrine of abuse of rights. Indeed, it is hard to see what is gained by using the expression ‘abuse of rights’. Ranieri’s example is the use of voting rights by minority partners or shareholders of a company to hamper or distort decision-making. Such behaviour is impermissible in Germany, where it is called an abuse of rights, and in Italy, France and Belgium, where it is not.66 These cases of voting rights are an example of what Günther Roth, commenting on the German case law, has called ‘misuse of right in the narrow sense’. Roth has made one of the most successful efforts to sort the German cases into categories or, as the Germans put it, Fallgruppen. Roth has divided cases of the ‘misuse of right in the narrow sense’ into three Fallgruppen. The voting rights cases belong to the first of these categories which he calls the ‘violation of duty’ (Pflichtverletzung). A minority that uses their voting rights to obstruct the operations of their company violates a ‘duty of loyalty’.67 In other cases, the duty violated is one of ‘care’. For example, a landlord or an employer terminates a lease or fires an employee without due regard for an acute danger to his health or to a psychological crisis he was undergoing.68 Ranieri may well be right that courts throughout Europe give relief in such cases, though they may not use the expression ‘abuse of rights’. But, one must ask, what advantage is gained in using it? In these cases, to know whether a right has been abused, one must know whether a duty has been violated. To make that determination, one has to consider what duties a minority shareholder or a landlord or an employer might have. If one decides that a duty has been violated, it is hard to see what is gained by calling the violation an abuse of rights. A physician would violate his duty of loyalty to a patient if he used medical records for blackmail, and his duty of care if he removed the wrong organ during an operation. What would be gained by saying 64  F Ranieri, ‘Verbot des Rechtsmissbrauchs und Europäisches Gemeinschaftsprivatrecht: Entscheidung des Gerichtshofs der Europäischen Gemeinschaften vom 12. Mai 1998’ (2001) 9 Zeitschrift für Europäisches Privatrecht 165, 173. 65  Ranieri (n 64 above). 66  Ranieri (n 64 above) 174–76. 67  Roth (n 7 above) § 242, paras [214–16]. 68  Roth (n 7 above) § 242, para [215].

44  James Gordley that he thereby abused his right to consult the patient’s medical records or to operate on him? The same can be said of cases that Roth includes in another category of ‘misuse of right in the narrow sense’. A right is misused, not because a duty was violated, but because of the manner in which a right was exercised. For example, the right to terminate an employee is misused if a notice to terminate is given at an inappropriate time.69 Perhaps similar cases are decided in a similar way throughout Europe even though courts outside Germany do not use the expression ‘abuse of rights’. Again, however, one must ask what is gained by using it. To know if there was an abuse of rights, one must first consider whether the employer had the right to terminate at will, or to do so only at an appropriate time. The expression ‘abuse of rights’ does not help to answer that question. Once the question is answered, one does not need the expression ‘abuse of rights’. According to Roth, in a third group of cases, courts have found an ‘abuse of rights’ because of ‘the absence of an interest worthy of protection’. In the cases he cites, the purpose for which a person acted was objectionable. There would have been no ‘abuse’ if he had acted in the same way but for a proper purpose. Sometimes, the purpose was improper because the act was done solely to harm someone else. For example, as in the Roman texts on aemulatio, a landowner used his property merely to bother a neighbour.70 In another case, to vex a tenant, a landlord forbid him to acquire ordinary household appliances.71 A classic example is an old case in which, to distress his son, a father refused him access to the land where he could visit his mother’s grave.72 In other cases, the purpose was improper for a different reason. It was retaliatory, as when a landlord terminated a lease because his tenant exercised his right to complain about noise or asked for a reduction in rent.73 It was discriminatory, as when an employer fired an employee because he was homosexual.74 It was immoral, as when a husband tried to dissolve one bigamous marriage in order to contract another.75 In these cases, to know if a right has been ‘abused’, one must first know whether a purpose is improper. If one determines that the purpose is improper, it is hard to see what is gained by speaking of an ‘abuse of rights’. A supervisor would act improperly if he told a joke to a female employee in order to embarrass her, or if he discriminated in promoting her, or if he fired her for complaining about discrimination, or if he offered her money for sex. What would it add to say that he has abused his right to tell jokes, to promote, to fire, or to spend his own money? As one might expect, if it is not helpful to speak of an abuse of rights in those cases, it is even less so when the expression is used in a broad sense. Roth includes cases in which a person sought to acquire a right by improper conduct, such as non-disclosure to another contracting party, or murdering his testator;76 cases such as Verwirkung in which a person forfeited a right by sleeping on it or letting another party think it would not be asserted;77 cases of venire contra factum proprium in which, as in common law estoppel, one person  Landesarbeitsgericht München, 6 July 1949 [1950] Neue Juristische Wochenschrift 399.  Roth (n 7 above) § 242, para [215]. 71   BGH, 8 May 1963 [1963] Neue Juristische Wochenschrift 1539. 72  RG, 3 December 1909, RGZ 72, 251. 73  Roth (n 7 above) § 242, para [215]. 74   BAG, 23 June 1994 [1995] Neue Juristische Wochenschrift 275. 75   BGH, 29 April 1959, BGHZ 30, 140. 76  Roth (n 7 above) § 242, paras [219], [226–30], [240]. 77  Roth (n 7 above) § 242, para [238]. 69 70

The Abuse of Rights in the Civil Law Tradition  45 misled another who then changed his position;78 and cases in which a person lost his right to the performance of another party’s duty by failing to perform his own.79 In these cases, there is a reason why a party either failed to acquire a right or subsequently lost one: he failed to disclose, or committed a murder, or waited too long, or misled someone, or failed to perform. One can call it an ‘abuse’ for him then to assert a right which he never had or now has lost. But the reason he cannot assert the right is that he does not have it. That point has been made by Jürgen Schmidt. When a person is said to make an ‘impermissible use of right’, he loses, ‘not because of his claim of right . . . but because he doesn’t have a right’. ‘Correct doctrine’ according to Schmidt, would reformulate statements about when a right is impermissibly used as statements about when a right is acquired or lost, independent of its use.80 Looking at the German cases, Antonio Gambaro reached a similar conclusion. ‘[U]nder the name of abuse of rights, many different rules can be collected together. [C]lassifications are superfluous since in no way can they give a guideline for future decisions exceeding the general criterion of stare decisis.’81 Ranieri may be right that similar cases are decided throughout Europe without speaking of an ‘abuse of rights’. But the reason may be, not because such a doctrine is immanent in the cases, but that the cases can be decided perfectly well without it. Even if the doctrine had been more useful in private law, however, it would not be useful in EU law unless the problems are sufficiently similar. But they are not. In the cases that have come before the European Court of Justice, a person acted for the sole purpose of escaping taxation or regulation. In one case that we have described, that of the bigamous marriage, a person acted for an immoral purpose. In the other cases, a person’s conduct was objectionable because it harmed someone else, whether by the violation of a duty, or the improper exercise of a right, or by an action done solely to inflict harm, or to discriminate or to retaliate. The problem in the EU cases is not to determine whether a person acted for an immoral purpose. In and of itself, it is not immoral to try to escape taxation or regulation. Neither is the problem to determine the boundary between one person’s freedom to act and another person’s security against harm. The problem is to accommodate the national schemes of taxation and regulation of the Member States with principles of free movement of labour, capital and goods of the European Union. It is hard to see how the private law doctrine of abuse of rights could help with those problems even if it could tell us when a person is acting immorally or unjustifiably harming someone else. Antonio Gambaro has called ‘abuse of rights’ a ‘doctrine without a future’ in private law.82 One should hope this is right, given the doctrine’s past and present. The French jurists who formulated it overgeneralised cases like the Roman aemulatio in order to build a critique of nineteenth-century formalism. The doctrine found supporters in France and later in Germany, not because of the merits of this critique, but despite its flaws. It promised to be a vehicle for political and social change. The doctrine survived, not because this promise was fulfilled, but largely because it was forgotten. In the cases in which it has been applied, the doctrine is not grounded on any distinct notion of ‘abuse’.  Roth (n 7 above) § 242, para [235].  Roth (n 7 above) § 242, paras [355, 357].   J Schmidt, Staudingers Kommentar zum Bürgerlichen Gesetzbuch, 13th edn (Munich, Sellier/de Gruyter 1995) § 242, para [68]. 81   Gambaro (n 19 above) 569. 82   Gambaro (n 19 above) 570. 78 79 80

46  James Gordley ‘Abuse’ is a label applied to conduct when there is some independent reason for thinking the conduct is objectionable. Even if it were not so, and even if one could identify a concept of ‘abuse’ that explains when conduct is immoral or impinges unjustifiably on others, one wonders how much light it could shed on the right way to reconcile national systems of taxation and regulation with EU principles of free movement of goods, capital and persons. Should we conclude, then, that the European Union should not adopt a doctrine of abuse of rights? Certainly not. We can only conclude that the European Union should not borrow such a doctrine from private law. Because the problems that the European Union faces are quite different from those of private law, it may well be that some new and different doctrine of abuse of rights may be useful in resolving them. But it will have to be new and different.

5 Fitting ‘Abuse of Rights’ into EU Law Governing the Free Movement of Goods and Services Stephen Weatherill I. Introduction

T

he purpose of this chapter is, first, to explain the impact exerted by the free movement provisions of the TFEU on the competence of public authorities in the Member States to regulate commercial activity which is carried out on their territory; and, second, to consider where the notion of ‘abuse of rights’ might be accommodated within this legal framework. My concern is predominantly with the law governing the free movement of goods and services, which I analyse as a means to allocate regulatory competence over goods and services in the EU. That concerns both horizontal allocation of competence—between ‘home State’ and ‘host State’—and vertical allocation—between States and the EU as a source of harmonised rules. The chapter seeks to demonstrate how in the area of free movement of goods and services there is no need to develop an independent ‘abuse’ doctrine, for the anxieties that animate suppression of ‘abuse’ are already adequately embedded within the analytical framework of EU law, which puts to the test the claim of a ‘host State’ to assert regulatory competence in a way that will impede cross-border trade. II. The structure of the law governing the free movement of goods and services in the TFEU

The free movement provisions of the TFEU serve as instruments for the allocation of regulatory competence. Consider the trader based in State A who wishes to sell goods or services in State B. State A is the ‘home State’, State B the ‘host State’. Which State’s rules apply? If the trader is entitled to operate in State B solely by virtue of compliance with the rules laid down by State A, then State B’s regulatory expectations are completely undermined. Such a model of absolute ‘home State control’ is not envisaged by the Treaty. If, by contrast, the trader is entitled to operate in State B only on condition that State B’s rules are fully complied with, then any impetus towards the restructuring of markets in Europe and the intensification of inter-State competition is halted. Such a model of absolute ‘host State control’ is also not envisaged by the Treaty. The constitutional collision between home State and host State control is managed in a more sophisticated

50  Stephen Weatherill manner. Instead, the free movement provisions of the Treaty envisage a middle way—a conditional or non-absolute model of host State control. That means that State B may apply its rules to the activities of the trader based in State A wishing to operate in State B by, for example, requiring compliance with technical specifications applied to products or fulfilment of professional training as a pre-condition to supplying services—but only provided there is a sufficiently good reason for maintaining such rules. EU free movement law, then, puts host State regulatory choices to the test. It admits of limited regulatory autonomy. This model is visible in the TFEU. Article 34 is balanced by derogations envisaged by Article 36: Article 56 by Article 52 via Article 62. Free movement is driven hard by the Treaty, but limits are recognised within which the host State may demonstrate that its rules are justified despite their restrictive effect on inter-State trade.1 Both ends and means must be shown to be justified, in the sense that, in particular, if a Member State has a choice between various measures to attain the same objective it should choose the means which least restricts the free movement of goods.2 But the fact that the applicable model is one of conditional or non-absolute host State control ensures that EU free movement law is not a charter for remorseless deregulation—it does not force down standards of regulatory protection in the EU to the lowest level. This means that the fact that one Member State imposes less strict rules than another Member State does not necessarily mean that the latter’s rules are disproportionate and hence incompatible with EU law.3 A regulating State may set tougher standards than its competitors even if that causes an obstacle to free movement—if it can justify them. The institutional implication is then that if such obstacles to inter-State trade are to be eliminated it shall not be by judicial application of the free movement rules but rather by legislative intervention pursuant to the Treaty provisions authorising the harmonisation of laws. III. The development of the law by the Court OF JUSTICE

The Treaty therefore recognises free movement law as a principle of allocation of regulatory competence, placing conditional or non-absolute regulatory responsibility in the hands of the host State. But it is the case law of the Court of Justice which has fleshed out what is meant. The Court’s case law dealing with the Treaty rules on free movement is more than merely illustrative of what is at stake. The Court has gone so far as to develop a set of applicable legal principles which are certainly inspired by and faithful to the basic allocation of regulatory competence foreseen by the Treaty, but which are much fuller and more sophisticated than the skeletal terms of the Treaty. It is to the Court’s case law that one inevitably must turn to understand what constitutes a barrier to trade which requires justification; and it is to the Court’s case law that one must turn to grasp the nature of the conditional or non-absolute principle of host State control. The Court has converted the Treaty rules into a subtle, flexible and intrusive set of instruments for advancing market 1  The burden is placed on the regulator, eg Case 227/82 Van Bennekom [1983] ECR 3883; Case C-14/02 ATRAL SA v Belgian State [2003] ECR I-4431. 2  Eg, Case 261/81 Walter Rau Lebensmittelwerke v De Smedt PVBA [1982] ECR 3961; Case C-189/95 Harry Franzén [1997] ECR I-2471. 3  Eg, Case C-294/00 Deutsche Paracelsus Schulen für Naturheilverfahren GmbH v Kurt Gräbner [2002] ECR I-6515; Case C-3/95 Reisebüro Broede v Gerd Sanker [1996] ECR I-6511.

Fitting ‘Abuse of Rights’ into the Free Movement of Goods and Services  51 integration while seeking to take account of legitimate national interests in regulatory protection. Dassonville was a landmark ruling in this vein, so too Van Binsbergen.4 But Cassis de Dijon deserves its reputation as the most elegant expression of how and why national regulatory autonomy is subjected to the requirements of EU law insofar as the exercise of such autonomy damages market integration.5 That famous judgment has provided a structure for the debate about the proper scope of free movement law for fully thirty years. French-made blackcurrant liqueur could not be sold in Germany. Germany did not target its rules against imports. The problem was that German rules governing alcohol content were simply different from French rules. The French-made product did not contain enough alcohol for German requirements. This is the classic case in which conformity with the rules of the ‘home State’ (France) is insufficient to allow penetration of the market of the ‘host State’ (Germany) where the regulatory regime is different. Inter-State regulatory diversity causes a barrier to inter-State trade. The Court began its ruling by insisting on the primacy of host State control in the absence of common EU rules: ‘It is for the Member States to regulate all matters relating to the production and marketing of alcohol and alcoholic beverages on their own territory.’ But this initial concession to the host State is deceptive. The Court proceeded to qualify this statement and to subject the permissible application of host State rules to a significant threshold requirement. It added that: ‘Obstacles to movement in the Community resulting from disparities between the national laws in question must be accepted in so far as those provisions may be recognised as being necessary in order to satisfy mandatory requirements.’ The host State is therefore permitted to apply its rules only on condition that they carry a sufficient justification in the public interest to prevail over the interest in the integration of markets. The Court added that these ‘mandatory requirements’, a most inelegant phrase better rendered as ‘compelling interests’ or simply as ‘the public interest’,6 which may be advanced as counterweights to the interest in securing trade integration included ‘in particular the effectiveness of fiscal supervision, the protection of public health, the fairness of commercial transactions and the defence of the consumer’. They therefore go beyond the closed list of exceptions recognised by Article 36 TFEU. The core of the Cassis ruling inverts the initial embrace of host State control. The Court concludes that: There is therefore no valid reason why, provided they have been lawfully produced and marketed in one of the Member States, alcoholic beverages should not be introduced into any other Member State; the sale of such products may not be subject to a legal prohibition on the marketing of beverages with an alcohol content lower than the limit set by the national rules.7

Home State control applies in the absence of a sufficiently compelling basis for host State control. And Germany, the host State, had no good reason to insist that such 4  Case 8/74 Procureur du Roi v Benoît and Gustave Dassonville [1974] ECR 837; Case 33/74 Johannes Henricus Maria Van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299. 5  Case 120/78 Rewe Zentrale v Bundesmonopolverwaltung für Branntwein [1979] ECR 649. 6  On the question of mistranslation, see S Weatherill and P Beaumont, EU Law, 3rd edn (London, Penguin Books 1999) 575, fn 35. In fact, more recently, the Court has commonly simply adopted the language of public interest in referring to the availability of justification. 7   Cassis de Dijon (n 5 above), para [14].

52  Stephen Weatherill products shall contain more alcohol than required by France, the home State. So host State regulation of the product could not be justified. In the case itself, free movement law exerted a deregulatory impact, transferring power over the future commercial success of the French-made liqueur in Germany away from the German regulator into the hands of the German consumer. An appealing consequence of this model is that the competence to regulate is not left exclusively in the hands of either the home or the host State—nor is it transferred to the EU’s legislature. Instead, the model of conditional or non-absolute host State control involves a distribution of regulatory competences consequent on scrutiny of the merits of the competing claims. The host State regulates the matter—as long as it shows it has a sufficient reason to do so. If not, home State control is treated as adequate. And regulatory diversity, promoting choice and variety in the product and service markets of the EU, persists free of the imposition of common rules ‘from above’, by the EU as (harmonised) lawmaker.8 Cassis is a vintage that is now thirty years old but its taste is still readily discerned in the Court’s consistent approach today. In fact, over that thirty-year period, the principal subsequent changes in the flow of the mainstream of free movement case law have not concerned the Court’s ruling in Cassis itself but rather later perversions of it. The Keck ruling9 famously addressed the problem that even matters of purely local market regulation, lacking any connection with the quest to integrate markets, had been swept within the embrace of Article 34 TFEU and consequently had required justification. But Keck was not an assault on the essence of Cassis, rather, a retreat from its more enthusiastic though ill-advised extensions, among which the most striking was the Court’s pair of Sunday Trading cases.10 In fact, within the analytical framework favoured by this chapter, Keck could be read as an attempt by the Court to strip out from this area of the law instances where there is no need to pursue a host State/home State analytical framework because, in fact, the matter can be left for regulation by the host State without any damage being inflicted on cross-border trade patterns. But once a barrier to inter-State trade is identified—still a matter of persisting controversy11—the (logically subsequent) question of the circumstances in which the impugned measure may be treated as justified is still today addressed by the Court in a manner which reveals no material distinction from that elaborated thirty years ago in Cassis de Dijon.

    8   Cf S Weatherill, ‘Pre-emption, Harmonisation and the Distribution of Competence to Regulate the Internal Market’ in J Scott and C Barnard (eds), The Law of the Single European Market (Oxford, Hart Publishing 2002); N Bernard, ‘La libre circulation des marchandises, des personnes et des services dans le Traite CE sous l’angle de la competence’ (1998) 34 Cahiers de droit européen 11.     9   Joined Cases C-267 and C-268/91 Bernard Keck and Daniel Mithouard [1993] ECR I-6097. 10  Case 145/88 Torfaen BC v B&Q plc [1989] ECR 765; Case C-169/91 Stoke on Trent and Norwich City Councils v B & Q plc [1992] ECR I-6635. 11   Cf Case C-110/05 Commission v Italy; Case C-142/05 Åklagaren v Percy Mickelsson and Joakim Roos, both pending. See L Gormley, ‘Silver Threads among the Gold . . . 50 Years of the Free Movement of Goods’ (2008) 31 Fordham International Law Journal 1637; P Oliver and W-H Roth, ‘The Internal Market and the Four Freedoms’ (2004) 41 Common Market Law Review 407; S Enchelmaier, ‘The Awkward Selling of a Good Idea, or a Traditionalist Interpretation of Keck’ (2003) 22 Yearbook of European Law 249; A Dawes, ‘Importing and Exporting Poor Reasoning: Worrying Trends in Relation to the Case Law on the Free Movement of Goods’ (2007) 8 German Law Journal 761.

Fitting ‘Abuse of Rights’ into the Free Movement of Goods and Services  53

IV. The sensitivity of the Court’s role

This approach to the distribution of competences between home and host States in the quest to construct an integrated market covering the territory of all the Member States of the Union does not go so far as to dictate that either one State or the other but never both shall be permitted to regulate a particular product or service sector. The space allowed to a host State to justify its rules means that it is at least possible that dual regulation will occur. But these circumstances are limited by the rules of the Treaty, boldly interpreted by the Court. Over time, the Court has been invited to widen the scope of justifications which may permissibly be advanced by a host State seeking to maintain the application of rules which serve to restrict inter-State trade. It has normally been prepared in principle to do so. Plenty of cases illustrate the approach to assessing whether national rules that are claimed to protect the economic interests of consumers may be maintained insofar as they obstruct inter-State trade. The Court commonly concludes that the consumer is able to look after him- or herself and consequently finds in favour of the interest in integrating markets at the expense of national (host State) regulation12 but on occasion, usually where protective rules are targeted at the interest of vulnerable groups, it is persuaded that the deregulatory impetus of EU law should be halted. So the host State’s regulatory competence, where justified, is respected.13 Beyond the case of consumer protection, a broad spectrum of regulatory objectives have been accommodated within the Cassis de Dijon formula and, in cases involving restrictions to trade in both goods and services, their worth has been duly assessed on the facts of particular cases. The underlying question has not altered—has the host State shown a sufficient reason to justify application of its rules even where they hinder inter-State trade? But the type of national measures under scrutiny has moved far beyond the type of obvious trade-related and frequently witlessly protectionist measures that litter the early years of the European Court Reports. So trade integration may collide with protection of press diversity pursued by national rules,14 with protection of freedom of expression and freedom of assembly (which may be directed against importation of goods or services),15 with rules protecting human dignity,16 with protection of the child,17 with concern for animal welfare,18 and with rules designed to protect workers and the right to take collective action.19 12  Eg, Case C-470/93 Verein gegen Unwesen in Handel und Gewerbe Köln e.V. v Mars GmbH [1995] ECR I-1923; Case 178/84 Commission v Germany [1987] ECR 1227; Case C-154/89 Commission v France [1991] ECR I-659. Cassis (n 5 above) itself is also of this type. 13  Eg, Case 382/87 R Buet and Educational Business Services (EBS) v Ministère public [1989] ECR 1235; Case C-441/04 A-Punkt Schmuckhandels GmbH v Claudia Schmidt [2006] ECR I-2093. 14  Case C-368/95 Vereinigte Familiapress Zeitungsverlags- und vertriebs GmbH v Heinrich Bauer Verlag [1997] ECR I-3689. 15  Case C-112/00 Schmidberger v Austria [2003] ECR I-5659. 16  Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-9609. 17  Case C-244/06 Dynamic Medien Vertriebs GmbH v Avides Media AG [2008] ECR I-505. 18  Case C-219/07 Nationale Raad van Dierenkwekers en Liefhebbers VZW, Andibel VZW v Belgische Staat [2008] ECR I-4475. 19  Case C-438/05 International Transport Workers’ Federation and Finnish Seamen’s Union v Viking Line ABP and OÜ Viking Line Eesti [2007] ECR I-10779; Case C-341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet, Svenska Byggnadsarbetareförbundets avdelning 1, Byggettan and Svenska Elektrikerförbundet [2007] ECR I-11767.

54  Stephen Weatherill These cases demonstrate the sensitive nature of the adjudicative role that the Court has claimed for itself (and for its national judicial allies) in its extended reading of the scope of the law of free movement. The host State enjoys conditional regulatory autonomy— the precise determination of whether or not it has met the necessary conditions belongs with the Court. Frequently the Treaty offers little, if any, help in defining just what EU law should make of, for example, the place of press diversity or animal welfare. In particular, in connection with the cases that have set economic rights embedded in the TFEU against social and political rights and freedoms guaranteed under national law (and recognised within EU law), concerns have been raised about the Court’s strained legitimacy in assuming a task—weighing matters of social constitutional law against economic considerations—which, it may be argued, the Treaty does not authorise and for which the Court consequently lacks valid credentials.20 This tension is endemic in EU trade law given the absence of ‘hard’ limits to its reach dictated by the European Treaties. It is in this regard noticeable that at least rhetorically the Court has accepted the importance of balancing the several interests at stake. Free movement does not have any automatic priority and in this sense the Court is in such cases particularly sensitive to protection of the host State’s (admittedly conditional) regulatory autonomy, declaring it to enjoy a margin of discretion in such circumstances.21 V.  ‘Abuse of rights’ and the law of free movement: the basic framework

That the Court’s role in determining the scope and nature of the law of free movement is significant and sensitive is beyond dispute. Deeper exploration, however, belongs outwith this chapter’s remit. The current concern is with ‘abuse of rights’. Where does ‘abuse of rights’ fit into the model of the law of free movement? The essence of an allegation of ‘abuse’ is that host State control should be permitted to apply without restraint. Put another way, successfully depicting a claim to rely on the free movement rules as an ‘abusive’ claim entails that the home State’s regulatory choices are dismissed as simply not relevant. The host State would enjoy unconditional regulatory competence. That might, in turn, engage the connected argument that free movement law is, in the circumstances, inapt to drive forward integration and that instead, the route to opening up cross-border competition should lie in the development of harmonised rules agreed at EU level. 20   Viking Line and Laval (n 19 above) are especially fertile sources. See eg, A Davies, ‘One Step Forward, Two Steps Back?’ (2008) 37 Industrial Law Journal 126; P Syrpis and T Novitz, ‘Economic and Social Rights in Conflict: Political and Judicial Approaches to Their Reconciliation’ (2008) 33 European Law Review 411; P Rodiere, ‘Les arrêts Viking et Laval, le droit de grève et le droit de négociation collective’ (2008) 44 Revue trimestrielle de droit européen 47; L Azoulai, ‘The Court of Justice and the Social Market Economy: The Emergence of an Ideal and the Conditions for its Realisation’ (2008) 45 Common Market Law Review 1335; D Wyatt, ‘Horizontal Effect of Fundamental Freedoms and the Right to Equality after Viking and Mangold, and the Implications for Community Competence’ (2008) 4 Croatian Yearbook of European Law and Policy 1; C Joerges and F Rödl, ‘Informal Politics, Formalised Law and the “Social Deficit” of European Integration: Reflections after the Judgments of the ECJ in Viking and Laval’ (2009) 15 European Law Journal 1. 21  Eg, Schmidberger (n 15 above), paras [81–82]; Dynamic Medien (n 17 above), para [44]. Views are admittedly mixed on how appropriate the Court’s ‘concessions’ in this vein really are: cf C Kombas, ‘Fundamental Rights and Fundamental Freedoms: A Symbiosis on the Basis of Subsidiarity’ (2006) 12 European Public Law 433; J Sweeney, ‘A Margin of Appreciation in the Internal Market: Lessons from the European Court of Human Rights’ (2007) 34 Legal Issues of Economic Integration 27; J Malmberg and T Sigeman, ‘Industrial Actions and EU Economic Freedoms: The Autonomous Collective Bargaining Model Curtailed by the European Court of Justice’ (2008) 45 Common Market Law Review 1115.

Fitting ‘Abuse of Rights’ into the Free Movement of Goods and Services  55 But does ‘abuse’ mean that the host State rules do not constitute a true barrier to interState trade at all or does it mean that they do constitute a barrier to inter-State trade but that they are justified in the circumstances? Or—can it cover both? In principle, this question matters a great deal. It is a matter of jurisdiction. If the matter is not a barrier to inter-State trade in the first place, then the regulatory autonomy of the host State is preserved unaffected by Articles 34 and 56 TFEU. If, however, this is a trade barrier, but one that is justified, then the matter is squarely within the scope of EU competence. This choice has several significant implications, among them the potential consequences for the scope of EU legislative competence. In practice, by contrast, this question may not matter. In determining whether in the particular circumstances that have arisen the host State rules may be set aside by the alleged ‘abuser’, it may not be important which of these two analytical routes is chosen. In either event, the surrounding circumstances must be examined and a successful claim of ‘abuse’ results in the host State being free to apply its rules. Put another way, the maxim ‘the law ends where the abuse begins’22 holds that there is no right on which to rely once the matter is deemed an abuse, rather than that the right is somehow suppressed as a consequence of its abusive character. But the inquiry in practice is not much assisted by such a conceptually pure formulation. One must in any event determine just what is involved in the notion of an ‘abuse’ and whether the particular circumstances disclose one. VI.  ‘Abuse’, ‘evasion’ and ‘circumvention’ in the Court’s case law on the free movement of goods and services

One of the earliest free movement cases in which the notion of ‘abuse’ features in the Court’s judgment is Knoors.23 Knoors was a Dutch national resident in Belgium where he worked as a plumber. The Dutch authorities refused to permit him to work in the Netherlands because he lacked the relevant Dutch qualifications. He was informed that as a Dutch national claiming against the Dutch public authorities, he could not take advantage of any right under the Treaty to have his experience in Belgium recognised. The Court, in receipt of a preliminary reference made by a Dutch court, disagreed. Knoors was a migrant worker. That he was seeking to rely on the Treaty against the State of which he was a national could not detract from that fact nor from the associated entitlements guaranteed by EU law. Before the Court, the Netherlands argued that such an approach created the risk that individuals might evade local regulatory requirements concerning matters such as training, by skipping across a border and then targeting their economic activities at the State they have quit. And the Court did not deny that this anxiety carried weight. It observed that: [I]t is not possible to disregard the legitimate interest which a Member State may have in preventing certain of its nationals, by means of facilities created under the Treaty, from attempting wrongly to evade the application of their national legislation as regards training for a trade.24

  M Planiol, cited by AG La Pergola in Centros (n 36 below).  Case 115/78 J Knoors v Staatssecretaris van Economische Zaken [1979] ECR 399. 24   Knoors (n 23 above), para [25]. 22 23

56  Stephen Weatherill This it described as a ‘risk of abuse’.25 But it was not present in the current case. A Directive ensured that adequate training had to have been undergone by an individual in the position of Knoors. And the Court added that it remained open to the EU legislature— at the time, the Council—to ‘remove the causes of any abuses of the law by arranging for the harmonisation of the conditions of training for a trade in the various Member States’.26 Knoors demonstrates the Court’s recognition that the law of free movement is capable of being abused. Specifically, it nods at the phenomenon of the sham cross-border transaction, where an individual crosses a border in order to escape local rules and take advantage of another State’s rules—a case of acquiring a sham ‘home State’, thereby converting the true home State into a host State and then seeking to rely on the Treaty to set aside those (sham) host State rules. But in Knoors there was no such abuse. And therefore there was no reason to restrain an assertion of (true!) home State control. The Court’s suggestion in Knoors that such matters could and would be dealt with by legislative solutions is typical of its time. In the late 1970s, harmonisation was seen as a good deal more politically feasible and economically desirable than it is today. But even in that decade the Court had demonstrated a vigorous readiness to employ the free movement provisions to tackle obstacles to inter-State trade without waiting for the careful crafting of legislative solutions.27 We know today that a great many obstacles to inter-State trade caused by diversity between national laws that might have been expected to be tackled by legislative harmonisation have instead been left in the hands of the judiciary, at national and EU level. That is the vigour and the richness of the evolved law of free movement. Already in Knoors the Court had hinted that it might place limits on the cutting-edge of the law of free movement—where reliance upon it involves attempting wrongly to evade the application of national rules. When might one find such a sham? A simple but uncommon example is provided by Leclerc v Au Blé Vert.28 This concerned French rules requiring books to be sold at the retail price fixed by the publisher. This discouraged the marketing of books published in France but sold elsewhere and then re-imported into France because it prevented the importer from passing on any advantage resulting from a lower price in the exporting State. It was consequently treated as a barrier to inter-State trade in goods. However, this did not apply to books exported from France ‘for the sole purpose of re-importation in order to circumvent legislation of the type at issue’.29 What seems to be at stake here is the crossing of a border as a means to seek to evade the rules of what was the home State and which the migration has converted into the host State—but with no subjection to the rules of the ‘new’ home State. This is a sham: it is not a cross-border situation at all. This, however, will be rare. And it cannot occur where the undertaking seeking to rely on EU law of free movement needs to set up some distribution or production base, however limited, in another Member State.

  Knoors (n 23 above), para [26].   Knoors (n 23 above), para [27]. 27  Eg, Case 2/74 Jean Reyners v Belgian State [1974] ECR 631; and plainly Cassis (n 5 above) itself is of this type. 28  Case 229/83 Association des Centres distributeurs Édouard Leclerc et al v SARL ‘Au Blé Vert’ et al [1985] ECR 1. 29   Au Blé Vert (n 28 above), para [27]. 25 26

Fitting ‘Abuse of Rights’ into the Free Movement of Goods and Services  57 TV10 v Commissariaat voor de Media provides an illustration.30 TV10 was a public limited company governed by Luxembourg law. It acted as a commercial broadcasting undertaking. It obtained authorisation from the Luxembourg authorities to transmit its programmes via the Astra satellite, which directed them to the territory of the Netherlands. The Dutch Commissariaat took the view that since day-to-day management of TV10 was largely in the hands of Dutch nationals and that most of its target audience was Dutch, TV10 was seeking improperly to evade the Dutch rules governing broadcasting. It therefore denied it the right to transmit its programmes by cable. The Court made plain that as far as it was concerned, the company’s choice of location in Luxembourg might perfectly well have been a means to escape the Dutch legislation, but that did not preclude the broadcasts being regarded as services within the meaning of the Treaty. That, the Court observed, ‘is distinct from the question of what measures a Member State may take to prevent a provider of services established in another Member State from evading its domestic legislation’.31 The Court then sited its analysis in the context of previous case law which accepts that the relevant Dutch regime is ‘intended to establish a pluralist and non-commercial radio and television broadcasting system and thus forms part of a cultural policy whose aim is to safeguard the freedom of expression in the audiovisual sector of the various components, in particular social, cultural, religious and philosophical ones, of the Netherlands’.32 Such rules could properly be applied to TV10 in order to prevent reliance on the freedoms guaranteed by the Treaty permitting it ‘wrongfully to avoid obligations under national law, in this case those designed to ensure the pluralist and non-commercial content of programmes’ (emphasis added).33 The Court held that it could not be regarded as incompatible with the Treaty for the Dutch authorities to treat TV10 as a domestic undertaking. This could be interpreted to mean that the Court regarded this as a sham—that there was in reality no true crossborder element to the case. But this seems implausible. After all, TV10 was established in Luxembourg. It seems more persuasive to interpret the Court’s decision as based on acceptance that the Dutch rules were barriers to inter-State trade in services but in the circumstances justified—that is, the Netherlands, as host State, could lawfully apply its rules to the provider of cross-border services just as it could apply those rules to Dutch firms. Whether or not one characterises TV10’s conduct as ‘abusive’ or ‘wrongful’ avoidance of Dutch law, the result is that it is unable in the circumstances to displace the competence of the host State to apply its rules on cultural preservation to undertakings operating on Dutch territory. The Opinion of Advocate-General Lenz in TV10 is helpful. He explained that the concern to avoid the rules of the host State had been described in its written observations by the Commission as a subjective criterion, but that the Commission had ceased to advance this view at the hearing. A subjective element made little sense to AdvocateGeneral Lenz in the context of provision of services by a legal person, and in any event, he plainly preferred an objective assessment of notions of wrongful avoidance of host State rules.  Case C-23/93 TV 10 v Commissariaat voor de Media [1994] ECR I-4795.   TV 10 (n 30 above), para [15]. 32   TV 10 (n 30 above), para [18], citing Case C-288/89 Collectieve Antennevoorziening Gouda v Commissariaat voor de Media [1991] ECR I-4007, paras [22, 23]; Case C-353/89 Commission v Netherlands [1991] ECR I-4069, paras [3, 29 and 30]; and Case C-148/91 Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487, para [9]. 33   TV 10 (n 30 above), para [21]. 30 31

58  Stephen Weatherill He also insisted on the need to exclude the relevance of the nationality of the founders of the company. Drawing on Factortame he stated that a link with the nationality of the natural persons behind the company is contrary to EU law where those persons are nationals of a Member State.34 It is submitted that Mr Lenz is correct. Moreover his approach elucidates that taken in earlier cases by the Court. The issue here is not nationality, but cross-border economic activity. That Knoors was a Dutch national subject to Dutch regulation was without significance. Even had he been Belgian, precisely the same analysis would have been undertaken—namely, is it justified for the Netherlands to insist on compliance with its rules as the ‘host State’ as a pre-condition to permitting the migrant to gain access to its market? Precisely the same point may be made about the earlier decision in Van Binsbergen.35 This strongly suggests that the key issue here is not that ‘abuse’ involves denial of any true barrier to inter-State trade but rather that ‘abuse’ is relevant to assessing whether the host State may justifiably impose its rules on a resisting migrant. Put another way, even to employ the notion of ‘evasion’ in these circumstances is unhelpful. What is really at stake is simply exercise of the Treaty-conferred right to free movement and a claim—by Knoors, by Van Binsbergen, by TV10—to be free of the regulatory burden imposed by the host State. EU law then provides a framework within which to decide whether the host State has a sufficiently compelling reason to justify application of those regulations, even given their detrimental effect on inter-State trade. In Knoors and Van Binsbergen it did not, in TV10, where the Court was impressed by the cultural aims of the rule, it did. VII. A glance at the Centros case

Centros is an unavoidable point of reference in this debate, even if the case concerns the right of establishment rather than the free movement of goods or services.36 Centros Ltd was a private limited company registered in England. It had never traded in the United Kingdom since its formation. It existed purely because its two shareholders, Danish nationals resident in Denmark, wished to take advantage of the favourable provisions of company law in the United Kingdom, specifically the absence of any requirement relating to a minimum paid-up share capital for such a company. The Danish Trade and Companies Board refused to register a branch of Centros Ltd in Denmark, insisting that in reality this was an attempt to establish in Denmark not a branch, but a principal establishment. This, it decided, was a deft ruse designed to circumvent the Danish rules concerning, in particular, the paying-up of minimum capital. As it was. But was such a deft ruse protected by the Treaty rules on free movement? Centros relied on the right of establishment protected at the time by Articles 52 and 58 EC [now Articles 49 and 54 TFEU]. Denmark argued before the Court that the fact pattern—Danish nationals wishing to carry on business through a branch in Denmark— was purely internal to Denmark. This was, in short, a sham and no business of EU free movement law. The Court briskly disagreed. There was, after all, a company registered in another Member State, the UK, involved. It is ‘immaterial that the company was 34  Case C-221/89 The Queen v The Secretary of State for Transport, ex p Factortame Limited et al [1991] ECR I-3905. 35   Van Binsbergen (n 4 above). 36  Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459.

Fitting ‘Abuse of Rights’ into the Free Movement of Goods and Services  59 formed in the first Member State only for the purpose of establishing itself in the second, where its main, or indeed entire, business is to be conducted’.37 So this is a situation in which the free movement rules in principle apply, and the issue is only whether in the particular circumstances, restrictions placed upon them by the host State, Denmark, are to be permitted as a response to ‘attempts by certain of its nationals to evade domestic legislation by having recourse to the possibilities offered by the Treaty’.38 In short, is this ‘abuse of the freedom of establishment’, as Denmark submitted?39 The Court cited a number of previous rulings concerning alleged improper circumvention of national legislation, including TV 10 and Knoors and Leclerc v Au Blé Vert, and stated that: [A]lthough, in such circumstances, the national courts may, case by case, take account—on the basis of objective evidence—of abuse or fraudulent conduct on the part of the persons concerned in order, where appropriate, to deny them the benefit of the provisions of Community law on which they seek to rely, they must nevertheless assess such conduct in the light of the objectives pursued by those provisions.40

In Centros itself, choosing to register a company according to the regime of company law in the EU which is most favourable and then to set up branches in other Member States ‘cannot, in itself, constitute an abuse of the right of establishment’.41 Quite the contrary: such a choice ‘is inherent in the exercise, in a single market, of the freedom of establishment guaranteed by the Treaty’.42 And, in a tone that is noticeably different from the yearning expectation in Knoors that the legislature might take the strain, the Court added dismissively that ‘the fact that company law is not completely harmonised in the Community is of little consequence. Moreover, it is always open to the Council, on the basis of the powers conferred upon it by Article 54(3)(g) of the EC Treaty [now Article 50(2)(g) TFEU], to achieve complete harmonisation.’ Denmark had acted unlawfully by impeding exercise of the right freely to set up a secondary establishment which the Treaty is intended to guarantee. The Court also rejected other justifications for the Danish rules on paid-up share capital rooted in the need to reinforce the financial soundness of companies, and more generally, to protect all creditors by anticipating the risk of fraudulent bankruptcy due to the insolvency of companies whose initial capitalisation was inadequate. The Court did not exclude the possibility that in particular, targeted instances of malpractice on its market, Denmark could intervene. But an outright refusal to register Centros as a branch went too far. Centros does not assert complete transfer of regulatory responsibility away from the host State to the home State. The free movement rules established by the Treaty never in principle go so far. The free movement rules institute a system of regulatory competition but that competition is not absolute, rather it is conditional: which means that a State may exclude goods or services from its market even where the supplier has complied with the rules of its home State. But such host State intervention needs to be justified and in Centros, Denmark failed to do so. The case law on corporate mobility has taken several twists and turns since Centros, most recently visible in Cartesio in which the Court held that moving a company from   Centros (n 36 above), para [17].   Centros (n 36 above), para [18].   Centros (n 36 above), para [23]. 40   Centros (n 36 above), para [25]. 41   Centros (n 36 above), para [27]. 42   Centros (n 36 above), para [27]. 37 38 39

60  Stephen Weatherill one Member State to another, involving alteration of the applicable national law, is protected by the Treaty, but moving the seat while keeping the applicable national law unchanged is different, and may lawfully be resisted by the state of incorporation.43 The Court’s judgment exhibits deference to the legislative process as a means to alter the position—which, of course, has not always been the Court’s attitude.44 However, it is not the purpose of this chapter to track in detail the extent to which EU free movement law can be exploited to allow a selection among available corporate jurisdictions.45 It is here simply asserted that a golden thread remains visible. A ‘host State’ is able to impose its regulatory requirements on a natural or legal person based in another Member State, but it must be able to justify that intervention. Pinned down to its true function in free movement law, an allegation of ‘abuse’ invites an assessment of whether the host State rules are apt for application to the alleged abuser. The mere fact that a trader has chosen to operate in another jurisdiction and then to target its original jurisdiction is not of itself abusive: it does not automatically disentitle the trader from putting the obstructive rules of the original jurisdiction to the test set by the law of free movement. Put another way, EU law of free movement allows—encourages—regulatory arbitrage but it does not promote it unconditionally: ‘abuse’ (like ‘evasion’) is then merely one term which opens up consideration of when the cross-border trader’s claim to profit from regulatory arbitrage is insufficiently strong and where instead the host State may insist on compliance with its rules as a pre-condition to access to its market.46 VIII. Conclusion

An allegation of ‘abusive’ reliance on the Treaty rules of free movement seems best understood not as a denial that there is any obstacle to inter-State trade—except perhaps in an extreme ‘sham’ case such as Leclerc v Au Blé Vert 47—but rather as a claim that in the circumstances, the host State rules remain justified and may be applied to the alleged ‘abuser’ despite their obstructive effect on inter-State trade. The Court is then placed in a position of considerable power in determining whether an adequate justification is really present. An allegation of ‘abuse’ in this sense, if successfully advanced, has the consequence that regulatory competition is blocked: the host State’s regulatory choices are not undermined by entry on to its market of goods or services which comply only with the regulatory regime set by their home State. The case law can therefore be seen as a judicial contribution to a debate which is descriptively and normatively rich—how to allocate tasks in a multi-level system of governance.48 This is an inquiry with both  Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641.   Cartesio (n 43 above), see especially paras [108–14]. 45   See contributions by WG Ringe, ‘Sparking Regulatory Competition in European Company Law: The Impact of the Centros Line of Case Law and its Concept of “Abuse of Law”’, ch 8 below; H Eidenmüller, ‘Abuse of Law in the Context of European Insolvency Law’, ch 10 below; and T Tridimas, ‘Abuse of Rights in EU Law: Some Reflections with Particular Reference to Financial Law’, ch 12 below. 46   Cf P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351. 47   Au Blé Vert (n 28 above). 48  A topic attracting a vast literature! For particular helpful surveys of the issues at stake see eg, S Deakin, ‘Legal Diversity and Regulatory Competition: Which Model for Europe?’ (2006) 12 European Law Journal 440; W Kerber and R van den Bergh, ‘Unmasking Mutual Recognition: Current Inconsistencies and Future Chances’ 11-2007 Marburger Volkswirtschaftliche Beiträge: www.isnie.org/assets/files/papers2007/kerber.pdf. 43 44

Fitting ‘Abuse of Rights’ into the Free Movement of Goods and Services  61 horizontal implications—where a host State rule is treated as unjustified, regulatory competence belongs with the home State, whereas if the host State is permitted to apply its rules, its competence is preserved at the expense of home State control—and, as is evident in several of the cases mentioned, there emerge vertical implications too insofar as where trade integration is impeded because host State rules survive the test of justification, the emphasis shifts to the possibility of legislative intervention by the EU to set centralised common rules (albeit that in the EU these are then, in the main, implemented and enforced locally). This is not to exclude the possibility that there is a general EU law principle of abuse of rights, nor even a principle of abuse of EU law rights.49 The concern of this chapter has been to fit the ‘abuse cases’ into the orthodox law governing the free movement of goods and services. The case has been made that they do so fit. I am not convinced that the quest to uncover ‘wholly artificial’ arrangements which animates the current decade’s key decisions on abuse in the fields of export refunds on agricultural products,50 valueadded tax (VAT)51 and corporate taxation52 has any close parallel in the field of free movement of goods and services. Perhaps Leclerc v Au Blé Vert53 is a comparable case, where the inter-State element is purely formal and lacks any genuine economic character. This could not be said of Knoors or TV10 or Centros. Perhaps the general notion that the rules of EU law shall not be exploited in a manner that is incompatible with their purpose serves to bind together all these cases in which ‘abuse’ (or evasion, or circumvention) is identified—but that is a rather loose binding. In conclusion, this chapter does not exclude the possibility that the way that ‘abuse of rights’ is treated in free movement cases may be connected conceptually and/or functionally with phenomena elsewhere in EU law for which the label ‘abuse’ is or may be appropriate. But, written from a standpoint that is ‘internal’ to free movement law, this chapter does not seek to make that broader case and prefers to leave it to others. It is submitted that the law of free movement of goods and services requires no special principle designed to guard against ‘abuse of rights’. It already accommodates adequate means for reviewing the strength of a host State’s claim to retain a competence to regulate traders which target its market for goods or services. The allegation of an abuse of rights aimed at a trader simply involves an assertion of host State regulatory competence, and is judged as such under the normal approach to assessing the matter of justification for host State rules that impede cross-border trade. Cf as examples of sector-specific inquiries, D Benson and A Jordan, ‘Understanding Task Allocation in the European Union: Exploring the Value of Federal Theory’ (2008) 15 Journal of European Public Policy 78; C Barnard, ‘Social Dumping and the Race to the Bottom: Some Lessons for the European Union from Delaware?’ (2000) 25 European Law Review 57; R Drury, ‘The Delaware Syndrome: European Fears and Reactions’ [2005] Journal of Business Law 709; N Moloney, ‘Innovation and Risk in EC Financial Market Regulation: New Instruments of Financial Market Intervention and the Committee of European Securities Regulators’ (2007) 32 European Law Review 627. 49  For a successful exploration of these issues, with comprehensive bibliographic references, see R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. 50  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 51  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 52  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 53   Au Blé Vert (n 28 above).

6 Free Movement of Broadcasting Services and Abuse of Law Dimitrios Doukas* I. Introduction

T

he field of broadcasting has provided a fertile ground for practices involving an abuse of the right to free movement within the European Union for the purpose of circumventing the less favourable legislation of a Member State concerning broadcasting activities. Broadcasters can engage in forum shopping with a view to choosing the Member State with the most beneficial legal framework.1 This is due to the technical facilities (satellite, cable and digital) available to crossborder broadcasting services and the regulatory competition2 resulting from divergences between national laws, and the minimum harmonisation at EU level brought about by the ‘Television Without Frontiers’ (TWF) Directive 89/552 and its successor, the Audiovisual Media Services (AVMS) Directive 2010/13.3 More than ten per cent of the approximately 7,000 TV channels currently originating or available in the EU target non-domestic markets.4 Apart from so-called pan-European broadcasters in single (eg, BBC World) or multiple linguistic versions (eg, Eurosport), these include broadcasters which are established in one Member State but target another, or which serve their Member State of origin but have ad hoc advertising and/or programme windows designed for   * Many thanks to Professor Catherine Barnard, Dr Rita de la Feria, and all the other participants at the Joint Symposium on ‘Prohibition of Abuse of Law: A New General Principle of EU Law?’ (Oxford, 3–4 October 2008) for their valuable comments and the constructive discussion. 1  See K Ottersbach, Rechtsmissbrauch bei den Grundfreiheiten des Europäischen Binnenmarktes (Baden-Baden, Nomos 2001) 78; and I Katsirea, ‘The Transmission State Principle: The End of the Broadcasting Sovereignty of the Member States?’ (2003–04) 6 Cambridge Yearbook of European Legal Studies 105, 108. 2  See T Gibbons, ‘The Impact of Regulatory Competition on Measures to Promote Pluralism and Cultural Diversity in the Audiovisual Sector’ (2006–07) 9 Cambridge Yearbook of European Legal Studies 239, 252–54 and 257; and LH Hansen, ‘The Development of the Circumvention Principle in the Area of Broadcasting’ (1998) 25(2) Legal Issues of Economic Integration 111, 113. 3  Council Directive 89/552/EEC of 3 October 1989 on the coordination of certain provisions laid down by law, regulation or administrative action in Member States concerning the pursuit of television broadcasting activities [1989] OJ L298/23, as amended by Directive 97/36/EC of the European Parliament and of the Council of 30 June 1997 [1997] OJ L202/60, and by Directive 2007/65/EC of the European Parliament and of the Council of 11 December 2007 [2007] OJ L332/27. Dir 89/552 was codified and repealed by Directive 2010/13/EU of the European Parliament and of the Council of 10 March 2010 on the coordination of certain provisions laid down by law, regulation or administrative action in Member States concerning the provision of audiovisual media services (Audiovisual Media Services Directive) [2010] OJ L95/1 (as corrected in [2010] OJ L263/15). 4  See European Audiovisual Observatory of the Council of Europe (EAVO), Press Releases of 15 October 2008 and 13 January 2010: www.obs.coe.int/about/oea/pr/mipcom2008.html and www.obs.coe.int/about/oea/ pr/mavise_end2009.html.

64  Dimitrios Doukas (an)other Member State(s).5 The latter two groups of broadcasters might be prone to circumvention.6 Evasion of national broadcasting laws was first examined by the European Court of Human Rights in the light of Article 10 of the European Convention on Human Rights. It was acknowledged as a ground for justifying restrictions or bans imposed by a State on the freedom of broadcasting of a foreign broadcaster that could reasonably be held to operate from foreign territory in order to circumvent the legislation of that State, with which it had strong links and towards which its programmes were primarily directed.7 Within the EU the case law of the Court of Justice concerning allegedly abusive practices on the part of providers of broadcasting services has built on the circumvention or evasion doctrine8 of Van Binsbergen,9 and has paved the way for the emergence of an EU law principle of prohibition of abuse of law.10 Although the jurisprudence has not been unambiguous, it cannot be disputed that the Court has applied the principle of prohibition of abuse in two different manners. On the one hand, as a principle governing the interpretation of the scope of the relevant EU law provisions—ie Article 56 TFEU [ex Article 49 EC] or Article 2 of the TWF Directive on the transmitting State principle; this, as is well known, makes the Member State which has jurisdiction over a broadcaster primarily responsible to ensure that the latter complies both with its domestic audiovisual media legislation and the provisions of the Directive.11 On the other hand, and more importantly, the Court has applied the prohibition of abuse as a principle providing a ground for the justification of restrictions on the free movement of broadcasting services.12 The relevant case law can be classified into two main categories: those cases which preceded and those which followed the implementation of the TWF Directive. The former (including Veronica,13 TV10,14 and the Flemish cable network15 cases) are characterised by the relatively broad discretion left to Member States to adopt measures to combat circumvention, with the prohibition of discriminatory and protectionist measures constituting the red line. In the post-Directive jurisprudence, the Court of Justice, first,     5  In 2004, more than 200 of the 1,100 TV channels originating or available in the EU targeted non-domestic markets. See EAVO, ‘Transfrontier Television in the European Union: Market Impact and Selected Legal Aspects’ (Background Paper, Conference on Broadcasting, Dublin and Drogheda 1–3 March 2004): www.obs. coe.int/online_publication/transfrontier_tv.pdf.en 1, 9–10.    6  See I Katsirea, ‘The Circumvention Principle: An Effective Defence of National Broadcasting Systems Against Abuse?’ (2005) 10 Communications Law 130.    7   Groppera Radio AG et al v Switzerland (App no 10890/84) (1990) Series A no 173, para [73]. The (former) European Commission of Human Rights followed suit by its decisions of 29 November 1993 Cable Music Europe Ltd v Netherlands (App no 18033/91), and of 11 January 1994, X SA v Netherlands (App no 21472/93) DR 76. See D Doukas, Werbefreiheit und Werbebeschränkungen (Baden-Baden, Nomos 2005) 185–88, 214–17, 222.    8  See J Harrison and L Woods, European Broadcasting Law and Policy (Cambridge, Cambridge University Press 2007) 177.    9  Case 33/74 Van Binsbergen v Bedrijfsvereniging Metaalnijverheid [1974] ECR 1299, para [13]. See also Case 205/84 Commission v Germany (insurance) [1986] ECR 3755, para [22]. 10  See R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 403. 11  See (n 3 above) re: Arts 2(1) and 3(6) (ex-Art 3(2)) of Dir 89/552, as amended by Art 1(3) and (5) of Dir 2007/65. 12  See KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 428–30 and 432–33. 13  Case C-148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487. 14  Case C-23/93 TV10 SA v Commissariaat voor de Media [1994] ECR I-4795. 15  Case C-211/91 Commission v Belgium [1992] ECR I-6757.

Free Movement of Broadcasting Services and Abuse of Law  65 interpreted the jurisdiction of a Member State over a broadcaster by setting out such criteria as to prevent or reduce the risk of abuses of the freedom to provide services and the transmitting State principle (Belgian cable network,16 UK satellite services,17 Denuit,18 and VT419). Second, although the Court appeared to accept the application of the principle of abuse in the field covered by the TWF Directive, it adopted a more restrictive interpretation, which curtailed the discretion of national authorities, so that it could not impair the essence or the effectiveness of the transmitting State principle (Belgian cable network, VT4, and Advocate-General Jacobs in De Agostini20). This narrower interpretation came more into line with the similarly strict approach of the Court in the field of establishment21 and proved the forerunner of the Centros line of authority,22 according to which the fact that a company is formed in a Member State for the sole purpose of benefiting from more favourable legislation does not in itself constitute an abuse of the freedom of establishment, even if the company pursues its activities entirely or mainly through a branch in another Member State.23 The strict construction of the principle of abuse is supported by more recent case law in the field of audiovisual media (United Pan-Europe24). This chapter is structured in four parts. The first part examines the case law that recognised the principle of prohibition of abuse in the field of broadcasting services before the TWF Directive took effect (II). The second part critically analyses the impact of this principle on the construction of the TWF Directive and the Court’s gradual shift towards a more restrictive interpretation (III). The jurisprudence led to the amendment of the TWF Directive, which has integrated some of the core elements of the principle of abuse, as summarised in the third section (IV). The final section discusses the significance of the case law on broadcasting services and its integration into secondary law for the emergence of a general principle of prohibition of abuse of European Union law (V). II. The Principle of Abuse before Implementation of the TWF Directive

Before the entry into force of the TWF Directive, the Court of Justice set out the basic conditions necessary to trigger the application of the prohibition of abuse principle, but in general, allowed Member States a relatively broad discretion to adopt measures to prevent a circumvention of their national laws. It is interesting that initially the Court appeared to regard the prohibition of abuse as a principle governing the interpretation of the freedom to provide services, which pre-empted the application of what is now  Case C-11/95 Commission v Belgium [1996] ECR I-4115.  Case C-222/94 Commission v UK [1996] ECR I-4025. 18  Case C-14/96 Criminal Proceedings against Paul Denuit [1997] ECR I-2785. 19  Case C-56/96 VT4 Ltd v Vlaamse Gemeenschap [1997] ECR I-3143. 20  AG Jacobs in Joined Cases C-34 to C-36/95 Konsumentombudsmannen (KO) v De Agostini (Svenska) Förlag AB (C-34/95) and TV-Shop i Sverige AB [1997] ECR I-3843. 21  Case 79/85 DHM Segers v Bestuur van de Bedrijfsvereniging voor Bank- en Verzekeringswezen, Groothandel en Vrije Beroepen [1986] ECR 2375, para [16]. 22  See C Barnard, The Substantive Law of the EU—The Four Freedoms, 3rd edn (Oxford, Oxford University Press 2010) 391–392; and M Varney, ‘European Controls on Member State Promotion and Regulation of Public Service Broadcasting and Broadcasting Standards’ (2004) 10 European Public Law 503, 516–17. 23  Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459, paras [25–27, 29], and Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155, para [96]. 24  Case C-250/06 United Pan-Europe Communications Belgium SA et al v Belgian State [2007] ECR I-11135. 16 17

66  Dimitrios Doukas Article 56 TFEU [ex Article 49 EC] (Veronica).25 However, it subsequently relied on the principle as a ground for justifying restrictions on this fundamental freedom (TV10).26 This appeared more in line with the outer limits set to the principle in the earlier Flemish cable network case. A.  The Principle of Abuse as a Ground for Precluding the Application of Article 56 TFEU: Veronica In Veronica,27 the Court of Justice examined whether there was a breach of what is now Article 56 TFEU [ex Article 49 EC] by Dutch legislation, which—in the context of a national public service broadcasting system—prohibited all the broadcasting organisations that were authorised to broadcast on the national network from setting up or investing in and assisting a (commercial) broadcasting company in another Member State.28 Veronica was a domestic broadcasting organisation which had its authorisation temporarily suspended because it financed and legally assisted the setting-up of a commercial TV station in Luxembourg that was intended for reception particularly in the Netherlands.29 The Court underlined the aim of the legislation in question, which was to preserve the pluralistic and non-commercial character of the audiovisual sector in the Netherlands, and especially to prevent domestic broadcasting organisations, which were essentially financed from public funds, from using the financial resources available to them for purely commercial ends.30 This constituted a cultural policy objective intended to safeguard freedom of expression for the various currents (social, cultural, religious or philosophical), and thus a public interest, which allowed a Member State to formulate the statutes of its own broadcasters in an appropriate manner.31 Surprisingly, although the Court referred to its case law that recognises media pluralism as a public interest ground for justifying restrictions on the free movement of services, it regarded the provision in question as escaping the net of what is now Article 56 TFEU [ex Article 49 EC].32 The Court of Justice relied on the right acknowledged to a Member State in Van Binsbergen33 to take measures to prevent the exercise of the freedoms guaranteed by the Treaty by a person whose activity is entirely or principally directed towards its territory for the purpose of avoiding the professional rules of conduct which would be applicable 25  See also AG Léger in Case C-55/94 Reinhard Gebhard v Consiglio dell’ Ordine degli Avvocati e Procuratori di Milano [1995] ECR I-4165, para [27] and fn 19. 26  See also Hansen (n 2 above) 122, 127. 27   Veronica (n 13 above). 28  Art 57(1) and (4) of Mediawet prohibited the pursuit by broadcasters of any activities other than the production of their programmes or those authorised by the Dutch media regulator, and the use of the available financial resources for any other purpose. 29  AG Tesauro in Veronica (n 13 above), paras [2–3]. 30   Veronica (n 13 above), paras [9] first limb, and [11]. 31   Veronica (n 13 above), paras [9] second limb, and [10], with reference to Case C-353/89 Commission v Netherlands [1991] ECR I-4069, paras [29–30] and [41–42], and Case C-288/89 Stichting Collectieve Antennevoorziening Gouda v Commissariaat voor de Media [1991] ECR I-4007, paras [22–24]. 32   Veronica (n 13 above), paras [14, 16]. Cf AG Tesauro in Veronica (n 13 above), paras [6–7] and [11] first limb, who however relied on the Preamble to the TWF Directive, which guaranteed the national competence over organisation, financing and content of TV programmes. See W Hins, ‘Case Note on Veronica’ (1994) 31 Common Market Law Review 901, 906; and CA Jones, ‘Television Without Frontiers’ (1999–2000) 19 Yearbook of European Law 299, 303. 33   Van Binsbergen (n 9 above), para [13].

Free Movement of Broadcasting Services and Abuse of Law  67 to him if he were established within that State.34 It concluded that the legislation in question could not be deemed incompatible with the free movement of services (or capital) inasmuch as it prohibited domestic broadcasters from engaging in activities directed towards the establishment of a commercial TV station in another Member State, the broadcasts of which were intended for reception, in particular, in the territory of the first Member State. That prohibition was also held to be necessary to ensure that the pluralistic and non-commercial character of the audiovisual system introduced by that legislation could not be improperly evaded, and thus to safeguard the exercise of the freedoms guaranteed by the Treaty.35 Interestingly, the Court established a link between the prohibition of an improper evasion of national legislation pursuing a legitimate public interest objective and the safeguarding of the (proper) exercise of the fundamental freedoms.36 It thereby partly endorsed the Opinion of its Advocate-General that the latter could not be relied upon to avoid or circumvent the application of national rules which are compatible with EU law.37 However, in stark contrast to the Court, Advocate-General Tesauro had advocated the application of the Van Binsbergen doctrine not as a ground for excluding a provision from the net of what is now Article 56 TFEU [ex Article 49 EC], but instead as a ground for justifying a national measure, should the latter be classified as a restriction on the free movement of services in the first place.38 In this context, he proposed a rigorous proportionality test and held that the specific commercial activities exercised by the broadcaster in question were, taken as a whole, liable to compromise the legitimate aim of the relevant legislation.39 In his Opinion, a situation where a national broadcasting organisation set up or participated and maintained significant financial interests in a TV broadcaster established in another Member State, which broadcast programmes principally directed towards the Netherlands, was likely to affect the balance created between the various national broadcasting organisations by Dutch legislation. AdvocateGeneral Tesauro pointed to the unlawful advantages that the former organisation would gain over the other domestic broadcasters, both in terms of air time and financing, and particularly in advertising revenues.40 There was no evidence that the aim of ensuring pluralism could be attained by less restrictive means.41 If we set aside the question of the effect, pre-emptory or justificatory, of the principle of abuse in the field of broadcasting services, the judgment in Veronica set out two basic conditions necessary to trigger its application: (a) an objective element of abuse, consisting in a U-turn construction (ie the setting up in a Member State of a broadcaster, the programmes of which are specifically intended for reception in another Member State); and (b) a subjective element, consisting in the intention to improperly use the free movement of services in order to evade national rules of the host Member State, which aim (and are necessary) to protect a public interest objective legitimate under EU   Veronica (n 13 above), para [12].   Veronica (n 13 above), paras [13–15]. 36  Against Katsirea (n 6 above) 131. 37  AG Tesauro in Veronica (n 13 above), para [6], subpara [4] second limb. See also Ottersbach (n 1 above) 98. 38  AG Tesauro in Veronica (n 13 above), para [10]. See also Hins (n 32 above) 909. 39  AG Tesauro in Veronica (n 13 above), para [9]. 40  AG Tesauro in Veronica (n 13 above), para [10], subparas [1–2], citing Van Binsbergen (n 9 above), para [13]; Case 115/78 J Knoors v Staatssecretaris van Economische Zaken [1979] ECR 399; Commission v Germany (n 9 above), para [22]; and Case 130/88 C Van de Bijl v Staatssecretaris van Economische Zaken [1989] ECR 3039, para [26]. 41  AG Tesauro in Veronica (n 13 above), para [10], subpara [3]. 34 35

68  Dimitrios Doukas law. Such use of the fundamental freedom is considered incompatible with the Treaty objectives.42 B.  The Principle of Abuse as a Ground for Justifying Restrictions within the Meaning of Article 56 TFEU: TV10 Both the conditions and the effect of the principle of abuse were further elaborated in the case of TV10.43 This was a commercial broadcasting corporation established and authorised in Luxembourg to transmit its programmes via satellite. Although the production of its programmes and the organisation and implementation of its broadcasting activity were to take place in Luxembourg,44 TV10 was denied access to the cable network in the Netherlands on the ground that it was largely managed by Dutch nationals, most of its employees were Dutch, and its target audience was the Dutch public. It was therefore presumed to be a domestic broadcaster that had established itself in another Member State with the manifest intention of benefiting from the laxer rules applicable to foreign broadcasters under Dutch law.45 This presumption was confirmed by the Dutch Council of State, which relied on Van Binsbergen but decided to request a preliminary ruling on the interpretation of ‘activities directed from another Member State by a broadcaster constituted under the laws of that State, where it is formally established’.46 The reason was the judgment in Segers, where the Court of Justice had held that a company enjoys the protection of the freedom of establishment provided it is formed in accordance with the law of a Member State and has its registered office, central administration or principal place of business within the Union; even if it conducts its business through an agency, branch or subsidiary solely in another Member State.47 The Court clarified two important issues: on the one hand, it held that any alleged intention to make an abuse of EU law for the purpose of circumventing national law was not an obstacle to the application of the free movement provisions on services;48 instead, it could only justify restrictions. This marked a shift away from the pre-emptory effect of the principle of abuse accepted in Veronica.49 Thus, the fact that TV10 was established in Luxembourg, in accordance with that Member State’s law, and intended to broadcast to the Netherlands was regarded as sufficient to establish a cross-border element and trigger the application of what is now Article 56 TFEU [ex Article 49 EC].50 This was distinct  See Ottersbach (n 1 above) 97, 100.   TV10 (n 14 above). 44  AG Lenz in TV10 (n 14 above), para [17]. 45  Art 66(1) of Mediawet. TV10 (n 14 above), paras [6–7]. As in Veronica, the facts of the case had taken place before the Dutch media legislation was amended so as to allow domestic commercial broadcasters. AG Lenz in TV10 (n 14 above), para [9]. 46   TV10 (n 14 above), paras [9–10]. 47   Segers (n 21 above), para [16]. 48   TV10 (n 14 above), para [16]. See also A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of the European Court of Justice on Circumvention, Fraud and Abuses of Community Law’ in M Andenas and W-H Roth (eds), Services and Free Movement in EU Law (Oxford, Oxford University Press 2002, reprint 2004) 245, 260. 49  Against the justificatory effect of the principle: P Wattel, Case Note on TV10: ‘Circumvention of National Law—Abuse of Community Law?’ (1995) 32 Common Market Law Review 1257, 1269. 50   TV10 (n 14 above), para [14]. See also AG Lenz in TV10 (n 14 above), paras [15], [18–20] and [32], on the broadcaster’s actual and permanent connection with the economy of the Member State, in which it was incorporated and established. 42 43

Free Movement of Broadcasting Services and Abuse of Law  69 from the question of what measures a Member State might take to prevent a service provider established in another Member State from evading its domestic legislation.51 By the same token, Advocate-General Lenz referred to a doctrine permeating the case law, ‘which is characterised by refusal to allow the person concerned to invoke fundamental freedoms in the event of the avoidance of national provisions or the abuse of the freedoms guaranteed by Community law’.52 He explained that, from a Union law perspective, there were, in principle, no objections to the free choice of one’s place of establishment in terms of its aim.53 Instead, abuse or circumvention should exceptionally allow the host Member State to (additionally) subject the particular cross-border activity of the service provider to the rules of professional conduct applicable on its territory as if the provider were established there.54 Such an exception should be narrowly construed so as to cover only the application of the rules which the service provider seeks to circumvent.55 On the other hand, and in the context of examining the compatibility with EU law of restrictions on the free movement of services, the Court of Justice seems to have established a presumption for ascertaining the intention of abuse of Union law on the part of a broadcaster for the purpose of circumventing the legislation of a Member State (subjective element): such an intention can be presumed if the activities of the broadcaster established in another Member State are wholly or principally directed to the territory of the first Member State (objective element).56 The Court took the lead from AdvocateGeneral Lenz, who—whilst admitting that it is for the national court to make findings of fact as to whether circumvention is involved, albeit subject to the review of the Court of Justice57—raised the question whether such an assessment should be based on subjective factors such as intention and motive, or solely on objective criteria.58 His basic premise was the Van Binsbergen test, which has both an objective and a subjective component.59 As regards the objective element of abuse, Advocate-General Lenz affirmed that the activities of the broadcaster from Luxembourg were entirely or principally directed towards the Netherlands on the basis of objective factors, such as the content or the language of programmes and advertisements, and the suspension of broadcasting when the broadcaster was refused authorisation to feed its programmes into the Netherlands cable network.60 However, as regards the subjective element, Advocate-General Lenz suggested that a corporate intent to circumvent could only be assessed by means of objective criteria.61 He explained that, in contrast to the conduct of natural persons, 51   TV10 (n 14 above), para [15]. Cf Case C-46/08 Carmen Media Group Ltd v Land Schleswig-Holtstein, judgment of 8 September 2010, nyr, para [50]. 52  AG Lenz in TV10 (n 14 above), paras [50–53], citing Case 229/83 Leclerc v Au blé vert [1985] ECR 1, para [27] (goods); Case 39/86 Sylvie Lair v Universität Hannover [1988] ECR 3161, para [43] (workers); Knoors (n 40 above), para [25] (establishment and services); Van Binsbergen (n 9 above), para [13]; Case 39/75 R-G Coenen et al v Sociaal-Economische Raad [1975] ECR 1547, para [20]; Commission v Germany (n 9 above), para [22]; Van de Bijl (n 40 above), para [26]; and Veronica (n 13 above), para [12] (services). 53  AG Lenz in TV10 (n 14 above), para [48]. 54  AG Lenz in TV10 (n 14 above), paras [31, 33 and 35]. 55  AG Lenz in TV10 (n 14 above), para [34]. 56   TV10 (n 14 above), para [26]. 57  AG Lenz in TV10 (n 14 above), paras [23] and [58]. 58  AG Lenz in TV10 (n 14 above), paras [22] and [56–57]. 59  Objectively, the activity of the service provider must be entirely or principally directed towards the territory of the host Member State; and subjectively, the provider must use the freedom to provide services for the purpose of avoiding the professional rules of conduct which would be applicable to him if he were established in that State. Van Binsbergen (n 9 above), para [13]. 60  AG Lenz in TV10 (n 14 above), para [55]. 61  AG Lenz in TV10 (n 14 above), para [61].

70  Dimitrios Doukas motives were especially difficult to evaluate in connection with the activity of a legal person, not least because they were scarcely susceptible of proof. Besides, no uniform rule of EU law could be derived from the Member States’ legal traditions on imputing acts of natural persons to the responsibility of legal persons.62 This difficulty was highlighted by the decision of the Court to use the objective element as a presumption for ascertaining the subjective element, rather than distinguishing between the two components as independent requirements for applying the principle of abuse.63 Nevertheless, the Court did not specify which objective circumstances could be used to establish an intended circumvention because of the centre of gravity of a broadcaster’s activities.64 By contrast, Advocate-General Lenz suggested that such objective factors should not include the nationality of persons standing behind or acting through a company, as this would involve unlawful discrimination.65 Instead, criteria such as the time of the commencement, the substance and the orientation of a company’s business operations, and possibly corporate relationships should be taken into account.66 In the present case, such factors should include the objective impossibility of broadcasting in the manner chosen by the broadcaster concerned in the Netherlands, the organisation and orientation of its programmes, the suspension of broadcasting following the denial of its access to the Netherlands cable network, and possibly the production of its programmes in the Netherlands by a company to which certain links might exist.67 The Court concluded that what are now Articles 56 and 57 TFEU [ex Articles 49 and 50 EC] do not preclude a Member State from treating as a domestic broadcaster a broadcasting body which establishes itself in another Member State in order to provide services (wholly or principally) directed towards the territory of the first State, insofar as the following two conditions are met:68 (a) the broadcasting body can be presumed— because of the centre of gravity of its activities—to have intended to evade the rules that would be applicable to it if it were established within the first State;69 and (b) the aim of the measure adopted by the first Member State is to prevent such a broadcasting organisation ‘from being able, by exercising the freedoms guaranteed by the Treaty, wrongfully to avoid obligations under national law, in this case those designed to ensure the pluralist and noncommercial content of programmes’.70 While the first condition refers to the combined (objective and subjective) requirements for triggering the application of the principle of abuse, the second condition sets a limit to the host Member State’s discretion to apply that principle, albeit of a rather general and vague nature: any restriction on the freedom of a broadcaster from another Member State to provide services can only be justified if it specifically aims to prevent a wrongful avoidance by that broadcaster of the legitimate, public interest objective(s) of the relevant national legislation through the exercise of the fundamental freedoms. In this regard, the legitimacy per se of a Member State’s decision to maintain a non-commercial  AG Lenz in TV10 (n 14 above), paras [22, 57] and [59–61].   TV10 (n 14 above), para [26]. Against Katsirea (n 6 above) 132. 64  See also Harrison and Woods (n 8 above) 179. 65  AG Lenz in TV10 (n 14 above), paras [62–64, 66]. Cf Case C-221/89 The Queen v Secretary of State for Transport, ex p Factortame Ltd et al [1991] ECR I-3905, para [33]. 66  AG Lenz in TV10 (n 14 above), para [65]. 67  AG Lenz in TV10 (n 14 above), para [67]. 68   TV10 (n 14 above), paras [18–20, 22], with reference to Van Binsbergen (n 9 above), para [13], and Veronica (n 13 above), para [9]. 69   TV10 (n 14 above), para [26]. 70   TV10 (n 14 above), para [21]. 62 63

Free Movement of Broadcasting Services and Abuse of Law  71 broadcasting system might justify its power to frustrate an attempt by an undertaking to evade its jurisdiction.71 In sum, in TV10 the Court of Justice, on the one hand, departed from Veronica, by classifying the principle of abuse as a ground for justifying restrictions on the freedom to provide services. On the other hand, it reaffirmed and refined the two requirements for the application of the principle: ie (a) the objective element of an activity blatantly incompatible with the purpose of the free movement of services, such as the establishment of a broadcaster in another Member State from the one where the centre of gravity of its activities is located; and (b) the subjective element, that is the broadcaster’s intention of abuse of a fundamental freedom for the purpose of circumventing rules of the host Member State which are legitimate under EU law.72 The Court held that the subjective component of abuse is not an independent requirement but is presumed if the objective element is satisfied.73 It also drew a general limit to the Member States’ discretion from the principle of proportionality: the abstract pursuit of a public interest by the national rules sought to be evaded does not suffice to justify any restrictions on the freedom to provide services. Instead, the principle of abuse can only justify such measures that specifically aim to prevent an evasion of the legitimate objectives of the rules in question by means of abuse of the fundamental freedom. However, this did not essentially alter the broad margin of discretion left to Member States nor did it involve a rigorous proportionality test of the anti-avoidance measures lest the effectiveness of the principle of abuse be undermined.74 It only confirmed the outer limits set to this principle in the earlier Flemish cable network judgment.75 C.  The Outer Limits to the Principle of Abuse: The Flemish Cable Network Case Legislation of the Flemish Community in Belgium prohibited TV broadcasters established in other Member States from transmitting their programmes on the Flemish cable network if the programmes were not in the language(s) of the Member State where the broadcasters were established. The Court of Justice found the prohibition to be a directly discriminatory obstacle to the free movement of services, not only because it did not apply to broadcasters established in Belgium, but mainly because it prevented broadcasters established in a Member State other than the neighbouring Netherlands from offering their programmes in Dutch to audiences in Flanders, while that possibility was open to domestic broadcasters.76 Because of its protectionist nature, the measure could not be justified by any of the express derogations of primary law or any overriding cultural policy objectives recognised in case law: its real purpose was to protect  See AG Lenz in TV10 (n 14 above), paras [45, 68].  See also Hansen (n 2 above) 125, and M-C Boutard-Labarde, ‘Case Note on TV10’ (1995) No 11 La Semaine Juridique—Édition générale 126. Against the limitation of the application of the evasion principle to national rules justified by an overriding public interest: L Woods and J Scholes, ‘Broadcasting: The Creation of a European Culture or the Limits of the Internal Market?’ (1997) 17 Yearbook of European Law 47, 57. 73   Cf Ottersbach (n 1 above) 41–44; and Harrison and Woods (n 8 above) 179. 74  See G Straetmans and C Goemans, ‘Case Note on TV10’ (1994–95) 1 Columbia Journal of European Law 319, 329–31. Cf AG Geelhoed in Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607, para [101]. 75   Cf Sørensen (n 12 above) 458. 76   Flemish cable network (n 15 above), paras [5–6]. 71 72

72  Dimitrios Doukas domestic broadcasters from genuine competition in order to maintain their advertising revenues.77 More importantly, although the Court reaffirmed the Van Binsbergen doctrine,78 it refused to give the host Member State carte blanche to adopt any measures whatsoever in order to prevent an abuse of EU law.79 The Court drew the outer limits of the principle of abuse, by emphasising that ‘it does not follow that it is permissible for a Member State to prohibit altogether the provision of certain services by operators established in other Member States, as that would be tantamount to abolishing the freedom to provide services’.80 Thus, reliance on this principle might justify restrictions but not an outright ban that would generally bar market access and infringe the essence of the right to free movement.81 The Court of Justice followed the Opinion of Advocate-General Tesauro that the prohibition in question was not conceived in such a manner as to prevent a circumvention of domestic provisions on establishment; instead, it applied to the transmission of all programmes not broadcast in the language of the Member State where a broadcaster was established, and regardless of any ‘quantitative’ assessment of the programmes transmitted in Dutch by a particular broadcaster established in another Member State.82 III. The Principle of Abuse after Implementation of the TWF Directive

The case law that followed the implementation of the TWF Directive is characterised by two main trends. On the one hand, the Court used the prohibition of abuse as a principle governing the interpretation of the Directive, and in particular the provision on the subjection of a broadcaster to the jurisdiction of a Member State (Article 2). It thus developed such criteria as to prevent or reduce the risk of an abuse of the freedom to provide services and the transmitting State principle (Belgian cable network, UK satellite services, Denuit and VT4). On the other hand, although the Court did not preclude the application of the principle of abuse in the field covered by the Directive as a ground for justifying exceptions, it adopted a narrower interpretation, which constrained the Member States’ discretion, and preserved the effectiveness of the transmitting State principle (Belgian cable network, VT4, and Advocate-General Jacobs in De Agostini).83 More recent case law concerning the ‘must-carry’ status of certain broadcasters on domestic cable networks lends support to the strict construction of the principle of abuse (United Pan-Europe).

77   Flemish cable network (n 15 above), paras [9–10]. Cf AG Tesauro, in Flemish cable network (n 15 above), para [3]. 78   Flemish cable network (n 15 above), para [12] first and second limb, citing Van Binsbergen (n 9 above), para [13]. 79  Contrast AG Lenz in TV10 (n 14 above), fn 54, who inaccurately interprets the judgment in Flemish cable network (n 15 above) as one where the application of the circumvention principle was considered but refused. 80   Flemish cable network (n 15 above), para [12] second limb. 81   Cf Sørensen (n 12 above) 458. 82  AG Tesauro in Flemish cable network (n 15 above), para [4]. 83  See J Hörnle, ‘Country of Origin Regulation in Cross-Border Media: One Step Beyond the Freedom to Provide Services?’ (2005) 54 International & Comparative Law Quarterly 89, 115–18.

Free Movement of Broadcasting Services and Abuse of Law  73 A.  The Dual Function of the Principle of Abuse in the Field of the TWF Directive: The Belgian Cable Network Case Belgian legislation subjected to express prior authorisation the retransmission, via the cable network in the French Community, of TV programmes broadcast by broadcasters from other Member States which contained commercial advertisements and teleshopping programmes especially intended for viewers in Wallonia. The grant of authorisation was conditional on the involvement of those broadcasters in supporting pluralism in the French Community’s domestic broadcasters and press.84 The Court of Justice rejected the argument of the Belgian government that, because of the circumvention doctrine, TV broadcasters from other Member States specifically targeting the viewers in the French Community should not be regarded as falling under the jurisdiction of their Member State of establishment within the meaning of Article 2(2) of the (original version of the) TWF Directive.85 In particular, the Court dismissed the assumption that the emission by broadcasters from other Member States of commercial advertisements and teleshopping programmes intended for viewers in the Belgium French Community constituted in itself a circumvention of the legislation of the receiving State.86 The Court implied that the provisions in question were incompatible with the transmitting State principle, by seeing no need to examine whether, in the light of the Directive, a Member State is still entitled, on the basis of Article 56 TFEU [ex Article 49 EC], to take measures to prevent a provider of services whose activity is entirely or principally directed towards its territory from exercising the freedoms guaranteed by the Treaty for the purpose of avoiding the rules which would be applicable to him if he were established within that State.87 It added that the circumvention principle ‘do[es] not in any event authorise a Member State generally to exclude provision of certain services by operators established in other Member States, since that would entail abolition of the freedom to provide services’.88 Advocate-General Lenz proposed a narrow interpretation of the circumvention doctrine, as an exception to the fundamental freedoms, which should also apply to the field covered by the TWF Directive.89 Accordingly, the transmitting State principle precludes a Member State from presuming that all the programmes broadcast by TV broadcasters from other Member States, which are specifically intended for the public in that State, constitute in themselves an abuse of law;90 the cable retransmission of such broadcasts could not on that account be generally subjected to a system of prior authorisation.91 A receiving Member State could exceptionally treat individual broadcasters from another Member State as if they were subject to its own jurisdiction only if they were found in concreto to have acted abusively.92 84  Arts 26 and 26(b), Decree of the Council of the French Community of 17 July 1987 on audiovisual communications (Moniteur Belge, 22 August 1987, 12505), as amended by the Decree of 19 July 1991 (Moniteur Belge, 2 October 1991, 21671). Belgian cable network (n 16 above), paras [59–61]. 85   Belgian cable network (n 16 above), para [62]. 86   Belgian cable network (n 16 above), paras [63–64]. 87   Belgian cable network (n 16 above), para [65] first limb. 88   Belgian cable network (n 16 above), para [65] second limb, citing Van Binsbergen (n 9 above), para [13], and TV10 (n 14 above), para [20], and reaffirming Flemish cable network (n 15 above), para [12]. 89  AG Lenz in Belgian cable network (n 16 above), paras [73–76]. 90  AG Lenz in Belgian cable network (n 16 above), para [74] subpara [2]. 91  AG Lenz in Belgian cable network’(n 16 above), para [75]. 92  AG Lenz in Belgian cable network (n 16 above), para [76].

74  Dimitrios Doukas The Court also declared incompatible with EU law Belgian legislation which subjected to prior authorisation the retransmission via the cable network in the Flemish Community of any programmes broadcast by broadcasters from other Member States; this authorisation was conditional on compliance of those programmes with the requirements of Belgian law, which allowed the Flemish executive to monitor whether the originating Member State had properly performed its duties under the TWF Directive with regard to the control of broadcasters falling under its jurisdiction.93 The Court rejected the argument that the Directive does not preclude the power of Member States to verify whether TV programmes broadcast from another Member State have a real connection with and are in fact controlled by the Member State of origin; and to prevent broadcasters without any such link from improperly relying on the freedom to provide services.94 The Court reaffirmed that it is solely for the Member State of origin to monitor compliance with the Directive by broadcasters subject to its jurisdiction. It added that, even if the aim of the Belgian legislation was merely to introduce a system of verification, the requirement of prior authorisation constituted a disproportionate obstacle to the free movement of services.95 In sum, although the Court of Justice did not rule out the application of the principle of abuse in the field covered by the TWF Directive, it adopted a narrow interpretation, consistent with the transmitting State principle. Such interpretation precludes the subjection by a Member State of TV programmes broadcast by broadcasters established in another Member State to prior authorisation, even if those programmes are especially intended for the viewers in the receiving State. The Court implied that such a finding could not suffice to establish an abuse of the free movement of services.96 B. The Impact of the Principle of Abuse on the Construction of the TWF Directive: The UK Satellite Services and Denuit Cases The impact of the principle of abuse on the interpretation of the TWF Directive should not be underestimated even in cases which did not on their facts directly involve broadcasters circumventing the law of the receiving Member State. The Court’s effort to strike a balance between the transmitting State principle and the prohibition of abuse of the freedom to provide services is illustrated in the criteria that it developed for the construction of the original wording of Article 2 of the Directive, which generally subjects broadcasters to the jurisdiction of the Member State in which they are established.97 In the UK satellite services case98 the Court of Justice declared incompatible with the TWF Directive UK legislation, which applied, with regard to satellite television services requiring authorisation, criteria other than that of establishment for determining which 93  Arts 3 and 5 of the Decree of the Flemish Council of 4 May 1994 on radio and television networks and the authorisation necessary for the establishment and operation of such networks and on promoting the broadcasting and production of television programmes (Belgisch Staatsblad, 4 June 1994, 15434/15440). Belgian cable network (n 16 above), paras [79–80, 82]. 94   Belgian cable network (n 16 above), paras [83–85]. 95   Belgian cable network (n 16 above), paras [86, 88]. Cf AG Lenz in Belgian cable network (n 16 above), para [97], citing Recital 13 of the Preamble to Dir 89/552 (n 3 above). See also Barnard (n 22 above) 386–87. 96   Cf Varneys (n 22 above) 511, 515–16. 97   Cf Sørensen (n 12 above) 428. 98   UK satellite services (n 17 above).

Free Movement of Broadcasting Services and Abuse of Law  75 broadcasters fell within UK jurisdiction; the latter covered all broadcasters transmitting their broadcasts from UK territory even if they were established in another Member State and used solely a satellite capacity or uplink in the UK.99 The Court dismissed both the criterion of the place of transmission and that of the place of reception of the TV programmes of a broadcaster as irrelevant for the purpose of determining the competent Member State.100 On the basis of a literal interpretation, it construed Article 2(1) of the Directive as referring to the ratione personae jurisdiction of a Member State over a TV broadcaster. This could only be based on the broadcaster’s connection to the legal system of that State, which essentially overlapped with the concept of establishment within the meaning of Article 56(1) TFEU [ex Article 49(1) EC].101 As regards the second indent of Article 2(1), which originally provided for the competence of the Member State that granted a frequency or a satellite capacity or in which a satellite up-link was situated, its transmission-based criteria were held to apply only to broadcasters which are not otherwise subject to the jurisdiction of any Member State.102 As far as the efficacy of the criterion of establishment is concerned, the Court of Justice acknowledged that such an interpretation of the TWF Directive might give rise to a conflict of jurisdiction in cases where a broadcaster was established in more than one Member State.103 To avoid such conflicts, it endorsed the Commission’s suggestion that the Member State of establishment should be that in which lies the centre of gravity of the broadcaster’s activities—in particular, the place where decisions concerning programme policy are taken and the programmes to be broadcast are finally put together.104 Such a construction would considerably reduce the risk of abuse inherent in the choice of a Member State of establishment, insofar as a broadcaster could try and avoid the legislation of a particular Member State by moving its seat to another country within the EU.105 The Court emphasised that, in any event, a delimitation of jurisdiction based on the Member State from which broadcasts are transmitted might, in the current state of technology, entail an even greater risk of abuse.106 In Denuit,107 the Court had a further opportunity to reaffirm the concept of establishment within the meaning of the Treaty as the basic criterion for determining a Member State’s jurisdiction over a broadcaster under Article 2(1) of the TWF Directive.108 Denuit, the managing director of the City of Brussels cable network, was criminally prosecuted for distributing the programme ‘TNT and Cartoon Network’; this was broadcast via satellite by Turner, a broadcaster established and authorised in the UK, which used a satellite frequency allocated to Luxembourg.109 The retransmission of that programme via the Brussels cable network was prohibited by the Belgian government as it was broadcast by  Art 43 of the Broadcasting Act 1990, UK satellite services (n 17 above), paras [10] and [20].   UK satellite services (n 17 above), paras [61, 63–65] and [77–79].   UK satellite services (n 17 above), paras [40–42]. 102   UK satellite services (n 17 above), paras [25, 28–29, 31, 34, 38] and [48]. 103   UK satellite services (n 17 above), paras [55–57]. 104   UK satellite services (n 17 above), para [58]. See Woods and Scholes (n 72 above) 63. 105   UK satellite services (n 17 above), paras [59–60]. 106   UK satellite services (n 17 above), para [60] second limb. See also AG Lenz in UK satellite services (n 17 above), paras [72–73]. 107   Denuit (n 18 above). 108   Denuit (n 18 above), para [23] citing UK satellite services (n 17 above), para [42]. 109  It was regarded as ‘non-domestic satellite service’ under UK law. Art 43(2) Broadcasting Act 1990, UK satellite services (n 17 above), para [12]. See also AG Lenz in UK satellite services (n 17 above), para [7]. See below (n 116).     99 100 101

76  Dimitrios Doukas a commercial broadcaster and was, for the most part, produced outside the EU (in the US).110 As the UK was the competent State for the broadcasts of Turner, the Court of Justice regarded as immaterial both the use of a satellite frequency allocated to another Member State and the origin of the productions broadcast or their conformity with the Directive.111 Therefore, the transmitting State principle precluded a Member State (such as Belgium) from prohibiting or restricting the retransmission on its territory of broadcasts of a TV broadcaster subject to the jurisdiction of another Member State, even if those broadcasts did not comply with the requirements of Articles 4 and 5 of the TWF Directive, which prescribed quotas for European productions.112 Interestingly, the Belgian government argued that failure by a broadcaster to comply with Articles 4 and 5 of the Directive constituted in itself an abuse of EU law and should allow the receiving Member State to impede the retransmission of that broadcaster’s programmes on its territory.113 The question was not examined by the Court but was dismissed by its Advocate-General. Advocate-General Lenz confirmed that the circumvention doctrine remained applicable notwithstanding the implementation of the TWF Directive, but reiterated that it called for a restrictive interpretation. In the present case, there was nothing to indicate an abuse. This was self-evident, not least because of the absence of a broadcasting activity entirely or principally directed towards Belgium. Non-compliance with the Directive could not constitute an abuse as such, since this would allow the receiving Member State to exercise back-door secondary control over a TV broadcaster subject to the jurisdiction of the transmitting State.114 C. Towards a Narrower Interpretation of the Principle of Abuse: VT4 The principle of abuse was further refined in the VT4 case,115 where the Court of Justice was called to examine the compatibility with Article 2 of the TWF Directive of a decision of the Flemish executive to bar access to the cable network in Flanders by the programmes broadcast by the eponymous broadcaster. VT4 was incorporated under English law and established in London, where it held a UK non-domestic satellite licence.116 Its sole shareholder was a Luxembourg company, and its programmes, including commercials, were aimed at the Flemish public, as they were either produced or subtitled in Dutch. VT4 had a subsidiary in Belgium, where it was in contact with advertisers and production companies, and where it collected information for its news programmes.117 VT4 was regarded by  Ministerial Decree of 17 September 1993. Denuit (n 18 above), paras [13–14] and [17].   Denuit (n 18 above), paras [26–27]. 112  See now Arts 16 and 17 of AVMS Dir 2010/13 (n 3 above). Denuit (n 18 above), paras [32–36], citing Belgian cable network (n 16 above), paras [34–37]. See also Jones (n 32 above) 324–25. 113  See AG Lenz in Denuit (n 18 above), para [35]. 114  AG Lenz in Denuit (n 18 above), paras [34–35], citing his Opinion in Belgian cable network (n 16 above), paras [73–79]. 115   VT4 (n 19 above). 116  The Broadcasting Act 1990 exempted the holders of a UK non-domestic satellite licence (ie broadcasters with headquarters in the UK but broadcasting from satellites abroad) from foreign and cross-media ownership restrictions applicable to domestic broadcasters and domestic satellite services. Since the adoption of the Broadcasting Act 1996, there has no longer been a distinction of satellite licences into ‘domestic’ and ‘nondomestic’. Both Acts were amended by Communications Act 2003. See A Harcourt, ‘Institution-Driven Competition: The Regulation of Cross-Border Broadcasting in the EU’ (2004) European University Institute Working Paper RSCAS No 2004/44: www.iue.it/RSCAS/publications, 5–7. 117   VT4 (n 19 above), paras [7–8]. 110 111

Free Movement of Broadcasting Services and Abuse of Law  77 the Flemish executive as a domestic broadcaster which had established itself in the UK in order to circumvent the application of the Flemish Community legislation; as such, it lacked a licence as a private commercial broadcaster in Flanders.118 The sole such licensed broadcaster eligible for access to the Flemish cable network was the domestic company VTM,119 which held a monopoly in commercial TV broadcasting.120 Alternatively, the Flemish executive argued that even if VT4 were classified as a UK broadcaster, this would not suffice to satisfy the requirements of the Flemish Community legislation, which made access to the cable network by broadcasters from another Member State conditional on the exercise by their home State of proper supervision in terms of compliance with EU law.121 The Court of Justice, on the one hand, used the prohibition of abuse as a principle governing the interpretation of the (original) jurisdiction clause of the TWF Directive.122 It reaffirmed that, where a broadcaster has more than one place of establishment, it falls within the jurisdiction of the Member State in which lies the centre of the broadcaster’s activities. Thus, it left it to the national court to determine which Member State is competent, considering in particular the place where decisions concerning programme policy are taken and the programmes to be broadcast are finally put together.123 Should the national court find it impossible to reach a clear conclusion even after the most scrupulous application of the above test, Advocate-General Lenz suggested that the broadcaster concerned be subject to the jurisdiction of the Member State in which it starts transmitting in the technical sense.124 On the other hand, the Court of Justice, building on the Belgian cable network case,125 appeared to further restrict the scope of the prohibition of abuse as a principle justifying restrictions on the freedom to provide broadcasting services. It held that the mere fact that all the programmes and advertisements broadcast by the broadcaster in question were aimed exclusively at the Flemish public in Belgium did not suffice as such to treat that broadcaster as if it were not established in the UK. The Court emphasised that the Treaty does not prohibit an undertaking from exercising the freedom to provide services if it does not offer any services in the Member State in which it is established.126 This suggests that neither the existence of a secondary establishment in the host Member State nor the direction of all broadcasts exclusively towards the host Member State would suffice to presume a broadcaster’s intention to make an abuse of EU law for the purpose of circumventing the legislation of that State.127   VT4 (n 19 above), para [9].   Decrees of the Flemish Executive of 28 January 1987 on licensing of private TV broadcasters (Belgisch Staatsblad, 19 March 1987) and of 12 June 1991 on radio and television advertising and sponsoring (Belgisch Staatsblad, 14 August 1991), and Art 10(1) No 2 of the Decree of the Flemish Executive of 4 May 1994 on TV and radio cable networks (Belgisch Staatsblad, 4 June 1994). VT4 (n 19 above), paras [4–5]. 120  AG Lenz in VT4 (n 19 above), para [6]. 121  Art 10(2) No 4 of the Decree of the Flemish Executive of 4 May 1994 on TV and radio cable networks (Belgisch Staatsblad, 4 June 1994). VT4 (n 19 above), paras [6] and [9]. 122   VT4 (n 19 above), para [18]. See also UK satellite services (n 17 above), paras [42, 58] and [61], and Denuit (n 18 above), para [23]. Cf Sørensen (n 12 above) 428. 123   VT4 (n 19 above), paras [19] and [23]. See K Trammel, ‘VT4 Ltd v Vlaamse Gemeenschap: Court of Justice Ruling Heralds New Age of European Broadcast and Advertising Deregulation’ (1998–99) 27 Denver Journal of International Law and Policy 553, 556–57 and 561. 124  AG Lenz in VT4 (n 19 above), paras [33–34]. 125  See Belgian cable network (n 16 above), para [65]. 126   VT4 (n 19 above), para [22]. 127   Cf Barnard (n 22 above) 391–92; J Schwarze, ‘Medienfreiheit und Medienvielfalt im Europäischen Gemeinschaftsrecht’ in J Schwarze and A Hesse (eds), Rundfunk und Fernsehen im digitalen Zeitalter (BadenBaden, Nomos 2000) 87, 97; and Varney (n 22 above) 515–16. 118 119

78  Dimitrios Doukas It is noteworthy that the Court of Justice did not elaborate on the justificatory function of the prohibition of abuse.128 Given the establishment of the broadcaster in the present case in more than one Member State, the Court placed the emphasis on the interpretative function of the principle with regard to the jurisdiction clause of Article 2(1) of the TWF Directive. This explains its suggestion that, for the purposes of the Directive, the broadcaster in question should be treated as subject to the jurisdiction of the Member State in which its essential activities are centred, particularly its programme policy decisionmaking and the final processing or compilation of programmes.129 Nevertheless, while such an interpretation seems to considerably limit, it does not pre-empt all possibilities of abuse, and therefore still leaves room for the application of the prohibition of abuse as a principle justifying the subjection of a broadcaster falling within the jurisdiction of another Member State to specific restrictions imposed by the receiving Member State; for instance, in cases where a broadcaster has only one establishment but is found to be abusively centring its main activities in the territory of the host Member State. In any event, the Court seems to have shifted from its TV10 judgment by setting out stricter requirements for the establishment of the objective element of abuse,130 which serves as a presumption for ascertaining its subjective component. In TV10 a very broad criterion, ie the exclusive or principal direction of the activities of a broadcaster to the territory of the host Member State, was deemed sufficient to trigger the application of the principle of abuse.131 By contrast, in VT4 the Court implied that the location in the host Member State of essential activities of the relevant broadcaster should be a decisive factor for ascertaining abuse.132 That is the same criterion used to determine which Member State has jurisdiction over a broadcaster with more than one place of establishment within the EU.133 Interestingly, Advocate-General Lenz adopted an equally restrictive but somewhat different approach to the prohibition of abuse.134 In his Opinion, an abuse could not be inferred from the mere exercise by a person of his freedom of establishment. Thus, although VT4 might have established itself in the UK simply to evade the Flemish broadcasting legislation, this did not as such constitute abuse. Advocate-General Lenz suggested that the emphasis should be laid on the condition inherent in the subjective requirement necessary to affirm abuse, as set out in Veronica and TV10: that is, the national provisions sought to be evaded must protect an important public interest which is legitimate in EU law. On that account, he distinguished the present case from TV10 on the facts: the latter concerned national legislation intended to establish and maintain a non-commercial and pluralistic audiovisual system. The Luxembourg-based broadcaster TV10 would have had the right to broadcast its programmes from the Netherlands if it had satisfied the requirements of Dutch law. On the contrary, this would not have been 128   Cf Woods and Scholes (n 72 above) 65–67, who argue that the judgment did not deal with the circumvention principle at all. 129   VT4 (n 19 above), para [23]. See BJ Drijber, ‘The Revised Television without Frontiers Directive: Is It Fit for the Next Century?’ (1999) 36 Common Market Law Review 87, 94–95, who notes that, unlike TV10, where the referring court had established circumvention, in VT4 the national court had not made a similar finding of fact. 130  See Sørensen (n 12 above) 444. Against Katsirea (n 6 above) 133. 131   TV10 (n 14 above), para [26]. 132  See VT4 (n 19 above), para [23]. Cf Drijber (n 129 above) 93–94. 133   VT4 (n 19 above), para [19] and UK satellite services (n 17 above), para [58]. 134  AG Lenz in VT4 (n 19 above), paras [36–37], citing his Opinion in Belgian cable network (n 16 above), paras [73–79].

Free Movement of Broadcasting Services and Abuse of Law  79 possible in the case of VT4 because no other commercial broadcaster could operate in Flanders by reason of the statutory monopoly held by the domestic company VTM. This was sufficient to rule out any question of abuse.135 In stark contrast to its Advocate-General, the Court of Justice did not enter into the legitimacy of the objectives pursued by the Flemish Community legislation and appeared to leave open the question of the existence of an abuse, should the centre of the relevant broadcaster’s essential activities be found to lie in effect in Belgium. This might raise the question whether or not the public interest rationale inherent in the abuse test adopted in Veronica and TV10 remains relevant.136 Despite the absence of any explicit reference to those two cases, it seems difficult to read the VT4 judgment as a departure from the public interest rationale. This should rather be seen as a self-evident requirement, as it would be unreasonable to establish abuse of Union rules on the basis of circumvention of national provisions which are not compatible with EU law.137 This rationale has also been referred to by commentators to explain the Court’s shift from the broad approach in TV10, which concerned a national public service broadcasting system designed to ensure media pluralism, to the stricter approach in VT4, which concerned a national legal system that authorised the creation of a private commercial broadcasting monopoly. The latter is difficult to accommodate with the regulatory competition encouraged by the TWF Directive and its successor, the AVMS Directive.138 Thus, there appears to be only a difference in focus between the Advocate-General’s Opinion and the Court of Justice. The latter placed the emphasis on curtailing the discretion enjoyed by Member States in cases of alleged abuses of the freedom to provide broadcasting services. Such a reading of the VT4 case seems to have been confirmed by its aftermath. Following the declaration of the statutory monopoly in question as incompatible with what are now Articles 106(1) and 49 TFEU [ex Articles 86(1) and 43 EC],139 the Flemish Community amended its legislation to allow the licensing of more commercial broadcasters. In 2002, VT4 voluntarily moved its place of establishment from the UK to Belgium, where it was awarded a licence as a domestic broadcaster.140 D.  The Principle of Abuse between Strict Interpretation and Non-application: De Agostini The principle of abuse in its narrow interpretation was reaffirmed by Advocate-General Jacobs in De Agostini,141 although it was impossible to apply to the facts. While the 135  AG Lenz in VT4 (n 19 above), para [38]. Cf S Nikoltchev, ‘Right of Intervention to Address Abuses of the Fundamental Freedoms: The Role of the Case-law’ in Institute of European Media Law (ed), Regulating the New Media Landscape: A Directive for Audiovisual Media Services without Frontiers (Baden-Baden, Nomos 2008) 25, 30. Against Katsirea (n 6 above) 133. 136   Cf Woods and Scholes (n 72 above) 77. 137  See Hörnle (n 83 above) 116; Hansen (n 2 above) 137; and Kjellgren (n 48 above) 273–74. Cf Katsirea (n 6 above) 133. 138  See Gibbons (n 2 above) 254–56; Woods and Scholes (n 72 above) 76–77; and AG Geelhoed in Akrich (n 74 above), para [101]. 139  Commission Decision 97/606/EC of 26 June 1997 on the exclusive right to broadcast television advertising in Flanders [1997] OJ L244/18, and Case T-266/97 Vlaamse Televisie Maatschapij NV v Commission [1999] ECR II-2329. 140  EAVO, ‘Transfrontier Television in the European Union’ (n 5 above) 13, 19. 141  AG Jacobs in De Agostini (n 20 above), para [51].

80  Dimitrios Doukas Court of Justice did not enter into the question of a possible abuse,142 the contribution of Advocate-General Jacobs’ Opinion to the construction of the prohibition of abuse of EU law as an application of the general principle of abuse of rights, common to most legal systems,143 should not be underestimated.144 The case concerned, as is well known, the obstacles raised by the Swedish bans on TV advertising addressed to children and on misleading advertising to the retransmission in Sweden of commercials broadcast via satellite by TV3, a broadcaster established in the UK. Given the jurisdiction of the UK, Advocate-General Jacobs dismissed the argument that TV3 should be treated as a Swedish broadcaster inasmuch as its programmes were all produced in Sweden, dubbed or subtitled in Swedish, and included advertisements exclusively designed for the Swedish market.145 In order to prevent the application of the TWF Directive to advertisements included in TV broadcasts specifically directed to the receiving State, both the Swedish Consumer Ombudsman and the Swedish government relied on the circumvention doctrine.146 Advocate-General Jacobs suggested that the latter could not be invoked in a case ‘where the interests protected by the rules [on advertising] allegedly sought to be avoided are within the scope of the Directive;’ otherwise the transmitting State principle would be fundamentally undermined.147 In this regard, he contrasted the TWF Directive with the European Convention on Transfrontier Television,148 which expressly provided that advertisements ‘which are specifically and with some frequency directed to audiences in a single Party other than the transmitting Party shall not circumvent the television advertising rules in that particular Party’.149 This seems to be an even narrower interpretation of the principle of abuse than the one inherent in the Court’s case law, and is not totally consistent with the recognition of the applicability of the principle in the field of the TWF Directive, as the risk of abuse has not vanished despite the minimum harmonisation at EU level.150 More accurate appears Advocate-General Jacobs’ assertion that there was nothing to suggest that TV3 had in fact established itself in the UK in order to avoid the Swedish legislation. In accordance with the transmitting State principle, he stressed that the mere fact that a broadcaster transmits programmes intended for reception in another Member State than the one where it is established could not suffice to trigger the application of the principle of abuse; instead, the broadcaster concerned ‘must be acting wrongfully or improperly’. The burden of proof of such impropriety falls on ‘the Member State seeking to avail itself of the exception’.151 Nevertheless, the Advocate-General did not elaborate  See also Kjellgren (n 48 above) 262.  See M Byers, ‘Abuse of Rights: An Old Principle, A New Age’ (2002) 47 McGill Law Journal 389. 144  AG Jacobs in De Agostini (n 20 above), para [45], citing Belgian cable network (n 16 above), para [65] and AG Lenz in Belgian cable network (n 16 above), paras [74–75]. Cf N Burrows and R Greaves, The Advocate General and EC Law (Oxford, Oxford University Press 2007) 7. 145  AG Jacobs in De Agostini (n 20 above), paras [32–34], citing UK satellite services (n 17 above), paras [35–42]. 146  AG Jacobs in De Agostini (n 20 above), para [40]. 147  AG Jacobs in De Agostini (n 20 above), para [46]. 148  European Convention on Transfrontier Television (ECTT) of 5 May 1989, Council of Europe, European Treaty Series No 132, amended by Protocol of 1 October 1998 (entered into force on 1 March 2002), ETS No 171. 149  Art 16 ECTT (n 148 above). AG Jacobs in De Agostini (n 20 above), para [46]. 150  See now AVMS Dir 2010/13 (n 3 above). See also Hansen (n 2 above) 133–34. 151  AG Jacobs in De Agostini (n 20 above), para [47], citing TV10 (n 14 above), para [18] and Veronica (n 13 above), para [17], where the stronger terms abusivement/de manière abusive are used in the French version. 142 143

Free Movement of Broadcasting Services and Abuse of Law  81 on the impropriety requirement.152 Instead he relied on the aim of the principle, which is ‘to pre-empt reliance on Community law by a provider of services whose activity is entirely or principally directed towards the Member State seeking to invoke it’, and ruled out its application in De Agostini, as the programmes of TV3 were also broadcast to other Scandinavian countries.153 With regard to the academic question whether the principle of abuse presupposes a significant difference between the laws sought to be evaded and the laws to which the evading entity is subject, Advocate-General Jacobs admitted that there is no scope for the principle ‘where the laws in question are to all intents and purposes the same’. The principle could only apply where differences in legislation were sufficient to warrant the establishment of an undertaking in another Member State for the sole purpose of exploiting those differences. However, Advocate-General Jacobs suggested that it was unfeasible to lay down a general rule on the degree of parity required as a matter of law in order to preclude or trigger the application of the principle.154 Finally, Advocate-General Jacobs noted that the principle of abuse would be unlikely to assist the advertisers who invoked it. In contrast to Veronica and TV10, the national rules in De Agostini were not sought to be enforced against a broadcaster established in another Member State but against advertisers based in Sweden. He said that the principle could not apply to such circumstances unless it was further developed. In any event, any argument that the advertisers had used the services of a broadcaster established in another Member State for the sole purpose of avoiding the legislation of their home State would founder because the commercials in question had also been broadcast on domestic TV channels.155 Advocate-General Jacobs ruled out not only the application of the principle of abuse in this case, but also the compatibility with the TWF Directive of restrictions imposed by a Member State on advertisements transmitted by TV broadcasters from another Member State insofar as they aimed to protect children or prohibit misleading advertising.156 This view was partly followed by the Court of Justice—solely with regard to the ban on TV advertising addressed to children—subject to the two celebrated conditions set out in its judgment.157 Unlike the ECJ in De Agostini, the EFTA Court158 did examine the question of abuse in the similar case Mattel and Lego.159 This concerned commercials broadcast via satellite in Norway, once again by the UK-based broadcaster TV3 on behalf of a Norwegian toy company, in violation of the Norwegian ban on TV advertising aimed at children. The EFTA Court admitted that the TWF Directive might encourage advertisers to avail themselves of the services of broadcasters subject to a jurisdiction with the least restrictive rules.  See Hansen (n 2 above) 134–35.  AG Jacobs in De Agostini (n 20 above), para [48].  AG Jacobs in De Agostini (n 20 above), paras [49–50]. Cf Schwarze (n 127 above) 97, who suggests that blatant differences between national laws should establish a presumption for abuse. 155  AG Jacobs in De Agostini (n 20 above), para [52]. 156  AG Jacobs in De Agostini (n 20 above), paras [53–54, 61–62] and [86–87]. 157  Namely: (a) prohibition of secondary control over broadcasts from other Member States, in addition to the one carried out by the transmitting State; and (b) prohibition of prevention of retransmission as such of the TV programmes from other Member States. De Agostini (n 20 above), paras [34 and 38]. See also Barnard (n 22 above) 389. 158  The Court of the European Free Trade Association (EFTA) States which are parties to the European Economic Area (EEA) Agreement (currently Iceland, Liechtenstein and Norway). 159   Joined Cases E-8/94 and E-9/94 Forbrukerombudet v Mattel Scandinavia and Lego Norge [1994–95] EFTA Court Reports 113: www.eftacourt.int/index.php/cases/case_e_8_94/. 152 153 154

82  Dimitrios Doukas However, it held that, while this result might be viewed as a circumvention of domestic law from a national perspective, it was in fact a logical and necessary consequence of the principles underlying this Directive. Accordingly, transfrontier broadcasts satisfying the standards of the transmitting State and those of the Directive should not be impeded.160 In contrast to Advocate-General Jacobs in De Agostini, the EFTA Court dismissed not only the concrete application but also the very existence of a principle of abuse either as a principle governing the interpretation of the TWF Directive or as a ground for justifying restrictions. This is not consistent with the case law of the Court of Justice and has been overtaken by the subsequent amendment(s) of the Directive.161 E.  The Narrow Interpretation of the Principle of Abuse Consolidated: United Pan-Europe The strict interpretation of the principle of prohibition of abuse could find support in a recent case concerning the ‘must-carry’ obligation imposed by Belgian legislation on the cable network operators in the bilingual region of Brussels-Capital; ie the obligation to relay the programmes transmitted not only by all domestic public service broadcasters but also by certain private broadcasters established in Belgium.162 The Court of Justice held that such a restriction on the free movement of services could be justified by the general interest in preserving the pluralist nature of the range of TV programmes available in that bilingual region, and thus formed part of a cultural policy aiming to safeguard freedom of expression in the audiovisual sector.163 The Court acknowledged the suitability of the ‘must-carry’ obligation to ensure that TV viewers in that region would not be deprived of access, in their own language, to local and national news, and to programmes which are representative of their culture.164 While the Court of Justice left it to the national court to decide on the compatibility of the legislation in question with EU law, it set out limits to the discretion of national authorities by suggesting that the award of ‘must-carry’ status should be based on criteria which are neither disproportionate in relation to the aim pursued nor discriminatory.165 This precluded making the award of that status, either in law or in fact, conditional on the requirement that a broadcaster be established in the national territory.166 This limit becomes more important if it is read with the obiter dictum that the Belgian government stated that the fact that one of the private broadcasters possessing ‘must-carry’ status recently decided to transfer its head office to another Member State (Luxembourg) was a factor which would be taken into account in the evaluation of the need to maintain that status, even if the content of the programmes transmitted by that broadcaster had not changed.167 While the Court generally referred to the Flemish cable network case,168   Mattel and Lego (n 159 above), paras [46–49].  See criticism by Hansen (n 2 above) 129–32, 135. Cf Sørensen (n 12 above) 436–37. 162   United Pan-Europe (n 24 above), paras [3–5]. 163   United Pan-Europe (n 24 above), paras [41–42]. 164   United Pan-Europe (n 24 above), para [43]. 165   United Pan-Europe (n 24 above), paras [44–47] and [49]. Cf Art 31(1) of Dir 2002/22/EC of the European Parliament and of the Council of 7 March 2002 on universal service and users’ rights relating to electronic communications networks and services (Universal Service Directive) [2002] OJ L 108/51, as amended by Dir 2009/136/EC [2009] OJ L337/11. 166   United Pan-Europe (n 24 above), para [48]. 167   United Pan-Europe (n 24 above), para [35]. See also AG Poiares Maduro in United Pan-Europe (n 24 above), para [11]. 168   Flemish cable network (n 15 above), para [12]. 160 161

Free Movement of Broadcasting Services and Abuse of Law  83 Advocate-General Poiares Maduro relied on the judgment in VT4 as authority that a broadcaster may provide TV programmes (including news), the contents of which are tailored to the public in one Member State, while being established in another Member State.169 This implies that if a broadcaster satisfies, on the basis of its programmes, the essential requirements prescribed by the national ‘must-carry’ legislation for the attainment of the public interest objective pursued, it should not be denied the ‘mustcarry’ status on the mere ground that it has its head office in another Member State while its programmes specifically target the public in the first Member State.170 IV. The Codification of the Principle of Abuse in the TWF Directive

A.  The Codification of the Interpretative Function of the Principle of Abuse The principle of abuse was taken into account by the two revisions of the TWF Directive in 1997 and 2007.171 Under the first amending Directive 97/36, the function of the prohibition of abuse as a ground for justifying restrictions was only recognised in the Preamble. This acknowledged the right of a Member State: [T]o take measures against a TV broadcasting organisation that is established in another Member State but directs all or most of its activity to the territory of the first Member State if the choice of establishment was made with a view to evading the legislation that would have applied to the organisation had it been established on the territory of the first Member State.172

This was obviously not sufficient to remove the legal uncertainty about the specific conditions attached to the principle.173 On the other hand, the prohibition of abuse was integrated primarily as a principle governing the interpretation of Article 2 of the TWF Directive.174 This was amended to include four new paragraphs ((2)–(5)) concerning  AG Poiares Maduro in United Pan-Europe (n 24 above), para [16].  Note the complaint launched by the media regulator (CSA) of the Belgium French Community on 1 February 2006 against the decision of the broadcaster TVI, a Belgian subsidiary of the Luxembourg-based company CLT-UFA, not to apply for a renewal of its authorisation by the French Community claiming that its parental company had taken over the editorial responsibility for its services. According to Art 2(7) of the Broadcasting Decree of 27 February 2003 of the French Community, the latter has authority over editors of broadcasting services established in another EU or EEA Member State, whose activities—as observed by the CSA on consultation with the European Commission—are entirely or mainly directed at the general public in the French Community, and who have established themselves in that other State with a view to evading the rules that would apply to them if they were established in Wallonia. See EAVO, IRIS 2006-3:10/16: merlin. obs.coe.int/iris/2006/3/article16.en.html. At the November 2006 meeting of the Contact Committee established under Art 23(a) of the TWF Dir (n 3 above)—now Art 29 of the AVMS Dir 2010/13 (n 3 above)—the Luxembourg and Belgian delegations agreed to cooperate in order to find a solution which would guarantee respect by CLT of the requirements prescribed by the French Community legislation concerning audiovisual media productions. See Sixth Report from the Commission on the application of Dir 89/552, COM (2007) 452 final, ss 2.1 and 2.6. On 15 January 2009 the Belgian Conseil d’ État held that CLT and TVI had been duly licensed in—and thus subjected to the jurisdiction of—Luxembourg since 2006. Seventh Report from the Commission on the application of Dir 89/552, COM (2009) 309 final, s 2.1. 171   Dir 97/36 and Dir 2007/65 (n 3 above). 172  Recital 14 of Preamble to Dir 97/36 (n 3 above). 173  See Hansen (n 2 above) 111, 115–17. 174  See Recital 19 of Preamble to Dir 97/36 (n 3 above), emphasising the need ‘to make arrangements within a Community framework, in order to avoid potential legal uncertainty and market distortions and to reconcile free circulation of television services with the need to prevent the possibility of circumvention of national measures protecting a legitimate general interest’. 169 170

84  Dimitrios Doukas jurisdiction over broadcasters.175 Accordingly, a broadcaster falls within the jurisdiction of the Member State in which it is established on the basis of the criteria set out in Article 2(3).176 Should the latter not apply, a Member State has jurisdiction over a broadcaster which uses a frequency granted by that State, or a satellite capacity appertaining to it, or a satellite up-link situated in its territory.177 Under Article 2(3), a broadcaster is deemed established in the Member State in which it has its head office and the editorial decisions about programme schedules are taken.178 If the head office and the place of those decisions lie in different Member States, the broadcaster is considered established in the State in which a significant part of the workforce involved in the pursuit of the TV broadcasting activity operates.179 If that is true for both countries, the broadcaster is presumed established in the Member State where its head office is located.180 If the broadcaster does not have a significant part of its workforce operating in either of the two Member States, it is deemed established in the Member State where it first began broadcasting in accordance with the legal system of that State, provided that it maintains a stable and effective link with its economy.181 If its head office lies in a Member State but decisions on programme schedules are taken in a third country, or vice versa, the broadcaster is presumed to be established in the Member State, provided that a significant part of the workforce involved in the pursuit of the broadcasting activity operates there.182 In case the question of jurisdiction cannot be determined in accordance with the criteria set out in Article 2(3) and (4), the competent Member State shall be the one in which the broadcaster is established within the meaning of Article 49 TFEU [ex Article 43 EC].183 One can only refer to the interpretation of freedom of establishment as involving the ‘actual pursuit of an economic activity through a fixed establishment in another Member State for an indefinite period’.184 Against this backdrop, the Commission instituted infringement proceedings against Belgium due to the decision of the Flemish media regulator to oblige VT4 to apply for a licence in Belgium even though it fell under UK jurisdiction.185 Similar proceedings were launched against the Netherlands following the decision of its media regulator to grant itself jurisdiction over the channels RTL4 and RTL5, which were transmitted via satellite by a broadcaster subject to the jurisdiction of Luxembourg and directed their services to the Dutch market; the Dutch media regulator alleged that responsibility for those channels lay with the Dutch broadcaster RTL/Veronica de Holland. The case 175  See also Katsirea (n 1 above) 111–14; Varney (n 22 above) 516; and C Farrar, ‘EC Broadcasting Law Clarified: The Paul Denuit and VT4 Cases and the New ‘Television without Frontiers’ Directive’ (1998) 9 Entertainment Law Review 16, 19. 176  Art 2(2) of Dir 89/552, as amended by Dir 97/36 (n 3 above). See now Art 2(2) of AVMS Dir 2010/13 (n 3 above). 177  Art 2(4)(a) to (c) of Dir 89/552 (n 3 above). See now Art 2(4) of AVMS Dir 2010/13 (n 3 above), which has left out the first of those three conditions, ie the use of a frequency granted by a Member State. 178  Art 2(3)(a) of Dir 89/552 (n 3 above). See now Art 2(3)(a) of AVMS Dir 2010/13 (n 3 above). 179  Art 2(3)(b) first limb of Dir 89/552 (n 3 above). See now Art 2(3)(b) first limb of AVMS Dir 2010/13 (n 3 above). 180  Art 2(3)(b) second limb of Dir 89/552 (n 3 above).See now Art 2(3)(b) second limb of AVMS Dir 2010/13 (n 3 above). 181  Art 2(3)(b) third limb of Dir 89/552 (n 3 above). See now Art 2(3)(b) third limb of AVMS Dir 2010/13 (n 3 above). 182  Art 2(3)(c) of Dir 89/552 (n 3 above). See now Art 2(3)(c) of AVMS Dir 2010/13 (n 3 above). 183  Art 2(5) of Dir 89/552 (n 3 above). See now Art 2(5) of AVMS Dir 2010/13 (n 3 above). 184   Factortame II (n 65 above), para [20]. See also Jones (n 32 above) 312–13. 185  See Press Release IP/99/455 of 5 July 1999, and Third Report from the Commission on the application of Dir 89/552, COM (2001) 9 final, s 4.1.

Free Movement of Broadcasting Services and Abuse of Law  85 was closed after the Dutch Council of State annulled the decision of the Dutch media regulator, stating that the latter could not claim jurisdiction and trigger a situation of dual jurisdiction in conflict with the TWF Directive.186 B.  The Codification of the Justificatory Function of the Principle of Abuse The criteria introduced into the TWF Directive to delimit jurisdiction over broadcasters considerably reduced, but did not remove completely, the risk of abuse of the freedom of movement within the EU.187 This was recognised in the second revision of the Directive in 2007, which extended its scope to all providers of audiovisual media services,188 including both broadcasters and providers of on-demand audiovisual media services or audiovisual commercial communication.189 While this latest amendment has kept intact the criteria determining jurisdiction,190 it has introduced into Article 3 of the Directive a new set of provisions which have codified the function of the principle of prohibition of abuse as a ground for justifying restrictions on the free movement of broadcasting services.191 Accordingly, a two-step procedure is available to any Member State which: (a) applies to domestic broadcasters more stringent or detailed rules than those of the Directive, which serve the protection of a public interest objective in compliance with EU law; and (b) assesses that a broadcaster under the jurisdiction of another Member State provides a TV broadcast wholly or mostly directed towards its territory.192 That Member State should first contact the competent Member State ‘with a view to achieving a mutually satisfactory solution’. The latter shall, on receipt of such a substantiated request, ask the broadcaster to comply with the rules of general public interest in the receiving State, and must inform that State of the results achieved within two months.193 If the receiving Member State is not satisfied with the outcome of the first step of the procedure, it may take appropriate measures against the broadcaster which has allegedly tried to circumvent the stricter rules applicable in its territory by establishing itself in another Member State. However, it must first notify the Commission and the home Member 186  See Fourth and Fifth Reports from the Commission on the application of Dir 89/552, COM (2002) 778 final, s 3.2, and COM (2006) 49 final, s 3.2. 187  See Drijber (n 129 above) 91–98; Farrar (n 175 above) 20; and Woods and Scholes (n 72 above) 79. See also the criticism of the criterion of editorial responsibility for programme schedules by Harrison and Woods (n 8 above) 185–86, and T Gibbons, ‘Jurisdiction over (Television) Broadcasters—Criteria for Defining “Broadcaster” and “Content Service Provider”’ in Institute of European Media Law (ed), Die Zukunft der Fernsehrichtlinie— The Future of the ‘Television without Frontiers’ Directive (Baden-Baden, Nomos 2005) 53, 58–59. 188   Dir 2007/65 (n 3 above). See generally, M Burri-Nenova, ‘The New Audiovisual Media Services Directive: Television without Frontiers, Television without Cultural Diversity’ (2007) 44 Common Market Law Review 1689. 189  Art 1(a) and (d) of Dir 2007/65 (n 3 above). 190  Art 2(2) to (5) of Dir 89/552, as amended by Art 1(3) of Dir 2007/65 (n 3 above). See now Art 2(2) to (5) of AVMS Dir 2010/13 (n 3 above). 191   Art 3 of Dir 89/552, as amended by Art 1(5) of Dir 2007/65 (n 3 above). See also 31–33 of Preamble to Dir 2007/65 (n 3 above) and Recitals 40–42 of Preamble to AVMS Dir 2010/13 (n 3 above) Art 3 of Dir 89/552 now Art 4 of AVMS Dir 2010/13 (n 3 above) does not include cases of circumvention of stricter national rules by providers of on-demand audiovisual media services, to which, however, the case law of the Court of Justice should apply. See Seventh Report from the Commission on the application of Dir 89/552 (n 170 above), s 2.1. 192  Art 3(2) read with Art 3(1) of Dir 89/552, as amended by Art 1(5) of Dir 2007/65 (n 3 above). See now Art 4(1) and (2) of AVMS Dir 2010/13 (n 3 above). 193  Either Member State may also invite the Contact Committee established under Art 23(a) of the Dir—now Art 29 of AVMS Dir 2010/13 (n 3 above)—to examine the case. Art 3(2) of Dir 89/552, as amended by Art 1(5) of Dir 2007/65 (n 3 above). See now Art 4(2) of AVMS Dir 2010/13 (n 3 above).

86  Dimitrios Doukas State of its intention, and substantiate the grounds on which its assessment is based.194 The Commission must decide within three months whether the measures intended by the receiving Member State comply with EU law. In particular, it must determine whether the assessment made by that State is well-founded, and whether the measures are objectively necessary and proportionate, and can apply in a non-discriminatory manner. The measures cannot be adopted unless the Commission declares them compatible with EU law.195 This procedure is designed to strike a balance between the freedoms of establishment and services, on the one hand, and legitimate Member State concerns over circumvention of national rules of public interest (including rules on the protection of consumers or minors and cultural policy) on the other hand. This should not, however, call into question ‘the proper application of the country of origin principle’.196 It is therefore suggested that, in order to assess whether a circumvention can be established in an individual case where a broadcast is wholly or mostly directed to its territory, a Member State should consider indicators such as the origin of advertising and/or subscription revenues, the main language of the service or the existence of commercial communication or programmes specifically targeted at the public in the receiving State.197 It cannot be disputed that, despite their vagueness and incompleteness,198 the above provisions consolidate the case law on the application of the prohibition of abuse as a principle justifying restrictions.199 They also confirm its narrow interpretation, with an eye to preserving the effectiveness of the free movement provisions and, especially, the country of origin principle as the cornerstone of the Internal Market.200 V. The Foundations of a General Principle of Prohibition of Abuse of (EU) Law

The significance of the case law on broadcasting services and its integration into secondary law for the emergence of a general principle of prohibition of abuse of EU law cannot be overestimated. The Court of Justice further developed the circumvention doctrine originally established in Van Binsbergen, and recognised the prohibition of abuse, at first 194  Art 3(3) and (4)(a) of Dir 89/552, as amended by Art 1(5) of Dir 2007/65 (n 3 above). See now Art 4(3) and (4)(a) of AVMS Dir 2010/13 (n 3 above). 195  Art 3(3) in fine, (4)(b) and (5) of Dir 89/552, as amended by Art 1(5) of Dir 2007/65 (n 3 above). See now Art 4(3) in fine, (4)(b) and (5) of AVMS Dir 2010/13 (n 3 above). 196  Recitals 31–32 of Preamble to Dir 2007/65 (n 3 above), citing VT4 (n 19 above), Centros (n 23 above), Belgian cable network (n 16 above), Denuit (n 18 above), Van Binsbergen (n 9 above), and TV10 (n 14 above). See also Recitals 40–41 of Preamble to AVMS Dir 2010/13 (n 3 above). 197  Recital 33 of Preamble to Dir 2007/65 (n 3 above), and Recital 42 of Preamble to AVMS Dir 2010/13 (n 3 above). 198   Cf Harrison and Woods (n 8 above) 183. 199   Cf Art 1(2)(b) of Annex II to the Agreement between the EC and the Swiss Confederation in the audiovisual field, establishing the terms and conditions for the participation of the Swiss Confederation in the Community programmes MEDIA Plus and MEDIA Training—Final Act—Declarations [2006] OJ L90/23. This gives Switzerland the right: [T]o take action against a television broadcasting organisation which is established in the territory of a Member State of the Community but whose activity is entirely or mainly directed towards Swiss territory, if the place of establishment was chosen with a view to evading the rules which would have applied to the organisation had it been established on Swiss territory. These conditions will be interpreted in the light of the relevant case law of the Court of Justice. 200   Cf the criticism by Gibbons (n 2 above) 255–56, that the strict scope of the principle leaves Member States very little room to defend their national media and cultural policies against negative effects of regulatory competition.

Free Movement of Broadcasting Services and Abuse of Law  87 as a principle which could pre-empt the application of what is now Article 56 TFEU [ex Article 49 EC], and subsequently as a principle which could justify specific restrictions imposed by a Member State on the free movement of broadcasting services.201 While the Court left the assessment of abuse in each particular case to the national courts, it set out two basic conditions necessary to trigger the application of the principle of abuse: (a) an objective element, consisting in the direction of the activities of a broadcaster totally or principally towards the territory of a Member State other than that in which it is established; and (b) a subjective element, ie the broadcaster’s intention to improperly rely on the freedom to provide services in order to circumvent national rules of the host Member State which are justified by public interest objectives legitimate under EU law (such as cultural policy and media pluralism). These two requirements, the application of which is subject to review by the Court of Justice,202 may be regarded as the forerunner of the two-step abuse test that was introduced in the field of free movement of goods in EmslandStärke.203 The latter made the establishment of abuse conditional, first, on a combination of objective circumstances in which, despite formal observance of the conditions laid down by the rules of EU law, the purpose of those rules has not been achieved; and second, a subjective element consisting in the intention to obtain an advantage from the Union rules by creating artificially the conditions laid down for obtaining it.204 The case law on broadcasting services demonstrates that the Court of Justice has not treated the subjective prerequisite to abuse as an independent requirement but has established a presumption for ascertaining an intention of abuse. This can be presumed if the objective condition is satisfied—that is, if a broadcaster is found to be acting in a manner blatantly incompatible with the purpose of the free movement of services, by centring all or most of its activities in a Member State other than the one in which it is established.205 Nevertheless, there has been a significant shift in the way the Court has interpreted the objective (and concomitantly, the subjective) element of abuse. Before the implementation of the TWF Directive, the Court allowed Member States a relatively broad discretion by suggesting that the mere fact that all the programmes of a broadcaster were directed to the territory of a Member State other than that of its establishment might suffice to establish abuse. The only limit set by the Court to that discretion was the exclusion of outright bans or measures of a protectionist and discriminatory nature, which would be tantamount to a denial of the freedom to provide services. Since the implementation of the TWF Directive, the Court appears to have adopted a much narrower construction of the principle of abuse so as not to undermine the essence or the effectiveness of the free movement of services, and the country of origin principle inherent in the Directive. The basic premise is that the Treaty does not prohibit an undertaking from exercising the freedom to provide services if it does not offer any 201  See Hansen (n 2 above) 127–28, 133, 136–37, who however supports only the pre-emptory effect of the principle. 202  See also Ottersbach (n 1 above) 54–56. 203  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 204   Emsland-Stärke (n 203 above), paras [52–53]. See de la Feria (n 10 above) 402–03, 408–11; D Weber, ‘Abuse of Law’ (2004) 31 Legal Issues of Economic Integration 43; and P Koutrakos, ‘The Emsland-Stärke Abuse of Law Test in the Law of Agriculture and Free Movement of Goods’, ch 14 below. 205   Cf AG La Pergola in Joined Cases C-147 and 148/97 Deutsche Post AG v Gesellschaft für Zahlungssysteme mbH and Citicorp Kartenservice GmbH [2000] ECR I-825, para [15], fn 22, who cites Veronica (n 13 above) and TV10 (n 14 above) as examples of cases where the service providers ‘acted in an unreasonable and unconscionable manner, for the sole purpose of obtaining, to the detriment of others, improper benefits which were manifestly alien to the purpose of the Treaty rules on freedom of establishment and freedom to provide services’.

88  Dimitrios Doukas services in the Member State in which it is established. This more restrictive interpretation of the principle of abuse seems more in line with the similarly strict approach originally adopted by the Court in the field of establishment in Segers,206 and has paved the way for the Centros line of authority.207 Thus, neither the mere direction of all the broadcasts transmitted by a broadcaster to the public in the host Member State nor the existence of a secondary establishment in that State would suffice any more to demonstrate an intended abuse of EU law. Instead, the case law suggests that the location of essential activities of a broadcaster in the receiving Member State, such as activities relating to the programme policy decision-making and the final processing of programmes, could be a strong indication for ascertaining abuse. Interestingly, this strict construction of the principle of abuse has been used by the Court of Justice in a twofold manner. First and foremost, it has been used to interpret the scope of the TWF Directive and especially the provisions on Member States’ jurisdiction over broadcasters. This has been the first step in the Court’s effort to strike the right balance between a broadcaster’s objective establishment in a Member State and a possible intention of abuse of the freedom of movement.208 Secondly, should the tension between those two parameters be impossible to solve by way of interpretation of the Directive, the Court has not ruled out the application of the principle of abuse as a ground for justifying potential restrictions on a broadcaster’s freedom to provide services. The former use of the principle has a pre-emptory effect, consisting in the subjection of a broadcaster only to the legal system of one Member State. The latter use of the principle has a justificatory effect, allowing the subjection of a broadcaster to a second set of (legitimate) rules in the host Member State, in addition to the legislation of its Member State of origin.209 Regardless of the difference in effect, both applications of the principle of abuse could be seen as two sides of the same coin, ie the interpretation of EU law. They both have in common the prohibition of reliance on primary or secondary EU law for the attainment of improper advantages inconsistent with the objectives of the relevant provisions.210 It is true that, despite the integration of the case law on the prohibition of abuse into the TWF Directive and its successor, the AVMS Directive, there are still ambiguities, both as to the effect and the conditions of the application of the principle to the field of audiovisual media services, which need to be further elaborated by the Court of Justice.211 This cannot, however, undermine the contribution of the case law to the emergence of a general principle   Segers (n 21 above), para [16].  See Centros (n 23 above). See also V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205, 210–12; AG Alber in Inspire Art (n 23 above), paras [119–20]; and AG Geelhoed in Akrich (n 74 above), para [100]. Cf AG Poiares Maduro in Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641, paras [32–35]. 208  See Straetmans and Goemans (n 74 above) 320, 328–29. 209  Against Ottersbach (n 1 above) 101–02, 104–07, who argues that the application of the principle should only serve to distinguish the freedom to provide services from the freedom of establishment and lead to the application of one set of rules. 210  See AG Poiares Maduro in Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd and County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, paras [68–69]; and in Case C-419/02 BUPA Hospitals Ltd and Goldsborough Developments Ltd v Commissioners of Customs & Excise [2006] ECR I-1685, paras [67–68]; and AG Stix-Hackl in Case C-452/04 Fidium-Finanz AG v Bundesanstalt für Finanzdienstleistungsaufsicht [2006] ECR I-9521, paras [81–82]. 211  See R Craufurd Smith, ‘European Community Media Regulation in a Converging Environment’ in N Nic Shuibhne (ed), Regulating the Internal Market (Cheltenham, Edward Elgar Publishing 2006) 105, 112. 206 207

Free Movement of Broadcasting Services and Abuse of Law  89 of prohibition of abuse of EU law212 by developing some of its core elements. These include: first, the two main conditions for abuse—objective and subjective—and their interrelation; secondly, the strict construction of the scope of the principle in the light of the objectives of the relevant Union rules and, in particular, the Internal Market, which is premised on the idea of free movement to the Member State with the greatest comparative advantage;213 and last but not least, the dual function of the principle, either to pre-empt the application of EU law or to justify restrictions. These core elements have transcended the boundaries of the free movement of services and the broadcasting sector, and have been applied to other (albeit not all) fundamental freedoms or fields of EU law.214 These same essential features lend to the prohibition of abuse, on the one hand, the degree of generality and the dimension of weight inherent in a principle as defined by Dworkin. Accordingly, principles are standards which must be observed as requirements of justice or fairness, and which, unlike rules, do not apply in an all-or-nothing fashion (setting out legal consequences that follow automatically when the conditions provided are met), but as a consideration inclining in one direction or another.215 On the other hand, the typical features of the prohibition of abuse as developed by the European Union’s judicature can also satisfy Alexy’s definition of principles as optimisation requirements; that is, as norms which require that something be realised to the greatest extent possible, and which can be satisfied to varying degrees, with the appropriate degree of satisfaction depending on what is factually and legally possible.216 Admittedly, the case law of the Court of Justice and its gradual integration into the European Union’s legislative framework for the field of audiovisual media services cannot provide any support to the emergence of a Union’s general principle of prohibition of abuse of (any type of) law capable of striking down conflicting domestic legislation.217 They can only be seen as milestones in the development of a principle of prohibition of abuse of EU law. This is not a systemic principle inherent in the constitutional structure of the Union and autonomous to the EU legal order. Nor does it constitute a principle deriving from the rule of law, which requires that the exercise of public power be subject to substantive or procedural limitations and judicial review.218 Unlike the latter category 212  See Case C-367/96 A Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I-2843, para [20]; Case C-373/97 D Diamantis v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [2000] ECR I-1705, para [33]. See also Sørensen (n 12 above) 423–24, 427, 450–52. 213   Cf C Barnard, ‘Social Dumping and the Race to the Bottom: Some Lessons for the European Union from Delaware?’ (2000) 25 European Law Review 57, 68. 214  See de la Feria (n 10 above) 433–37; D Triantafyllou, ‘L’ interdiction des abus de droit en tant que principe général du droit communautaire’ [2002] Cahiers de droit européen 611, 618–19, 621, 628; and AG Stix-Hackl in Fidium-Finanz (n 210 above), paras [92–98]. Cf Sørensen (n 12 above) 432–33; Edwards and Farmer (n 207) 208–15, 224–26, contrasting the pre-emptory with the justificatory effect of the application of the principle in Emsland-Stärke (n 203 above), para [56], and Centros (n 23 above), para [18], respectively. 215  R Dworkin, Taking Rights Seriously (London, Duckworth 1977, 2005 impression) 22–27. Cf S DouglasScott, Constitutional Law of the European Union (Oxford, Pearson-Longman 2002) 452–53; TC Hartley, The Foundations of European Union Law, 7th edn (Oxford, Oxford University Press 2010) 141. 216  R Alexy, A Theory of Constitutional Rights (Oxford, Oxford University Press 2002) 47–48; R Alexy, Theorie der Grundrechte, 3rd edn (Frankfurt, Suhrkamp 1996) 75–76. 217  See de la Feria (n 10 above) 437–40; Triantafyllou (n 214 above) 628, 630–32; and J-P Maublanc, ‘Réévaluation de l’ abus de droit à la lumière de la jurisprudence communautaire sur les pratiques abusive’ (2007) 507 Revue du Marché commun et de l’ Union européenne 267, 268–69. 218  See T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press 2006) 4; A Arnull, A Dashwood, M Ross, D Wyatt, E Spaventa and M Dougan (eds), Wyatt and Dashwood’s European Union Law, 5th edn (London, Sweet & Maxwell 2006) 236.

90  Dimitrios Doukas of general principles of EU law in the strict sense, which are also the product of judicial development based on sources lying outside the Union’s legal order, the prohibition of abuse of EU law has the primary effect of depriving private parties of benefits that they could otherwise enjoy under Union rules. It is typically used by the administration to delimit the scope of rights or other provisions of EU law and determine whether restrictions or exceptions can be justified.219 Nonetheless, the mainly interpretative function of the prohibition of abuse of EU law cannot preclude its classification into the distinct category of general principles of EU law, which underlie the fundamental freedoms or other fields of Union law.220

VI. Conclusions

Due to the technical facilities available and the regulatory competition furthered by the differences between the laws of the Member States and the minimum harmonisation at EU level, the field of broadcasting services has provided a fertile ground for practices of forum shopping with a view to choosing the most beneficial jurisdiction. The case law of the Court of Justice concerning allegedly abusive practices on the part of providers of broadcasting services has further developed the Van Binsbergen doctrine of circumvention, and paved the way for the evolution of a general principle of prohibition of abuse of EU law. While leaving the ultimate decision on abuse to the national courts, the Court of Justice has introduced for the first time a twofold test necessary to trigger the application of the principle. This consists of the objective requirement that a broadcaster direct all or most of its activities towards the territory of a Member State other than that in which it is established; and the subjective prerequisite of an intention of abuse of EU law for the purpose of circumventing legitimate national rules. From the outset, the two components have been interconnected so that the motive of abuse can be presumed if the objective requirement—of an activity blatantly incompatible with the objectives of the freedom to provide services—is satisfied. The principle of abuse in the field of broadcasting services has been accorded a dual effect in the case law: first, a pre-emptory one—pre-empting the application of Article 56 TFEU [ex Article 49 EC] or the jurisdiction of a Member State under the TWF Directive; and, second, a justificatory effect, providing a ground for restrictions on free movement. Both applications should be regarded as two sides of the same coin; ie the prevention of economic operators from benefiting from the Treaty or secondary EU law for the attainment of purposes inconsistent with the objectives of the respective provisions. Following the implementation of the TWF Directive, the Court of Justice seems to have restrained the relatively broad discretion originally allowed to Member States to combat abuse by gradually shifting towards a narrower construction of the principle. This should not undermine the effectiveness of either the free movement of broadcasting services or the country of origin principle inherent in the Directive. The basic premise has 219  See A Arnull, ‘What is a General Principle of EU Law?’, ch 2 above. Cf Edwards and Farmer (n 207) 205–06. 220  See Tridimas (n 218 above) 4; A Arnull, The General Principles of EEC Law and the Individual (Leicester, Leicester University Press 1990) 1–2; K Lenaerts and P Van Nuffel, Constitutional Law of the European Union, 2nd edn (London, Sweet & Maxwell 2005) 138, 715; Sørensen (n 12 above) 451–52, 458–59; and AG Poiares Maduro (n 210 above). Against Ottersbach (n 1 above) 186, and Kjellgren (n 48 above) 276–77.

Free Movement of Broadcasting Services and Abuse of Law  91 been that a broadcaster should not be precluded from exercising the freedom to provide services if it does not offer any services in the Member State where it is established. Accordingly, the Court has declared the location of the centre of gravity of a broadcaster’s essential activities in the host Member State as a reliable indicator of abuse. The relevant jurisprudence has led to revisions of the Directive which have integrated some of the essential features of the principle of abuse and are now codified in the AVMS Directive. Despite this codification, there are still ambiguities, both as to the effect and the conditions of the application of the principle, which need to be further elaborated by the Court of Justice. Nevertheless, the core elements of this principle have had a spillover effect on other fundamental freedoms or fields of EU law, including the free movement of goods and the freedom of establishment. They can thus be seen as milestones in the emergence of a general principle of prohibition of abuse of EU law, with a primarily interpretative function.

7 The Notion of Abuse and the Freedom to Provide Services: A Labour Lawyer’s Perspective Catherine Barnard I. Introduction

A

s Dimitrios Doukas’ chapter1 shows, in the early days of the European Union, the Court of Justice was sensitive to the issue of companies using what is now Article 56 TFEU [ex Article 49 EC] to evade obligations of national law.2 Once a Directive was adopted, the Court took a more restrictive approach to the notion of abuse and, where the parties complied with the Directive, their conduct was unlikely to constitute abuse. In my comments I would like to draw parallels with the developments outlined by Dr Doukas in respect of the Audiovisual Media Services (AVMS) Directive3 and the Court’s approach to the Posted Workers’ Directive 96/714 in Laval, Rüffert and Luxembourg.5 I want to suggest that while the Court’s stance on the interpretation of the Posted Workers’ Directive and the Treaties, as well as on the concept of abuse, has much to commend it from a Single Market perspective, its emphasis on ensuring freedom to provide services paves the way for a substantial erosion of national labour laws and industrial relations. To trade unionists, this erosion of hard-won domestic employment rights amounts to abuse, at least in the lay sense of the term. II. A Brief Outline of the Posted Workers’ Directive

The Posted Workers’ Directive 96/71, adopted under the Single Market legal bases  D Doukas, ‘Free Movement of Broadcasting Services and Abuse of Law’, ch 6 above.  Case 33/74 Johannes Henricus Maria van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299 as applied in eg, Case C-148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487; Case C-23/93 TV10 SA v Commissariaat voor de Media [1994] ECR I-4795. 3  Originally Council Directive 89/552/EEC of 3 October 1989 on the coordination of certain provisions laid down by law, regulation or administrative action in Member States concerning the pursuit of television broadcasting activities (Television Without Frontiers (TWF) Directive) [1989] OJ L298/23, now Dir 2010/13/ EU [2010] OJ L95/1. 4  Directive 96/71/EC of the European Parliament and of the Council of 16 December 1996 concerning the posting of workers in the framework of the provision of services (Posted Workers’ Directive) [1997] OJ L18/1. 5   Case C-341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet et al [2007] ECR I-11767; Case C-346/06 Dirk Rüffert v Land Niedersachsen [2008] ECR I-1989; and Case C-319/06 Commission v Luxembourg [2008] ECR I-4323. 1 2

94  Catherine Barnard (Articles 47(2) and 55 EC—now Articles 53(1) and 62 TFEU), was introduced to ensure a ‘climate of fair competition’.6 In essence, it was intended to facilitate the posting of workers by companies to fulfil contracts in other Member States. For our purposes, three provisions of the Directive are particularly significant: Articles 3(1), 3(7) and 3(10). Article 3(1) lays down a ‘nucleus of mandatory rules’7 which the host State must apply to posted workers working in their territory. These include rules on working time, health and safety, equality and, most importantly for our purposes, pay, but only minimum rates of pay.8 However, Article 3(1) does not cover other key aspects of labour law such as protection against unfair dismissal and redundancy. Article 3(7) contains a minimum standards clause.9 This provides: ‘Paragraphs 1 to 6 shall not prevent the application of terms and conditions of employment which are more favourable to workers.’ Finally, Article 3(10), first paragraph allows Member States to apply terms and conditions of employment on matters other than those referred to in Article 3(1) in the case of so-called ‘public policy provisions’. The question for the Member States, and ultimately the Court of Justice, is how to reconcile these provisions with the practice of a number of Member States, including the UK, of applying national labour laws to all those working in their territory. These Member States justify this so-called territorial approach to their labour laws on the grounds of equality, fairness and good industrial relations and they seemed to have assumed that Articles 3(7) and 3(10) ensured that they could continue to do this. As we shall see, the Laval line of case law casts serious doubt on these assumptions. These cases suggest that the Court of Justice sees the application of national labour law by a host State as a barrier to the provision of services under Article 56 TFEU [ex Article 49 EC] and therefore presumptively unlawful. III.  Laval

All three provisions of the Posted Workers’ Directive outlined above were at issue in the ground-breaking case of Laval.10 Laval, a Latvian company, had a wholly-owned subsidiary, L&P Baltic Bygg, a company incorporated under Swedish law. Baltic won a contract to refurbish and extend a school in the Stockholm suburb of Vaxholm; Laval posted thirtyfive of its own Latvian workers to fulfil the contract. These workers earned about forty per cent less per hour than comparable Swedish workers. Although Laval had entered into   5th Recital to the Directive.   13th Recital to the Directive.   Laval (n 5 above), para [70].     9   Confirmed in Recital 17. 10   For a full discussion of this case see eg A Davies, ‘One Step Forward, Two Steps Back? Laval and Viking at the ECJ’ (2008) 37 Industrial Law Journal 126; J Malmberg and T Sigeman, ‘Industrial Action and EU Economic Freedoms: The Autonomous Collective Bargaining Model Curtailed by the European Court of Justice’ (2008) 45 Common Market Law Review 1115; N Reich, ‘Free Movement v Social Rights in an Enlarged Union—The Laval and Viking Cases Before the ECJ’ (2008) 9 German Law Journal 125; P Syrpis and T Novitz, ‘Economic and Social Rights in Conflict: Political and Judicial Approaches to Their Reconciliation’ (2008) 3 European Law Review 411; C Barnard, ‘Viking and Laval: An Introduction’ (2007–8) 10 Cambridge Yearbook of European Legal Studies 463 and the articles in the same volume. For a discussion of the background to the dispute, see C Woolfson and J Sommers, ‘Labour Mobility in Construction: European Implications of the Laval un Partneri Dispute with Swedish Labour’ (2006) 12 European Journal of International Relations 49.     6     7     8

The Notion of Abuse and the Freedom to Provide Services  95 a collective agreement in Latvia with the Latvian building sector’s trade union, Byggnads, the major Swedish construction trade union, wanted Laval to apply the Swedish national agreement. This agreement covered a range of matters, including an obligation to pay a ‘special building supplement’ to an insurance company to finance group life insurance contracts. Pay was, however, left to be negotiated at local level between the local trade union (Byggettan) and the employer on a project-by-project basis.11 After some prevarication, Laval refused to sign the national collective agreement for the building sector (ie the agreement with Byggnads) because ‘it was not possible for it to know in advance what conditions would be imposed on it in relation to wages’.12 This meant that the peace obligation (ie no striking for the duration of the collective agreement), triggered by Swedish law once a collective agreement had been entered into, was not engaged. Byggettan therefore asked Byggnads to institute collective industrial action. This led to a union picket at the school site and a blockade by construction workers preventing the delivery of goods onto the site and prohibiting Latvian workers and vehicles from entering the site.13 Subsequently, the electricians’ unions took sympathy industrial action, preventing Swedish undertakings belonging to the organisation of electricians’ employers from providing services to Laval.14 Other trade unions followed suit and announced sympathy industrial action, boycotting all of Laval’s sites in Sweden.15 Laval brought proceedings in the Swedish Labour Court, claiming that the industrial action and blockade were illegal, as was the sympathy strike, and that the trade unions should stop the action and pay compensation.16 Its case was funded by the Swedish Employers Association. In an interim ruling, the Labour Court said that such action, aimed at pushing through an add-on to the leading collective agreement within the industry, could not be seen as going against good labour market practices.17 Subsequently, the ten million-strong International Federation of Building and Wood Workers launched a solidarity campaign and the trade unions intensified their boycott. Eventually, Baltic went bankrupt because of the blockade and the Latvian workers returned home.18 The Labour Court referred questions on whether Articles 12 and 49 EC [now Articles 18 and 56 TFEU] and the Posted Workers’ Directive precluded trade unions from attempting, by means of collective action, to force a foreign undertaking posting workers to Sweden to apply a Swedish collective agreement. The Court ruled that Article 49 EC [now Article 56 TFEU] could apply to trade unions, that collective action constituted a restriction on free movement and so breached Article 49 EC, and that the collective action could not be justified because the requirements that the trade unions were trying to enforce through collective action were not only lacking in transparency but also, more significantly, as we shall see below, were not compatible with the Posted Workers’ Directive. Therefore, the collective action could not be justified. In respect of the Posted Workers’ Directive (PWD), the Court found that because Sweden had failed to comply with the technical requirements of Articles 3(1) and (8) of the PWD   Laval (n 5 above), paras [24–25].   Laval (n 5 above), para [36]. 13   Laval (n 5 above), para [34]. 14   Laval (n 5 above), para [37]. 15   Laval (n 5 above), para [38]. 16   Laval (n 5 above), para [39]. 17   Woolfson and Sommers (n 10 above) 56. 18  Laval has now ceased its activities in the construction industry, focusing instead on the food industry: ‘EU Court: Swedish Union Breached Latvian Company’s Rights’ International Herald Tribune, 18 December 2007. 11 12

96  Catherine Barnard it could not apply its collective agreements on wages to the Latvian workers,19 a view also confirmed in the German context in Rüffert. Article 3(1) and (8) are complex. Matters listed in Article 3(1) must be laid down by (1) law, regulation, or administrative provision; and/or (2) in respect of activities referred to in the Annex (ie all building work relating to the construction, repair, upkeep, alteration or demolition of buildings), collective agreements or arbitration awards which have been declared universally applicable within the meaning of Article 3(8). If they are not laid down in one of these ways, then they cannot be applied to the posted workers. The trade unions in Laval and in the subsequent German case of Rüffert20 learned this the hard way. Because there is no law on the minimum wage in Sweden (or Germany),21 Laval focused on Article 3(8) dealing with collective agreements. Article 3(8) has two limbs. The first deals with those systems which do have a doctrine of extension (also known as the erga omnes effect) of collective agreements. It explains that collective agreements, or arbitration awards which have been declared universally applicable, are those which must be observed by all undertakings in the geographical area and in the profession or industry concerned. This limb of the definition was at issue in Rüffert.22 Germany has a system for declaring collective agreements universally applicable. However, in the particular situation in Rüffert, the collective agreement setting pay in the building industry had not been declared universally applicable,23 and so the collectively agreed rules on pay rates could not be applied to the posted workers,24 no matter that the German state law on the award of public contracts required contractors and their subcontractors to pay posted workers at least the remuneration prescribed by the collective agreement in force at the place where those services were performed. The second limb of Article 3(8) deals with those systems which do not have a procedure for extending collective agreements to all workers (systems such as those in the UK and Sweden). In these situations, Member States may, if they so decide, base themselves on: collective agreements or arbitration awards which are generally applicable to all similar undertakings in the geographical area and in the profession or industry concerned; and/or collective agreements which have been concluded by the most representative employers’ and labour organisations at national level and which are applied throughout national territory.

The Court required Member States positively to opt for either of these possibilities.25 Because Sweden had not taken advantage of the second paragraph of Article 3(8)26 the Swedish trade unions could not impose on Baltic/Laval enterprise-level collective bargaining.27 19  Sweden wished to apply matters not listed in Article 3(1) to posted workers (eg special insurance supplement). This, the Court also found, contravened the PWD. 20   Rüffert (n 5 above). 21   Laval (n 5 above), para [8]. All other terms and conditions laid down by Art 3(1) of the Posted Workers’ Directive have been implemented by Swedish law: para [63]. 22   Rüffert (n 5 above). 23   Rüffert (n 5 above), para [26]. 24   Rüffert (n 5 above), para [31]. Cf AG Bot in Rüffert (n 5 above). 25   Cf AG Mengozzi in Laval (n 5 above), especially paras [179–81]. He concludes at para [187]: ‘It is therefore beyond doubt, in my view, that the right to take collective action granted by Swedish law to trade unions to enable then to impose the wage conditions laid down or governed by Swedish collective agreements provides a suitable means of attaining the aim of protecting posted workers laid down in Article 3 of Directive 96/71.’ 26   Laval (n 5 above), para [67]. 27   Laval (n 5 above), para [71].

The Notion of Abuse and the Freedom to Provide Services  97 Why did the Court take such a strict line in interpreting the provisions of the Directive? One explanation is that Article 3(1) is a ‘derogation’28 from the principle of home State control laid down in Article 6(2)(a) of the Rome Convention (which says that it is the law of the home State which governs the employment relationship of temporary workers)29 and derogations must be narrowly construed. Another explanation is that the Posted Workers’ Directive is not primarily a worker protection measure but a measure to facilitate freedom to provide services, as its legal basis (Articles 47(2) and 55 EC—now Articles 53(1) and 62 TFEU) demonstrates. According to the Court, it should therefore be interpreted in a way that facilitates service provision which means that any potential barriers to freedom to provide services, particularly employment laws of the host State, should be narrowly construed. Therefore, if the detailed terms of the Directive are not complied with, in particular Article 3(1), there will be a breach of Article 56 TFEU [ex Article 49 EC]. This point is made explicitly by Advocate-General Mengozzi in Laval: ‘A measure that is incompatible with Directive 96/71 will, a fortiori, be contrary to Article 49 EC, because that Directive is intended, within its specific scope, to implement the terms of that Article.’30 The Posted Workers’ Directive thus plays a key role in interpreting Article 49 EC: if the detailed terms of the Directive are not complied with to the letter there will be a breach of this Article. The interdependency between the Directive and Article 49 EC is confirmed in paragraph [36] of Rüffert where the Court said that its interpretation of Directive 96/71 was ‘confirmed by reading it in the light of Article 49 EC, since that Directive seeks in particular to bring about the freedom to provide services, which is one of the fundamental freedoms guaranteed by the Treaty’. So the Directive and Article 56 TFEU [ex Article 49 EC] are mutually reinforcing: the restrictive interpretation of the Directive is derived from this Article and the substance of this Article is derived from the Directive.31 With the Article 3(1) route closed, the Swedish trade unions turned to Article 3(7) and argued that since the Directive laid down only minimum standards they were entitled to require posting employers to respect the higher standards of the host State. However, the Court rejected this argument. It held that under Article 3(7) the host State could not impose terms and conditions of employment which went ‘beyond the mandatory rules [in Article 3(1)] for minimum protection’.32 Rather, the ECJ continued, Article 3(7) applied: (1) to the situation of out-of-state service providers voluntarily signing a collective agreement in the host State which offered superior terms and

  Per AG Mengozzi in Laval (n 5 above), para [132].  This point is made more explicit by Article 8(2) of the Regulation (EC) No 593/2008 of the European Parliament and of the Council of 17 June 2008 on the law applicable to contractual obligations (Rome 1) [2008] OJ L177/6: To the extent that the law applicable to the individual employment contract has not been chosen by the parties, the contract shall be governed by the law of the country in which or, failing that, from which the employee habitually carries out his work in performance of the contract. The country where the work is habitually carried out shall not be deemed to have changed if he is temporarily employed in another country. (Emphasis added to show the major changes from the Rome Convention.) 30   Laval (n 5 above), para [149]. 31  See by analogy, the Court’s interpretation of the Equal Pay Directive 75/117 and its relationship with Article 141 EC [now Article 157 TFEU]. In Case 96/80 Jenkins v Kingsgate [1981] ECR 911, the Court said that Art 1 ‘is principally designed to facilitate the practical application of the principle of equal pay outlined in Article 157 TFEU [and] in no way alters the content or scope of that principle as defined in the Treaty’. 32   Laval (n 5 above), para [80]; Rüffert (n 5 above), para [33]. 28 29

98  Catherine Barnard conditions to their employees; and (2) to the situation where the home State laws or collective agreements were more favourable and these could be applied to the posted workers. So, under this interpretation, British rules on, for example, the minimum wage could legitimately be extended to British workers posted to Latvia. Likewise, a German worker temporarily posted to Portugal would continue to enjoy German terms and conditions. (It was this scenario that prompted the German government to support, in early drafting, what is now the Court’s understanding of Article 3(7) of the Posted Workers’ Directive (PWD) 96/71, namely that the home State can apply higher standards to its posted workers, thereby protecting German workers working in Portugal.) However, the Court’s interpretation meant that Sweden could not rely on Article 3(7) to extend all of its collectively agreed rules to Latvian workers posted to Sweden. The final route available for the Swedish trade unions was to rely on Article 3(10). However, in Laval the Court said that if a State wished to take advantage of a derogation it had to do so expressly.33 In the absence of a State act, Article 3(10) was closed to trade unions. Subsequently, in Commission v Luxembourg the Court closed the door still more firmly on Article 3(10). It held that the public policy exception, as a derogation from the fundamental principle of freedom to provide services, had to be interpreted strictly.34 In support, it cited Declaration No 10 on Article 3(10) recorded in the Council minutes which provides that the ‘public policy provisions’ should be construed as covering ‘those mandatory rules from which there can be no derogation and which, by their nature and objective, meet the imperative requirements of the public interest. These may include, in particular, the prohibition of forced labour or the involvement of public authorities in monitoring compliance with legislation on working conditions.’35 Later in the judgment the Court tightened up the criteria still further: ‘[W]hile the Member States are still, in principle, free to determine the requirements of public policy in the light of national needs, the notion of public policy in the Community context . . . may be relied on only if there is a genuine and sufficiently serious threat to a fundamental interest of society.’36 For good measure, the Court added ‘that the reasons which may be invoked by a Member State in order to justify a derogation from the principle of freedom to provide services must be accompanied by appropriate evidence or by an analysis of the expediency and proportionality of the restrictive measure adopted by that State, and precise evidence enabling its arguments to be substantiated.’37 The test for public policy set out in Commission v Luxembourg replicates the wellknown test for public policy developed by the Court in Bouchereau38 in the context of free movement of persons (and now enshrined in Article 27 of the Citizens Rights Directive 2004/38), to consider whether a Member State is justified in deporting a migrant. It is extremely surprising that the Court has transplanted this test to the very different context of labour law. With the exceptions of laws against slavery (referred to in Declaration No 10), it is difficult to see how any labour laws, however fundamental to the system of employment protection, satisfy the extraordinarily high standard laid down in   Laval (n 5 above), para [84].   Commission v Luxembourg (n 5 above), paras [30] and [49].   Commission v Luxembourg (n 5 above), para [33]. 36   Commission v Luxembourg (n 5 above), para [50]. 37   Commission v Luxembourg (n 5 above), para [51]. 38   Case 30/77 R v Pierre Bouchereau [1977] ECR 1999. 33 34 35

The Notion of Abuse and the Freedom to Provide Services  99 Luxembourg. The Court’s interpretation of public policy therefore risks interpreting Article 3(10) PWD almost out of existence. With the Bouchereau test as its yardstick, it is not at all surprising that the Court then found that specific aspects of Luxembourg labour law that were applied to posted workers could not be justified under Article 3(10). For example, Luxembourg law, like UK law, required workers to be provided with a written contract or document established pursuant to Directive 91/533 on the written statement, a matter not covered by Article 3(1) PWD. The Court pointed out that since employers were required by the law of the home State to provide a written statement, the Luxembourg law was likely to dissuade undertakings established in another Member State from exercising their freedom to provide services. The Court therefore concluded that the Luxembourg rule did not ‘comply with the first indent of Article 3(10) of Directive 96/71, in so far as it is not applied in compliance with the EC Treaty’.39 Luxembourg also applied its own laws implementing the Part-Time Work Directive 97/81 and the Fixed-Term Work Directive 99/70 to posted workers. Once again, the Court noted that the requirement ‘concerns a matter which is not mentioned in the list in the first subparagraph of Article 3(1)’.40 This is a surprising observation since Article 3(1)(g) talks of ‘equality of treatment between men and women and other provisions on nondiscrimination’. Both Directives 97/81 and 99/70 have the principle of non-discrimination at their core. Nevertheless, the Court continued that the application of such rules was likely to hinder the exercise of freedom to provide services by undertakings wishing to post workers to Luxembourg,41 and that since the host law already ensures the protection of these rights, Luxembourg again could not rely on the public policy exception.42 IV. The Notion of Abuse

A.  The Single Market Perspective The analysis of the cases so far suggests that a territorial approach to labour law is no longer possible. For Single Market lawyers, this makes sense. It has long been established that, under what is now Article 56 TFEU [ex Article 49 EC], providers of services should not be subject to compliance with all the conditions required for establishment, since this ‘deprives of all practical effectiveness’ the Treaty provisions on services.43 As with the later broadcasting/advertising cases discussed by Dr Doukas, the Court has read the Posted Workers’ Directive and Article 49 EC so as to facilitate free movement of services and to endorse the principle of home State control. Labour lawyers were wrong to think of the Posted Workers’ Directive as a worker protection measure; it is predominantly a Single Market measure. This view is reinforced by the Court’s observations in Laval that the Directive is ‘firstly’ intended to ‘ensure a climate of fair competition between national undertakings and undertakings which provide services transnationally’,44 and   Commission v Luxembourg (n 5 above), para [44].   Commission v Luxembourg (n 5 above), para [57].   Commission v Luxembourg (n 5 above), para [58]. 42   Commission v Luxembourg (n 5 above), para [60]. 43   Case C-164/99 Portugaia Construções Lda [2002] ECR I-787. 44   Laval (n 5 above), para [74]. 39 40 41

100  Catherine Barnard that the mandatory rules for minimum employment protection prevent a situation from arising where the out-of-state provider competes unfairly against the host State.45 Only ‘secondly’ does the Court refer to the worker protection element of the Directive.46 It could therefore be argued that the Court has reached a careful compromise in Laval and Rüffert: posted workers will enjoy the terms and conditions of employment in the host State but only if the host State has complied with the letter of the Directive; if it has not, then any attempt to apply the host State rules will breach both the Directive and Article 49 EC [now Article 56 TFEU]. This solution could be seen as a particularly good compromise for the new Member States. One of the principal reasons why the A-8 countries surrendered at least part of their recently acquired sovereignty on accession to the EU was to gain access to the markets of the EU-15. Laval and the other cases make it clear that it does not amount to abuse of European law to make use of free movement of services to exploit the comparative advantage of cheaper labour.47 Many would say hurrah to that. There were signs in Laval that the trade unions did try to run arguments based on abuse, albeit that the issue was couched in terms of jurisdiction of the Court of Justice to hear the case rather than points of substance. They argued that there was no link between the questions referred and the facts of the case in the main proceedings, pointing out that the national court had asked the ECJ to interpret provisions relating to freedom to provide services and Directive 96/71, when in fact Laval was established in Sweden, under Article 43 EC [now Article 49 TFEU], through its subsidiary, Baltic, in which it held 100 per cent of the share capital until the end of 2003. The unions argued that since the share capital of Laval and of Baltic was held by the same persons, and those companies had the same representatives and used the same trademark, they should be regarded as one and the same economic entity from the point of view of European law, even though they constituted two separate legal persons. Therefore, the unions argued, Laval was under an obligation under Article 43(2) EC [now Article 49(2) TFEU] to pursue its activity in Sweden under the same conditions as Sweden lays down for its own nationals.48 The trade unions also argued that the purpose of the dispute was to enable Laval to circumvent Swedish law and, for that reason, the dispute, at least in part, was artificial. They said that: Laval, whose activity consists of placing, on a temporary basis, staff of Latvian origin with companies which carry on their activities on the Swedish market, is seeking to escape all the obligations under Swedish legislation and rules relating to collective agreements and, by relying on the provisions of the Treaty on services and on Directive 96/71, is making an improper attempt to take advantage of the possibilities offered by Community law.49

The Court rapidly disposed of these potentially serious issues. As is usual, it noted that ‘in proceedings under Article 234 EC [now Article 267 TFEU], which are based on a clear separation of functions between the national courts and the Court of Justice, any   Laval (n 5 above), para [75].   Laval (n 5 above), para [76]. Cf AG Mengozzi who reverses the priority (para [171]): ‘Article 3 of Directive 96/71 has a twofold aim of providing minimum protection for posted workers and equal treatment as between service providers and domestic undertakings in similar circumstances. Those two requirements must be pursued concurrently.’ 47   For full discussion on the abuse case law, see R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. 48   Laval (n 5 above), para [43]. 49   Laval (n 5 above), para [44]. 45 46

The Notion of Abuse and the Freedom to Provide Services  101 assessment of the facts in the case is a matter for the national court’. Likewise, it was for the national court to determine in the light of the particular circumstances of the case both the need for a preliminary ruling, in order to enable it to deliver judgment, and the relevance of the questions which it submits to the Court. The Court concluded that, where the questions submitted concern the interpretation of European law, the Court was in principle bound to give a ruling.50 Exceptionally, the Court noted, it would refuse to answer a question referred for a preliminary ruling by a national court: where it is quite obvious that the interpretation of Community law that is sought bears no relation to the facts of the main action or its purpose, where the problem is hypothetical, or where the Court does not have before it the factual or legal material necessary to give a useful answer to the questions submitted to it.

This was not the case in Laval. In Laval, the national court sought an interpretation of Articles 12 EC and 49 EC [now Articles 18 and 56 TFEU], and Directive 96/71: [I]n the context of the dispute between Laval and Byggnads, Byggettan and Elektrikerna concerning collective action taken by the latter following Laval’s refusal to sign the collective agreement for the building sector, that the dispute concerns the terms and conditions of employment applicable to Latvian workers posted by Laval to a building site in Sweden, the work being carried out by an undertaking belonging to the Laval group, and that, following collective action and suspension of the work, the posted workers returned to Latvia.

In this way the Court ducked the single economic entity issue. The ECJ continued: ‘Because the questions referred do have a bearing on the subject-matter of the case in the main proceedings, as described by the national court, and that the factual context in which the questions put to it are set does not support the view that the dispute in question is artificial.’ Thus, the arguments about abuse were rejected at the jurisdictional stage. It is not surprising the Court did not engage with the issue of abuse at the substantive stage, given it had already rejected similar arguments based on abuse in the more dramatic case of Centros.51 In Centros the Danish husband and wife team deliberately searched across Europe for the most favourable legal regime in which to incorporate their company. They settled on the UK. When they tried to set up a branch in Denmark, the Danish registrar of companies refused to register the branch on the grounds that they were deliberately trying to circumvent Danish law. The Court held:52 [T]he fact that a national of a Member State who wishes to set up company chooses to form it in the Member State whose rules of company law seem to him the least restrictive and to set up branches in other Member States cannot, in itself, constitute an abuse of the right of establishment. The right to form a company in accordance with the law of a Member State and to set up branches in other Member States is inherent in the exercise, in a single market, of the freedom of establishment guaranteed by the Treaty.53   Laval (n 5 above), para [45].   Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459. 52   Centros (n 51 above), para [27]. 53   Centros (n 51 above), para [27]. At para [29] the Court added: [T]he fact that a company does not conduct any business in the Member State in which it has its registered office and pursues its activities only in the Member State where its branch is established is not sufficient to prove the existence of abuse or fraudulent conduct which would entitle the latter Member State to deny that company the benefit of the provisions of Community law relating to the right of establishment. 50 51

102  Catherine Barnard Apart from the trade union’s arguments about single economic entity, there was no suggestion by the Court that Laval was using the services provisions of the Treaty deliberately to circumvent the rules that would have applied had it been established in Sweden. Laval was a Latvian company and was operating out of Latvia. The fact that Latvia offered a more favourable labour law regime was Laval’s comparative advantage. Taking advantage of this was, using the Centros language, ‘inherent in the exercise, in a single market’, of the freedom to provide services guaranteed by the EC Treaty. In other words, Laval was at most engaging in ‘legitimate circumvention’ of the Swedish rules on employment regulation and certainly not abuse. Moving beyond the facts of Laval, is it possible to imagine a situation where legitimate circumvention becomes abuse? Trade unions are aware of examples of companies saying to a British contractor: ‘hire us because we are not bound by all of British law’. How does this relate to the Court’s approach to abuse? What about the situation where a Latvian company is set up and posts workers to the UK for, say, twelve months (no one is clear how long a posted worker can be posted for), then sends those workers back home for a couple of months before posting them to the UK again? What happens when the company fails and, phoenix-like, emerges from the ashes as a new company and sends the same workers? Are any of these situations ‘abusive’? The most comprehensive, modern test for abuse can be found in the well-known case of Emsland-Stärke: A finding of an abuse requires, first, a combination of objective circumstances in which, despite formal observance of the conditions laid down by the Community rules, the purpose of those rules has not been achieved. It requires, second, a subjective element consisting in the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it. The existence of that subjective element can be established, inter alia, by evidence of collusion between the Community exporter receiving the refunds and the importer of the goods in the non-member country. 54

The Court concluded (at para [54]) that it is for ‘the national court to establish the existence of those two elements, evidence of which must be adduced in accordance with the rules of national law, provided that the effectiveness of Community law is not thereby undermined’. It seems unlikely that the first situation outlined above (‘hire us, we’re out-of-state and so cheaper’) satisfies the Emsland-Stärke test. In essence, the out-of-state company is 54   Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, paras [52– 54]. Cf the slight change in language in Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, paras [74–76]: 74. It would appear that, in the sphere of VAT, an abusive practice can be found to exist only if, first, the transactions concerned, notwithstanding formal application of the conditions laid down by the relevant provisions of the Sixth Directive and the national legislation transposing it, result in the accrual of a tax advantage the grant of which would be contrary to the purpose of those provisions. 75. Second, it must also be apparent from a number of objective factors that the essential aim of the transactions concerned is to obtain a tax advantage. As the Advocate General observed in point 89 of his Opinion, the prohibition of abuse is not relevant where the economic activity carried out may have some explanation other than the mere attainment of tax advantages. 76. It is for the national court to verify in accordance with the rules of evidence of national law, provided that the effectiveness of Union law is not undermined, whether action constituting such an abusive practice has taken place in the case before it.

The Notion of Abuse and the Freedom to Provide Services  103 merely taking advantage of its competitive position. The other situations come closer to abuse under Emsland-Stärke and this may be a further line of argument for trade unions in the future. B.  The Labour Law Perspective So far, we have focused on an interpretation of Laval from a Single Market perspective. This views it as a ‘good thing’ for different systems of labour law regulation to be put into competition through the operation of the free movement provisions. Some of the effects of this regulatory competition have been mitigated by the application of the Posted Workers’ Directive but, as we have seen, the Court has insisted on a strict construction of the Directive. In extreme cases, the application of the free movement provisions might be mitigated by the doctrine of abuse but, as already suggested, this will occur only in exceptional circumstances. I now want to consider the situation not from the perspective of a Single Market lawyer but of a domestic labour lawyer. This might give a rather different take on what freedom to provide services means to individuals actually working on the ground. Consider the following situation. At the 2012 Olympics site in London, some of the builders are British workers working directly for a British contractor or subcontractor on the site. Assuming they fall within the personal and temporal scope of British employment law (eg they are employees with the relevant period of service) they will benefit from the full application of British labour law. Working alongside them are Bulgarian builders employed by a Bulgarian company which has won a contract to provide certain services on the Olympic site. Following the Laval line of case law, the Bulgarian workers will enjoy only those British employment laws listed in Article 3(1) of the Posted Workers’ Directive (albeit that at present, the UK insists on applying all of its employment laws to anyone on its territory, probably in contravention of European law55). The rest of the Bulgarian workers’ rights are protected by Bulgarian law. Is it conducive to good industrial relations (and equality of treatment) to have two workers working side by side on the Olympics site, employed on significantly different terms and conditions? While this might not amount to abuse under the Emsland-Stärke definition, it sits uncomfortably with the Memorandum of Agreement between the Olympic Delivery Authority (ODA) and the Trades Union Congress on Principles of Cooperation56 which requires the ODA to: Ensure a safe working environment, with effective industrial relations and fair terms and conditions of employment. Make fair employment standards a key consideration in the procurement process.

What happens when the Bulgarian posted worker joins a British trade union in order to seek the union’s help to raise his employment standards in line with the Memorandum of Agreement? Would it then be an abuse for the employer to raise Article 56 TFEU [ex Article 49 EC], read in conjunction with the PWD, to obstruct the British trade union from assisting the Bulgarian workers to raise their employment standards above those 55   C Barnard, ‘The UK and Posted Workers: The Effect of Commission v Luxembourg on the Territorial Application of British Labour Law’ (2009) 38 Industrial Law Journal 122. 56  See: www.london2012.com/news/media-releases/2008-09/london-2012-and-uk-unions-agree-to-worktogether-for-inspirational-games.php.

104  Catherine Barnard laid down by Article 3(1), possibly by taking strike action? Laval would suggest that an employer could raise Article 56 TFEU and the PWD. But if the trade union cannot call its members out on strike to enforce their rights, where does that leave the ‘fundamental right’ to take collective action,57 traditionally an area left to the Member States to regulate, as the 27th Recital of the Rome II Regulation 864/2007 points out: The exact concept of industrial action, such as strike action or lock-out, varies from one Member State to another and is governed by each Member State’s internal rules. Therefore, this Regulation assumes as a general principle that the law of the country where the industrial action was taken should apply, with the aim of protecting the rights and obligations of workers and employers.58

To British trade unionists it feels like an abuse, at least in the lay sense of the term, that European law may well prevent them from enforcing employment rights for their members. V. Conclusions

Dimitrios Doukas’ discussion of the evolving case law on abuse in the broadcasting sector shows just how far the Court has come: from recognising the doctrine of abuse in the early case law, in situations which today might not be considered abusive, to a more nuanced use of the abuse doctrine. Some of the attempts to prevent abuse have been incorporated into revised text of the AVMS Directive. In the context of posted workers, Laval shows us how much the Court has learned from case law in related areas, in particular broadcasting. The Laval case law, too, retains a strong endorsement of the country of origin principle and an implicit statement that it is not abusive to take advantage of the country of origin principle when providing services. Despite calls from the trade union movement to amend the Posted Workers Directive59 to address some of the difficulties that the Scandinavian trade unions are having with the judgment,60 there is no appetite in the Commission for proposing any changes. As a result, the trade union movement is left with facing the reality that despite all the talk about the need to balance the economic with the social dimension of the EU,61 in cases of conflict, the economic freedoms will generally prevail and that arguments based on abuse of European law will usually not assist them.

57   Case C-438/05 International Transport Workers’ Federation, Finnish Seamen’s Union v Viking Line ABP, OÜ Viking Line Eesti [2007] ECR I-10779, para [44]: Although the right to take collective action, including the right to strike, must therefore be recognised as a fundamental right which forms an integral part of the general principles of Union law the observance of which the Court ensures, the exercise of that right may none the less be subject to certain restrictions. As is reaffirmed by Article 28 of the Charter of Fundamental Rights of the European Union, those rights are to be protected in accordance with Union law and national law and practices. 58  See also the preambular paragraph in Directive 96/71 that ‘this Directive is without prejudice to the law of the Member States concerning collective action to defend the interests of trades and professions’. 59  See: www.etuc.org/IMG/pdf_ETUC_Viking_Laval_-_resolution_070308.pdf. 60  See the Swedish government’s response, ‘The Laval Enquiry: Action in Response to the Laval Judgment’: www.sweden.gov.se/sb/d/8311/a/117722. 61  In both Viking and Laval, the Court began by recognising the need to reconcile the competing objectives of the Union: on the one hand the completion of the Internal Market, and on the other ‘a policy in the social sphere’ (Viking (n 57 above), para [78]; Laval (n 5 above), para [104]).

8 Sparking Regulatory Competition in European Company Law: The Impact of the Centros Line of Case Law and its Concept of ‘Abuse of Law’ Wolf-Georg Ringe

I. Introduction

T

he concept of ‘abuse of law’ has attained a certain prominence at European level not only in tax law, the discipline in which it was originally conceived, but also in the case of company law.1 The first remarkable ‘abuse’ case, Daily Mail,2 was in reality a tax case, but was discussed by company lawyers and interpreted as a company law case.3 It was, however, not until the Centros case4 and from 1999 onwards, that the debate on abuse and misuse of fundamental freedoms in company law created a sudden awareness of the possibilities that the Internal Market offered, provoking a storm of academic

1   V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205. See on the concept of abuse in a European and comparative perspective generally, D Anderson, ‘Abuse of Rights’ [2006] Judicial Review 348; KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423; R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395; H Fleischer, ‘Der Rechtsmißbrauch zwischen Gemeineuropäischem Privatrecht und Gemeinschaftsprivatrecht’ [2003] Juristenzeitung 865; A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of the European Court of Justice on Circumvention, Fraud and Other Misuses of Community Law’ (2000) 11 European Business Law Review 179; P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351. 2   Case 81/87 R v HM Treasury and Commissioners of Inland Revenue, ex p Daily Mail and General Trust plc [1988] ECR 5483. 3   CM Schmitthoff, ‘Daily Mail Loses in the European Court’ [1988] Journal of Business Law 454; S Frommel, ‘EEC Companies and Migration: A Setback for Europe’ [1988] Intertax 409; L Cerioni, ‘The Barriers to the International Mobility of Companies within the European Community: A Re-reading of the Case Law’ [1999] Journal of Business Law 59; A Gill, ‘Migration of Companies and the Right of Establishment in EC Law’ [1989] Irish Law Times and Solicitors’ Journal 59. 4   Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459.

108  Wolf-Georg Ringe literature5 and corresponding business behaviour that quite quickly adapted to the situation.6 The following chapter presents the concept of ‘abuse’ in relation to the freedom of establishment for companies7—as developed in Centros and subsequent cases—and analyses the consequences of the Court’s liberal approach in this field. Allowing the possibility to choose the state of incorporation for setting up a company, empirical research suggests that this case law has led to regulatory competition between Member States, following a substantial business migration to more flexible company law systems within the European Union. The peculiarity with this result is that the Court declared a course of conduct, which had previously been considered to be abusive, as being covered by the freedom of establishment. This chapter does not consider abuse of law in respect of secondary legislation (ie directives and regulations), which became relevant in company law arising in several Greek cases in the mid-1990s.8 Instead, it deals with abuse of law in relation to the fundamental freedoms themselves.9 The remainder of this chapter is structured as follows: Part II illustrates the concept of ‘abuse of law’ in the Court’s company law cases since Centros. It is then shown that the liberal attitude towards abuse has led to increased company mobility in the European Union (Part III), before examining how this exerts pressure on individual Member States to adapt their company law system to meet business needs (Part IV). Finally, a prospective assessment of future challenges is given—questions remain as to the relevance of any comparison with the United States and the future developments for corporate reincorporation (Part V). II.  Centros and abuse of law in company law

A.  Case Law on Company Law Four major cases illustrate the Court’s attitude toward allegedly ‘abusive’ behaviour in company law. In the Centros case in 1999, the Court was called upon to decide on the behaviour of a Danish couple which had set up a company in the United Kingdom only to subsequently form a branch in Denmark in order to circumvent the strict Danish capital requirements 5  See for example WH Roth, ‘From Centros to Ueberseering: Free Movement of Companies, Private International Law, and Community Law’ (2003) 52 International and Comparative Law Quarterly 177; J Rickford, ‘Current Developments in European Law on the Restructuring of Companies: An Introduction’ (2004) 15 European Business Law Review 1225; J Armour, ‘Who Should Make Corporate Law? EC Legislation Versus Regulatory Competition’ (2005) 58 Current Legal Problems 369; WG Ringe, ‘No Freedom of Emigration for Companies?’ (2005) 16 European Business Law Review 621; W Schön, ‘The Mobility of Companies in Europe and the Organizational Freedom of Company Founders’ [2006] European Company and Financial Law Review 122; M Siems, ‘SEVIC: Beyond Cross-Border Mergers’ (2007) 8 European Business Organization Law Review 307. 6   See Part III below. 7   Arts 49 and 54 TFEU [ex Arts 43 and 48 EC]. 8   Case C-441/93 Panagis Pafitis et al v Trapeza Kentrikis Ellados AE et al [1996] ECR I-1347; Case C-367/96 Kefalas et al v Greece [1998] ECR I-2843; Case C-373/97 Dionysios Diamantis v Greece and Organismos Ikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [2000] ECR I-1705. 9  On the distinction between abuse in relation to primary law as opposed to secondary law, see Sørensen (n 1 above) 435; P Léger et al, ‘Libertés communautaires et abus de droit’ [2007] Droit Fiscal 978.

Sparking Regulatory Competition in European Company Law  109 for companies.10 Danish authorities had refused to recognise the foreign company. Advocate-General La Pergola pointed out that the concept of abuse of law is one of the fundamental principles of EU law, thereby relying on earlier case law of the Court.11 The Advocate-General quoted the famous formula of the French scholar Planiol, that ‘law ceases where abuse begins’ (le droit cesse où l’abus commence) and stated that this tenet still holds true. The Court agreed with this starting point, affirming that abusive or fraudulent exercise of Community law would not be permissible.12 However, the Court went on to say that ‘although, in such circumstances, the national courts may, case by case, take account—on the basis of objective evidence—of abuse or fraudulent conduct . . . they must nevertheless assess such conduct in the light of the objectives pursued by those provisions’.13 In the case at stake, the Court found that the provisions of Danish law which the parties sought to avoid concerned the formation of companies,14 an activity with which the Treaty provisions of freedom of establishment are particularly concerned.15 Finally, the Court concluded with its famous assertion that it follows from these arguments that where someone chooses to set up a company in the Member State the company law of which seems to him the least restrictive and then sets up branches in other Member States, this ‘cannot, in itself, constitute an abuse of the right of establishment . . . but is inherent in the exercise, in a single market, of the freedom of establishment guaranteed by the Treaty’.16 The Centros ruling turned out to be a landmark decision on the interpretation of the freedom of establishment and produced substantial reverberations in legal doctrine.17 This is hardly surprising, since a course of conduct that comes close to abuse—and was widely believed to be a circumvention of national law—was not only declared legal,

10   A Arnull, A Dashwood, M Ross, D Wyatt, E Spaventa and M Dougan, Wyatt & Dashwood’s European Union Law, 5th edn (London, Sweet & Maxwell 2006) para [19-031]. 11  Inter alia, AG La Pergola cited Kefalas (n 8 above), para [20], and Case C-19/92 Kraus v Land BadenWürttemberg [1993] ECR I-1663, para [34]. 12   Centros (n 4 above), para [24], citing Case 33/74 Van Binsbergen v Bedrijfsvereniging Metaalnijverheid [1974] ECR 1299, para [13]; Case C-148/91 Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487, para [12]; and Case C-23/93 TV 10 v Commissariaat voor de Media [1994] ECR I-4795, para [21]; regarding freedom of establishment, Case 115/78 Knoors [1979] ECR 399, para [25], and Case C-61/89 Bouchoucha [1990] ECR I-3551, para [14]; regarding the free movement of goods, Case 229/83 Leclerc et al v Au Blé Vert et al [1985] ECR 1, para [27]; regarding social security, Case C-206/94 Brennet v Paletta [1996] ECR I-2357, para [24]; regarding freedom of movement for workers, Case 39/86 Lair v Universität Hannover [1988] ECR 3161, para [43]; regarding the common agricultural policy, Case C-8/92 General Milk Products v Hauptzollamt Hamburg-Jonas [1993] ECR I-779, para [21]; and regarding company law, Kefalas (n 8 above), para [20]. 13   Centros (n 4 above), para [25]. 14  The Court distinguished this activity from ‘the carrying on of certain trades, professions or businesses’, for which arguably a different outcome might have been possible. 15   Centros (n 4 above), para [26]. 16   Centros (n 4 above), para [27]. 17   See amongst others the references given (n 5 above) and E Wymeersch, ‘Centros: A Landmark Decision in European Company Law’ in T Baums, KJ Hopt and N Horn (eds), Corporations, Capital Markets and Business in the Law: Liber amicorum Richard M Buxbaum (Deventer, Kluwer 2000) 629; F Munari and P Terrile, ‘The Centros Case and the Rise of an EC Market for Corporate Law’ in G Ferrarini, KJ Hopt and E Wymeersch (eds), Capital Markets in the Age of the Euro: Cross-Border Transactions, Listed Companies and Regulation (The Hague, Kluwer Law International 2002) 529; E Werlauff, ‘Using a Foreign Company for Domestic Activities’ (1999) 10 European Business Law Review 306; WH Roth, ‘Case Note on Centros’ (2000) 37 Common Market Law Review 147; H Xanthaki, ‘Centros: Is This Really the End for the Theory of the Siege Reel?’ (2001) 22 Company Lawyer 2.

110  Wolf-Georg Ringe but even acknowledged as the very essence of the freedom of establishment.18 From this moment on, it seemed very tempting for businesses to set up a company in the Member State with the most flexible company law system—a development which we will deal with later.19 Of course, things did not develop as smoothly in the beginning as there was much legal uncertainty as to the exact borderlines between permissible and abusive behaviour. This was largely due to the question as to whether the so-called ‘real seat’ theory was in any way impinged upon by this judgment. The real seat theory is a conflict-of-laws rule, determining the law applicable to companies (the lex societatis) according to the company’s head office or ‘real seat’, which was subscribed to by a number of EU Member States.20 Without going into the details here,21 this theory was applied in a way detrimental to corporate mobility and made it de facto impossible for foreign companies to register in a Member State, but do business exclusively in another Member State which followed the real seat theory. Requests for clarification were soon made in the form of references to the Court of Justice under Article 234 EC Treaty (now Article 267 TFEU), one of which was the reference in Überseering. In the Überseering case, the German Federal Supreme Court (Bundesgerichtshof) sought a ruling as to whether the German real seat theory was compatible with the freedom of establishment. The ECJ declared that the German case law pursuant to which a company that had transferred its ‘real seat’ or head office into Germany could not have legal personality and therefore did not have standing in court, was incompatible with Articles 43 and 48 EC Treaty [now Articles 49 and 54 TFEU].22 Moreover, it ruled that the host Member State has to accept the legal capacity and the capacity to be a party to legal proceedings, which the foreign company enjoys under the law of its home Member State. Concerning the question of justification, the Court stated that overriding requirements relating to the general interest might, in certain circumstances and subject to certain conditions, justify restrictions on freedom of establishment; however, this could not justify denying the legal capacity and the capacity to be a party to legal proceedings of a company in another Member State.23 In its third important judgment, Inspire Art, the Court had to rule upon the Dutch law on foreign companies (Wet op de Formeel Buitenlandse Vennootschappen, ‘Law on Formally Foreign Companies’). According to this act, companies registered abroad, but doing business only in the Netherlands had to add a suffix to their name, indicating their status as a pseudo foreign company.24 Quite significant consequences followed from this status, such as the application of Dutch minimum capital rules and the imposition of personal liability on the directors for sufficient minimum capital (ie an amount equal to the difference between the Dutch minimum capital mandated amount and the capital 18  P Craig and G de Búrca, EU Law: Text, Cases, and Materials, 4th edn (Oxford, Oxford University Press 2008) 809. 19   See Part III below. 20  RR Drury, ‘Migrating Companies’ (1999) 24 European Law Review 354, 356; J Lowry, ‘Eliminating Obstacles to Freedom of Establishment: The Competitive Edge of UK Company Law’ (2004) 63 Cambridge Law Journal 331, 332. 21   See on this Ringe (n 5 above) 626. 22   Case C-208/00 Überseering BV v Nordic Construction Company Baumanagement GmbH [2002] ECR I-9919. 23   Überseering (n 22 above), para [93]. 24   See on this act RR Drury, ‘The “Delaware Syndrome”: European Fears and Reactions’ [2005] Journal of Business Law 709, 720.

Sparking Regulatory Competition in European Company Law  111 actually paid up to the company). The Court followed its previous reasoning, stating that Articles 43 and 48 of the EC Treaty prevent national legislation from imposing certain conditions on companies formed in accordance with the law of another Member State for the exercise of secondary establishment, for example in respect of company formation relating to minimum capital, or directors’ liability.25 The Court repeated its Centros formula that: [I]t is immaterial . . . that the company was formed in one Member State only for the purpose of establishing itself in a second Member State, where its main, or indeed entire, business is to be conducted. The reasons for which a company chooses to be formed in a particular Member State are, save in the case of fraud, irrelevant with regard to application of the rules on freedom of establishment.26

Nevertheless, the Court maintained a possible scope for justification of measures by the host Member State in order to protect certain interests. However, this reserved area seems to be rather narrow and cannot be fixed in general terms but can only be stated on a caseby-case basis. The fourth and latest judgment in this series was delivered in 2008.27 The hope that this case would remove outstanding doubts as to the question of emigration rather than immigration of companies28 has not been fulfilled. In Cartesio, a Hungarian court referred the question to the ECJ as to whether a Hungarian rule forbidding the relocation of a company’s head office from Hungary to another Member State infringes the freedom of establishment.29 Ultimately, the Court favoured the right of the Member State of origin to veto the trans-border migration, however thereby leaving a back door open.30 More importantly for the issue under consideration here, Advocate-General Poiares Maduro in his Opinion of 22 May 200831 took the opportunity to set out his perspective on potential abuse of the freedom of establishment: [D]espite what the rulings in Inspire Art and Centros suggest, it may not always be possible to rely successfully on the right of establishment in order to establish a company nominally in another Member State for the sole purpose of circumventing one’s own national company law. In its recent judgment in Cadbury Schweppes, the Court reiterated that ‘the fact that [a] company was established in a Member State for the purpose of benefiting from more favourable legislation does not in itself suffice to constitute abuse of [the freedom of establishment]’. However, it also emphasised that Member States may take measures to prevent ‘wholly artificial arrangements, which do not reflect economic reality’ and which are aimed at circumventing national legislation. In particular, the right of establishment does not preclude Member States from being wary of ‘letter box’ or ‘front’ companies. In my view, this represents a significant qualification of the rulings in Centros and Inspire Art, as well as a reaffirmation of established case-law on the   Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155.   Inspire Art (n 25 above), para [95] (references omitted). 27   Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9614. 28   See on this problem Ringe (n 5 above). 29   Case C-210/06 Reference for a preliminary ruling from the Szegedi Ítélotábla ˝ (Court of Appeal Szeged) lodged on 5 May 2006—Cartesio Oktató és Szolgáltató Betéti Társaság [2006] OJ C165/17. 30   See on this case A Johnston and P Syrpis, ‘Regulatory Competition in European Company Law after Cartesio’ (2009) 34 European Law Review 378; R Szudoczky, ‘How Does the European Court of Justice Treat Precedents in Its Case Law? Cartesio and Damseaux from a Different Perspective’ (2009) 37 Intertax 346; T Biermeyer, ‘Bringing Darkness into the Dark: European Corporate Cross-Border Mobility in Re Cartesio’ (2009) 16 Maastricht Journal of European and Comparative Law 251. 31  See on this WG Ringe, ‘Keine Beschränkungen des Wegzugs von Gesellschaften innerhalb der EU („Cartesio“)’ [2008] Zeitschrift für Wirtschaftsrecht (ZIP) 1067, 1072. 25

26

112  Wolf-Georg Ringe principle of abuse of Community law, even though the Court continues to use the notion of abuse with considerable restraint—and rightly so.32

B. Analysis Advocate-General Poiares Maduro’s Opinion in Cartesio is the most recent of the statements discussed above. His Opinion attempts to reconcile the entire previous case law and to give a sense of what the Court considers as abusive in the context of freedom of establishment. As he underlines, the distinction between legal behaviour and abuse can be found in the comparison between the setting up of a company for the mere purpose of benefiting from more favourable legislation on the one hand and ‘wholly artificial arrangements’ on the other.33 Both of the cases that he cites, Centros and Cadbury Schweppes,34 seem to be examples of the former situation. Thus, the quaestio famosa is: what constitutes a ‘wholly artificial arrangement’? It is clear now that the mere fact that a company is incorporated in Member State A solely to avoid laws which would otherwise apply were it incorporated in Member State B does not constitute an ‘abuse’ of the freedom of establishment.35 This was first decided in Centros. Subsequently, Cadbury Schweppes curbed the broad scope of this rule by excluding arrangements that are ‘wholly artificial’.36 To evade categorisation as a ‘wholly artificial arrangement’, the Court requires the ‘actual pursuit of an economic activity through a fixed establishment in that State for an indefinite period’.37 Consequently, it presupposes actual establishment of the company concerned in the host Member State and the pursuit of genuine economic activity there. This requirement may be further supported by the previous ‘General Programme’ for the freedom of establishment, which was enacted by the Council in 1961,38 and which became obsolete with the entering into force of the Treaty of Amsterdam. This programme, which had no specific legal status,39 required an undertaking to have a ‘real and continuous link’ with the economy of a Member State in order to benefit from the freedom of establishment. When this ‘General Programme’ was phased out of force, the EC Treaty gave full effect to the freedom of establishment with no mention of such a requirement, which implies that such a ‘real link’ is definitely no longer necessary.40 The distinction between the setting up of a company for the mere purpose of benefiting from more favourable legislation on the one hand and ‘wholly artificial arrangements’   AG Poiares Maduro in Cartesio (n 27 above), para [29] (references omitted).   Cartesio (n 27 above). Cf Léger (n 9 above).   Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 35   Centros (n 4 above), paras [27–29]; Inspire Art (n 25 above), para [96]. 36   Cadbury Schweppes (n 34 above), para [51]. Cf Edwards and Farmer (n 1 above) 218. 37   See already Case C-221/89 Factortame [1991] ECR I-3905, para [20] and Case C-246/89 Commission v United Kingdom [1991] ECR I-4585, para [21]. 38  European Council of Ministers, General Programme for the abolition of restrictions on the freedom of establishment of 16 December 1961 [1974] OJ Spec Ed IX 7. 39   Arnull et al (n 10 above), para [19-010]. 40   See on the ‘General Programme’, A Verhoeven, ‘Privatisation and EC Law: Is the European Commission “Neutral” with Respect to Public Versus Private Ownership of Companies?’ (1996) 45 International and Comparative Law Quarterly 861, 879. 32 33 34

Sparking Regulatory Competition in European Company Law  113 on the other may sound rational in theory. However, if situations such as Centros and Cadbury Schweppes are considered to be not ‘wholly artificial’, then the scope for actual abusive behaviour is very small, at least in company law. In company law cases, there will most certainly always be an economic motivation to profit from a more favourable jurisdiction; this explains why Centros Ltd was considered not to have behaved abusively. In other words: if the mere setting up of a letter-box company, as was the case in Centros, suffices for an ‘economic activity through a fixed establishment for an indefinite period’, then it is difficult to imagine that the Court will ever find a company law situation that is ‘wholly artificial’.41 This is a very broad conception of the freedom of establishment, and correspondingly a very narrow construction of ‘abuse’, which seems at odds with the general concept of ‘abuse of law’.42 What can the national lawmaker do in order to combat abusive behaviour? It has long been recognised that there is some scope for national legislators to enact anti-abuse rules.43 Over the years, the Court has progressively specified the necessary requirements for these rules. Most importantly, the Court has repeatedly emphasised that the anti-abuse legislation needs to be designed in a way that takes account of a ‘case-by-case analysis’ on the basis of ‘objective evidence’ of abusive behaviour.44 In other words, a general rule that applies irrespective of the specific facts of the situation will most certainly be contrary to the freedom of establishment. The reason for this attitude is to be found in two main principles of EU law: the principle of supremacy of EU law over national law45 and the principle of effet utile,46 ie the Member States’ duty to give full effect to European law. Both principles are instrumental in understanding the Court’s approach towards ‘abuse’. It is of the utmost importance for the Court that the application of anti-abuse rules by national legislatures or courts should not detract from the full effect and uniform application of EU law in the Member States.47 National measures against abuse may not ‘entail an alteration of the provision’

41  On the debate how and if at all to reconcile Centros and Cadbury Schweppes, compare Edwards and Farmer (n 1 above) 218 (critical) with O Rousselle and HM Liebman, ‘The Doctrine of the Abuse of Community Law: The Sword of Damocles Hanging over the Head of EC Corporate Tax Law?’ [2006] European Taxation 559, 563 (seeing no contradiction). Cf also T O’Shea, ‘The UK’s CFC Rules and the Freedom of Establishment: Cadbury Schweppes plc and its IFSC Subsidiaries—Tax Avoidance or Tax Mitigation?’ (2007) 16 EC Tax Review 13, 18. 42   M Rehberg, ‘Inspire Art—Freedom of Establishment for Companies in Europe between “Abuse” and National Regulatory Concerns’ (2004) 4 European Legal Forum 1, 2, 4. It is questionable, however, if the Court will make a distinction between the setting up and relocating of a company, as Rehberg suggests. 43  See Centros (n 4 above), para [24] with further references. 44   We find this requirement in various cases. See eg, Leclerc (n 12 above), para [27]; Case C-28/95 A LeurBloem v Inspecteur der Belastingdienst/Ondernemingen Amsterdam 2 [1997] ECR I-4161, para [41]; Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I-4695, para [26]; Centros (n 4 above), para [25]; Case C-110/99 Emsland-Stärke GmbH and Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, para [39]; Case C-324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I-11779, para [37]; AG Alber in Inspire Art (n 25 above), paras [102, 106]; Inspire Art (n 25 above), paras [105, 143]. Cf Kjellgren (n 1 above) 183; Sørensen(n 1 above) 453; W Schön ‘Der „Rechtsmissbrauch“ im Europäischen Gesellschaftsrecht’ in R Wank et al (eds), Festschrift für Herbert Wiedemann (Munich, CH Beck 2002) 1271, 1281, 1285; G Kraft and J Bron, ‘Implikationen des Urteils in der Rechtssache „Cadbury Schweppes“ für die Fortexistenz der deutschen Hinzurechnungsbesteuerung’ [2006] Internationales Steuerrecht 614. 45  For a detailed explanation see Craig and de Búrca (n 18 above) 344. 46   Art 4(3) TEU [ex Art 10 EC]. See Craig and de Búrca (n 18 above) 200. 47   Pafitis (n 8 above), para [68] (Cf AG Tesauro in Pafitis (n 8 above), para [27]); Kefalas (n 8 above), para [22]; Diamantis (n 8 above), para [34]; Centros (n 4 above), para [24].

114  Wolf-Georg Ringe of EU law.48 This is why the Court has taken the approach of limiting the application of anti-abuse rules and to accept the reproach of abuse only in a very limited number of exceptional cases.49 In sum, we observe two major issues that are specific to the concept of abuse of law in company law. First, we can generally speak of a liberal approach towards abusive behaviour. If an extreme case, such as the conduct at hand in Centros, is declared not only as not abusive, but rather as the paradigm of taking advantage of the Internal Market, little scope remains for ‘real’ abusive behaviour. The Court has described unacceptable scenarios as ‘wholly artificial arrangements’, but so far no situation in company law has been found to correspond to this classification. Second, the lesson that the national lawmaker has to learn is that general, abstract anti-abuse rules are deemed to violate the freedom of establishment of companies; only rules which allow for an individual assessment that takes account of the specific situation on a case-by-case analysis could pass this test. C. Technique In the company law cases that were described above the Court’s technique is not explicitly based on a specific doctrine of ‘abuse of law’. Despite calls from commentators for an independent concept and the development of a general principle of EU law50 (which may well rest on solid ground when considering the recent tax law cases) the Court’s approach in Centros and its progeny cannot be said to be based on a systematic principle. Rather, the Court employs a technique of interpreting the scope of the fundamental freedom and analyses whether the national measure at stake (which is designed to combat abuse) infringes the freedom of establishment of the respective company. This approach can be described as construing the ‘internal’ limit of the law in itself, rather than the ‘external’ right to make use of the law.51 ‘In other words, it is claimed that to determine whether or not a right is actually being exercised in an abusive manner is simply to define the material scope of the right in question.’52 Other commentators have stated that in regard to this aspect of the abuse of law, there appears to be a certain affinity between the concept of abuse and the principle of proportionality as a criterion for limiting the exercise of power, and further, the concept has even been compared to the English law concept of ‘reasonableness’.53 Abuse of law carries with it some kind of excess: taking advantage of an otherwise well-respected right must be characterised as ‘abusive’ where it is exercised in an unreasonable, disproportionate manner. Similarly, Member States may only disallow truly ‘abusive’ behaviour; all other restrictive national legislation ceases to be compatible with the fundamental freedoms.   AG Saggio in Diamantis (n 8 above), para [26].   Schön (n 44 above) 1281. 50   See for example, de la Feria (n 1 above) 436; Fleischer (n 1 above) 871. Contrast this with the Opinion of AG Tesauro in Kefalas (n 8 above), paras [22–23]. 51   See on this differentiation, Fleischer (n 1 above) 871. 52   AG La Pergola in Centros (n 4 above), para [20], with reference to C Nizzo, ‘L’abuso dei “diritti comunitari”; un quesito non risolto’ [1997] Diritto del commercio internazionale 766, 770. Cf Schön (n 44 above) 1282; Kjellgren (n 1 above) 186. 53   AG La Pergola in Centros (n 4 above), para [20]; LN Brown, ‘Is There a General Principle of Abuse of Rights in European Community Law?’ in D Curtin and T Heukels (eds), Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers, vol II (Dordrecht, Martinus Nijhoff 1994) 511, 521. 48 49

Sparking Regulatory Competition in European Company Law  115 This analysis may be considered as formalistic. It is certainly true that the Court has as yet failed to pronounce a scholarly, well-founded concept of abuse of law, at least in the context of company law. However, the judgments and opinions discussed above all deal with de facto abusive behaviour, and although in each case the Court did not explicitly state that it was contemplating whether the behaviour in question was abusive or not, as a matter of substance, this is precisely the enterprise in which the Court was engaged. It is against this background that we now turn to the reaction that the Court’s case law has provoked. III. Impact of the case law on company activity

The Centros case law has had a significant impact on company activity in Europe. With every free movement judgment handed down from Luxembourg, enterprises became increasingly assured that the freedom of establishment indeed allowed them to register in Member State A, while conducting their business exclusively in Member State B. In this manner, the Court of Justice has created a market for corporate forms within the European Union, granting de facto a choice between the legal forms of now twenty-seven Member States. This holds true especially for entrepreneurs, but is becoming increasingly relevant also to pre-existing businesses.54 Over this period, English company law has proved to be the most competitive company law jurisdiction in the EU.55 The English private company limited by shares can be set up in a comparatively fast and easy way and does not require any minimum capital or the involvement of a public notary. Because of these advantages, many companies have been set up in the years following Centros, Überseering and Inspire Art by entrepreneurs from outside the UK who conduct business exclusively in their home Member States. Special agencies sprang up like mushrooms in continental Europe, helping entrepreneurs to set up an English company seeking to further promote this legal form and to profit from its popularity.56 These marketing agencies hail the advantages of English company law and convince inexperienced businessmen on the continent of the degree of ease with which an English law company can be established. According to one of these agencies, which claims to have set up over 35,000 English companies in Germany, every third company set up in Germany is now an English company.57 This estimate may well be tainted by a certain degree of self-interest. However, academics and government officials have now begun to work on empirical data, providing 54  The possibility of a relocation of an existing business is still to some degree uncertain, due to lacunae in the Court’s case law. Unfortunately, the Commission officially dropped the proposed Fourteenth Directive of the transfer of the seat from one Member State to another (proposal from 1997: Doc No XV/D2/6002/97-EN REV.2). Commissioner McCreevy announced in October 2007 that he would not proceed with the Directive (Speech at the European Parliament’s Legal Affairs Committee, 3 October 2007, Speech/07/592). Yet, the European Parliament is currently trying to revive the project (see n 114 below). So far, the Merger Directive 2005/56/EC of 26 October 2005 on cross-border mergers of limited liability companies [2005] OJ L310/1, and Art 8 of the European Company Statute (SE) provide for the only means to move an existing business across the border without being dissolved or having to re-register. 55   BR Cheffins, Company Law: Theory, Structure and Operation (Oxford, Oxford University Press 1997) 441. 56  Some examples are: www.insolution-ltd.co.uk, www.easyentrepreneur.com, www.go-limited.de, www. Limited4you.de, www.limited24.de, www.france-offshore.fr and www.adcomp.de. 57   See the leaflet by Go-Ahead Limited, www.go-limited.de/fileadmin/redakteur/Info-_und_ Seminarbroschuere.pdf.

116  Wolf-Georg Ringe a more accurate approximation of the number of these ‘foreign’ limited companies.58 The German government, responding to an official question in Parliament, estimated the number of English companies set up in Germany between January and August 2005 to be around 3,200.59 A more detailed survey was compiled by several German academics, whose estimations however diverged significantly. One short study suggested without much evidence, that the total number of English limited companies in Germany amounted to 30,300 in December 2005 and to 46,000 in November 2006.60 Another—more detailed— study by Wilhelm Niemeier offers various calculation methods and suggests a number of between 7,100 and 7,600 at the end of 2005 when collecting data from German registration sources.61 However, Niemeier points out that presumably the real number is significantly higher, since not all foreign companies in Germany fulfil their duties to register a branch in Germany and to apply for a trade concession.62 Furthermore, he points to the fact that those studies analysing data from the Companies House (the British company register) come to a different outcome. A similar approach is put forward in a recent study by Horst Eidenmüller.63 Not only the German, but also the English-speaking academic literature has tried to quantify this German interest in a foreign legal form. In a first, impressionistic overview, John Armour arrived at a figure of over 500 ‘German’ English companies at the end of 2004.64 One of the major studies that followed the methodology of collecting data from the English Companies House and then filtering out those companies apparently conducting business exclusively in other countries is the study conducted by Becht, Mayer and Wagner, which had been discussed as a working paper since 2006 and was finally published in 2008.65 This detailed study counts a total of 25,866 ‘German’ English companies set up between 1997 and 2005 and an additional 15,633 in 2006 alone, totalling up to 41,499 between 1997 and 2006.66 Of course, these figures do not take into account the number of firms that were wound up during the same period.67 Arguably, this number is relatively high, since ‘foreign’ English companies suffer from an increased risk of being shut down during the period of their early existence.68

58   See S Deakin, ‘Legal Diversity and Regulatory Competition: Which Model for Europe?’ (2006) 12 European Law Journal 440, 450. 59  Reply by the Federal Government to the question by the deputies Otto Fricke, Dr Max Stadler, Jens Ackermann, further deputies and the FDP parliamentary party (Drucksache 16/134) of 16 December 2005, Bundetagsdrucksache 16/283. 60   AO Westhoff, ‘Die Verbreitung der englischen Limited mit Verwaltungssitz in Deutschland’ [2007] GmbHRundschau 474, building on AO Westhoff, ‘Die Verbreitung der limited mit Sitz in Deutschland’ [2006] GmbHRundschau 525. 61  W Niemeier, ‘GmbH und Limited im Markt der Unternehmensrechtsträger’ [2006] Zeitschrift für Wirtschaftsrecht (ZIP) 2237. 62  Niemeier (n 61 above) 2242. 63  H Eidenmüller, ‘Die GmbH im Wettbewerb der Rechtsformen’ [2007] Zeitschrift für Unternehmens- und Gesellschaftsrecht (ZGR) 168, 170. 64   Armour (n 5 above) 386. This data was identified by searching Companies House for companies with largely German-language names, but ending with the word ‘Limited’. 65   M Becht, C Mayer and H Wagner, ‘Where Do Firms Incorporate? Deregulation and the Cost of Entry’ (2008) 14 Journal of Corporate Finance 241. 66   Becht et al (n 65 above) 248, table 3, panel B. 67   See Niemeier (n 61 above) 2244. 68  Niemeier (n 61 above) 2241.

Sparking Regulatory Competition in European Company Law  117 Table 1:  Newly Set Up English Private Limited Companies Doing Business Exclusively in Germany69 New English ltd companies in Germany 1997-2006 45,000 40,000 35,000

number

30,000 25,000

newly set up

20,000

total

15,000 10,000 5,000 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year

Source: adapted from Becht et al (n 65 above).

This is not the place to comment on the correctness or the flaws of these different empirical studies and methodologies. For our purposes, it is sufficient to point to the one common element to all these studies: that the number of companies incorporated in England but managed from abroad has been steadily rising over the last years, and that this appears to be a direct result of the Court’s case law that was discussed above.70 For the purposes of our enquiry here, we can thus conclude that the Court’s liberal approach toward abuse and the fact that almost all national ‘anti-abuse’ regulation has been struck down, has caused company migration to surge. Setting up a company in Member State A and conducting business exclusively in Member State B, the phenomenon that was considered to be abusive before Centros, has become a blueprint for many entrepreneurs all over Europe. This leads us to the next task: to analyse how the Member States reacted to this phenomenon. IV. Impact on national company law

The study by Becht, Mayer and Wagner mentioned above goes beyond merely calculating the number of ‘foreign companies’ incorporating in England. It also analyses the reasons for the high number of these foreign companies. The study shows that amongst the main reasons are the substantial differences regarding minimum capital requirements, 69  The ‘total’ number is only the sum of the previous numbers and does not take into account the number of companies that were wound up over the period. 70  This latter fact is proven by Becht et al (n 65 above) 247.

118  Wolf-Georg Ringe set-up costs and set-up speed.71 Further supplementary reasons may be qualitative aspects in company law or insolvency law including distribution or finance rules, or a flexible restructuring law.72 Other aspects might include the rules of company organisation, including the co-determination right of employees.73 The national lawmakers responded to the challenge posed by English company law that was triggered by the Court’s case law. In France and Spain, new, deregulated forms of limited liability companies were introduced soon after Centros. Since 2003, it has been possible in France to found the Société à Responsabilité Limitée (SARL) within twenty-four hours and with a nominal minimum capital of €1 (previously, the minimum capital for the SARL used to be €7,500).74 The formalities of incorporation were reduced to a minimum—even applications for registration via the internet became possible. Founders can also seek the assistance of a Centre de Formalités des Entreprises. The French reform legislation included further facilitative measures in other areas: the newly-founded companies enjoy certain tax and social contribution reductions within the first years of their existence and may use the private address of the founder as the corporate seat even where this would otherwise be contrary to rental contracts or general city planning legislation. In sum, the main purpose of this new legislation is to encourage small start-up enterprises that have their real seat and their main field of activity in France, to choose the French legal form.75 Largely similar legislation has existed in Spain since April 200376 with the adoption of the Sociedad Limitada Nueva Empresa (SLNE) as a special form of the traditional Sociedad de Responsabilidad Limitada (SL or SRL).77 In contrast to French law, the SLNE must have a minimum nominal capital of €3,012.78 Like the new French SARL, the nueva empresa is designed to encourage and support small and medium sized start-up enterprises located in Spain.79 One special feature is a quick set-up procedure: according to Article 134 of the amended version of the Ley de Sociedades de Responsabilidad Limitada, the SLNE can be founded within forty-eight hours.80 Arguably, these two ‘early’ reforms have already significantly reduced the business exits in these two countries, as compared to other EU countries.81 But other countries are quickly following suit. In the Netherlands, a fundamental review of the private limited (BV) law is underway in the Dutch Parliament.82 The consultation   Becht et al (n 65 above) 250.  See on the latter aspect WG Ringe, ‘Forum Shopping under the EU Insolvency Regulation’ (2008) 9 European Business Organization Law Review 579. 73   Cheffins (n 55 above) 442. 74  Drury (n 24 above) 729. From the French literature, see T Massart, ‘Une grande réforme à petit budget: la SARL au capital de un euro’ [2002] Bulletin Joly Sociétés 1361; S de Vendeuil, ‘Les dispositions de droit des sociétés de la loi numéro 2003-721 du 1er août 2003 pour l’initiative économique’ [2003] La Semaine Juridique Entreprise et Affaires 1401. 75  The new form of SARL has not been introduced with a view to attract incorporations from abroad. 76  Ley 7/2003 of 1 April 2003, de la Sociedad Limitada Nueva Empresa por la que se modifica la Ley 2/1995, de 23 de marzo, de Sociedades de Responsabilidad Limitada, BOE núm 79, de 2 de abril, Sec 1, 12679. 77  JM Embid Irujo, ‘Eine spanische „Erfindung“ im Gesellschaftsrecht: Die „Sociedad limitada nueva empresa“—die neue unternehmerische GmbH’ [2004] Recht der Internationalen Wirtschaft 760. 78  The new legislation also fixes a maximum nominal capital of €120,200. 79  Embid Irujo (n 77 above) 761. 80  Embid Irujo (n 77 above) 763. 81   Becht et al (n 65 above) 252. 82   See on this H Boschma, L Lennarts and H Schutte-Veenstra, ‘The Reform of Dutch Private Company Law: New Rules for the Protection of Creditors’ (2007) 8 European Business Organization Law Review 567; JA McCahery, EDG Kiersch and L Timmerman, Private Company Law Reform: International and European Perspectives (The Hague, Asser Press 2007) part III. 71 72

Sparking Regulatory Competition in European Company Law  119 documents explicitly state that ‘the [reformed] Dutch private limited company must take on the competition with foreign legal forms’.83 The draft reform bill was submitted to the Lower House of Parliament on 31 May 2007, where it is still pending.84 The planned reform provides for more freedom for entrepreneurs in that businesses will have more options in their articles of association to depart from the default provisions of law on matters such as voting rights, board member appointment and shareholder resolutions. Most importantly, the existing minimum capital of €18,000 will no longer be required. Furthermore, the set-up speed for businesses, which was in urgent need of reform, as this was an important ground for Dutch entrepreneurs to seek incorporation abroad, is set to be improved.85 Similarly, Germany has also introduced a major reform.86 At first, it was uncertain whether the country would take the Spanish route of adopting a new legal form alongside the existing ones,87 or rather follow the French/Dutch example of reforming the existing legal forms. Finally, a compromise was found: the existing Gesellschaft mit beschränkter Haftung (GmbH, private limited liability company) was reformed and at the same time, a variant of the GmbH was introduced. Whereas the existing GmbH still requires the minimum capital of €25,000, the new younger brother of the GmbH, the Unternehmer­ gesellschaft (‘UG’—start-up company),88 is integrated into the existing GmbH statute and can be set up without minimum capital.89 However, the new code imposes some restrictions: the UG is not allowed to fully distribute its profits until its level of capital has reached the threshold of the regular GmbH (€25,000).90 One principal concern of the German reform is to facilitate and accelerate the establishment of a business. Thus, for instance, model articles of association are made available for standard set-ups, without having to consult a public notary. Furthermore, the set-up process is accelerated. In the meantime, France has already adopted a second reform. After the 2003 reform concerning the SARL, the Société par Actions Simplifiée (SAS) was reformed in 2008.91   Becht et al (n 65 above) 252.  Dutch Ministry of Justice, ‘Simplified Formation of Private Limited Liability Companies’, Press Release of 31 May 2007. 85  Drury (n 24 above) 730, 740, pointing to the facts of Case 79/85 DHM Segers v Bestuur van de Bedrijfsvereniging voor Bank- en Verzekeringswezen, Groothandel en Vrije Beroepen [1986] ECR 2375. 86   Gesetz zur Modernisierung des GmbH-Rechts und zur Bekämpfung von Missbräuchen (MoMiG) of 23 October 2008, [2008] Bundesgesetzblatt I 2026. From the (already) vast literature, see W Goette, Einführung in das neue GmbH-Recht (Munich, CH Beck 2008); H Hirte, ‘Die „Große GmbH-Reform“—Ein Überblick über das Gesetz zur Modernisierung des GmbH-Rechts und zur Bekämpfung von Missbräuchen (MoMiG)’ [2008] Neue Zeitschrift für Gesellschaftsrecht 761. 87   Various suggestions were made to this end: the Bavarian idea of a Kaufmann (single trader) with limited liability (www.justiz.bayern.de/imperia/md/content/stmj_internet/ministerium/ministerium/gesetzgebung/ gesetzentwurf_kaufmann_mbh_16_05_2007.pdf) or the North Rhine-Westphalian suggestion of a ‘BasisGmbH’ (BGmbH) (www.justiz.nrw.de/JM/justizpolitik/gesetzgebung/gesetzgebungsvorhaben/gmbh_recht/ inhalt_gesetzentwurf/gesetzentwurf.pdf). The Green Party developed the concept of a ‘Personengesellschaft mit beschränkter Haftung (PmbH)’ (partnership with limited liability), see M Berninger and H Schnittker, Eckpunktepapier zur Schaffung der Gesellschaftsform einer Personengesellschaft mit beschränkter Haftung (June 2006). Finally, an academic came up with the ‘Leipzig draft’ of a Kommanditgesellschaft with limited liability (KmbH), see T Drygala, ‘Für eine alternative Rechtsform neben einer reformierten GmbH— Leipziger Entwurf einer Kommanditgesellschaft mit beschränkter Haftung (KmbH)’ [2006] Zeitschrift für Wirtschaftsrecht (ZIP) 1797. 88  The full name of the legal form is ‘Unternehmergesellschaft (haftungsbeschränkt)’. 89   Section 5(1) of the new GmbH-Gesetz (act on the private limited liability company). 90   Section 5a of the new GmbH-Gesetz. 91  Loi no 2008-776 du 4 août 2008 de modernisation de l’économie [2008] Journal officiel de la République française no 0181, 12471. See: www.modernisationeconomie.fr/. 83 84

120  Wolf-Georg Ringe The SAS is a legal form between the SARL and the Société Anonyme (SA), the latter being the French public limited company. Amongst the various amendments to the SAS, the most remarkable element of the reform was without doubt the waiving of the previous minimum capital requirement of €37,000.92 One of the explicit goals was again the desire to make the French company legal form more attractive.93 But it is not only continental Europe that is reacting to enhanced regulatory competition, as the UK itself also attempts to further modernise its company law. One of the explicit ambitions of the new Companies Act 2006 was to make the UK even more desirable as a corporate home for overseas companies.94 Of course, while the company law reform in the UK started well before the ECJ handed down its judgment in Centros, nevertheless, the UK seems eager to take up the global challenge and to safeguard its leading position in Europe. All in all, the Court’s case law has created a level playing field for a corporate market in Europe. The liberal approach towards abuse has facilitated the usage of foreign legal forms. EU Member States are responding to the success of UK company law by so-called ‘responsive lawmaking’. Now that EU Member States have to recognise each other’s companies, some Member States are enacting corporate law preferred by their shareholders (especially entrepreneurs), managers or both, and are thus luring companies away from other Member States which have a less attractive company law system. Some authors suggest that the European development after Centros and Inspire Art may in some ways resemble the US developments in connection with the ‘Delaware effect’95 in corporate law in the nineteenth century.96 Others stress the difference between the EU and the US situations.97 Even though the regulatory competition that we are currently witnessing concerns start-up companies more than already existing enterprises, we should point out that cross-border relocations for existing companies are becoming increasingly common and will eventually lead to regulatory arbitrage on this level. This is on account of the ECJ decisions in Sevic98 and Cartesio99 which are sweeping away the remaining restrictions on corporate mobility, as well as secondary EU legislation such as the CrossBorder Merger Directive and the European Company Statute.

92  On this, R Mortier, ‘La modernisation du droit des sociétés’ [2008] La Semaine Juridique Entreprise et Affaires 2233; JF Barbièri, ‘La SAS revisitée par la LME’ [2008] Bulletin Joly Sociétés 560; T Massart, ‘La modernisation de la SAS ou comment apporter moins pour gagner plus’ [2008] Bulletin Joly Sociétés 632. 93   Mortier (n 92 above) 2233. 94  Department of Trade and Industry, Company Law Reform—White Paper (London, The Stationery Office 2005) Cm 6456, 9. 95   At the end of the nineteenth century, the states of New Jersey and Delaware, concerned about entrepreneurs’ decisions as to the place of their companies’ incorporation, adopted modernised statutes on incorporation. Eventually, Delaware’s statute made it the leading state for incorporations in the United States from the 1920s onwards, presently serving as the state of incorporation for nearly half of the corporations listed on the New York Stock Exchange and as the major destination for reincorporations. See R Romano, The Genius of American Corporate Law (Washington, AEI Press 1993) 6. 96   See for example, C Kirchner, RW Painter and WA Kaal, ‘Regulatory Competition in EU Corporate Law After Inspire Art: Unbundling Delaware’s Product for Europe’ [2005] European Company and Financial Law Review 159; D Martin and FG Alogna, ‘New Delaware’ Wall Street Journal (New York, 20 December 2007). 97  EM Kieninger, ‘The Legal Framework of Regulatory Competition Based on Company Mobility: EU and US Compared’ (2004) 6 German Law Journal 741, 769; WW Bratton, JA McCahery and EPM Vermeulen, ‘How Does Corporate Mobility Affect Lawmaking? A Comparative Analysis’ (2009) 57 American Journal of Comparative Law 347. 98   Case C-411/03 Sevic Systems AG [2005] ECR I-10805. 99   Cartesio (n 27 above).

Sparking Regulatory Competition in European Company Law  121 Table 2:  Company Law Reforms in Several EU Countries 2003–2009 Feature/ Country

Legal Form

Pre-reform

Post-reform

Alteration

Minimum Capital (in €) Denmark

Anpartselskab (ApS)

16,800

10,800

–35.7%

Société à Responsa­bilité Limitée (SARL)

7,500

1

–99%

Germany

Gesellschaft mit beschränk­ter Haftung (GmbH)

25,000

Netherlands

Besloten Vennootschap (BV)

18,000

1

–99%

Sociedad de Responsabilidad Limitada (SRL)

3,010

3,012

+ 0.07

ApS

6,715

0

–100%

France

Spain

25,000

1

+/–0

–99%

Set-up Costs (in €) Denmark France

SARL

450

310

–13%

Germany

GmbH

1,500

1,500

+/–0

Netherlands

BV

1,750

n.a.

Spain

SRL

600

450

–25%

ApS

23

6

–74%

Set-up speed (in days) Denmark France

SARL

10

7

–30%

Germany

GmbH

16–18

16–18

+/–0

Netherlands

BV

10

n.a.

Spain

SRL

47

2

–96%

Source: adapted from Bratton et al (n 97 above); Becht et al (n 65 above); www.doingbusiness.org; www.gruenderstadt.de/Infopark/spanische_slne.html

V. Assessment

Thus far this chapter has revealed that there is ample evidence that: (i) the forgiving stance of the Court of Justice towards conduct alleged to constitute an abuse of law has granted companies the opportunity to register in a Member State of their choice while conducting their entire business in another; and (ii) EU Member States are indeed responding to this phenomenon by amending their corporate laws to remain attractive in the face of competition from other jurisdictions. In considering the merits of these two developments, there are two key questions. First: are these developments detrimental or beneficial for our European economy? Second: how far will these developments go?

122  Wolf-Georg Ringe A.  Beneficial Effect In response to the first question, reference is often made to the famous debate in US regulatory thinking on whether regulatory competition provokes a ‘race to the bottom’, ie a contest where the country with the laxest standards and the most depressed protection levels will prevail, or whether it will eventuate in a ‘race to the top’, meaning a situation where the most developed and efficient company law jurisdiction will triumph.100 In general, while the claim that Delaware corporate law is efficient remains much disputed,101 it is generally agreed that regulatory competition need not, necessarily, imply a degradation of standards. Of course it is apparent that there are major differences between the EU and the US situations. So far, regulatory competition in Europe concerns only entrepreneurs when setting up a company, and as yet does not touch upon the relocation or reincorporation of existing businesses.102 Relocation of a business still faces significantly greater legal obstacles and expense (both in real investment, and also in information costs103) in Europe than in the United States. Moreover, the nascent European regulatory competition only concerns private limited companies and does not yet extend to the legal forms of public limited companies.104 This omits a significant and important part of the market which can be attributed largely to the fact that the existing company law harmonisation at the EU level primarily addresses publicly held companies, so that there is little scope for companies to make use of disparities between the national systems.105 Moreover, a public limited company is most often founded through conversion from an existing private limited company rather than original incorporation; and therefore the persisting obstacles to relocation indirectly affect the mobility of public limited companies as well. Despite the aforementioned differences,106 many similarities between the US and EU situations cannot be denied. As was the case in the United States, one state is taking the lead and attracting companies from other states and, correspondingly, this will have the consequence of improving proficiency in company law and court practice in the existing well-established financial centre of London, which in turn becomes even more attractive, accelerating further its progression to pre-eminence (‘first mover advantage’).107 Further, 100   See eg sceptically, W Cary, ‘Federalism and Corporate Law: Reflections Upon Delaware’ (1974) 83 Yale Law Journal 663; speaking of a ‘race to the top’ see R Winter, ‘State Law, Shareholder Protection and the Theory of Corporation’ (1977) 6 Journal of Legal Studies 251; D Fischel, ‘The “Race to the Bottom” Revisited: Reflections on Recent Developments in Delaware’s Corporation Law’ (1982) 76 Northwestern University Law Review 913, 916, 920. See also, F Easterbrook, ‘The Economics of Federalism’ (1983) 26 Journal of Law and Economics 23, 28; and F Easterbrook and D Fischel, ‘Voting in Corporate Law’ (1983) 26 Journal of Law and Economics 395. For a more recent account, see Z Fluck and C Mayer, ‘Race to the Top or Bottom? Corporate Governance, Freedom of Reincorporation and Competition in Law’ (2005) 1 Annals of Finance 349. 101  L Bebchuck and A Ferrell, ‘Federalism and Takeover Law: The Race to Protect Managers from Takeovers’ (1999) 99 Columbia Law Review 1168; O Bar-Gill, M Barzuza and L Bebchuk, ‘The Market for Corporate Law’ (2006) 162 Journal of Institutional and Theoretical Economics 134. 102   See Armour (n 5 above) 381. This may well change now that new means such as the Merger Directive and the SE are available (see n 54 above). 103  Information costs are transactions costs that include the assessment of the investment merits of a relocation. 104   Becht et al (n 65 above) 247 show that corporate mobility does not concern public limited companies. 105   Kieninger (n 97 above) 769. 106  For other differences between the US and the EU being discussed that do not convince me entirely, see Becht et al (n 65 above) 252 and Kieninger (n 97 above) 769. 107   Martin and Alogna (n 96 above).

Sparking Regulatory Competition in European Company Law  123 even though the European Union does not recognise the US system of franchise taxes,108 EU Member States have sufficient incentives for implementing an attractive company law, for reasons relating to political control, as well as remaining attractive to business and securing employment for the legal profession. Be that as it may, in analysing the effects of the current regulatory competition in Europe, it is clear that the response by Member States thus far has produced certainly ‘better’ company laws. Since, as explained above, the regulatory competition in the EU over the last several years concerned only entrepreneurs (and not established companies), the main issues at stake were such elements of company law as minimum capital, set-up speed and set-up cost.109 When Member States adopt reform statutes reducing these startup burdens, this surely gives businessmen and investors the necessary legal framework for swift and uncomplicated practical implementation of their entrepreneurial ideas and is thus beneficial for all. The TFEU mandates in its Article 120 [ex Article 98 EC] that ‘Member States and the Union shall act in accordance with the principle of an open market economy with free competition, favouring an efficient allocation of resources’. This tenet is the basis for the application of fundamental freedoms, and is one of the reasons why the full effectiveness of fundamental freedoms is of the utmost importance, notwithstanding the need to curb abusive behaviour.110 If we believe in the instrumental role of the fundamental freedoms in facilitating an efficient allocation of resources, the case law of the Court has certainly proved successful: the consequent and steady application of the freedom of establishment in recent years, from Centros to Cartesio, has definitely lowered the direct and indirect costs of incorporation. Rigid formalities had locked the evolution of company law in a certain path, and so thwarted the emergence of more flexible legislation. Increased mobility has removed some of the obstacles and opened up opportunities for reformminded lawmakers. B. Outlook As explained above, the current regulatory competition concerns only start-up companies and has not yet impinged upon the relocation of existing companies. Scholars are divided as to what the future will bring. Some are sceptical, believing that the level of regulatory competition will slow down, once continental lawmakers have caught up with the UK.111 Others hope that the Court will remove the remaining obstacles to cross-border mobility, which would enhance competition. I am optimistic in this sense for various reasons. First, the latest indications point towards the availability of full regulatory competition in the not all too distant future. The Cartesio case was only partly a disappointment, and might in the long term turn out to be more supportive of regulatory competition than at first glance imagined.112 Besides this, an increasingly burgeoning amount of tools for cross-border movement has become available,113 and even the plans for the adoption of 108   See on this R Romano, ‘Empowering Investors: A Market Approach to Securities Regulation’ (1998) 107 Yale Law Journal 2359, 2388. 109   See Part IV (text with n 71 above). 110   Schön (n 44 above) 1280. 111   Becht et al (n 65 above) 256 and Kieninger (n 97 above) 770. 112   Cartesio (n 27 above), esp paras [111]–[113]; Cf Biemeyer (n 30 above) 253. 113  Especially the (Tenth) Cross-Border Merger Directive and the European Company Statute (see n 54 above).

124  Wolf-Georg Ringe the Fourteenth Company Law Directive on the cross-border transfer of the registered office of limited companies might be revived.114 Second, even though harmonisation of company law suggests that incentives for (public) companies to migrate could be reduced, it is obvious that many different company law regimes still exist, with differences that are sufficiently significant to still be worth overcoming—one need think only of German co-determination.115 Furthermore, harmonisation on the European level has been characterised as ‘trivial’, meaning that it does not cover the most important areas of company law and still allows for much divergence.116 These two factors suggest that enough incentives will continue to exist—even for public companies—to compare legal systems and to choose between them. Thirdly, even if we remain in the current (limited) state of regulatory competition, it is not yet certain that continental Europe will indeed eventually catch up with the UK. Ambitious reform projects have often in the past suffered a last-minute setback, and the UK has just made its own company law more attractive.117 And fourthly, the European Union itself is now taking up the gauntlet: after successfully introducing the European Company (SE) in 2004, the European Private Company (SPE) has recently been officially proposed.118 Both legal entities rely on one important feature: they both can transfer their registered office across the border without having to reregister or to be wound up.119 Thus, these European legal forms will further fuel the degree of company mobility in Europe and increase the pressure on national lawmakers. Therefore, I believe that we are not at the end of a (limited) regulatory competition process for start-up companies in Europe, but perhaps at the end-point of a first wave. This first wave will hopefully be followed by more to come, and eventually encompass a sea of free reincorporation and a genuine level playing field between the different company law systems that the Member States and the EU itself have on offer.

114  Regrettably, the plans had been dropped (see n 54 above). Now, however, the European Parliament wants to reopen the legislative process. See Committee of Legal Affairs, ‘Draft Report with recommendations to the Commission on cross-borders transfers of company seats’ 2008/2196 (INI) of 17 October 2008. 115  The confederation of German industry (BDI) has been trying to reform the German system of codetermination for years. See BDA and BDI, ‘Mitbestimmung modernisieren—Bericht der Kommission Mitbestimmung’, Berlin 2004 (available at: www.arbeitgeber.de/www/arbeitgeber.nsf/res/Bericht_der_ Kommission_Mitbestimmung.pdf/$file/Bericht_der_Kommission_Mitbestimmung.pdf). On German codetermination in general, see K Pistor, ‘Co-Determination in Germany: A Socio-Political Model with Governance Externalities’ in M Blair and MJ Roe (eds), Employees and Corporate Governance (Washington, Brookings Institution Press 1999) 163; J Du Plessis and O Sandrock, ‘The Rise and Fall of Supervisory Codetermination in Germany?’ (2005) 16 International Company and Commercial Law Review 67. Historically, compare H Wiedemann, ‘Codetermination of Workers in German Enterprises’ (1980) 28 American Journal of Comparative Law 79. On an economic account, see FR FitzRoy and K Kraft, ‘Economic Effects of Codetermination’ (1993) 95 The Scandinavian Journal of Economics 365; B Frick and E Lehmann, ‘Corporate Governance in Germany: Ownership, Codetermination and Firm Performance in a Stakeholder Economy’ in H Gospel and A Pendleton (eds), Corporate Governance and Labour Management (Oxford, Oxford University Press 2004) 122. 116  L Enriques, ‘EC Company Law Directives and Regulations: How Trivial Are They?’ (2006) 27 University of Pennsylvania Journal of International Economic Law 1. 117   See for example, P Davies and J Rickford, ‘An Introduction to the New UK Companies Act’ [2008] European Company and Financial Law Review 48. 118  European Commission, ‘Proposal for a Council Regulation on the Statute for a European Private Company’ of 25 June 2008 (COM(2008) 396/3). See on this RR Drury, ‘The European Private Company’ (2008) 9 European Business Organization Law Review 125. 119   Art 8 of the SE Regulation 2157/2001; Art 35 of the proposed SPE Regulation.

Sparking Regulatory Competition in European Company Law  125

vI. Conclusion

This chapter has analysed the Court’s case law on abuse of law in the company law arena. This case law has repeatedly touched on the question of abuse, most notably in the situation where a company was set up in a Member State only to do business exclusively in another. Beginning with the landmark case of Centros in 1999, the Court has repeatedly stressed that it employs a liberal approach towards abuse in this field. According to the Court, making use of the disparities between different legal standards when setting up a company is not abuse, but explicit use of the freedom of establishment. This case law has provoked a sizeable entrepreneurial migration from many Member States towards the United Kingdom. Empirical studies show that many companies are now set up in England and are registered there, but conduct their entire business in other Member States. This behaviour in turn has stirred national lawmakers in continental Europe to adapt their company law to make it more attractive for those starting up a company. It has been argued that, at least so far, the (limited) competition between Member States has been beneficial: registration time and cost was or is being reduced, and minimum capital requirements at least partly abolished. It is argued that this competition will continue in various areas, and eventually encompass reincorporations and relocations. The Court of Justice has changed the landscape of company law in Europe with just a few judgments. The Commission is taking up its job to supplement this case law with secondary legislation. Exciting times for the corpus of European company law lie ahead.

9 Sparking Regulatory Competition in European Company Law: A Response John Vella I. Introduction

C

entros1 and the cases that followed it2 have spawned a remarkable volume of academic writing. This contribution is limited to comments on the points made in Georg Ringe’s chapter,3 which can be divided into three parts. It starts by considering Centros and the concept of ‘abuse of law’ in company law. In essence, it notes that Centros has opened the door for regulatory arbitrage in company law and contends that, effectively, this has not been altered by recent tax cases dealing with abuse. The second part describes the effect Centros has had in practice, noting, in particular, the spate of incorporations of UK private companies by nationals of other Member States and the legislative response by some of these States. In the third and final part, it considers some of the obstacles and uncertainties that prevent full regulatory competition in company law in the EU, and predicts that we have just come to the end of a first phase and that more encompassing regulatory competition will ensue. Whilst brief comments are offered on the points made in each of these three parts, this response will focus on those made in the first part. II.  Centros and ‘abuse of law’ in company law

Centros affirmed that ‘the fact that a national of a Member State who wishes to set up a company chooses to form it in the Member State whose rules of company law seem to him the least restrictive and to set up branches in other Member States cannot, in itself, constitute an abuse of the right of establishment’.4 This allowed for regulatory arbitrage in company law. It meant that an individual from Member State A was free to form a company in Member State B which operated exclusively in Member State A, perhaps by means of a branch. The fact that this was done with the intention of circumventing  Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459.  Particularly Case C-208/00 Überseering BV v Nordic Construction Company Baumanagement GmbH [2002] ECR I-9919 and Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155. 3   G Ringe, ‘Sparking Regulatory Competition in European Company Law: The Impact of the Centros Line of Case Law and its Concept of “Abuse of Law”’, ch 8 above. 4   Centros (n 1 above), para [27]. 1 2

128  John Vella the laws of State A was of no consequence.5 Indeed, in Centros, the incorporators did not dispute that such an arrangement was undertaken to avoid burdensome national company law rules. It was also clear that the company was never intended to trade, and never did trade, in its state of incorporation.6 The legitimacy of avoiding uncongenial national legislation by establishing a company or subsidiary in another Member State was confirmed in the tax context in Cadbury Schweppes.7 Crucially, however, a further test was added. Circumvention of national legislation is possible as long as it is not carried out by means of ‘wholly artificial arrangements which do not reflect economic reality’.8 National measures restricting the freedom may be justified in such circumstances. What was the legal reasoning that led to the development of this additional test? To answer this question we must consider Centros and Cadbury Schweppes side by side.9 This exercise will help us determine whether the adoption of this test in the company law field is justified from a doctrinal perspective. In both Centros and Cadbury Schweppes the parties involved wished to make use of their freedom of establishment to circumvent national law. In Centros, a Danish couple wished to set up a company to operate in Denmark, but were eager to avoid Danish legislation on the formation of private companies—in particular, the onerous minimum capital requirements. They thus incorporated a company in the UK and sought to register a branch in Denmark to conduct its entire business there. The Danish authorities refused to register the branch. In Cadbury Schweppes, a UK parent company set up two subsidiaries in Ireland to benefit from a lower rate of corporation tax. The UK authorities sought to apply Controlled Foreign Companies legislation to the subsidiaries, thus effectively taxing the parent for the profit made by one of the subsidiaries.10 In Centros, the Court of Justice recognised the principle that ‘a Member State is entitled to take measures designed to prevent certain of its nationals from attempting, under cover of the rights created by the [EC] Treaty, improperly to circumvent their national legislation . . . or to prevent individuals from improperly or fraudulently taking advantage of provisions of Community law’ (emphasis added).11 Clearly, the key word here is improperly, because, as noted, circumvention per se was not deemed objectionable. Circumvention would be improper if it ran counter to the objective of the provisions on freedom of establishment.12 The identification of the objective of these provisions is thus of crucial importance as it determines whether circumvention is improper or not.   Centros (n 1 above), paras [18, 27].   Centros (n 1 above), paras [3, 11 and 18].     7  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995, paras [37–38].     8   Cadbury Schweppes (n 7 above), para [51].     9  For an excellent and detailed consideration of these two cases see V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205. 10  A simple explanation of the relevant parts of the legislation is found in Cadbury Schweppes (n 7 above), paras [3–12]. 11   Centros (n 1 above), para [24]. 12   Centros (n 1 above), para [25]: However, although, in such circumstances, the national courts may, case by case, take account—on the basis of objective evidence—of abuse or fraudulent conduct on the part of the persons concerned in order, where appropriate, to deny them the benefit of the provisions of Community law on which they seek to rely, they must nevertheless assess such conduct in the light of the objectives pursued by those provisions. (Emphasis added.)     5     6

Sparking Regulatory Competition in European Company Law: A Response  129 Having clearly established the importance of the objectives of the provisions on freedom of establishment in determining whether circumvention was improper, one might have expected a thorough examination of said objectives. Instead, the Court simply stated: The provisions of the [EC] Treaty on freedom of establishment are intended specifically to enable companies formed in accordance with the law of a Member State and having their registered office, central administration or principal place of business within the Community to pursue activities in other Member States through an agency, branch or subsidiary.13

This led the Court to conclude that choosing to form a company in a Member State with the least restrictive company law rules while operating in another via a branch is not abusive, but merely inherent in the exercise of this freedom.14 The question of abuse was revisited two paragraphs down, when the Court, relying on a paragraph in Segers,15 held that the fact that a company does not carry out any business in its state of incorporation is not sufficient to constitute abuse in such cases.16 No mention is made of the all-important objective in reaching this conclusion on abuse. Two points can be made here. First, one can state that the objective of the provisions on the freedom of establishment identified by the Court in Centros is, at best, narrow and immediate. Indeed, rather than identifying the objective behind Articles 43 and 48 EC [now Articles 49 and 54 TFEU], the Court practically restated them. Secondly, the argument’s progression from identification of the objective to the conclusion is not without difficulty. The objective of enabling companies to pursue activities in other Member States does not naturally lead to the conclusion that it is inherent in the exercise of this freedom for an individual to first set up a company in a Member State other than his own and then have that company set up a branch in his Member State. Centros thus approved circumvention subject to an extremely narrow abuse test which relied on the identification of an equally narrow objective behind the freedom. In Cadbury Schweppes the Court confirmed the finding in Centros to the effect that circumvention of national law was possible,17 however, it then employed a more exacting abuse test. Indeed, by the time Cadbury Schweppes was decided, the Court’s approach to instances of alleged abuse had developed further through cases such as Emsland-Stärke18 and Halifax.19 In Cadbury Schweppes, the Court thus found that circumvention of national laws is still possible as long as it is carried out by means of an actual establishment in the host State that carries out a ‘genuine economic activity’ there and not by ‘wholly artificial arrangements’.20 This latter test is to be ascertained from objectively verifiable factors such as premises, staff and equipment.21 Crucially, the adoption of this test was predicated on the identification of a more general objective behind the freedom of establishment than that identified in Centros: ‘[allowing] a Community national to participate, on a stable   Centros (n 1 above), para [26]. The same objective was identified in Inspire Art (n 2 above), para [137].   Centros (n 1 above), para [27]. 15  Case 79/85 DHM Segers v Bestuur van de Bedrijfsvereniging voor Bank- en Verzekeringswezen, Groothandel en Vrije Beroepen [1986] ECR 2375. 16   Centros (n 1 above ), para [29]. 17   Cadbury Schweppes (n 7 above), paras [37–38]. 18  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 19  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 20   Cadbury Schweppes (n 7 above ), paras [51, 54 and 55]. 21   Cadbury Schweppes (n 7 above), para [67]. 13 14

130  John Vella and continuing basis, in the economic life of a Member State other than his State of origin and to profit therefrom’.22 The approach adopted in Centros and Cadbury Schweppes was thus similar but its application in the two cases had a decisive difference. In both cases, the Court acknowledged that a Member State is entitled to take measures against attempts to ‘improperly . . . circumvent their national legislation’ by relying on the rights created by the EC Treaty.23 In both cases the Court noted that circumvention would be improper if it ran counter to the objective of the provisions on freedom of establishment.24 But here, the application of the approach in the two cases diverges. In Centros the Court identified a narrow, immediate objective, whilst in Cadbury Schweppes it identified a more general objective. This led to the creation of the more exacting test in Cadbury Schweppes (the ‘Cadbury Schweppes test’). The difference here could thus be said to arise due to the multiplicity of purposes or objectives behind a provision—a well-known problem in the literature on purposive interpretation.25 This explains how the additional test was developed. Three questions now arise. First, do the facts in Centros fall foul of this test? Secondly, is the test applicable in the company law sphere? Thirdly, what is the likely effect of this test on regulatory arbitrage in company law? The first question to consider is whether the facts in Centros fall foul of this test. Dr Ringe recognises the test but assumes that the facts in Centros do not fall foul of it. He thus concludes: [I]f situations such as Centros and Cadbury Schweppes are considered to be not ‘wholly artificial’, then the scope for actual abusive behaviour is very small, at least in company law. In company law cases, there will most certainly always be an economic motivation to profit from a more favourable jurisdiction; this explains why Centros Ltd was considered not to have behaved abusively. In other words: if the setting up of a letter-box company, as was the case in Centros suffices for an ‘economic activity through a fixed establishment for an indefinite period’, then it is difficult to imagine that the Court will ever find a company law situation that is ‘wholly artificial’.26

We should not assume that Centros satisfied the additional test introduced in Cadbury Schweppes on the ground that the ECJ found that the arrangement in Centros was legitimate. The ECJ did find the arrangement in Centros to be legitimate; however, it applied a different test. Whilst the Court could be said to have adopted the same approach in allowing the circumvention of national laws as long as it was not ‘improper’, in Cadbury Schweppes it adopted a more stringent interpretation of the term ‘improper’. It employed a more general objective as a lever to slip in a more onerous test. It is suggested that the facts in Centros do not satisfy the Cadbury Schweppes test. The test, we should recall, concerns ‘wholly artificial arrangements aimed at circumventing   Cadbury Schweppes (n 7 above), para [53].  One notes that in Cadbury Schweppes (n 7 above), para [35], the ECJ referred to Centros (n 1 above), para [24] on this point. 24  One notes that in Cadbury Schweppes (n 7 above), para [52], the ECJ referred to Centros (n 1 above), para [25] on this point. 25  DR Miers and AC Page, Legislation, 2nd edn (London, Sweet & Maxwell 1990) 188–91; J Bell and G Engle, Cross on Statutory Interpretation, 3rd edn (London, Butterworths 1995) 35, 57; RS Summers, Essays in Legal Theory (Dordrecht, Kluwer Academic Publishers 2000) 240; FAR Bennion, Statutory Interpretation, 4th edn (London, Butterworths 2002) 813. 26  Ringe (n 3 above). 22 23

Sparking Regulatory Competition in European Company Law: A Response  131 the application of the legislation of the Member State concerned’.27 A national measure restricting freedom of establishment in these specific circumstances may be justified. In Cadbury Schweppes the Court described a ‘wholly artificial arrangement’ in the context of the facts before it, as arising if a controlled foreign company (ie the subsidiary) ‘is a fictitious establishment not carrying out any genuine economic activity in the territory of the host Member State’.28 Leaving aside for the moment the question as to whether the test applies when the circumvention is created by means of a primary as opposed to a secondary establishment,29 the focal point here is the need for a genuine economic activity. The company in Centros did not carry out any trading in the UK. How can that be said to be a genuine economic activity? In Cadbury Scwheppes the Court pointed to premises, staff and equipment as objective signs that the company carried out a genuine economic activity in a Member State.30 Given its size and the fact that it did not carry out any trading in the UK, one assumes that the company in Centros had very limited, if any, premises, staff and equipment. One cannot but conclude that there was no genuine economic activity in the sense intimated by the Court in Cadbury Schweppes. Furthermore, the existence of an economic motivation for incorporation in a particular Member State does not equate to the performance of a genuine economic activity there. The company in Centros was an archetypal letter-box company, and the Court in Cadbury Schweppes singled out ‘letter-box’ or ‘front’ companies as potentially falling foul of its test.31 The facts in Centros thus seem to fall foul of the Cadbury Schweppes test. The second question we must ask is whether the Cadbury Schweppes test is applicable in the company law sphere. One could perhaps argue that the test does not apply to the Centros fact pattern because it was developed in the tax field. One could also argue that Cadbury Schweppes involved an allegedly artificial secondary establishment whilst Centros involved an artificial primary establishment. Indeed, in Centros, the establishment that failed to carry out the genuine economic activity and thus could fall foul of the Cadbury Schweppes test was the primary establishment (the UK company) and not the secondary one (the Danish branch). Neither of these two arguments is fully convincing. As the Court reminded us in both Centros and Cadbury Schweppes, to determine whether articles granting a freedom are abused, one has to look at their objective. In Centros, the Court seems to have restated the articles rather than identify the objective behind them. In Cadbury Schweppes, on the other hand, the Court did consider the objective behind these articles at some length, and certainly, in a more satisfactory manner.32 If the objective is that identified in Cadbury Schweppes, namely allowing a ‘Community national to participate, on a stable and continuing basis, in the economic life of a Member State other than his State of origin’,33 then the setting up of wholly artificial establishments which do not participate in the economic life of the Member State must surely constitute an abuse of that freedom. It should not make a difference if this is carried out by means of an artificial primary or secondary establishment, or if it is carried out to avoid company   Cadbury Schweppes (n 7 above ), para [51].   Cadbury Schweppes (n 7 above), paras [66–68]. 29  One notes that whilst discussing the characteristics of a wholly artificial arrangement in para [66], the ECJ states that these characteristics are ‘apparent’ from the case law quoted earlier on in the judgment at paras [52–54]. This case law includes Centros. 30   Cadbury Schweppes (n 7 above), para [66]. 31   Cadbury Schweppes (n 7 above), para [68]. 32   Cadbury Schweppes (n 7 above), paras [53–54]. 33   Cadbury Schweppes (n 7 above), para [53]. 27 28

132  John Vella rather than tax law. In either case, the artificial establishment would be contrary to the objective of the freedom. The Court reminded us in Cadbury Schweppes that: [T]he concept of establishment within the meaning of the [EC] Treaty provisions on freedom of establishment involves the actual pursuit of an economic activity through a fixed establishment in that State for an indefinite period . . . Consequently, it presupposes actual establishment of the company concerned in the host Member State and the pursuit of genuine economic activity there.34

Can the use of letter-box companies ever be in line with this objective? It is thus suggested that the Cadbury Schweppes test is applicable to cases where individuals seek to circumvent national company law by setting up a company (primary establishment) in another Member State. Indeed, Dr Ringe assumes that it is applicable. Significantly, Advocate-General Poiares Maduro seems to have reached the same conclusion: [D]espite what the rulings in Inspire Art and Centros suggest, it may not always be possible to rely successfully on the right of establishment in order to establish a company nominally in another Member State for the sole purpose of circumventing one’s own national company law. In its recent judgment in Cadbury Schweppes, the Court reiterated that ‘the fact that [a] company was established in a Member State for the purpose of benefiting from more favourable legislation does not in itself suffice to constitute abuse of [the freedom of establishment]’. However, it also emphasized that Member States may take measures to prevent ‘wholly artificial arrangements, which do not reflect economic reality’ and which are aimed at circumventing national legislation. In particular, the right of establishment does not preclude Member States from being wary of ‘letter box’ or ‘front’ companies. In my view, this represents a significant qualification of the rulings in Centros and Inspire Art, as well as a reaffirmation of established case-law on the principle of abuse of Community law, even though the Court continues to use the notion of abuse with considerable restraint—and rightly so.35 (Emphasis added.)

It is this author’s impression that Advocate-General Poiares Maduro’s weighty words, particularly those emphasised in the above quotation, are not being given sufficient consideration, if at all, in discussions on these issues. Given that the legitimacy or otherwise of circumvention of national laws here ultimately rests on the objective of the Treaty provisions on freedom of establishment, it follows that to dispute the introduction of the Cadbury Schweppes test and its application to company law cases from a doctrinal perspective, one has to challenge the objective identified there as incorrect or somehow incomplete. This is not the place to engage in a discussion of the objectives of the freedoms, and given the clear identification of the freedom in Cadbury Schweppes, undertaking a successful challenge appears to be an uphill task. Nevertheless, it might not be impossible. After stating that the ‘freedoms are not aimed at facilitating arbitrage or indeed regulatory/tax competition’, Dr Schammo makes the following comment, which, he writes, is owed to Professor Derrick Wyatt: Admittedly, the tasks of the Community and with those tasks, the aims of the internal market have not stood still. Think, for instance, of the increased role of competitiveness in the tasks of the Community after the Maastricht and Amsterdam amendments. Thus, if global competitiveness is one of the aims of the internal market, as the Commission appears to assume in recent years, 34   Cadbury Schweppes (n 7 above), para [54]. See the interesting comment in A Arnull, A Dashwood, M Ross, D Wyatt, E Spaventa and M Dougan, Wyatt & Dashwood’s European Union Law, 5th edn (London, Sweet & Maxwell 2006) 846–47. 35  AG Poiares Maduro in Case C-210/06 Cartesio Oktató és Szolgáltató Betéti Társaság [2008] ECR I-9641, para [29].

Sparking Regulatory Competition in European Company Law: A Response  133 the task of achieving a high degree of competitiveness and convergence of economic performance can arguably be said to plead in favour of interpreting the fundamental freedoms in a way which maximises access of market actors to the most ‘efficient’ regulatory regimes. If so, it could also be argued that the contrast between the aims of the fundamental freedoms and regulatory competition may not be as clear cut as it is suggested above.36

Finally, we must ask what the likely effect of the Cadbury Schweppes test will be in the company law field. Determining what constitutes a ‘genuine economic activity’ or a ‘wholly artificial arrangement’ will not be a simple task, both in the tax and the company law field. However, letter-box companies are extreme cases, and they must surely constitute a manageable target for Member States. Any legislative action taken must, of course, be correctly calibrated.37 In particular, it must allow for the determination of abuse on a case-by-case basis subject to objective evidence.38 The successful targeting of letter-box companies could have repercussions on the regulatory arbitrage observed over the past few years. Evidence suggests that most of the private companies set up in the UK by continental entrepreneurs are small, having only one or two directors, and many appear to be mere letter-box companies.39 It is important to point out, however, that as is discussed further on, the legislative response by continental states to this arbitrage could slow it down considerably. The extent to which non-letter-box companies can be successfully targeted is less clear. It will all depend on the ability of companies to comply with the ‘genuine economic activity’ condition in terms of premises, staff and equipment. The fact that the Court referred to ‘wholly artificial arrangements’ and not merely ‘artificial arrangements’ suggests that the bar will not be set too high. Companies on the smaller side of the scale might still have some difficulties complying with these requirements, but they should not constitute an insurmountable obstacle for companies of a certain size. This is not to say that doing so will not entail some careful thought. One would have to ensure, for example, that in setting up a company in another Member State and carrying out a genuine economic activity there would not have negative repercussions on the tax treatment of the company. Also, the cost of compliance with these requirements will have to be taken into account when assessing the net benefit to be gained. III. THE Impact of case law on company activity and national legislation

Centros generated excitement about the possibility of regulatory competition in company law between Member States. Once Centros allowed nationals of Member State A to set up a company in Member State B to carry out business exclusively in Member State A, that gave rise to the possibility of individuals engaging in regulatory arbitrage by choosing to set up their companies in the Member State that offered the company law that best suited their needs. Once there were ‘buyers’ of company law shopping around, ‘vendors’ 36  P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351, 361, fn 59. 37   See for example, Cadbury Schweppes (n 7 above), para [57]. 38   See for example, Centros (n 1 above), para [25]. 39   M Becht, C Mayer and H Wagner, ‘Where Do Firms Incorporate? Deregulation and the Cost of Entry’ (2008) 14 Journal of Corporate Finance 241, 242.

134  John Vella would respond. Therefore, it was predicted, or hoped, that Member States would react by carrying out changes to their company laws so as to attract incorporations. Let us consider what has taken place to date. Dr Ringe cites a number of well-known studies which show a considerable increase in incorporations of private companies in the UK by continental entrepreneurs intending to operate in their Member States. Continental legislators have responded to this phenomenon. As Dr Ringe explains, France, Spain and Germany have introduced deregulated forms of limited liability companies or deregulated their existing forms; the Netherlands is in the process of following suit. In all these cases, the primary characteristics of the changes made appear to be enhanced speed and cost of incorporation, and lower or no minimum capital requirements. This is unsurprising given that, according to a major study, the main drivers behind the incorporation of ‘foreign’ UK private companies are direct and indirect costs of national incorporation procedure.40 One cannot but agree with Dr Ringe that the impact of Centros has been beneficial in terms of inducing national legislators to reduce or eliminate minimum capital requirements, and to reduce set-up time and costs.41 Minimum capital requirements do not serve any useful purpose42 and reductions in costs and time for the setting up of companies must be welcomed. Despite this, a few sobering observations can be made on the effect of Centros thus far. The available evidence suggests that the choice to incorporate in the UK was not actually made on the basis of a preference for the substantive company law of the UK.43 Consequently the legislative response was not aimed primarily at changing substantive aspects of company law. One might thus hesitate to call the result ‘better’ company law. Indeed, as has been argued, ‘there has been little or no sign of high quality legislative or case law reform’.44 Secondly, given the narrow reasons for these UK incorporations, this initial hemorrhaging of incorporations could be significantly halted once the legislative response from these Member States is in place.45 Indeed, it has been claimed that speeding up the incorporation process in Denmark alone led to a twenty-five per cent drop in the use of UK private companies.46 If the UK incorporations were a function of a deeper appreciation of UK company law they would not have been countered with such ease. Thirdly, the response of continental legislators seems to have been aimed primarily at stopping national entrepreneurs from incorporating in the UK and not at attracting foreign nationals to incorporate in their states.47 This suggests a limited appetite for regulatory competition on the part of these Member States. Fourthly, and unsurprisingly perhaps, the survival rate for these foreign UK companies has been low. One study claims that fifty per cent of German limiteds fail within one year and more than ninety per cent  Becht, Mayer and Wagner (n 39 above).  Ringe (n 3 above).   See for example, J Armour, ‘Share Capital and Creditor Protection: Efficient Rules for a Modern Company Law?’ (2000) 63 Modern Law Review 355; J Rickford (ed), ‘Reforming Capital: Report of the Interdisciplinary Group on Capital Maintenance’ (2004) 15 European Business Law Review 919; L Enriques and JR Macey, ‘Creditors Versus Capital Formation: The Case Against the European Legal Capital Rules’ (2001) 86 Cornell Law Review 1165. 43   ‘The evidence supports a simple model of choice of legal form dictated by relative costs of incorporation in different jurisdictions rather than a broader set of non-price considerations.’ Becht et al (n 39 above) 242. 44   WW Bratton, JA McCahery and EPM Vermeulen, ‘How Does Corporate Mobility Affect Lawmaking? A Comparative Analysis’ [2008] ECGI Working Paper 91/2008, 7. 45  Becht et al (n 39 above) 252. See also Bratton, McCahery and Vermeulen (n 44 above) 33. 46  Bratton et al (n 44 above) 30–31. 47   See Bratton et al (n 44 above) 33–35. 40 41 42

Sparking Regulatory Competition in European Company Law: A Response  135 are dissolved after two years of trading.48 This might shed light on the type of company arising out of the post-Centros liberalisation. IV. Assessment and outlook

The effect of Cadbury Schweppes on regulatory competition is, at present, unknown. Even if one were to assume that it will have no effect at all, for full regulatory competition to ensue, the remaining legal barriers must be removed and the other necessary conditions must be in place. Dr Ringe is optimistic on both counts. At present, regulatory arbitrage is only fully available for companies at the incorporation stage. Arbitrage for existing companies is still subject to a number of difficulties. One of the main difficulties is the ability of the state of incorporation to place restrictions on the emigration of companies, as permitted by the well-known Daily Mail case.49 Companies can relocate under the Merger Directive50 or by the use of the Societas Europea, yet this is not ideal. Daily Mail has been criticised, yet the Court did not take the opportunity to comprehensively remove this obstacle in Cartesio.51 Even if this had been done, however, and there were no legal impediments to regulatory arbitrage within the EU, uncertainty remains as to whether the necessary conditions for regulatory competition are in place. Regulatory competition in company law is a wellknown and much-discussed phenomenon in the United States. A number of commentators have looked at the conditions in the US giving rise to regulatory competition and found them lacking in the EU. Dr Ringe recognises some differences between the situation in the US and the EU, but is unconvinced by others. He also argues that ‘many similarities cannot be denied’ and, ultimately, concludes: I believe that we are not at the end of a (limited) regulatory competition process for start-up companies in Europe, but perhaps at the end-point of a first wave. This first wave will hopefully be followed by more to come, and eventually encompass a sea of free reincorporation and a genuine level playing field between the different company law systems that the Member States and the EU itself have on offer.52

The debate on the existence or otherwise of the necessary conditions for regulatory competition to take place in the EU has thrown up a large number of arguments which, due to the nature of this contribution, cannot be revisited here. Without taking sides in this debate, one can still note that the adoption of a distinctly optimistic outlook necessitates addressing, if not refuting, the arguments made by less optimistic commentators.53 48  W Niemeyer, ‘Die “Mini GmbH” (UG) trotz Marktwende bei der Limited?’ [2007] Zeitschrift für Wirtschaftsrecht (ZIP), cited in Bratton et al (n 44 above) 27. 49  Case 81/87 R v HM Treasury and Commissioners of Inland Revenue, ex p Daily Mail and General Trust plc [1988] ECR 5483. 50   Merger Directive 2005/56/EC of 26 October 2005 on cross-border mergers of limited liability companies [2005] OJ L310/1. 51   Cartesio (n 35 above). 52  Ringe (n 3 above). 53   See for example, M Gelter, ‘The Structure of Regulatory Competition in European Corporate Law’ (2005) 5 Journal of Corporate Law Studies 247; and L Enriques, ‘EC Company Law and the Fears of a European Delaware’ (2004) 15 European Business Law Review 1259. John Armour, a generally optimistic commentator, after providing counter-arguments to those made by skeptics, made more qualified claims than those by Dr Ringe. See J Armour, ‘Who Should Make Corporate Law? EU Legislation versus Regulatory Competition’ (2005) 58 Current Legal Problems 369.

136  John Vella Will problems over litigation, concerns about legal advice and resistance to reincorpora­ tion in a different State by incumbent lawyers inhibit regulatory arbitrage and hence competition?54 These are examples of possible problems on the demand side. Regulatory competition also requires a supply side and even here questions arise as to whether Member States will have the necessary incentives to enter this competitive process. Franchise taxes and potential revenue for the local bar are the two most commonly cited incentives in US literature for making incorporations in a state desirable.55 As Dr Ringe acknowledges, franchise taxes are not permitted under EU law. He argues, however, that ‘EU Member States have sufficient incentives for implementing an attractive company law, for reasons relating to political control, as well as remaining attractive to business and securing employment for the legal profession’. It is not clear, without more, why the first two of these three reasons would provide sufficient incentives for Member States to carry out changes to their company law in order to attract incorporations or reincorporations. Different commentators have offered arguments supporting opposing views as to whether the local lawyer lobby is strong enough to drive regulatory competition.56 The matter is certainly not clear-cut. Finally, it is also worth pointing out that when considering whether regulatory competition in company law will take place in the EU, and if it does, what the likely outcome of this process will be, one must also consider the impact of other factors on the process, such as tax and securities laws. When considering where to incorporate or whether to relocate a company, company law will be taken into consideration, though these other considerations will undoubtedly weigh heavily on the decision too. V. Conclusion

Cadbury Schweppes, to paraphrase Advocate-General Poiares Maduro’s words in Cartesio, seems to have significantly qualified Centros.57 This could have a significant effect on the use of letter-box companies for regulatory arbitrage. In particular, it could affect the well-documented arbitrage undertaken by continental entrepreneurs setting up UK private companies. This phenomenon produced a positive result in that some Member States responded by making changes in the direct and indirect costs for incorporating companies. At the same time, however, this legislative response might considerably slow down the arbitrage observed thus far. Non-letter-box companies above a certain size should be able to comply with the additional requirements laid down in Cadbury Schweppes, but care will have to be taken when doing so. Finally, legal impediments to full regulatory arbitrage remain, but more importantly perhaps, uncertainty also remains as to whether the necessary conditions are in place for regulatory competition to take place in the EU.

54  For differing views on such issues see Armour (n 53 above), Gelter (n 53 above), Enriques and Macey (n 42 above) and JC Dammann, ‘Freedom of Choice in European Company Law’ (2004) 29 Yale Journal of International Law 477. 55   Gelter (n 53 above) 259. 56  Armour (n 53 above) and Gelter (n 53 above). 57  AG Poiares Maduro in Cartesio (n 35 above).

10 Abuse of Law in the Context of European Insolvency Law Horst Eidenmüller*

Q

uestions of abuse of law have gained increasing importance in the case law of the Court of Justice.1 From a business law perspective, the Centros jurisprudence of the Court stands out in particular (Centros, Überseering and Inspire Art).2 With this jurisprudence, the door has been opened for free choice of the applicable company law regime and regulatory competition in the area of company law in Europe. At the same time, the Court has made it very clear that the national courts may, on a case-by-case basis, take abuse or fraudulent conduct into account when assessing the reach of freedom of establishment.3 Hence, Member States may apply their own national law to a company established in another Member State if this ensures that abusive behaviour is kept in check. However, the mere fact that a company has no real connection to the Member State in which it was established (apart from its registered office) does not, in the Court’s opinion, constitute abuse.4 Whereas in the field of company law pioneering developments are associated with the Court’s interpretation of the fundamental freedoms, it is a statute that has in the past dominated legal practice and academic discourse in the field of insolvency law and continues to do so: the European Insolvency Regulation (EIR),5 which entered into force on 31 May 2002. Its main objective is to provide a stable framework for the conduct of cross-border insolvency proceedings in Europe. To achieve this objective, the EIR contains rules on jurisdiction for insolvency proceedings, the recognition of decisions

  * This chapter originates from a paper that I presented during a conference on ‘Prohibition of Abuse of Law: A New General Principle of EC Law?’ which was held at the Oxford University Centre for Business Taxation and the Institute of European and Comparative Law on 3 and 4 October 2008 in Oxford University’s Saïd Business School. The original character of the oral presentation has been retained and selected footnotes were added. An earlier version of the paper appeared in the European Company and Financial Law Review. I should like to thank John Armour, Andreas Engert, Heribert Hirte for helpful comments on an earlier draft and Tilman Turck for excellent research assistance. 1  For a discussion of the case law see H Eidenmüller (ed), Ausländische Kapitalgesellschaften im deutschen Recht (Munich, Verlag CH Beck 2004) § 3, paras 73–119. 2   Case C-212/97 Centros v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459; Case C-208/00 Überseering v Nordic Construction Company Baumanagement GmbH (NCC) [2002] ECR I-9919; Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155. The Court’s judgment in the Cartesio case (Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641) has not affected its Centros jurisprudence. In contrast to this jurisprudence, it dealt with an outward-bound shift of a company’s real seat. Moreover, the Court did not address abuse of law issues in Cartesio. 3   Centros (n 2 above), para [25]. 4   Centros (n 2 above), para [26]; Inspire Art (n 2 above), para [139]. 5   Council Regulation (EC) 1346/2000 of 29 May 2000 on insolvency proceedings [2000] OJ L160/1.

138  Horst Eidenmüller with respect to such proceedings, and the coordination of multiple proceedings involving a single debtor. The ECJ meanwhile handed down two judgments on the EIR: the Susanne Staubitz-Schreiber case was decided on 17 January 20066 and the Eurofood case on 2 May 2006.7 Both cases concern jurisdictional issues arising under the regulation, questions of forum shopping, and, as a consequence, are also relevant to abuse of law issues. However, until now there has been no judgment specifically dealing with abuse of law in the context of insolvency law. This is the background against which the following chapter looks at abuse of law as a concept in the insolvency law field. The goals of the chapter are twofold. First, I would like to offer a roadmap that could guide the Court when having to decide an abuse of law case in the area of insolvency law. Hence, I am not so much interested in making sense of the existing Court’s jurisprudence with respect to abuse of law in other fields or in interpreting or clarifying this jurisprudence. I rather aim to sketch the outlines of an appropriate abuse of law concept with a focus on insolvency law questions. Second, I would like to explore a potential reform of the EIR that would spare the Court the difficult task of developing a detailed abuse of law concept designed to apply to insolvency law questions. Could the EIR be changed in a way that would make abuse of law as a concept superfluous or at least less important? The chapter is divided into six parts. In Part I, I will lay out the context of the debate on abuse of law in the field of insolvency law. Part II will deal with abuse of law as a general doctrinal concept. Part III will be on the abuse of freedom of establishment (primary EU legislation) with respect to insolvency issues, and in Part IV, I will discuss abuse of the EIR (secondary EU legislation). Part V will present a proposal to amend the EIR to decrease its susceptibility for potential abuse. Part VI summarises the chapter’s main results and offers a conclusion. The main focus of the chapter will be on companies and not on natural persons. Abuse of law issues in the field of insolvency law are not limited to the former, though. Hence, Parts IV and V will also examine abuse of law questions associated with the insolvency of natural persons. I. THE CONTEXT OF THE DEBATE IN INSOLVENCY LAW

Abuse of law issues with respect to European insolvency law are closely linked to questions of jurisdiction arising under the EIR, forum shopping and regulatory competition. The EIR ties jurisdiction for main insolvency proceedings to a debtor’s ‘centre of main interests’ (COMI).8 There is a refutable presumption stating that the COMI of a company or legal person conforms to its registered office. The COMI was chosen as a criterion because it was considered advantageous to execute main insolvency proceedings at a location to which the creditors are accustomed, where they lived, transacted with the debtor and where most of the debtor’s assets are located—generally speaking: at a place where ‘the business was done’.9 The great importance of the COMI concept stems from the fact that   Case C-1/04 Susanne Staubitz-Schreiber [2006] ECR I-701.   Case C-341/04 Eurofood IFSC Ltd [2006] ECR I-3813.  Regulation 1346/2000 (n 5 above), Art 3(1). 9  See M Virgòs and E Schmit, ‘Report on the Convention on Insolvency Proceedings’, para [75], eg published as annex 2 to G Moss, IF Fletcher and S Isaacs (eds), The EC Regulation on Insolvency Proceedings: A Commentary and Annotated Guide (Oxford, Oxford University Press 2002). 6 7 8

Abuse of Law in the Context of European Insolvency Law  139 insolvency proceedings are subject to the rules of the forum in which they are opened.10 If, for example, main insolvency proceedings over the assets of a German company are opened in England, these proceedings are governed by English insolvency law. Since the early days of the EIR, the COMI concept occupied centre stage with respect to practical importance and scholarly debate.11 COMI is a fact-sensitive criterion, and the facts can be changed or manipulated. It took some time until the Court of Justice received an opportunity to clarify some issues surrounding the use of the COMI concept. In its Eurofood decision, the Court held that: The scope of that concept is highlighted by the 13th recital . . . which states that ‘the “centre of main interests” should correspond to the place where the debtor conducts the administration of his interests on a regular basis and is therefore ascertainable by third parties’. That definition shows that the centre of main interests must be identified by reference to criteria that are both objective and ascertainable by third parties. That . . . [is] necessary in order to ensure legal certainty and foreseeability concerning the determination of the court with jurisdiction to open main insolvency proceedings.12

The Court’s emphasis on ascertainability, certainty and foreseeability is clearly driven by the desire to restrict forum shopping and regulatory competition in bankruptcy law. Nevertheless: even ‘objective and ascertainable’ facts can be changed and/or manipulated, and it appears that the intensity of forum shopping has not decreased after the Eurofood decision.13 The United Kingdom in particular has established itself as a popular venue for the restructuring of failing enterprises, apparently because of the attractive features of the ‘Company Voluntary Arrangement’ (CVA) procedure.14 Cases such as DNick Holding plc and Schefenacker plc, in which big German companies were first transformed into English ones and then restructured in London, are not rare exceptions. A rising number of companies in distress, their creditors and advisors seem to have a different view on forum shopping from that of the Court. Is forum shopping and regulatory competition in insolvency law in Europe a problem? From an economic point of view, the answer to this question depends on goals that (international) insolvency rules are supposed to achieve.15 I submit that three such objectives are important. First, insolvency law rules should strive to maximise the net assets available to satisfy the creditors’ claims. A firm should be liquidated only if its liquidation value is greater than its reorganisation value. Reorganisation procedures should be designed so that this filtering function is correctly performed.16 A corollary of the aim to maximise the net assets available to satisfy creditors’ claims is that bankruptcy costs should be minimised. Procedures should be as speedy and simple as possible. Discrepancies between  Regulation 1346/2000 (n 5 above), Art 4.  See B Wessels, ‘The Place of the Registered Office of a Company: A Cornerstone in the Application of the EC Insolvency Regulation’ (2006) 3 European Company Law 183; K Pannen (ed), European Insolvency Regulation (Berlin, de Gruyter 2007) Art 3, paras 15–85. 12   Eurofood (n 7 above), paras [32–33]. 13  See the cases discussed in Parts II–IV of this chapter. 14  For details about the new CVA see IF Fletcher, ‘UK Corporate Rescue: Recent Developments—Changes to Administrative Receivership, Administration, and Company Voluntary Arrangements—The Insolvency Act 2000, The White Paper 2001, and the Enterprise Act 2002’ (2004) 5 European Business Organization Law Review 119, 130–33. 15  On these issues see H Eidenmüller, ‘Free Choice in International Company Insolvency Law in Europe’ (2005) 6 European Business Organization Law Review 423, 429–30. 16  On this issue see H Eidenmüller, Unternehmenssanierung zwischen Markt und Gesetz (Cologne, Verlag Dr Otto Schmidt 1999) 31–48. 10 11

140  Horst Eidenmüller the applicable insolvency and company law should be avoided. Such discrepancies create frictions in the insolvency administration and these translate into costs. Second, insolvency rules should contribute to fostering the efficiency of credit contracts. This efficiency is primarily a function of the correct pricing of credit risk. Of crucial importance here is the foreseeability of the bankruptcy forum and the applicable bankruptcy law. Hence, in emphasising the ‘legal certainty and foreseeability’ of the bankruptcy forum, the Court stresses a point that is of special economic relevance. Third, insolvency rules should strive to protect involuntary and other ‘non-adjusting’ creditors.17 Involuntary creditors such as tort creditors do not bargain for a claim and hence cannot protect themselves. ‘Nonadjusting’ creditors are creditors that, in principle, could bargain for protection but fail to do so because they lack the skills, information, or other resources, or who abstain from self-protection because they consider it to be uneconomical to invest resources to achieve that aim (eg because they only have very small claims). How does the COMI concept of the EIR fare against these criteria? On a theoretical level, arguments can be advanced both for positive and negative effects. On the one hand, the available reorganisation procedures in the UK may be particularly efficient and speedy, attracting failing companies who seek fast and economical restructuring. On the other hand, forum shopping on the eve of bankruptcy also causes some serious problems: discrepancies between the applicable company and insolvency law trigger (additional) bankruptcy costs (think of English insolvency proceedings over the assets of a German company), risk-inadequate credit contracts might be concluded and non-adjusting creditors exploited.18/19 Empirical evidence on the overall economic effects of insolvency forum shopping in Europe is missing. There are studies on bank recovery rates suggesting that these are high in the UK, lower in Germany, and even lower in France.20 However, these findings only tell us something about how banks fare in English insolvency proceedings, and they are not helpful for assessing the overall net assets available to satisfy creditors’ claims. Even more importantly, the effects of forum shopping on credit contracts and credit markets are not reflected in these results. 17  The term ‘non-adjusting’ creditors has been introduced in the literature by Bebchuk and Fried. See L Bebchuk and J Fried, ‘The Uneasy Case for the Priority of Secured Claims in Bankruptcy’ (1996) 105 Yale Law Journal 857, 882–91; L Bebchuk and J Fried, ‘The Uneasy Case for the Priority of Secured Claims in Bankruptcy: Further Thoughts and a Reply to Critics’ (1997) 82 Cornell Law Review 1279, 1295–1304. 18  It surely is debatable whether the exploitation of involuntary or other ‘non-adjusting’ creditors is a big risk/problem empirically in the European context. The possibility to initiate territorial insolvency proceedings in particular (Art 3(2)–(4) EIR) and the relevance of the law governing an employment contract also for insolvency issues (Art 10 EIR) seem to afford some protection. However, the former protection depends on an establishment being retained in the Member State of the ‘old’ COMI. In any event, the theoretical exploitation risk exists, and no studies have been undertaken yet to measure it empirically. 19  All in all, I believe that WG Ringe, ‘Forum Shopping under the EU Insolvency Regulation’ (2008) 9 European Business Organization Law Review 579, 602, underestimates the problems of forum shopping associated with the COMI standard. He writes: A company’s migration does not necessarily have to take place in the vicinity of insolvency. I would argue therefore that a creditor is never able to predict which insolvency law his claim will be subject to—as long as the Internal Market consists of different substantive insolvency regimes. At the least, it is the creditor’s responsibility to take protection against this type of risk as it is against every other type of business risk.

To be sure, COMI changes do occur also in the normal course of business. But opportunistic COMI changes on the eve of bankruptcy aggravate the problem: creditors expect the worst, and that may have very bad consequences for credit markets. 20   Cf S Davydenko and J Franks, ‘Do Bankruptcy Codes Matter? A Study of Defaults in France, Germany and the UK’ [2006] European Corporate Governance Institute—Finance Working Paper No 89/2005: ssrn.com/ abstract=647861.

Abuse of Law in the Context of European Insolvency Law  141 The situation in Europe is to some extent comparable to the situation in the United States where bankruptcy forum shopping has been around for many years.21 Delaware was the preferred bankruptcy venue until the end of the twentieth century when New York took over. Firms reorganised in Delaware show higher refiling rates than firms reorganised elsewhere (which is bad),22 but Delaware proceedings are faster than other proceedings (which is good).23 Hence, the available empirical evidence from the United States with respect to bankruptcy forum shopping is mixed. To conclude: the context of the debate on abuse of law in the insolvency field in Europe is characterised by forum shopping and regulatory competition with unclear overall economic effects. No conclusive empirical data exists on the total efficiency effects of forum shopping in the insolvency field. Both positive and negative consequences can be identified on a theoretical level, but the evidence to pass a final judgment on the phenomenon is lacking. II. ABUSE OF LAW AS A GENERAL CONCEPT

As already mentioned at the beginning of this chapter, the Court of Justice has recently handed down a series of judgments that are particularly important with respect to the issue of abuse of law as a general doctrinal concept.24 In European company law there are the cases Centros, Überseering and Inspire Art,25 and in European tax law there are the cases Halifax and Cadbury Schweppes.26 As mentioned above, the cases Susanne Staubitz-Schreiber and Eurofood in the insolvency law field touch issues bordering on abuse of law questions, but they do not discuss the concept directly. If one wishes to explore whether abuse of law is a general (European) legal concept and what the elements of this concept are, the existing case law of the Court is of only limited help. The Court must decide case by case. Moreover, its case law is characterised by the attempt to do justice to various fact patterns that arise in different areas of the law and are relevant for different fundamental freedoms. These circumstances are not conducive to 21  For an influential monograph on the topic see LM LoPucki, Courting Failure (Michigan, The University of Michigan Press 2005). 22  LM LoPucki and SD Kalin, ‘The Failure of Public Company Bankruptcies in Delaware and New York: Empirical Evidence of a “Race to the Bottom” ’ (2001) 54 Vanderbilt Law Review 231; LM LoPucki and JW Doherty, ‘Why are Delaware and New York Bankruptcy Reorganizations Failing?’ (2002) 55 Vanderbilt Law Review 1933. 23   KM Ayotte and DA Skeel, ‘Why Do Distressed Companies Choose Delaware? An Empirical Analysis of Venue Choice in Bankruptcy’ [2003] University of Pennsylvania Law School: Scholarship at Penn Law, Paper 20; RK Rasmussen and RS Thomas, ‘Whither the Race? A Comment on the Effects of the Delawarization of Corporate Reorganizations’ (2001) 54 Vanderbilt Law Review 283, 295–97. 24  On the question whether it even constitutes a general principle of European law, see R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 436–39, 440. 25  See n 2 above. Further, there are the ‘older’ cases Pafitis, Kefalas and Diamantis on the Second Directive: Case C-441/93 Panagis Pafitis v Trapeza Kentrikis Ellados AE [1996] ECR I-1347; Case C-367/96 Alexandros Kefalas et al v Elliniko Dimosio and Organismos Ikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I-2843; Case C-373/97 Dionysios Diamantis v Elliniko Dimosio, Organismos Ikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [2000] ECR I-1705. 26  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd and County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, paras [68]–[86]; Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995, paras [55]–[75].

142  Horst Eidenmüller developing an abuse of law concept ‘top down’ and in a general fashion. Hence, it comes as no surprise that the Court’s jurisprudence on abuse of law poses more questions than it provides clear-cut answers. First, the Court is unclear about the basis (the ‘location’) of the concept: does it derive from European law? Or is it a concept of the Member States’ laws? Second, there is no clear distinction drawn between abuse of law and fraud. In many cases, the Court seems to confuse the two concepts.27 Third, the Court is unclear on whether a subjective element is necessary for abuse of law to be ascertained, and if so, what sort of subjective element is required. Fourth and finally, the Court does not spell out precisely the consequences of abuse, once it has been ascertained. In the Centros judgment, for example, the ECJ held that Member States are justified in preventing abuse by ‘denying’ someone the benefit of the provisions of Community law on which he or she seeks to rely.28 What does this mean exactly? Hence, it is advisable to keep some distance from the Court’s jurisprudence when thinking about abuse of law as a general European legal concept. To begin with, the distinction between abuse of law and fraud is critical. Abuse of law is about correctly applying the law to a specific set of facts, whereas fraud is about correctly determining the facts to which the law then is applied.29 For instance, the Brochier cases that were litigated in European courts in recent years were fraud cases and not abuse of law cases.30 Brochier was an English limited company that clearly had its COMI in Nuremberg in Germany. However, an initial petition was made for the opening of main insolvency proceedings in London. That petition was supported by a distorted set of facts presented to the High Court.31 The facts were twisted to create the impression that Brochier’s actual head office was in London, whereas on an undistorted and complete set of facts there could be no doubt that the company’s COMI was in fact in Nuremberg. With respect to the elements of the abuse of law concept, there should be two requirements. First, there is only an abuse of law if the law is used according to its formal wording but contrary to its purpose.32 This implies that abuse of law in essence is about the proper interpretation of a particular legal provision. Such a proper interpretation does not stick to the wording of a rule but also takes into account its rationale. Abuse of law, therefore, is an interpretative concept, not a self-standing principle that is independent from the purpose of the provision in question.33 Second, abuse of law requires that the meeting of the formal requirements of a provision goes back to a deliberate action of  See for example, Centros (n 2 above), paras [24–25].   Centros (n 2 above), paras [24–25]. 29  This is also emphasised by H Fleischer, ‘Der Rechtsmissbrauch zwischen Gemeineuropäischem Privatrecht und Gemeinschaftsprivatrecht’ [2003] Juristenzeitung 865, 870. Contrast A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of The European Court of Justice on Circumvention, Fraud and Other Misuses of Community Law’ [2000] European Business Law Review 179, 180, who maintains that fraud can be conceived as a special form of abuse. 30   Hans Brochier Holdings Ltd v Exner [2006] EWHC 2594; Hans Brochier Holdings Ltd v Exner, High Court of Justice London, Order of 8 December 2006 [2007] Neue Zeitschrift für das Recht der Insolvenz und Sanierung 187; Amtsgericht Nuremberg, Resolution of 15 August 2006 [2007] Zeitschrift für Wirtschaftsrecht 81; Amtsgericht Nuremberg, Resolution of 1 October 2006 [2007] Zeitschrift für Wirtschaftsrecht 83. 31   Hans Brochier (n 30 above). 32  This is also stressed by W Schön, ‘Abuse of Rights and European Tax Law’ in JA Jones, P Harris and D Oliver (eds), Comparative Perspectives on Revenue Law: Essays in Honour of John Tiley (Cambridge, Cambridge University Press 2008) 75, 79–82. If the main purpose of a particular rule is to achieve legal certainty, then one has to look at other (secondary) purposes and ask the question whether the person relying on a particular legal position does so in contradiction to these purposes. 33  See A Teichmann, Die Gesetzesumgehung (Göttingen, Verlag Otto Schwartz & Co 1962) 64–65. However, Teichmann maintains that there are very few cases that cannot be solved by interpretation. 27 28

Abuse of Law in the Context of European Insolvency Law  143 the user. There must be some manufacturing of the facts—as opposed to the deception about facts—for an abuse of law to arise. What is not necessary, though, is a specific ‘abuse intention’ of the user—ie it is not necessary that the user has deliberately or at least knowingly used the law for a purpose which it has (objectively) not been intended to serve. All that is required is that the user establishes the formal elements of a legal provision and objectively makes use of it for a purpose that does not conform to its rationale. The interpretative character of the abuse of law concept has a bearing on the question whether the concept is part of EU law or part of the laws of the Member States. The answer to this question depends on the particular legal provision that has allegedly been abused. If abuse of EU law is the issue, the concept must be a European one. Conversely, if abuse of a legal provision of a Member State is the issue, the concept belongs to the law of that Member State. However, the EU and the national level are interrelated when it comes to European Directives. Technically speaking, the abuse of a legal provision of a Member State implementing a particular provision in a directive is a question of the law of that Member State. In assessing the purpose of the legal provision in question, however, the directive and its purpose may come into play: national laws implementing EU Directives must be interpreted in conformity with the purpose(s) of the directive.34 Finally, abuse of law is in principle not restricted to subjective entitlements (‘claims’) and private actors. The scope of the concept is broader. Although most of the relevant cases will centre around subjective entitlements being abused by private actors, there might be cases where other types of legal provisions are abused and where it is not a private actor but a public entity that engages in abusive conduct. Those cases are not only conceptually conceivable; they are, as we will see, potentially relevant especially in the field of insolvency law. Summing up the discussion on abuse of law as a general concept, we can conclude that it is an interpretative concept on the scope of a particular legal provision. Its two main elements are: (i) use of law contrary to its purpose and (ii) a deliberate action of the user in the establishment of the formal requirements of the provision in question. A specific ‘abuse intention’ is not required. Insofar as abuse of European law is the issue, the abuse of law concept is necessarily a European concept. III. ABUSING FREEDOM OF ESTABLISHMENT

It has already been emphasised that in the insolvency field, both freedom of establishment (Articles 49 and 54 TFEU—ex Articles 43 and 48 EC Treaty) and the EIR play a critical role with respect to abuse of law issues. Also, as we have just noted, abuse of law is an interpretative concept, at the heart of which lies the purpose of a particular legal provision. Hence, it is sensible to address separately abuse of law questions with respect to freedom of establishment on the one hand, and the EIR on the other hand. In this Part of the chapter, I will consider the former, while the next Part will discuss the latter. In the field of company law, two crucial questions arise with respect to the influence of Articles 49 and 54 TFEU [ex Articles 43 and 48 EC Treaty] on the options available to entrepreneurs. One can ask whether setting up a company in another Member State without there being any further connections to it apart from the company’s registered  See in general, W Brechmann, Die richtlinienkonforme Auslegung (Munich, Verlag CH Beck 1996).

34

144  Horst Eidenmüller office constitutes an abuse of freedom of establishment (the ‘Centros question’). A connected question is whether freedom of establishment forces the Member States to recognise a ‘free choice’ of the applicable company law. Similar questions can be asked in the insolvency law field. A.  Abusing Freedom of Establishment by COMI Shifts As a starting point, it is not doubtful that if a company shifts its COMI from one Member State to another, this activity is protected under the freedom of establishment. That is also true if the COMI shift is undertaken with an eye to the applicable insolvency regime. If a company moves its real seat or actual head office from Member State A into Member State B, this will always go hand in hand with a significant change in the regulatory environment affecting the company. A move that is motivated by a more attractive regulatory environment is commonplace. The decisiveness of the COMI for the applicable insolvency law reflects the consideration that it is at the COMI where the creditors are best protected. As a general principle, then, there is nothing wrong with COMI shifts undertaken with the new applicable insolvency regime in mind. However, exceptional circumstances are conceivable in which a COMI shift might be judged to amount to an abuse of freedom of establishment. If, for example, such a shift is undertaken exclusively in order to harm a (legal) person’s creditors as a group, it might be considered abusive. For all practical purposes, however, nothing turns on this judgment. As we shall see in Part IV, the EIR can and does take a more ‘critical’ stance with respect to COMI shifts. In order to achieve a higher level of creditor protection, COMI shifts are controlled—on the basis of the abuse of law concept—in a greater variety of circumstances. Hence, if the EIR is abused, this does not imply that there is a simultaneous abuse of freedom of establishment. Conversely, in all cases in which one could argue that freedom of establishment is abused, there will also be an abuse of the EIR. B.  Free Choice of the Applicable Insolvency Regime A more interesting question regarding the influence of freedom of establishment on the applicable insolvency regime is whether private actors might have a right to freely choose that regime. After all, this is the state of affairs in company law. An entrepreneur who is based in Member State A can freely choose to incorporate in Member State B to conduct his business activities, even though there is no material connection with Member State B apart from the formal act of the registration of the company. In essence, this amounts to a right to freely choose the company law regime applicable to a particular business. Hence, it is a sensible question to ask whether companies also have a right under Articles 49 and 54 TFEU [ex Articles 43 and 48 EC Treaty] to freely choose the insolvency law regime applicable in case of financial distress. This question seems all the more relevant as the boundaries between company and insolvency law are blurry, especially with regard to creditor protection issues. If, as a starting point, freedom of establishment protects COMI shifts, it is clear that there is a restriction on freedom of establishment if the company faces a new insolvency

Abuse of Law in the Context of European Insolvency Law  145 regime after a COMI shift. At the same time, the protection of workers and creditors are imperative requirements in the general interest that, in the Court’s jurisprudence, can justify such restrictions.35 These requirements are affected most strongly at the debtor’s COMI. Hence, the application of the insolvency law of the jurisdiction in which the COMI is located serves an imperative requirement in the general interest. Freedom of establishment, therefore, does not mandate a ‘free choice’ of the applicable insolvency regime.36 We can conclude Part III by stating that shifting the COMI of a company to another Member State is protected under the freedom of establishment and is in principle not abusive. However, freedom of establishment does not mandate a ‘free choice’ of the applicable insolvency regime—ie companies do not enjoy a right to pick the applicable insolvency law that best suits their needs irrespective of where the company’s real seat is located. IV. ABUSING THE EUROPEAN INSOLVENCY REGULATION

As the EIR is the legal framework for the administration of cross-border insolvencies in Europe, and as jurisdictional issues are dealt with by the EIR, abuse of law questions in the insolvency field arise primarily under the EIR. The phenomena to be explored further are COMI shifts. It has already been noted that abusing the EIR is an issue of real— not faked—COMI shifts (faked COMI shifts would be a fraud issue). COMI shifts are especially troublesome if they are not accompanied by a reincorporation and a change in the applicable company law. If a COMI shift is associated with a reincorporation under the Tenth Directive on cross-border mergers, the creditors of the participating companies receive protection under the applicable Member States’ rules (see Article 4(2) of the Directive).37 ‘Isolated’ COMI shifts without a change in the applicable company law are problematic for another reason: if a company thereafter is restructured in the Member State where the new COMI is located, a discrepancy between the applicable company and insolvency law results. This creates frictions and increases bankruptcy costs. Hence, the focus of the following will be on such ‘isolated’ COMI shifts without an associated change in the applicable company law.38   Überseering (n 2 above), para [92]; Inspire Art (n 2 above), para [135].  See on this issue, Eidenmüller (n 15 above) 446. Contrast Ringe (n 19 above) 613. However, Ringe does not sufficiently take into account the protection of workers and creditors in interpreting freedom of establishment and its reach. I agree that abuse of law can only be established on a case-by-case basis according to the Court of Justice. However, Member States are entitled to pass general laws in order to protect imperative requirements in the public interest (see only Inspire Art (n 2 above), para [133]). 37  To be sure, a change in the applicable company law can be achieved without making use of the mechanism of the Tenth Directive. As an example, consider an English limited company that joins a German partnership as a limited partner. Thereafter, the general partner and all other limited partners of the German partnership withdraw from it. The consequence of this is that the English limited company succeeds to all assets and liabilities of the firm as the only remaining partner and the German partnership vanishes. Such mechanisms are sometimes used in practice as, for example, in the Brochier case mentioned earlier in the text (n 30 above). If this is done, the creditor protection devices of the Tenth Directive obviously do not apply. However, this transaction only works if all shareholders agree on it. 38   Just take the case of a German GmbH or AG that tranfers its real seat and its COMI abroad: even under the real seat theory, it stays a German company (see Cartesio (n 2 above), paras [109]–[113] on the real seat theory and Arts 43, 48 EC Treaty—now Arts 49 and 54 TFEU), as long as it moves to a country like the UK or Ireland that applies the incorporation theory (which will lead to German law being applicable as the law of the statutory seat). German corporate law only requires a statutory seat, not a real seat, in Germany (section 4a GmbHG and section 5 AktG). 35 36

146  Horst Eidenmüller The contractual means at the disposal of creditors to prevent COMI shifts are limited. Meanwhile, syndicated loan agreements sometimes contain clauses with which the debtor represents that its COMI is in a particular Member State.39 However, such a clause refers only to the status quo and does not prevent a COMI shift. More effective in this respect is a covenant that requires the debtor to seek the assent of the lenders with respect to a planned COMI shift. Failure to do so results in an acceleration of the underlying payment obligation. However, such a loan covenant is found only in economically significant lending transactions. For trade creditors, it is normally simply uneconomical to seek and draft detailed covenants to obtain an enhanced level of protection. For them, retention of title clauses are a much more common and sensible means of self-help. Moreover, ‘anti-COMI-covenants’ certainly do not deter debtors that are determined to effectuate a COMI shift in order to benefit at the expense of their creditors. Hence, COMI shifts are a problem that creditors cannot solve on their own conclusively. A.  Goals of the Regulation and Prerequisites for Abuse If the principal requirement for an abuse of law is that a particular legal provision is used contrary to its purpose, the main task with respect to abuse of law questions in the context of the EIR is to identify precisely the goals of the Regulation. Two important goals are stated explicitly in the Regulation’s recitals. In Recitals 2, 8, 16, 19 and 20, the efficient and effective administration of cross-border insolvencies is emphasised. That goal conforms to the aim of maximising the net assets to satisfy creditors’ claims that was mentioned in Part I. Further, in Recital 4 the EIR explicitly states that forum shopping should be prevented. This goal conforms to the economic aim of fostering the efficiency of credit contracts, which was also mentioned in Part I. It is obvious that these two goals may conflict with each other. Under certain circumstances, an efficient and effective administration of a cross-border insolvency might require a change in the COMI. Imagine, for example, the simultaneous financial distress of many members of a corporate group. In such a situation, all insolvency proceedings over all the assets of all members of the corporate group should be administered in the same jurisdiction and possibly even by the same insolvency court and/or administrator.40 This way, it is much easier to coordinate the proceedings as opposed to a situation where different courts of different Member States and different insolvency administrators are in charge with respect to the various members of the corporate group. If, as a consequence, a COMI shift is undertaken with respect to some corporate group members, that will work against the goal of fostering the efficiency of credit contracts, however. Hence, the question that eventually must be answered is: which of the two goals of the EIR is more important? In my view, it is the first goal. There are a number of arguments that support this position. First, the EIR has deliberately chosen a changeable criterion with respect to 39  A clause in a loan document on file with the author reads as follows: ‘For the purposes of “The Council of the European Union Regulation No 1346/2000 on Insolvency Proceedings”, [the debtor’s] centre of main interest (as that term is used in Article 3(1) of the Regulation) is situated in England and it has no “establishment” (as that term is used in Article 2(h) of the Regulation) in any other jurisdiction.’ 40   H Eidenmüller, ‘Verfahrenskoordination bei Konzerninsolvenzen’ (2005) 169 Zeitschrift für das gesamte Handels- und Wirtschaftsrecht 528, 537–42, 560–63; H Hirte, ‘Towards a Framework for the Regulation of Corporate Groups’ Insolvencies’ (2008) 5 European Company and Financial Law Review 213, 218–21.

Abuse of Law in the Context of European Insolvency Law  147 jurisdiction for main insolvency proceedings, namely the COMI.41 Hence, it is both difficult and probably not sensible to make the COMI ‘stickier’ by interpreting it in a way that would make changes much more difficult. Second, the EIR does not establish a suspect period (ie it does not contain a rule to the effect that a former COMI remains decisive if the insolvency petition is filed shortly after a COMI shift). If such a suspect period existed in the EIR, it could be conceived as an ‘institutionalised anti-abuse rule’. The absence of such a rule indicates that the prevention of forum shopping is not the paramount goal of the EIR.42 Third, preventing forum shopping by applying the abuse of law concept introduces legal uncertainty into the handling of jurisdictional issues, and that—from an economic perspective—translates into costs. Fourth and finally, even though forum shopping is associated with undesirable economic effects, it is not the case that the insolvency regimes of the Member States are so different that one would have to fear that certain Member States might be attractive as an insolvency venue because they offer unique and extensive opportunities for the exploitation of certain creditors. Hence, preventing forum shopping is not the paramount goal of the EIR. The efficient and effective administration of a cross-border insolvency and thus maximising the net assets to satisfy creditors’ claims are more important. This has immediate consequences for the question under which circumstances a COMI shift undertaken to change the jurisdiction for main insolvency proceedings can be judged to be abusive. Everything turns on whether the COMI shift contributes to maximising the net assets available to satisfy creditors’ claims. If it does, it surely cannot be considered abusive. If it does not, it can, at least in principle. Given that the effect on the debtor’s net assets will be difficult to judge in many cases, abuse should be reserved for those cases where it is evident that this effect is not positive. Even then, it will often not be easy to determine whether this criterion is fulfilled. Hence, other factors must be considered with respect to the motives driving the COMI shift. COMI shifts that evidently benefit either the debtor at the expense of its creditors or some creditors at the expense of others are suspicious. Such COMI shifts are driven by distributive considerations and not by the goal of maximising the net assets available to satisfy creditors’ claims. If a COMI shift is evidently effected in order to enrich the person(s) initiating it at the expense of other stakeholders, the shift is abusive. Speaking very crudely, it all comes down to the question of whether a COMI shift is driven by considerations of efficiency (no abuse of law) or considerations of claiming value, ie distributive concerns (abuse of law). To conclude this section one may state that COMI shifts that evidently do not contribute to maximising the debtor’s net assets are abusive. This will be the case especially if a COMI shift is evidently only undertaken to benefit the debtor at the expense of its creditors or some creditors at the expense of others.

41  See also MP Weller, ‘Die Verlegung des Center of Main Interest von Deutschland nach England’ [2008] Zeitschrift für Unternehmens- und Gesellschaftsrecht 835, 850. 42  Suspect periods of the kind described in the text are in place in Italy and Spain (Cf Art 9(2) no 4 Legge Fallimentare and Art 10 no 1 Ley Concursal). In France, insolvency courts have apparently demanded that the debtor’s COMI be in France for about half a year, see G Hölzle, ‘Wege in die Restschuldbefreiung und Schuldenerlass im Exil—Oder: Lohnt die Flucht nach Frankreich wirklich?’ [2007] Zeitschrift für Verbraucherund Privatinsolvenzrecht 1, 4. Insofar as these provisions are meant to be applied also to cross-border COMI shifts, they cannot be reconciled with Art 3(1) EIR, ie they violate the Regulation and are to be disregarded.

148  Horst Eidenmüller B.  COMI Shifts Before the Insolvency Petition COMI shifts can be undertaken before or after the insolvency petition. The focus of this section will be on the former case, whereas the next section will address the latter. A further distinction can be made between COMI shifts by companies and shifts that are undertaken by natural persons. i. Companies The application of the general principle developed in the last section to COMI shifts effectuated by companies before the insolvency petition can be neatly illustrated by two recent cases that were decided by German courts. The first case concerns an inwardbound COMI shift43 and the second one an outward-bound COMI shift.44 In the first case, the ‘PIN Group’ holding company shifted its COMI from Luxembourg to Cologne for restructuring purposes. The PIN Group is a German enterprise with most of the operating companies having their registered office in Germany. The group is engaged in the mail delivery business. The holding company apparently had its registered office and its COMI in Luxembourg for financial reasons. The COMI shift occurred on the eve of insolvency, and the purpose of it was to facilitate the restructuring of the group by coordinating the proceedings over all of the group’s subsidiary companies. This was much easier with all proceedings being controlled by the same insolvency court. The Amtsgericht in Cologne held that even though the COMI shift occurred in the immediate vicinity of the insolvency petition, it was not abusive because it served the restructuring objective and was undertaken for that purpose. In this case, the COMI shift clearly was in the interests of all creditors of the PIN Group companies because it made the effective coordination of the proceedings easier and thus helped maximise the net assets available to satisfy the creditors’ claims. Hence, the COMI shift was in conformity with the main goal of the EIR. In the second case, the debtor company shifted its COMI from Germany to Spain on the eve of a bankruptcy petition by one of the (German) creditors in order to ‘bury’ the company there. The company’s owners apparently hoped that insolvency proceedings in Spain would not be initiated, or at least not opened, so that the company could be dissolved outside such proceedings and without court supervision. The company effectuated the COMI shift by appointing a new manager who was based in Spain. The Bundesgerichtshof decided that the COMI shift was abusive and therefore would not be recognised. Clearly, the shift was not triggered by the goal of maximising the net assets available to satisfy the creditors’ claims. Its only purpose was to transfer value from the creditors to the debtor and its owners—ie to enrich the latter at the expense of the former. Hence, the Court was correct in finding that the COMI shift was abusive. In fact, in this case there was not only an abuse of Article 3(1) EIR, but also an abuse of the freedom of establishment guaranteed by the EC Treaty [now the TFEU]. It is clearly not the purpose of Articles 49 and 54 TFEU [ex Articles 43 and 48 EC] to afford protection to a new primary establishment in another Member State which is set up only to prevent the enforcement of creditor rights.  Amtsgericht Cologne, Resolution of 19 February 2008 [2008] Zeitschrift für Wirtschaftsrecht 423.  Bundesgerichtshof, Resolution of 13 December 2007 [2008] Entscheidungen zum Wirtschaftsrecht 181.

43 44

Abuse of Law in the Context of European Insolvency Law  149 If—as in this case—an abuse of law with respect to a COMI shift prior to the insolvency petition can be ascertained, the ‘old’ COMI stays decisive on a purposive reading of the EIR. Jurisdiction to open main insolvency proceedings thus remains with the courts of the Member State in which the COMI was located before the abusive shift. What if months or even years pass after a COMI shift before an insolvency petition is filed? Denying jurisdiction at the new COMI is only justified if, on the facts of the particular case, the move and the insolvency petition are part of a strategic maneuver: the user establishes the formal elements of a legal provision and objectively makes use of it for a purpose that does not conform to its rationale. A judgment is required to ascertain this. However, that should not be too difficult in most situations in which the issue of abuse of law comes up at all: situations where the COMI shift and the insolvency petition appear to be part of a plan carried out in a short time span, as in the case decided by the Bundesgerichtshof. The two cases discussed in this section highlight the general principle developed in the previous one: COMI shifts undertaken before the insolvency petition in order to maximise restructuring value are not abusive; COMI shifts undertaken to transfer value from the creditors to the debtor do not contribute to maximising the debtor’s net assets and are abusive. ii.  Natural Persons COMI shifts before the insolvency petition can also occur with respect to natural persons. The COMI of individual entrepreneurs corresponds to their professional domicile. A consumer’s COMI coincides with his or her habitual residence. Both criteria can be changed quite easily and quickly: the individual simply has to move to another country with the intent to stay there permanently. As in the previous section, the focus here will again be on real—as opposed to faked—COMI shifts. There are various service providers who offer ‘professional’ help to those who wish to establish themselves in another Member State.45 France in particular has become very popular lately as a destination for forum shoppers. This may have something to do with its agreeable living conditions. In the insolvency context, however, another factor certainly is much more important: the discharge regime in operation under French insolvency law. Most jurisdictions worldwide give natural persons an opportunity to obtain a discharge from their debts by means of going through insolvency proceedings. In Germany, for example, such discharge can be obtained after a period of six years. Within that time, disposable income of the insolvent person above a certain retainable threshold is made available for the satisfaction of the creditors via a trustee.46 The system in France, particularly in the départements of Haut-Rhin, Bas-Rhin and Moselle, is much less rigid. In principle, a discharge is the immediate consequence of the ending of insolvency proceedings over the assets of a natural person.47 Given that such proceedings are handled within months and normally do not last longer than approximately a year,48 a discharge can be obtained in a much shorter time than in Germany.  See for example, the following websites: www.frankreich-insolvenz.com; www.sda-europe.de.  See sections 286–303 Insolvenzordnung.   Cf Arts L 622-32 and L 628-1 Code de Commerce. The insolvency law applicable in the above-mentioned regions differs from general French insolvency law so that the Code de Commerce is also applicable to natural persons that do not run a business. 48  See for example, Hölzle (n 42 above) 5. 45 46 47

150  Horst Eidenmüller The overall economic effects of shorter discharge periods are unclear.49 On the one hand, shorter discharge periods stimulate entrepreneurship, which is a positive economic effect. Also, such shorter periods contribute to preserving human capital because individuals are less prone to seek revenues in the shadow economy and more inclined to utilise and develop their talents ‘officially’. On the other hand, the prospect of shorter discharge periods may encourage entrepreneurs to engage in excessive risk-taking and/ or unprofessional management, which is a negative effect. Whether the positive effects outweigh the negative ones on an empirical level is an open question. However, with respect to applying the general principle developed in Part IV.A, the first two effects are irrelevant: the individual in question now is in financial distress, and any ex ante effects that might have been at work long before the onset of the crisis now no longer play a role. The only relevant effect now is the third one, the preservation of human capital. However, that effect clearly does not contribute to ‘enlarging the pie’ for the benefit of the creditors but is predominantly, if not exclusively, of advantage to the debtor. If that were otherwise, it would be hard to explain why creditors should forcefully object to a COMI shift of natural persons before the insolvency petition, which is what they regularly and unanimously do in practice. Hence, a COMI shift effectuated by a natural person before an insolvency petition in order to obtain a more favourable discharge regime evidently is not in the interests of that person’s creditors. It does not contribute to maximising the net assets to satisfy creditors’ claims. It thus is abusive. On the other hand, a COMI shift undertaken by a natural person before an insolvency petition for reasons unrelated to the imminent insolvency situation is unobjectionable. A distinction between these two scenarios can be made only on the basis of the fact pattern of the particular case. This pattern is relevant for judging whether the COMI (habitual residence) was shifted at all, and if so, whether the shift was abusive. A pertinent factor is the timing of the move: how close to the insolvency petition did it take place? And for how long after the insolvency petition was the new COMI maintained? Another relevant factor is whether the individual in question sought and/or found a new occupation/employment in the Member State into which he or she immigrated. Further, one has to consider whether assets have been shifted into the new Member State or not. One also must look at plausible other economic benefits that might have motivated the move (tax, health care, living expenses, etc).50 Finally, the magnitude of differences in the discharge provisions of the old and the new insolvency jurisdiction may play a role. Based on all these and possibly other factors surrounding the specific scenario, it must be determined whether the COMI was actually shifted at all, and if it was, whether the shift was effectuated primarily in order to obtain a more favourable discharge regime. If the habitual residence of the individual in question was not changed, or if a change was made predominantly to benefit from laxer discharge provisions, the ‘old’ COMI still governs jurisdictional issues.

49   Cf TH Jackson, ‘The Fresh-Start Policy in Bankruptcy Law’ (1985) 98 Harvard Law Review 1393; BE Adler, E Polak and A Schwartz, ‘Regulating Consumer Bankruptcy: A Theoretical Inquiry’ (2000) 29 Journal of Legal Studies 585; J Armour and D Cumming, ‘Bankruptcy Law and Entrepreneurship’ (2008) 10 American Law and Economics Review 303. 50  See the factors listed in Bundesgerichtshof, Resolution of 18 September 2001 [2002] Neue Juristische Wochenschrift 960, 961.

Abuse of Law in the Context of European Insolvency Law  151 C.  COMI Shifts After the Insolvency Petition COMI shifts not only occur before an insolvency petition. It is also conceivable, and in fact quite practical, that a natural person or a company shifts its COMI after an insolvency petition has already been filed. In this context, the Staubitz-Schreiber decision of the ECJ is of great importance. In this case, Susanne Staubitz-Schreiber moved to Spain after an insolvency petition had been filed in Germany. The question thus had to be decided whether this move had the consequence that the competent German court lost jurisdiction for main insolvency proceedings over the assets of Staubitz-Schreiber. Ultimately, the Bundesgerichtshof referred the case to the ECJ. The ECJ ruled51 that the COMI at the time of the insolvency petition was decisive. It held that acknowledging the COMI shift would go against the goals of the EIR: it would invite forum shopping, make it more difficult for the creditors to assess credit risks, and possibly force them to ‘chase’ the debtor around Europe; the debtor would otherwise be able to play ‘cat and mouse’ with his creditors by moving from one Member State to another. Moreover, acknowledging the validity of the COMI shift would, in the Court’s opinion, be incompatible with the powers of a provisional administrator already appointed after the first insolvency petition to seek protection under Article 38 EIR. Most of the arguments put forward by the ECJ to support its decision apply equally to COMI shifts that occur before an insolvency petition. However, as we have already seen in the previous sections, such COMI shifts may be driven by unobjectionable purposes and cannot always be characterised as abusive. Hence, some other factor must be present that could tip the balance in favour of the Court’s holding that COMI shifts undertaken after an insolvency petition should never be acknowledged. Indeed, one such other factor is present: if one were to acknowledge a COMI shift after an insolvency petition has already been filed, insolvency proceedings would be much less efficient. Once an insolvency petition is filed, costs will normally immediately be sunk in the first proceedings. A provisional administrator will usually be appointed, and temporary measures will be taken, etc. All these measures would be thwarted if a COMI shift after the petition resulted in courts of another Member State acquiring jurisdiction for main insolvency proceedings. Hence, the ECJ correctly held that such jurisdictional issues should always be governed by the COMI at the time of the first insolvency petition. This holding is also in line with the wide concept of the ECJ in Eurofood as to the ‘opening’ of insolvency proceedings. In Eurofood the Court held that ‘a decision to open insolvency proceedings . . . [is] a decision . . . based on the debtor’s insolvency . . . that . . . involves divestment of the debtor and the appointment of a liquidator [including a provisional liquidator] referred to in Annex C to the Regulation’.52 Based on that concept, the opening decision will often coincide with the decision to appoint a provisional liquidator,53 and this decision in practice is usually taken immediately after an insolvency petition has been filed. However, if insolvency proceedings have actually been opened in one Member State, that opening decision is to be recognised in all other Member States and a COMI shift would be irrelevant anyway.  See Susanne Staubitz-Schreiber (n 6 above), paras [21–29].   Eurofood (n 7 above), para [54]. 53  For an example, see Oberlandesgericht Innsbruck, Resolution of 8 July 2008 [2008] Neue Zeitschrift für das Recht der Insolvenz und Sanierung 700 with a comment by P Mankowski. 51 52

152  Horst Eidenmüller We can conclude that whereas COMI shifts before an insolvency petition warrant a differentiated analysis, shifts after an insolvency petition are not to be acknowledged jurisdictionally under the EIR. The Court has correctly held that the COMI is ‘frozen’ at the time of the insolvency petition. COMI shifts that occur after that point in time do not influence the allocation of jurisdictional powers between the Member States under the EIR. D.  Abuse by Member States/Courts When discussing the general elements of the abuse of law concept, it has already been mentioned that not only private actors but also public entities (authorities) might be those that abuse insolvency provisions. I would now like to return to this point by discussing two examples in which it is the Member States or their courts that arguably are the ones to engage in abusive conduct. The first example relates to changes in procedures listed in Annex A of the EIR and the second to the ‘mind of management’ theory adopted by some English courts to determine the COMI of a company. The EIR applies only to proceedings that presuppose the insolvency of the debtor. That is quite clear from Article 1(1) and Recital 10 of the Regulation. For reasons of legal certainty, Annex A contains a conclusive and decisive list of proceedings that are governed by the rules of the EIR.54 One need not, and must not, go back to the general rule in Article 1 EIR to determine whether the Regulation applies. Rather, a look at Annex A suffices. Now, suppose that Member State X changes proceedings listed in Annex A in such a way that, in the future, it will include restructurings that do not presuppose the insolvency of the debtor. Member State X does that in order to become Europe’s leading restructuring venue. Arguably, exactly this has happened with the new English administration procedure introduced by the Enterprise Act in 2002. Since then, holders of a ‘qualified floating charge’55 can apply to the court for an administration order, and the court is empowered to make the order without the need to be satisfied that the company is or is likely to become insolvent (paragraph 35(2)(a) Schedule B1 Insolvency Act 1986).56 It is true, however, that paragraph 35(2)(b) Schedule B1 Insolvency Act 1986 requires the court to be satisfied that the owner of the qualified floating charge could appoint an administrator under paragraph 14. According to paragraph 16, an administrator may not be appointed under paragraph 14 while a floating charge on which the appointment relies is not enforceable. A charge is not enforceable unless the debtor has failed to pay the underlying debt which it secures. Under English law, cash flow insolvency is established by a single unsatisfied demand for payment of an undisputed debt.57 Hence, in case of an undisputed debt, an insolvency according to Article 1(1) EIR seems to be present. However, even under English law insolvency is not a prerequisite if the claim of the holder of the charge has been disputed (which at least is conceivable). In addition, it is very doubtful whether the English definition of cash flow insolvency conforms to Article 1(1) 54  On this see H Eidenmüller, ‘Europäische Verordnung über Insolvenzverfahren und zukünftiges deutsches internationales Insolvenzrecht’ [2001] Praxis des Internationalen Privat- und Verfahrensrechts 2, 4. 55   Para 14(2) Schedule B1 Insolvency Act 1986. 56  See also Fletcher (n 14 above) 139. 57   Cornhill Insurance plc v Improvement Services Ltd [1986] BCLC 26.

Abuse of Law in the Context of European Insolvency Law  153 EIR. The EIR does not define the term ‘insolvency’. Nonetheless, it is clear that it only wants to deal with procedures that apply to companies in a financial crisis that threatens their existence.58 In this respect, the term ‘insolvency’ has an autonomous meaning that must be respected by the national legislators. An existence-threatening financial crisis can probably not be assumed in the case of a single unsatisfied demand for payment of an undisputed debt. Be that as it may, irrespective of the position of English law on that matter, the problem of ‘insolvency proceedings without insolvency’ can at least be posed on a theoretical level. In evaluating this case from an abuse of law perspective, one cannot stop at the primary purpose of Article 1 and Annex A of the Regulation, which is to achieve legal certainty. Rather, one has to look at further (secondary) purposes and ask whether the person or entity relying on a particular legal position does so in contradiction to these (secondary) aims. On that basis, it seems clear that a Member State effecting a change as described above abuses its powers under the EIR. It includes proceedings in the scope of the EIR to which the Regulation explicitly does not want to be applied. That surely will increase the number of COMI shifts on the eve of insolvency and, as a consequence, forum shopping. It is difficult to ascertain the consequences if a court of Member State X nonetheless opened ‘insolvency proceedings’ under these circumstances. Before the opening decision, the court must check whether the proceedings would fall within the reach of the EIR. Otherwise, the international jurisdiction of the court would not depend on Article 3(1) EIR. If the court determines that, because of the abuse just discussed, the EIR is not applicable, it must nevertheless open the proceedings on the basis of the national laws of Member State X. The sanction then comes with the reaction of the courts of other Member States: they would not need to recognise the proceedings as main insolvency proceedings under the EIR. Another example of a potential abuse of the EIR by public entities, the ‘mind of management’ theory, was in essence adopted early on by some English courts to determine the COMI of a company. The theory was developed especially with respect to insolvency cases in the context of corporate groups. The idea behind it was that one could establish an identical COMI for all members of the corporate group at the location where the holding company was based: the mind of management of the subsidiaries was alleged to be with the mind of management of the holding company.59 In the Daisytek and Rover cases, lip service was paid by English courts to the ascertainability criterion of the EIR. However, as a matter of fact, these courts still applied the ‘mind of management’ theory in basing their judgments mostly on internal factors between the companies that were not (easily) noticeable to outside creditors.60 Whilst the attempt to coordinate multiple proceedings by giving one single court jurisdiction over them is laudable, the ‘mind of management’ theory clearly violates the ascertainability criterion stated in Recital 13 of the Regulation. To be sure, maximising the net assets available for distribution to the creditors is the paramount goal of the EIR. Nevertheless, in pursuing this objective one is bound to the COMI criterion stipulated by   Virgòs and Schmit (n 9 above) 49.   Cf R Leithaus, ‘Veranstaltung zu grenzüberschreitenden Insolvenzen in der Insolvenzpraxis in Köln’ [2004] Neue Zeitschrift für das Recht der Insolvenz und Sanierung 194, 195 (reporting the view taken by S Taylor at a conference in Cologne on 6 February 2004). 60   Daisytek-ISA Ltd [2003] BCC 562; MG Rover Deutschland GmbH [2005] EWHC 874 (Ch). 58 59

154  Horst Eidenmüller the EIR. The ‘mind of management’ theory by far overstretches the interpretative leeway given by Article 3(1) EIR. If one finds this to be abusive, as I think one should, the opening decision nevertheless stays effective. Being an official promulgation of a public entity, it cannot qualify as nonexistent even if it is based on an abusive reading of the EIR. However, it is of course subject to being reversed on appeal by a superior court. V. REFORMING THE EUROPEAN INSOlVENCY REGULATION

In concluding this chapter, it seems worthwhile to explore whether the EIR could be amended so that the current problems with respect to forum shopping would be solved and at the same time the need to resort to the abuse of law concept eliminated or at least diminished. Even though the debate on regulatory competition in insolvency law in Europe is still unresolved on an empirical level, it is clear that forum shopping on the basis of the COMI concept creates certain significant problems: it is associated with discrepancies between the applicable company and insolvency law and thus high bankruptcy costs, and it is also associated with possible detrimental effects on non-adjusting creditors (see Part I). The abuse of law concept helps to ameliorate these problems, but it is only of limited help. It requires a case-by-case approach and thus creates legal uncertainty. Legal uncertainty, however, translates into economic costs. A different approach would be to tie insolvency proceedings under the EIR to the registered office of a company in financial distress.61 This proposal comes down to turning the rule that a company’s COMI conforms to its registered office from a refutable into an irrefutable presumption. Adoption of this proposal by the European legislator would improve matters compared to the status quo in various respects. First, in an insolvency situation, company law and insolvency law would always go hand in hand. A company would be put into insolvency proceedings in the Member State in which it has its registered office. These proceedings would be governed by the insolvency laws of that Member State, and the company law of that Member State would—on the basis of the incorporation theory—govern matters of company law. Hence, the frictions and bankruptcy costs associated with ‘isolated’ COMI shifts under the current regime could be avoided.62 Second, the proposal has advantages with respect to creditor protection. If a company wished to change the bankruptcy venue with respect to main insolvency proceedings, it would have to transfer its registered office. As we have already seen, this can be accomplished using the mechanism of the Tenth Directive. Such a reincorporation involves conforming to the creditor protection rules established by the laws of the affected Member States.63

61  See Eidenmüller (n 15 above) 438–40; J Armour, ‘Who Should Make Corporate Law? EC Legislation versus Regulatory Competition’ (2005) 58 Current Legal Problems 369. 62  This is also the decisive advantage of this approach compared to Ringe’s suggestion according to which companies would be put into insolvency proceedings at their COMI, but with the law of the incorporating Member State as the governing insolvency law (Cf Ringe (n 19 above) 615). The consequence would be that courts would have to apply foreign insolvency provisions with respect to their actual administration of an insolvency case. This is clearly not workable. 63  To be sure, under certain circumstances reincorporations may be achieved without using the framework of the Tenth Directive (see n 37 above). However, this cannot regularly be done.

Abuse of Law in the Context of European Insolvency Law  155 Would it be advisable to combine the proposal made here with a suspect period such that not the actual registered office is decisive with respect to the opening of main insolvency proceedings, but the registered office at some earlier point in time—eg a few months or a year before the insolvency petition? We find such a suspect period in some European jurisdictions,64 and also under US law (see 28 USC § 1408). Although arguments can be put forward in favour of it, especially its deterrent effect on last-minute forum shoppers and its clear-cut character (legal certainty), the length of such a suspect period is always arbitrary. Further, it prevents certain sensible late reincorporations. As an example, consider again the case of the Luxembourg holding company that was moved to Germany on the eve of bankruptcy. Finally, creditors are protected under the mechanisms of the Tenth Directive. Hence, I do not think that a suspect period with respect to changes in the registered office of a company on the eve of bankruptcy should be put into the EIR. The proposal just discussed applies to companies and not to natural persons. Only companies necessarily have a registered office (natural persons that are not running a sole proprietorship do not), and only companies are potentially subject to the mechanisms of the Tenth Directive on cross-border mergers. With respect to natural persons, there is probably no good alternative to retaining the status quo—ie to look for the habitual residence as criterion for international jurisdiction with respect to main insolvency proceedings. At the same time, it is clear that the problems discussed in Part IV.B.ii regarding forum shopping are serious and require a convincing solution. As opposed to the case of companies, relocations on the eve of insolvency are rarely, if ever, desirable from the creditors’ perspective in the case of natural persons. Also, the creditor protection provisions under the Tenth Directive do not apply to such persons. Therefore, a suspect period with respect to the habitual residence and, as a consequence, international jurisdiction to open main insolvency proceedings is the only available remedy. The EIR could stipulate, for example, that the habitual residence six months before the insolvency petition is decisive. To conclude, there are ways to improve the legislative framework of the EIR to reduce the negative effects of forum shopping on the eve of bankruptcy and render less pressing the need to employ the abuse of law concept in order to correct undesirable COMI shifts. For companies, the COMI should be replaced by the registered office as the only and decisive criterion with respect to international jurisdiction to open main insolvency proceedings. For natural persons, the habitual residence should be the relevant criterion, but it should be subject to a suspect period of a few months before the insolvency petition. VI. SUMMARY AND CONCLUSION

Questions of abuse of law in the context of insolvency law are closely linked to the debate about forum shopping on the eve of bankruptcy, and regulatory competition between the Member States of the European Union in the insolvency field. It has been shown that the abuse of law concept can play a role in that it can put a check on undesirable forms of forum shopping. At the same time, legislative reform would provide a better solution: the European Insolvency Regulation (EIR) could and should be amended to the effect that it is not the ‘centre of main interests’ (COMI) but the registered office of a company that is  See n 42 above.

64

156  Horst Eidenmüller decisive with respect to international jurisdiction to open main insolvency proceedings. The main findings of this chapter can be summarised as follows: 1. Forum shopping in the insolvency field has both positive and negative economic effects. It may lead to the selection of efficient restructuring procedures and, in a group insolvency context, it may facilitate the coordination of multiple proceedings (which is good). It may also lead to discrepancies between the applicable company and insolvency law, reduce the efficiency of credit contracts and be done at the cost of involuntary or other ‘non-adjusting’ creditors (which is bad). There is no conclusive empirical evidence on the overall economic effects of forum shopping in the insolvency field. 2. Abuse of law is a doctrinal concept on the scope of a legal provision. Its main element is use of law contrary to its purpose. A specific ‘abuse intention’ is not necessary. As an interpretative concept, abuse of law is a European concept insofar as the interpretation of a European legal provision is at issue. If a particular legal provision is abused, that provision cannot be relied upon by the (ab)user. Whoever wishes to rely on a particular legal provision is capable of potentially abusing it. Hence, not only private actors but also public entities might be challenged with abuse in a particular case (even though abuse by private actors surely is the much more common phenomenon). 3. Shifting the COMI from one Member State to another by a natural person or a company is protected under the freedom of establishment guaranteed by the TFEU and is in general not abusive. However, COMI shifts that evidently do not contribute to maximising the debtor’s net assets for the creditors’ benefit are abusive. COMI shifts that evidently benefit either the debtor at the expense of its creditors or some creditors at the expense of others are suspicious and fall into the category of abusive shifts: they are driven by distributive and not by efficiency concerns. As a consequence, COMI shifts by companies on the eve of bankruptcy that are undertaken in order to maximise restructuring value are not abusive; COMI shifts undertaken in order to enrich the debtor at the expense of its creditors are abusive, as are shifts in order to enrich certain creditors at the expense of others. COMI shifts by natural persons on the eve of bankruptcy effectuated to obtain a more favourable discharge regime are abusive. 4. The COMI of a company or natural person is ‘frozen’ at the time of the insolvency petition. As a consequence, COMI shifts that occur after the insolvency petition have no effect on the international jurisdiction of a court to open main insolvency proceedings under the EIR. 5. The negative effects of forum shopping on the eve of bankruptcy should be addressed by a change in the legislative framework of the EIR. With respect to companies, COMI should be replaced by the registered office. In the future, only the registered office should be decisive with respect to the international jurisdiction to open main insolvency proceedings. This solution is not available regarding natural persons. With respect to natural persons, the habitual residence should be the relevant criterion, as is the case under the current framework. However, a suspect period should be added to that framework, so that the habitual residence a few months before the insolvency petition determines international jurisdiction.

11 Abuse of European Insolvency Law? A Discussion John Armour*

I

i. Introduction

am delighted to have had the opportunity to discuss the interesting and thoughtful chapter written by Professor Eidenmüller.1 I shall begin my comments by outlining some contextual points on the role of insolvency law in general and the European Insolvency Regulation in particular. I shall then move on to an examination of the principal claims made by Professor Eidenmüller. As is inevitably the way with such exercises, my comments will focus on those aspects of his argument by which I am least persuaded: little value is added by a discussion that simply agrees with the original arguments. Let me be clear at the outset, however, that any points of disagreement suggested by this comment relate only to minor points: the points of agreement between us are far more general than any differences canvassed in this chapter. II. Insolvency Law and the European Insolvency Regulation

Insolvency law, for the main part, exists to preserve and enhance the value of assets of a distressed firm or individual. When a debtor becomes financially distressed, the creditors, fearing the debtor will be unable to pay them, may seek to execute individual enforcement against its assets. The ‘first come, first served’ aspect of individual enforcement proceed­ ings is likely to precipitate a ‘run’ on the debtor’s assets, with the result that they are dismembered and sold separately to pay multiple creditors. Where, as may commonly be the case with a business, the assets are worth more as a collection than if broken up and sold separately (that is, the assets have ‘synergies’) then the result of such a ‘race to collect’ will be loss of value for the creditors as a whole, and for society, through the loss of a good business. Insolvency proceedings, once triggered, replace creditors’ entitlements to individual enforcement proceedings with a right to participate in a collective scheme. Individual enforcement actions are stayed; in return, creditors receive entitlements to vote to elect an insolvency official who will oversee the resolution of the debtor’s affairs, and a right to payment from the pool of the debtor’s assets.2   * I am grateful to Horst Eidenmüller, Paschalis Paschalidis and participants at the Abuse of Law conference for helpful comments on an earlier version. The usual disclaimers apply. 1  H Eidenmüller, ‘Abuse of Law in the Context of European Insolvency Law’, ch 10 above. 2  TH Jackson, ‘Bankruptcy, Non-Bankruptcy Entitlements, and the Creditors’ Bargain’ (1982) 91 Yale Law Journal 857, 860–68.

158  John Armour In addition to enhancing the value of the debtor’s assets ex post, insolvency law also has ex ante effects. Anticipation of outcomes in insolvency may alter the behaviour of potentially affected parties in relation to companies that have not yet entered (and indeed may never enter) insolvency proceedings.3 Here, crucial questions concern how to ensure that companies have the greatest possible access to finance (altering priorities in insolvency may upset this),4 and that insolvency proceedings are triggered at a time that is not too early (harming good businesses) or too late (to the detriment of creditors).5 Where a debtor carries on operations in more than one country, then even if the relevant domestic insolvency laws respond to the foregoing problems effectively, the same issues may nevertheless manifest themselves at the cross-border level. This is because multiple territorial insolvency proceedings may be opened, which, if not coordinated effectively, may lead to the break-up of the debtor’s assets.6 Creditors may engage in a race to ‘grab’ assets under territorial laws which are most favourable to them, and the debtor’s business may be dismembered. Moreover, the anticipation of such consequences in times of financial distress may cause distortions in such firms’ ability to raise finance ex ante.7 The European Insolvency Regulation seeks to prevent these adverse consequences from resulting in the case of cross-border insolvencies occurring within the European Union.8 It seeks to identify a single jurisdiction which will provide both the forum and the law (lex loci concursus) which will govern the conduct of the debtor’s ‘main’ insolvency proceedings.9 This universal effect extends also to transaction avoidance, which must be determined according to the lex loci concursus.10 To give efficacy to this choice, other Member States must automatically recognise the opening of the main proceedings and give the same effect to any legal consequences thereby imposed—for example, a stay of enforcement— as is given in the jurisdiction where the main proceedings are opened.11 Moreover, in order to preserve certainty, courts of other Member States are required to respect the decision of the court first seized that it has jurisdiction to open main proceedings.12 Hence the only channel of appeal against such a decision, save for an objection based on ordre public,13 will be through the courts of that Member State and ultimately to the Court of Justice. The connecting factor linking the debtor to the jurisdiction where main proceedings are to be opened is therefore of the utmost importance to the Regulation’s scheme: it has significant consequences, and the opportunities for challenge of its initial determination are very limited.     3  See for example, RK Rasmussen, ‘The Ex Ante Effects of Bankruptcy Reform on Investment Incentives’ (1994) 72 Washington University Law Quarterly 1159; MJ White, ‘The Costs of Corporate Bankruptcy: A US–European Comparison’ in JS Bhandari and LA Weiss (eds), Corporate Bankruptcy: Legal and Economic Perspectives (Cambridge, Cambridge University Press 1996) 467, 479–83, 491.     4  See for example, J Armour, ‘Should We Redistribute in Insolvency?’ in J Getzler and J Payne (eds), Company Charges: Spectrum and Beyond (Oxford, Oxford University Press 2006) 189.     5  See for example, H Eidenmüller, ‘Trading in Times of Crisis: Formal Insolvency Proceedings, Workouts and the Incentives for Shareholders/Managers’ (2006) 7 European Business Organization Law Review 239.     6  See for example, LA Bebchuk and AT Guzman, ‘An Economic Approach to Transnational Bankruptcies’ (1999) 42 Journal of Law and Economics 775.     7   Bebchuk and Guzman (n 6 above).     8  Council Regulation (EC) No 1346/2000 of 29 May 2000 on insolvency proceedings [2000] OJ L160/1.     9  Regulation 1346/2000 (n 8 above), Art 3(1). 10  Regulation 1346/2000 (n 8 above), Art 4(2)(m); Case C-339/07 Christopher Seagon v Deko Marty Belgium NV [2009] ECR I-767. 11  Regulation 1346/2000 (n 8 above), Arts 16, 17. A limited exception exists to permit non-recognition or non-enforcement on grounds of public policy: Regulation 1346/2000 (n 8 above), Art 26. 12  Regulation 1346/2000 (n 8 above), Art 16(1); Preamble, Recital 22. 13  Regulation 1346/2000 (n 8 above), Art 26.

Abuse of European Insolvency Law? A Discussion  159 Unfortunately, the chosen connecting factor suffers from inherent vagueness.14 The crucial concept is the debtor’s ‘centre of main interests’ (‘COMI’).15 It appears from the preamble to the Regulation and the Court’s pronouncements on the question that the notion of COMI is intended to be allocated so as to make it as easy as possible for creditors to discover. Recital 13 explains that the COMI ‘should correspond to the place where the debtor conducts the administration of his interests on a regular basis and is therefore ascertainable by third parties.’ On first blush, this would appear to be a version of the Sitztheorie or siège réel, whereby the COMI is located at the place where in fact the debtor’s head office and principal administration of its business is conducted. However, it does not follow that this will always be easily ascertainable by third parties. Some indicia of the conduct of business (eg physical offices, addresses used in correspondence) will be more readily ascertainable than others (eg where board meetings are in fact conducted, where employees are in fact hired). The question arises whether the words ‘is therefore’ in Recital 13 are merely hortative, or whether they actually restrict the categories of factors which can be used in the determination of the COMI. An answer was given by the ECJ in Eurofood, who stated that the COMI must be ‘identified by reference to criteria that are both objective and ascertainable by third parties’, which was ‘necessary in order to ensure legal certainty and foreseeability’.16 Matters are further complicated in the case of corporate debtors by Article 3(1) of the Regulation, which provides that, in the ‘absence of proof to the contrary’ a company or legal person’s COMI shall be presumed to be the place of its registered office. This raises further questions concerning the strength of the presumption and the types of evidence relevant to its rebuttal. The ECJ in Eurofood commented, somewhat Delphically, that the Article 3(1) presumption could be rebutted if it were shown that the registered office were no more than a ‘mere letterbox’.17 However, there is a wide spectrum of possible activity levels in the jurisdiction of the registered office, ranging from a ‘mere letterbox’ (presumption clearly rebutted) to the administration of the debtor’s interests there on a regular basis (presumption clearly not rebutted). Whilst the consequences of a finding that the activity is at one of the outer ends of the spectrum are clear,18 uncertainty over what is required to rebut the presumption means that the consequences of a finding that the activity is somewhere in between these two polar positions remain unclear.19 Bearing in mind that the location of the registered office may well—particularly in today’s world of internet searches—be easier for creditors to ascertain than other indicia which may point to where the debtor regularly administers its interests, and the significance attached to ascertainable criteria, it is arguable that the presumption should be treated as rebutted only in extreme cases. In Eurofood, it was held that the mere fact that the company was part of a group, the parent company of which was in another jurisdiction, was insufficient to rebut the presumption.20 14  See generally, I Fletcher, ‘Scope and Jurisdiction’ in G Moss, I Fletcher and S Isaacs (eds), The EC Regulation on Insolvency Proceedings (Oxford, Oxford University Press 2002) 39–41. 15  Regulation 1346/2000 (n 8 above), Art 3(1). 16  Case C-341/04 Eurofood IFSC Ltd [2006] ECR I-3813, para [33]. 17   Eurofood (n 16 above), para [35]. 18  See for example, Hans Brochier Holdings Ltd v Exner [2006] EWHC 2594 (Ch) [2007] BCLC 127, paras [23–30] (little more than a letter-box at the registered office; COMI held to be in another jurisdiction). 19  See J Armour, ‘European Corporate Insolvencies: The Race Goes to the Swiftest?’ (2006) 65 Cambridge Law Journal 504, 506. 20   Eurofood (n 16 above), paras [36–37].

160  John Armour

III.  Abusive Shifting of COMI

The elasticity inherent in the concept of COMI creates a number of problematic incentives for strategic behaviour. Most obviously, debtors may seek to locate—or relocate—their COMI in such a way as to provide them with maximum protection against the claims of all or some part of their creditors. In response, Recital 4 of the Regulation’s Preamble refers to the necessity of avoiding incentives for forum-shopping in order to ensure the proper functioning of the Internal Market, and the Court has ruled that attempts by a debtor to alter her COMI after a petition to open insolvency proceedings has been lodged are ineffective.21 Yet at the same time, to prohibit a change of COMI outright—regardless of the timing of insolvency proceedings—would surely be an illegitimate restriction on freedom of establishment.22 This raises the question how a shift in COMI, undertaken before a petition for insolvency, but intended to undermine the position of creditors, should be treated. Professor Eidenmüller argues that such conduct should be subject to an overarching principle of European law restraining abuse of law, with the consequence that prima facie valid changes of a debtor’s COMI should be treated as invalid if they are ‘abusive’. He proposes two potentially cumulative bases. First, some transfers of COMI should be treated as abusing a corporate debtor’s freedom of establishment.23 Secondly, some transfers—even if they do not amount to an abuse of freedom of establishment—should be treated as abusive because they involve the use for an improper purpose of the European law rules established by the Regulation for the determination of COMI.24 A.  Abuse of Freedom of Establishment The first of these propositions seems quite orthodox, insofar as the specific implication is that Member State rules which restrict corporate freedom of establishment in order to protect creditors from debtors’ seeking to relocate so as to undermine creditors’ protection on insolvency should be viewed as permissible restrictions on freedom of establishment, provided that they are applied in a non-discriminatory manner, secure the attainment of their objective, and are proportionate in their effect.25 A ‘shift in COMI’ can in some cases be achieved by substitution of a new corporate entity in a different jurisdiction, effecting a change in the registered office. This can be done by entering into a merger into a newly-created company in the ‘target’ jurisdiction, relying on the Cross-Border Mergers Directive (Tenth Company Law Directive) to give effect to the transaction.26 The Directive does not specify any minimum standard  Case C-1/04 Susanne Staubitz-Schreiber [2006] ECR I-701.  See WG Ringe, ‘Forum Shopping Under the EU Insolvency Regulation’ (2008) 9 European Business Organization Law Review 579. 23  Eidenmüller (n 1 above), Part III. 24  Eidenmüller (n 1 above), Part IV. 25  Case C-55/94 Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano [1995] ECR I-4165. 26  Directive 2005/56/EC of the European Parliament and of the Council of 26 October 2005 on cross-border mergers of limited liability companies [2005] OJ L310/1. Similarly, in the case of a Societas Europea, the SE Regulation provides for change in the registered office: EC Council Regulation 2157/2001 of 8 October 2001 on the Statute for a European company (SE) [2001] OJ L294/1, Arts 7–8. 21 22

Abuse of European Insolvency Law? A Discussion  161 to protect creditors of the original entity in the ‘outbound’ jurisdiction, but leaves the determination of such safeguards to national law.27 Member States’ laws governing mergers must, however, be compliant with the Third Company Law Directive, which specifies that creditors of an entity merged into another who are not paid off under the transaction must be provided with ‘adequate safeguards’ where the financial situation of the merging companies makes such safeguards necessary and where they do not already have such safeguards.28 In some Member States, this equates to a right of veto; in others a right to be paid off or given security.29 In any event, such safeguards would preclude use of a cross-border merger to facilitate an abusive shift of COMI. The ECJ’s recent decision in Cartesio made clear that a corporate entity need not, however, rely upon a cross-border merger in order to effect a change in its registered office.30 Cartesio establishes that rules of Member States’ company laws which restrict the freedom of corporate entities to transfer their governing law to that of another Member State (that is, to ‘emigrate’) constitute prima facie infringements of corporate freedom of establishment.31 Examples of such restrictions might include requirements that such corporate emigration be accomplished only by the winding-up or dissolution of the entity established under the ‘outbound’ Member State’s law. Although only a dictum in this case, these observations make clear that existing corporations should be free to change their jurisdiction of incorporation without the need to rely upon any specific European legislation (such as the Tenth Directive or the currently-mothballed Fourteenth Directive on change of registered office) beyond the Treaty freedoms themselves.32 However, it seems equally clear that where such a shift has detrimental consequences for creditors in the ‘outbound’ Member State, that State would be justified in adopting measures that impede the corporate emigration, provided that they are proportionate to the goal of protecting the creditors.33 B.  Abuse of the European Insolvency Regulation Professor Eidenmüller’s second, and rather bolder, proposition is that some transfers of a debtor’s COMI—even if they do not amount to an abuse of freedom of establishment— should be treated as abusive under European law because they involve the use for an improper purpose of the European law rules established by the Regulation for the determination of COMI.34

 Directive 2005/56/EC (n 26 above), Art 4.  Council Directive (EEC) 78/855 of 9 October 1978 based on Article 54(3)(g) of the Treaty concerning mergers of public limited liability companies [1978] OJ L295/36, Art 13(2). 29  V Edwards, EC Company Law (Oxford, Oxford University Press 1999) 109–10. 30  Case C-210/06 Cartesio Okató és Szolgáltató bt [2008] ECR I-9641. 31   Cartesio (n 30 above), paras [110–12]. 32  This would appear to render English company law, which does not permit such an outward reincorporation, in contravention of the Treaty freedoms on this point. Such a power was recommended by the Company Law Review, but rejected by the Government on the basis of feared loss of tax revenue: see PL Davies, Gower and Davies Principles of Modern Company Law, 8th edn (London, Sweet & Maxwell 2008) 139–41. 33  See J Armour, ‘Who Should Make Corporate Law? EC Legislation versus Regulatory Competition’ (2005) 58 Current Legal Problems 369, 406; H Eidenmüller, ‘Free Choice in International Company Insolvency Law in Europe’ (2005) 6 European Business Organization Law Review 423, 446; Cf Ringe (n 22 above) 610–12. 34  Eidenmüller (n 1 above), Part IV. 27 28

162  John Armour This begs the question as to what might be viewed as a ‘proper’ purpose of the Regulation. Professor Eidenmüller’s answer, based on an exegesis of the Preamble to the Regulation, is that it exists primarily to ensure that creditors obtain maximum value from the debtor’s assets: that is, to prevent the sort of perverse consequences associated with cross-border insolvencies as were discussed in Part II.35 Whilst the preamble also expressly mentions the goal of restricting forum shopping in the field of insolvency law, this is argued to be subsidiary to the main goal of maximising value for creditors.36 Thus, a change of COMI in order to secure a better realisation of the assets for the creditors is consistent with the objectives of the Regulation. This analysis leads Professor Eidenmüller to conclude that a change of COMI may constitute an abuse of the Regulation if it evidently tends to decrease the overall value of the debtor’s net assets. The proposed formulation is in my view consistent with the basic purposes of insolvency law.37 Where the debtor is a natural person, moreover, it seems reasonably straight­forward to interpret. The personal insolvency laws of Member States differ widely, but the principal dimension of difference concerns the length of time before a debtor is permitted to be discharged from pre-insolvency debts (if ever).38 Clearly, moving from a jurisdiction with a lengthy discharge period to one with a shorter discharge period will harm the creditors to the benefit of the debtor. However, the practical application of the proposed principle seems much more problematic in relation to corporate debtors. European corporate insolvency laws also differ considerably, across a number of dimensions.39 They give differing weight to the interests of different constituencies. For example, secured creditors have traditionally found considerable favour in insolvency proceedings under English law.40 In contrast, the position of employees has traditionally been privileged in French insolvency law.41 Different codes also adopt widely different decision-making processes. For example, English law accords considerable weight to the business judgment of an insolvency office-holder, in particular over decisions about the future of the enterprise, and supplements this with creditors’ votes.42 In France, on the other hand, the court makes important decisions such as how the assets should best be realised.43 German law involves a combination of creditor and court input that might crudely be characterised as lying somewhere between these two cases. These differences mean that particular corporate insolvency regimes are likely to generate better results for particular constituencies. For example, Sergei Davydenko and  See text to n 6 above.  Eidenmüller (n 1 above), Part IV.A. 37  See text to nn 2–6 above. However, the reading of Recital 4 of the Preamble it necessitates is difficult. 38   For a summary of the position in a number of Member States, see J Armour and DC Cumming, ‘Bankruptcy Law and Entrepreneurship’ (2008) 10 American Law and Economics Review 303, 312–13; a fuller overview is available at the European Commission (DG Enterprise) website: ec.europa.eu/enterprise/entrepreneurship/ sme2chance/policy_legal_overview_en.html. 39  See generally, J Marshall (ed), European Cross-Border Insolvency: A Guide to Insolvency Procedures Across the European Union (London, Sweet & Maxwell 2004). 40  See for example, J Armour and S Frisby, ‘Rethinking Receivership’ (2001) 21 Oxford Journal of Legal Studies 73. 41  See for example, P Théry, ‘The Evolution of Insolvency Law in France’ in L Gullifer, WG Ringe and P Théry (eds), Current Issues in European Financial and Insolvency Law (Oxford, Hart Publishing 2009) 1. 42  Insolvency Act 1986 Sch B1, paras 3 and 49–52; Re Transbus International Ltd [2004] EWHC 932 (Ch), [2004] 2 BCLC 550. 43  Code de Commerce, Arts L 626-9 (plan de sauvegarde), 631-19 (plan de redressement) and 641-1 (liquidation judiciaire). 35 36

Abuse of European Insolvency Law? A Discussion  163 Julian Franks’ empirical study of returns to secured creditors in France, Germany and the UK finds that secured creditors fare best in the UK and least well in France, largely in accordance with intuitions about the structure of the insolvency codes.44 However, what is far less clear is whether these partial differences translate across into general differences. That is, does the fact that one group does better under a particular procedure mean that the creditors overall are disadvantaged? There is very little in the way of crosscountry empirical evidence on overall corporate insolvency outcomes. However, to the extent that one can extrapolate from existing studies (mainly relating only to individual jurisdictions), a clear impression emerges that there may be more than one way to reach a similar net result for creditors.45 For example, Edward Morrison, comparing outcomes achieved by bankruptcy judges approving Chapter 11 plans in the US, concludes that the results are indistinguishable from those which would be reached were the decision made by a private actor.46 John Armour, Audrey Hsu and Adrian Walters, comparing costs and realisations in English insolvencies before and after the coming into force of the Enterprise Act 2002 which reduced the power accorded to secured creditors, conclude that the net effect on outcomes is negligible.47 And Régis Blazy et al, in the first comprehensive study of outcomes under French insolvency law, report that the bias towards protection of employees apparently did not result in significant losses to creditors.48 On the basis of current knowledge, it is hard to see how a particular choice of corporate insolvency law could be stigmatised as failing to maximise net assets. This would result in a test that is too narrow in its practical impact to be of any utility in relation to corporate debtors. Sensitive to this difficulty, Professor Eidenmüller suggests rather that what should be categorised as abusive is not so much an attempt to shift to an insolvency law which is harmful to the value of the debtor’s net assets, but rather an attempt by a party (be they debtor or creditor) to determine the insolvency regime so as to secure a private advantage to that party. That is, abusive COMI shifts are those undertaken for distributional, rather than efficiency, considerations. The thinking is that the existence of such a distributional motive will be more readily observable than whether or not the change tends to increase the overall size of the pie. However, if the criterion for determining whether a transfer is abusive is simply whether it secures a private advantage for the party proposing it, then this might cast the net too wide. It is surely axiomatic that a party is unlikely to wish a particular venue to be the COMI unless it is in its private interests to do so. One possible resolution of this issue would be to move away from outcomes and to focus on the process employed for shifting COMI. Where the shift is effected consensually with (a majority of) creditors, then this gives prima facie evidence that it is value-increasing, as creditors have no incentive to agree to such a shift without there being some advantage for them.49 On the other hand, unilateral, or non-consensual shifts initiated by the debtor 44  S Davydenko and J Franks, ‘Do Bankruptcy Codes Matter? A Study of Defaults in France, Germany, and the UK’ (2008) 63 Journal of Finance 565. 45  See S Djankov, O Hart, C McLiesh and A Shleifer, ‘Debt Enforcement Around the World’ (2008) 116 Journal of Political Economy 1105. 46  ER Morrison, ‘Bankruptcy Decision Making: An Empirical Study of Continuation Bias in Small Business Bankruptcies’ (2007) 50 Journal of Law & Economics 381. 47   J Armour, A Hsu and A Walters, ‘The Costs and Benefits of Secured Creditor Control: Evidence from the UK’ [2006] University of Cambridge Centre for Business Research Working Paper No 332: srn.com/ abstract=912302. 48  R Blazy, B Chopard, A Fimayer and J-D Gigou, ‘Financial versus Social Efficiency: The French Dilemma?’ [2007] CREFI-LSF Working Paper 07-02: ssrn.com/abstract=1116681. 49  See Ringe (n 22 above) 604.

164  John Armour alone may be treated with greater suspicion, particularly where effected on the eve of insolvency. A second difficulty with the application of the proposed principle concerns timing. At what point in a company’s history is a transfer of COMI to be deemed abusive? Consider an example. Imagine that in the first instance X has its COMI in Member State A. It obtains credit from C1, C2 and C3 respectively, each premising their lending upon the assumption that the law of Member State A would govern any insolvency of X. X then moves its COMI to Member State B, without obtaining the consent of its creditors, and with the intention of harming them. Were X to default immediately after this move, then if the move prejudices C1, C2 and C3 to the benefit of X’s shareholders, it would appear to be a good case for decrying the shift as abusive under the proposed principle. However, consider the position if X does not default immediately following the shift in COMI, but continues to operate for some period of time, during which the loans from C1 and C2 are refinanced by new borrowings from C4 and C5. Both C4 and C5 advance funds on the assumption that the law of Member State B will govern any insolvency of X, pricing their lending accordingly. If the location of COMI must be determined now, are we still willing to characterise the transfer as ‘abusive’? Does its abusive character get rectified by the refinancing? C3, whose loan is still outstanding, is still worse off as a result of the shift in COMI. Yet C4 and C5 would prefer the law of Member State B to continue to govern, and a retransfer back to Member State A might actually prejudice their position. As we are dealing with a fixed pool of assets because of insolvency, then it may not be possible to make an order that restores C3 to its pre-shift priority without making C4 and C5 worse off. A more fundamental question is whether such a principle of ‘abuse of the European Insolvency Regulation’ of the variety proposed is necessary or desirable. First, as regards corporate debtors, many attempts to shift COMI will in fact involve a change of registered office, probably with the creation of a new entity. Domestic laws may respond to the problem of ‘corporate emigration’ calculated to undermine the position of creditors without this constituting an infringement of the debtor’s freedom of establishment.50 Any need for protection is therefore confined to cases where the debtor is a natural person or for corporate debtors where a shift in COMI is effected without a change in the applicable company law. Second, it is possible for many creditors to take steps to protect themselves against the risk of an abusive shift of COMI through contractual provisions. An appropriatelydrafted clause might specify that a change of COMI—or, more generally, any course of conduct calculated to or likely to lead to a change of COMI (for example, relocation of registered office, a cross-border merger, or relocation of head office)—would constitute an event of default. The widespread use of cross-default clauses means that only one such default need be triggered for the debtor to be in default with all, or the large part, of its creditors. The effect of such a clause is to force a debtor wishing to change its COMI to do so with the consent of its creditors—which, if obtained, would be a prima facie indicator that the shift was likely to yield a greater net recovery. Of course if the debtor is able to effect a change of COMI and relocate all its assets before the creditor becomes aware of this, then such a clause will leave the creditor with only the Pyrrhic victory of the default-triggering insolvency proceedings in the  See above, Part III.A.

50

Abuse of European Insolvency Law? A Discussion  165 jurisdiction of the debtor’s new COMI. However, in most cases triggering a default will have significant adverse consequences for the debtor. Any undrawn-down facilities may be terminated, and to the extent that any assets remain in the original jurisdiction, then creditors with security will still be able to rely on the insolvency consequences of that law,51 and unsecured creditors will be able to open territorial proceedings.52 For debtors of any size, the chance of a shift of COMI before sufficient creditors realise this to trigger injuctive relief is likely to be extremely slim. Any need for additional protection is therefore really only in issue where the debtor is a small private company or an individual, who may in theory be able to relocate their COMI and assets without creditors becoming aware. Third, it might be objected that so-called ‘non-adjusting’ creditors may be unable to rely on contractual self-protection of the variety just described.53 To the extent that the debtor wishes to move to a jurisdiction with an insolvency law that favours its position over its creditors, then it is in no creditor’s interests for the shift to occur. Under such circumstances, non-adjusting creditors may ‘free ride’ on the actions taken by voluntary creditors to prevent the debtor from taking such steps. More challenging for the nonadjusting creditors may be a situation where the debtor enters into an agreement with voluntary creditors to shift COMI in a way that may be harmful for the non-adjusting creditors. To some degree, non-adjusting creditors have protection against this built into the structure of the Regulation.54 In particular, Article 10 provides that the impact of insolvency on contracts of employment shall be governed solely by the law applicable to the contracts in question. More generally, the possibility of territorial secondary proceedings adds further protection for non-adjusting creditors.55 Imagine that the law of Member State A offers priority in insolvency to tort victims over general unsecured creditors. The debtor agrees with its voluntary creditors to relocate its COMI to Member State B, which does not offer such priority. However, if the debtor retains any part of its business activity in Member State A, it is likely that this will constitute an ‘establishment’,56 sufficient to open secondary proceedings under the law of Member State A to govern the realisation and distribution of the assets remaining in that jurisdiction. In this context, it is worth bearing in mind that a targeted host-country protection of tort victims—such as a mandatory insurance regime for those carrying on hazardous activities, for example— would be unlikely to constitute an unlawful impediment to freedom of establishment, and would dovetail with territorial proceedings. Fourth, there are also reasons for thinking that a European law prohibition on ‘abusive’ COMI shifts might generate significant costs. The application of any such test would be surrounded by considerable uncertainty.57 Given the inherent elasticity of the concept of COMI, it seems likely that courts may be tempted to elide—and confuse—the question of whether COMI has shifted at all with that of whether a prima facie shift should be struck down as being abusive. This would only exacerbate the problems surrounding the determination of COMI, and may make it more difficult for the sorts of procedural or  Regulation 1346/2000 (n 8 above), Art 5.  Regulation 1346/2000 (n 8 above), Art 3(2). 53  The terminology is derived from LA Bebchuk and JM Fried, ‘The Uneasy Case for the Priority of Secured Claims in Bankruptcy’ (1996) 105 Yale Law Journal 857. 54  See Armour (n 33 above) 408–10. 55  Regulation 1346/2000 (n 8 above), Art 3(2). 56  Defined as ‘any place of operations where the debtor carries out a non-transitory economic activity with human means and goods.’ Regulation 1346/2000 (n 8 above), Art 2(h). 57  See text to nn 39–48 above. 51 52

166  John Armour contractual self-protection we have discussed to be effective (for example, yielding greater opportunities for interpretive disputes over the triggering of ‘COMI clauses’). What is worse, it might even increase the possibilities for creditors to engage in jurisdictional arbitrage: they could draw attention to concerns about abusive behaviour by the debtor as an additional reason for their favoured jurisdiction to come to a determination that COMI lies within its territory. To recapitulate: the proposed principle would potentially be useful in the context of debtors who are natural persons. However, it may well generate additional uncertainty over the determination of COMI. Whether this is worthwhile in relation to natural persons may be a finely balanced exercise; it seems unlikely to be worthwhile in relation to corporate debtors. IV.  Abuse of the Regulation by Member States

The structure of the Regulation also creates perverse incentives across other dimensions. Creditors may use the elasticity of the COMI to persuade the courts of Member States with favourable insolvency regimes to open main insolvency proceedings, thereby triggering automatic recognition throughout the EU. The ECJ in Eurofood emphasised that the decision of the court first opening proceedings that the COMI is within its jurisdiction is not reviewable by other Member States’ courts.58 This may have the perverse consequence of encouraging creditors to engage in such forum shopping pre-emptively, for fear that if they do not, others may do so first. Member States’ courts may accede inappropriately to such requests, and their legislatures may be tempted to design insolvency rules that time the opening of proceedings as early as possible, so as to ensure their courts are likely to be first to open proceedings.59 Strategic behaviour of this sort is also clearly undesirable. Professor Eidenmüller suggests that over-ready acceptance of such claims by Member States’ courts, or legislative changes deeming insolvency proceedings to open at an earlier point of time under national law, might constitute an abuse of the Regulation by Member States actionable under European law. However, any such relief would need to be subject to the ‘mutual trust’ principle, such that it could only be sought by appeal from the decision of the Member State court in question, and thence to the Court of Justice.60 There are, of course, other channels through which such actions by Member State courts and legislatures could be curtailed. In particular, legislative changes are constrained by the implicit threat of renegotiation of the list of ‘insolvency proceedings’, in Annex A to the Regulation, which are entitled to recognition under Article 3(1).61 Unfortunately, as has been described,62 there may be a negative interaction between seeking to strike out abusive ‘finding’ of COMI by Member State courts and abusive ‘shifting’ of COMI by debtors and creditors. By introducing abuse restrictions into the ability of debtors to shift COMI, an additional weapon is given to the armoury of 58   Eurofood (n 16 above), para [42]. The ‘judgment’ which pre-empts other Member States’ courts is deemed to take effect at the point at which the debtor is divested of his assets and a liquidator appointed: Eurofood (n 16 above), para [54]. 59   Armour (n 19 above) 507. 60  Regulation 1346/2000 (n 8 above), Recital 22; Eurofood (n 16 above), para [42]. 61  This list has already been amended four times: see Council Regulations (EC) Nos 603/2005 [2005] OJ L100/1; 694/2006 [2006] OJ L121/1; 681/2007 [2007] OJ L159/1; 788/2008 [2008] OJ L213/1. 62  See text following n 57 above.

Abuse of European Insolvency Law? A Discussion  167 arguments that may be used by Member State courts to determine—inappropriately, and pre-emptively—that COMI is within their territory. V. Reform of the Regulation

Professor Eidenmüller’s final argument relates to potential reform of the Regulation. He suggests it would be desirable to amend the text so as to equate COMI exactly with the location of the registered office for corporate debtors. This is a suggestion with which I entirely agree.63 It would have three principal benefits. First, it would greatly increase certainty over the location of a debtor’s COMI. This would make it much easier for creditors to contract effectively with debtors in light of the insolvency procedure to be applied. Not only this, but it would actually implement the desiderata set out in Recital 13 and emphasised by the ECJ in Eurofood effectively. That is because the location of a debtor’s registered office is, in the age of internet searches, probably more objectively and readily ascertainable by third parties than any other single factor that can be considered. Second, it would reduce the scope for strategic behaviour ex post, both by debtors and creditors. As regards shifts of COMI by debtors, the ‘emigration’ of registered offices may be conditioned by domestic law on a principle of non-prejudice to creditors, so as to prevent a transfer simply to expropriate creditors.64 Such a principle could not be implemented where COMI depends on factual, rather than legal, matters. Third, linking COMI to a registered office would create the possibility for some form of regulatory competition over the selection of the appropriate insolvency law. By locating their registered office in a particular jurisdiction, debtor companies would signal to creditors and shareholders the law which would apply in their insolvency, and investors can then price this into the terms of their investment. Companies registering in jurisdictions with more efficient insolvency laws would be able to attract cheaper investment, hence giving them an incentive to gravitate towards such regimes. In an environment of uncertainty over which models are best for insolvency law, such freedom of choice may give Member States an incentive to innovate and experiment, and ultimately provide a stimulus for improved rules to develop. Furthermore, I believe that much could be achieved even without amending the text of the Regulation. This could be done by the Court treating the presumption established by Article 3(1) as a strong one, rebuttable only by the clearest evidence.65 This approach would remove any necessity for a European law principle of ‘abusive shift of COMI’ in relation to corporate debtors, which I have suggested would be problematic in any event. VI. Conclusion

The muddled conception of COMI under the European Insolvency Regulation as currently drafted seems to invite strategic behaviour by debtors, creditors, and even Member State courts and legislatures designed to secure benefits for themselves or for particular constituencies without increasing the overall returns to creditors. Efforts deployed in such   Armour (n 33 above) 407–08. See also Eidenmüller (n 33 above) 438–39.  This would surely be compatible with freedom of establishment: see text to nn 30–32 above. 65   Armour (n 33 above) 407–08. 63 64

168  John Armour activities are lost to society, and therefore the current position is problematic. Professor Eidenmüller starts from the premise that the central goal of the Regulation is to increase the overall size of the pie available for distribution to creditors—that is, to secure a more efficient resolution of the debtor’s affairs. This is a sound basis for insolvency policy, although as an interpretation of the Regulation, it is not clear to what extent the claim is normative as opposed to positive. From there, he argues for a principle of European law to restrict such ‘abusive’ COMI shifts—those which are done with a view solely to securing a larger slice of the pie for particular claimants, and not with a view to making the overall pie larger for everyone. I have suggested that a drawback of such a principle would be its potential to further complicate the determination of COMI. Moreover, in relation to corporate debtors, it is hard to see how such a principle could be operationalised; and in any event, corporate creditors typically have alternative protection available to them. This proposal is made, however, as nothing more than a stopgap measure as regards corporate debtors, subject to the limitations of the Regulation’s structure. I am in complete agreement with Professor Eidenmüller’s more comprehensive proposal: that the registered office should not only be a presumptive, but a conclusive, indicator of COMI. This would in one sweep remove both the incentives, and the opportunities, for the abuse we have discussed, in relation to corporate debtors. To this I would add that much could be achieved simply by the Court of Justice treating the presumption that COMI is the place of a corporate debtor’s registered office as hard to rebut. I am extremely grateful to Professor Eidenmüller and the conference organisers for having given me the opportunity to comment upon his chapter.

12 Abuse of Rights in EU Law: Some Reflections with Particular Reference to Financial Law Takis Tridimas I.  Abus de droit: Origins and development in EU law

T

he concept of abuse is invoked in diverse areas of law and in many legal systems to serve a variety of purposes. Its origins may be traced back to Roman law.1 A general distinction can be drawn between abuse which occurs where a person seeks to derive excessive advantages by relying on a right, thus causing disproportionate harm to others, and abuse which occurs in the improper use of a rule. Although both concepts have common elements, they are different in a number of respects. The first is primarily a concept of private law and derives from French law. It seeks to limit the exercise of private rights in the light of social objectives and, ultimately, curb the excesses of individualism. The second meaning is a broader one. Abuse is here understood as circumvention or evasion and seeks to prohibit a person from deriving a benefit which, although the benefit may result from formal compliance with a rule, in fact pursues ends which are beyond its objectives. Abuse of rights under the first meaning described above has found expression in many civil law systems. In French law, it found fertile ground in neighbour disputes.2 In its most basic form, it means that a right should not be exercised in a manifestly unreasonable manner with consequent harm to others.3 This definition stresses the objective elements of abuse and its affinity to the principle of proportionality. Whilst under some legal systems, objective harmful effects may be sufficient to constitute abuse,4 other systems require   ‘Nullus videtur dolo facere, qui suo iure utitur’ Gaius, Digest 50.17.55.   The so-called ‘troubles de voisinage’ cases: for relatively recent examples, see Cour de Cassation, 3rd Civ Ch, 3 November 1977, Recueil Dalloz et Sirey 1978.434; Cour de Cassation, 2nd Civ Ch, 3 January 1969, La semaine juridique (JCP) 1969 II 15920; and Cour de Cassation, 3rd Civ Ch, 4 February 1971, La semaine juridique (JCP) 1971 II 16781. 3  Commentators use a variety of definitions. For example, E Reid, ‘The Abuse of Rights and the Rule of Law’ (October 2004) vol 8.3 Electronic Journal of Comparative Law 2: www.ejcl.org, offers a more restrictive definition: ‘no one should exercise what is otherwise a legitimate right in a way which is solely motivated by the desire to cause annoyance to his or her neighbour’. 4   This is the case, for example, in Germany and Greece. Greek law provides a good illustration of the doctrine of abuse as used in civil law systems. Under Art 281 of the Greek Civil Code, the exercise of a right is prohibited if it manifestly exceeds the bounds provided by good faith, public morals or the social or economic purpose of the right. The doctrine of abuse is an objective one so that Art 281 may be violated even in the absence of fault. The same applies also under French law (see the cases referred to in n 2 above). 1 2

170 Takis Tridimas also a subjective element, such as an intention to harm.5 In the light of the judgment in Bradford v Pickles,6 the doctrine of abuse finds no place in English common law, although a more careful study of English law as a whole suggests a more nuanced approach.7 In European law, the doctrine of abuse was a latecomer.8 Traditionally, the ECJ did not recognise it as a general principle. The reason for this is a functional one. In filling the gaps of the Treaty, the Court developed a series of principles based on the rule of law with a view to protecting the individual vis-à-vis the exercise of public power.9 Abuse of right as a general principle of law does not have a public law ancestry. As stated above, it was developed, primarily, in the private laws of continental Member States to temper the unfettered exercise of property rights and mitigate the unfair results of contractual freedom. Although in national law, it found application in other areas such as procedural law and, in some forms, public law,10 in the formative years of Community law, it was not necessary for the ECJ to elaborate a general doctrine of abuse of rights. From the 1970s onwards, the case law made some references to the doctrine11 but they remained isolated. The Court opted for a broad definition of the fundamental freedoms suggesting that the doctrine had a limited role to play in the field of the Internal Market.12 This laissez-faire approach was understandable. In a judicial agenda of integration-building which sought to facilitate the completion of the Internal Market, the development of a doctrine of abuse of fundamental freedoms seemed to serve little purpose. As the Community legal order developed, a doctrine of abuse began to evolve. Two factors contributed to this. First, as litigants sought to stretch to the maximum the scope of Treaty freedoms, defendant governments resorted to the doctrine of abuse as a countervailing force. Secondly, the proliferation of Community legislation in areas such as indirect tax and subsidies inevitably increased opportunities for improper use of Community benefits. As a result, the Court began to articulate the elements of a general principle. Nowadays, references to the doctrine abound and have been made,

    5  This is the case, for example, under Italian law. See, the excellent analysis of R Cordara, ‘Abuse of Law: Origins and History in EU Law’, conference paper delivered 28 May 2006 at the Bar European Group Conference, Ljubljana, Slovenia.     6   Bradford Corporation v Pickles [1895] AC 587.     7  Note, for example, the importance of the concept of ‘clean hands’ in equity. More recently, the doctrine has found recognition in public law: see Art 17 of Pt 1, Sch 1 of the Human Rights Act 1998 which incorporates Art 17 of the European Convention on Human Rights (ECHR) prohibiting abuse of rights.     8  For bibliography, see among others, P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351; KE Sørensen, ‘Abuse of Right in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423; D Weber, ‘Abuse of Law’ (2004) 31 Legal Issues of Economic Integration 43; F Lagondet, ‘L’abus de droit dans la jurisprudence communautaire’ (2003) 95 Journal des tribunaux 8; L Neville Brown, ‘Is There a General Principle of Abuse of Rights in European Community Law?’ in D Curtin, T Heukels (eds), Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers, Vol II (Dordrecht, Nijhoff 1994) 511; V Karayannis, ‘L’abus de droits découlant de l’ordre juridique communautaire’ (1999) 1/2 Cahiers de droit européen 531.     9  See T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press 2006). 10  In the field of public law, Art 263 TFEU [ex Art 230 of the EC Treaty] recognises ‘misuse’ of powers as a ground of review of Union measures. Note also that the Charter of Fundamental Rights of the European Union, in line with the ECHR, prohibits abuse of rights. See Art 54 of the Charter and Art 17 ECHR. 11  See for the first reference, Case 33/74 Van Binsbergen v Bedrijfsvereniging Metaalnijverheid [1974] ECR 1299, para [13]. For subsequent cases, see for example, Case 115/78 Knoors v Secretary of State for Economic Affairs [1979] ECR 399, para [25]; Case 229/83 Leclerc v Au Blé Vert [1985] ECR 1, para [27]. 12  See for example, Case C-221/89 The Queen v Secretary of State for Transport, ex p Factortame Ltd et al [1991] ECR I-3905.

Abuse of Rights in EU Law with a Focus on Financial Law  171 among other areas, in agriculture,13 company law,14 tax,15 free movement of workers,16 free movement of establishment and inter-State corporate mobility,17 free movement of services,18 goods,19 social law, and procedural law.20 In Kofoed,21 building on its earlier judgment in Halifax,22 the ECJ referred for the first time to abuse as a general principle of Community law. One may distinguish two main categories of abuse. First, cases where a person seeks to circumvent national law by relying on a fundamental freedom and, second, cases where a person seeks to take improper advantage of a right granted by Union legislation. In both categories, the doctrine of abuse is understood with the second meaning given to it above, namely, as seeking to prevent a person from deriving a benefit which, although it may result from formal compliance with a rule, pursues ends which lie beyond its objectives. It is therefore a doctrine against the circumvention of law and not an equitable principle seeking to restrain excessive reliance on private rights as traditionally understood in civil law. To establish the existence of abuse of a right, one must first determine the outer boundaries of the right in issue. This involves, in turn, an inquiry into the objectives of the provision on which the right is based and a determination of the range of interests that it seeks to protect. Such an inquiry may be internalised, ie carried out covertly as part of the process of interpretation. The absence of an express reference to the principle does not necessarily mean that the court has failed to take into account the mix of interests involved where an abuse-centred inquiry takes place overtly.23 Finally, the principle of abuse may be concretised by legislation, namely, specific provisions which establish express prohibitions or determine the precise content of a right or the conditions under which it may be exercised. As we shall see, this is particularly relevant in the field of financial law where the concept of abuse of rights is highly concretised. Another example of such embodiment is provided by the Unfair Terms Directive which provides an indicative list of unfair contractual clauses.24 This chapter proceeds as follows. It first examines the development of the principle in the context of EU legislation starting with earlier cases in corporate law and agriculture and then looking at the field of taxation. It then looks at the development of the principle in the field of free movement by reference to the right of establishment and the freedom to  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569.  See for example, Case C-373/97 Diamantis v Elliniko Domosio [2000] ECR I-1705. 15  See for example, Halifax (n 22 below). 16  Case C-127/08 Blaise Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I-6241; Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607. See for example, Case 197/86 Brown v Secretary of State for Scotland [1988] ECR 3205. 17  Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459. 18  Case C-23/93 TV 10 v Commissariaat voor de Media [1994] ECR I-4795; Case C-148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487. 19   Leclerc (n 11 above). 20  Case 104/79 Foglia v Novello [1980] ECR 745; Case 244/80 Foglia v Novello [1981] ECR 3045. 21  Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795. 22  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 23  See for example, Case C-244/95 P Moskof AE v Ethnikos Organismos Kapnou (EOK) [1997] ECR I-6441. 24  See Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts [1993] OJ L95/29. For an express reference to abuse in secondary law, see for example Directive 2004/38 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States [2004] OJ L158/77, Art 35 of which provides that ‘Member States may adopt the necessary measures to refuse, terminate or withdraw any right conferred by this Directive in the case of abuse of rights or fraud, such as marriages of convenience’. 13 14

172 Takis Tridimas provide services. It finally looks, more specifically, at the field of financial law. By contrast, abuse of rights in the fields of social law and the free movement of workers falls beyond the scope of this chapter. iI.  The Road to the Establishment of a General Principle

At first, in Diamantis,25 the ECJ refused the invitation to establish a general principle of abuse of rights. The applicant was a former shareholder in a public company which, because of its indebtedness, had been placed under the provisional administration of the Organismos Ikonomikis Anasinkrotisis Epikhiriseon AE (OAE), a state body responsible for the restructuring of ailing undertakings. As part of the restructuring process, OAE had altered the capital of the company, effecting two increases and one reduction. The shareholders had been invited, but declined, to exercise their pre-emption rights when the first increase of capital took place. Once OAE’s intervention was successfully completed and the company resumed normal commercial operations, Mr Diamantis brought an action seeking a declaration that the alterations in the corporate capital were invalid on the ground that they had taken place without a resolution of the general meeting and were therefore contrary to Article 25(1) of the Second Company Law Directive which, in previous case law, the Court had held to be directly effective. The Greek Government and OAE counter-argued that, in the circumstances, the invocation of Article 25(1) was an abuse of right contrary to Article 281 of the Greek Civil Code.26 The Court first reiterated that Member States were not entitled to derogate from Article 25(1). It stated however that Community law cannot be relied on for abusive or fraudulent ends. That would be the case if a shareholder, in reliance on Article 25(1), brought an action for the purpose of deriving, to the detriment of the company, an improper advantage manifestly contrary to the objective of that provision. The Court then stated that national courts may deny a person the benefit of the provisions of Community law on the basis of objective evidence of abuse. It stressed however that the application of a national rule such as Article 281 must not detract from the full effect and uniform application of Community law in the Member States.27 The Court concluded that, in the circumstances of the case, the uniform application and full effect of Community law would not be compromised if the national court were to find that reliance on Article 25(1) was an abuse of rights. This is because, of the remedies available, the applicant in the main proceedings had chosen a remedy that would cause such serious damage to the legitimate interests of others that it appeared manifestly disproportionate. If the action for declaring the alteration of capital invalid was upheld, several operations that took place at the time when the company was under state administration could be affected. Moreover, the invalidity of the capital alterations would inevitably affect the rights of bona fide third parties.28 The judgment in Diamantis is interesting from a number of perspectives. Previously, in Pafitis and Kefalas the Court had insisted on the application of Article 25(1) and was 25  C-373/97 Dionisios Diamantis v Elliniko Dimosio (Greek State) and Organismos Ikonomikis Anasinkkrotisis Epikhiriseon AE (OAE) [2000] ECR I-1705. 26  See Diamantis (n 25 above). 27   Diamantis (n 25 above), paras [33–34]. 28   Diamantis (n 25 above), para [40].

Abuse of Rights in EU Law with a Focus on Financial Law  173 not particularly receptive to the argument on abuse of rights based on Article 281 of the Greek Civil Code.29 However, in successive actions, the detrimental effects of relying on the Second Company Law Directive became more apparent, and in Diamantis, the problems and issues involved became clearer. Diamantis is also notable because the Court accepted a limitation on the effects of directives on the basis of the doctrine of abuse of rights as it exists in Greek law rather than developing a Community general principle of abuse of rights. The latter would have, in fact, been preferable. Since the concept of abuse determines, in effect, the outer limits of a right, such determination in relation to Union rights is a matter for the Court to decide. In Emsland-Stärke,30 by contrast, the Court took the first steps towards defining the elements of abuse as a matter of EU law. It held that the concept of abuse entails two elements, an objective and a subjective one. It first presupposes a situation where, although the conditions laid down by EU rules have formally been observed, the purpose of these rules has not been achieved. Secondly, it requires an intention to obtain an advantage from the European rules by creating artificially the conditions laid down for obtaining it.31 In that case, a company had obtained subsidies (export refunds) for exporting products to Switzerland but, immediately after the completion of customs formalities, the products were reimported to Community territory and released for circulation there upon payment of the applicable import duties. At the time, the subsidies payable for exportation were twice as high as the subsidies payable for marketing within the Community. This, in combination with low import duties, made the detour to Switzerland profitable. The German authorities sought to recover the export refund paid and, in the ensuing litigation, the Bundesfinanzhof made a reference for a preliminary ruling. The Court held that all the formal conditions for the grant of the export refunds had been fulfilled. It raised, however, the question whether, in view of the circumstances of the case, a purely formal exportation for the sole purpose of benefiting from export refunds ran counter to the objectives of the applicable regulation. After referring to the objective and subjective elements of abuse, it held that it was for the national court to establish their existence but provided it with a number of guidelines. In effect, the Court applied a general principle of abuse of rights although it did not recognise its existence as such. III.  Tax cases

The Court had the opportunity to examine the doctrine of abuse in relation to the Sixth VAT Directive in Halifax v Commissioners of Customs and Excise.32 The question was raised in the context of a series of transactions between two subsidiaries of the Halifax Bank and independent contractors under a scheme designed to reduce the tax burden of the Halifax Group. The Inland Revenue refused value-added tax (VAT) relief to the subsidiaries, taking the view that only the contractors and not the subsidiaries themselves had actually supplied construction services to Halifax. On a reference for a preliminary 29  See Case C-441/93 Pafitis et al v Trapeza Kentrikis Ellados [1996] ECR I-1347, paras [68–70]; and Case C-367/96 Alexandros Kefalas et al v Elliniko Dimosio and Others [1998] ECR I-2843, para [20]. Note that in Diamantis the Court did not follow the Opinion of AG Saggio who considered that Art 25(1) of the Second Company Law Directive prohibited reliance on Art 281 of the Greek Civil Code. 30   Emsland-Stärke (n 13 above). 31   Emsland-Stärke (n 13 above), paras [52–53]. 32   Halifax (n 22 above).

174 Takis Tridimas ruling, the Court opted for an objective interpretation of the Sixth VAT Directive. It held that the question whether a given transaction is carried out for the sole purpose of obtaining a tax advantage is entirely irrelevant in determining whether it constitutes a supply of goods or services for the purposes of the Directive.33 But it then carried on to examine whether, under the Directive, a taxable person has no right to deduct input VAT where the transactions on which that right is based constitute an abusive practice. The Court held that the application of Community legislation cannot be extended to cover abusive practices, namely transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law.34 It saw the doctrine of abuse as a balance between, on the one hand, preventing the claiming of unjustified tax benefits and, on the other hand, the need to ensure certainty and foreseeability in the application of Community legislation.35 It pointed out that the Sixth Directive did not prohibit taxpayers from choosing to structure their business so as to limit their tax liability.36 On that basis, it came to the conclusion that in the sphere of VAT, abuse exists where two conditions are fulfilled: first, although the transactions in question comply formally with the conditions for the application of the VAT Directive, they result in the accrual of a tax advantage, the grant of which is contrary to the purpose of the Directive; secondly, it is clear on the basis of objective factors that the essential aim of the transactions is to obtain a tax advantage.37 The first requirement relates to the objectives of the law and stresses the need for misuse whilst the second requirement relates to the purposes of the transaction. Although the Court viewed those as separate requirements, they are in fact closely connected. The Court provided the national court with a number of guidelines, leaving it to the latter to determine the ‘real substance and significance’ of the transactions concerned.38 It attributed particular importance to the principle of fiscal neutrality which underlies the Sixth Directive. Whether the aim of the transaction is to obtain a tax advantage is not a subjective test but must be apparent from a number of objective factors and be verified in accordance with the rules of evidence of national law, provided that the effectiveness of EU law is not undermined.39 The burden imposed on the authorities here appears fairly high: abuse does not exist ‘where the economic activity carried out may have some explanation other than the mere attainment of tax advantages’.40 The Court then continued to determine the consequences of a finding of abuse. It held that abuse in the context of VAT does not itself entitle the Member State to impose penalties but only an obligation to repay any benefit unduly gained.41 A penalty may be  See Halifax (n 22 above), para [59].   Halifax (n 22 above), para [69]. See also Case 125/76 Peter Cremer v Bundesanstalt für landwirtschaftliche Marktordnung [1977] ECR 1593, para [21]; Case C-8/92 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas [1993] ECR I-779, para [21]. 35   Halifax (n 22 above), para [72]. 36   Halifax (n 22 above), para [73]. 37   Halifax (n 22 above), paras [74–75]. 38   Halifax (n 22 above), para [81]. 39   Halifax (n 22 above), paras [75–76]. 40   Halifax (n 22 above), para [75]. To establish artificiality, the national court may take account of links of a legal, economic and/or personal nature which may exist between the operators involved in the scheme for reduction of the tax burden: see para [81] of the judgment. 41  It follows that transactions involved in an abusive practice must be redefined so as to re-establish the situation that would have prevailed in the absence of the transactions constituting that abusive practice: see Halifax (n 22 above), paras [93–94]. 33 34

Abuse of Rights in EU Law with a Focus on Financial Law  175 imposed only if there is a clear and unambiguous legal basis in national law,42 and the principle of proportionality is respected.43 Halifax was followed by Kofoed.44 The case concerned the interpretation of Directive 90/434, the purpose of which is to eliminate fiscal barriers to cross-border restructuring. The Directive provides essentially that, where a company purchases the shares of another company in exchange for the allotment to its shareholders of shares in the acquiring company, such an allotment is not subject to taxation. The Directive incorporates an express anti-abuse provision. Article 11(1)(a) provides that, by way of exception and in specific cases, a Member State may refuse to apply or withdraw the benefit of the Directive where it appears that the exchange of shares has tax evasion or tax avoidance as its principal objective or as one of its principal objectives. The Court held that Article 11(1)(a) reflects the general Community law principle that abuse of rights is prohibited.45 Under that principle: Individuals must not improperly or fraudulently take advantage of provisions of Community law. The application of Community legislation cannot be extended to cover abusive practices, that is to say, transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law.46

In the circumstances of the case, the Court accepted that there was some evidence which might justify the application of the anti-abuse provision of Article 11(1)(a). Given, however, that the Directive had not been implemented into national law, it proceeded to examine whether that provision could apply in the absence of national implementing legislation. The Court’s answer was nuanced. It recalled that directives may not in themselves be relied upon by national authorities against individuals. It pointed out, however, that the transposition of a directive may, depending on its content, be achieved through a general legal context. What matters is that the legal situation arising from the national transposition measures is sufficiently precise and clear and that the persons concerned are put in a position to know the full extent of their rights and obligations. Where that is the case, transposition of a directive into national law does not necessarily require the adoption of specific provisions in national law.47 Furthermore, the Court pointed out the duty of consistent interpretation. Taking a narrow view of Arcaro,48 it held that while the requirement of consistent interpretation cannot reach the point where a directive by itself may create obligations for individuals or determine or aggravate criminal liability, a Member State may nevertheless impose a directive-compliant interpretation of national law on individuals. Thus, in Kofoed, whilst acknowledging that Article 11(1)(a) reflected a general principle of abuse of right, the ECJ stopped short of recognising that the principle was self-executing. This contrasts with its judgment in Mangold,49 where it established the   Halifax (n 22 above), para [93].   Halifax (n 22 above), para [92]. The Court also stressed that any penalty imposed must not have the effect of undermining fiscal neutrality, which is a fundamental principle of the common system of VAT. See Case C-454/98 Schmeink & Cofreth AG & Co KG v Finanzamt Borken and Manfred Strobel v Finanzamt Esslingen [2000] ECR I-6973, para [59]. 44   Kofoed (n 21 above). 45   Kofoed (n 21 above), para [37]. 46   Kofoed (n 21 above), para [37]. 47   Kofoed (n 21 above), para [44]. 48   Kofoed (n 21 above), paras [41–42]. 49  Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9981. 42 43

176 Takis Tridimas prohibition of discrimination on grounds of age as an unwritten general principle of law and accorded it full horizontal effect. This difference in approach may be explained by a number of reasons. In Mangold the dispute was between individuals whilst in Kofoed it was conversely vertical, ie involved a state authority relying on a directive against an individual. More importantly, Mangold involved equal treatment, which is a fundamental constitutional principle receptive to a dynamic interpretation. By outlawing the German legislation in issue, the ECJ took a specific ideological stance. Abuse of right, by contrast, is an exceptional principle restricting rights and is defensive in character. It is a corrective intervention seeking to ensure that economic benefits remain faithful to the objectives of the law. IV. Internal Market

The starting point of the case law is that the exercise of free movement rights to benefit from a more favourable legal regime in another Member State is entirely legitimate. Thus, the fact that a company exercises its right of establishment to take advantage of more favourable tax legislation50 or a preferential company law regime51 is not in itself an abuse. The case law also recognises, however, that individuals and undertakings may not improperly take advantage of provisions of EU law or circumvent their national legislation.52 A national measure restricting freedom of movement may be justified where it specifically relates to wholly artificial arrangements aimed at circumventing the application of the legislation of the Member State concerned.53 Determining whether a Union freedom has been exercised improperly is not an easy task. Overall, in the context of free movement, the Court has understood the concept of abuse narrowly. This approach is in keeping with the objectives of the Internal Market, and it is submitted that the case law has struck the right balance. On the one hand, freedom of inter-State trade and economic integration necessarily have a deregulatory effect. A degree of regulatory arbitrage is part of the rights guaranteed by the Treaty and can be seen as economically beneficial. It provides choice and encourages regulatory experimentation. Furthermore, the risk that deregulation may lead to a race to the bottom has not materialised. On the other hand, the case law incorporates safety valves to limit excesses. These include, for example, a skilful use of proportionality54 and the doctrine of abuse. 50  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995, para [36]; Case C-364/01 The heirs of H Barbier v Inspecteur van de Belastingdienst Particulieren/Ondernemingen buitenland te Heerlen [2003] ECR I-15013, para [71]. 51  Case 79/85 DHM Segers v Bestuur van de Bedrijfsvereniging voor Bank- en Verzekeringswegen, Groothandel en Vrije Beroepen [1986] ECR 2375; Centros (n 17 above), para [27]; Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155, para [96]. 52  See Knoors (n 11 above), para [25]; Case C-61/89 Marc Gaston Bouchoucha [1990] ECR I-3551, para [14]; Centros (n 17 above), para [24]. 53   Cadbury Schweppes (n 50 above), para [51]; Case C-324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I-11779, para [37]. 54   Thus, the less restrictive alternative test does not necessarily mean that a restriction imposed by a Member State will be judged to be disproportionate if another Member State imposes a less restrictive requirement: see for example, Case C-262/02 Commission v France [2004] ECR I-6569, para [37] (concerning the Loi Evin); Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-9609, para [38]. Proportionality is not a mechanical exercise but a balancing test influenced by multifarious considerations.

Abuse of Rights in EU Law with a Focus on Financial Law  177 We will examine here three leading cases which relate to the broader field of commercial law: Centros,55 Cadbury Schweppes56 and TV 10.57 The first celebrates freedom of incorporation, the second illustrates abuse in the field of tax law and represents an advanced attempt to deal with the doctrine, whilst the third provides a rare example where abuse was found to exist. The discussion is by no means exhaustive. Issues of abuse have been raised in a number of other cases, including cases in the field of free movement of workers and other individuals58 but these fall beyond the scope of this study. A.  Freedom of Incorporation The basic findings of Centros may be summarised as follows. The Court held that a host State may not refuse to register a branch of a company formed in accordance with the law of the State where its registered office is located on the ground that the branch is intended to enable the company to carry on all its economic activity in the host State without complying with the rules of the latter concerning the payment of share capital. The Court accepted that Member States may take measures in order to prevent their nationals from circumventing their national legislation under the pretext of the rights guaranteed by the Treaty. It pointed out, however, that the provisions of national law application of which the applicants sought to avoid, namely the Danish rules concerning the subscription of share capital, related to the formation of companies and not to the carrying on of a trade or profession.59 The Court thus drew a distinction between rules of the host State pertaining to access and rules pertaining to the exercise of an activity. Confirming the earlier judgment in Segers,60 it held that the fact that a national of a Member State chooses to set up a company in the Member State with the least restrictive rules, with the sole objective of carrying on business in another Member State through the establishment of a branch there, is not in itself an abuse of the right of establishment. In a Single Market, the right to form a company in accordance with the law of a Member State and to set up branches in other Member States is inherent in the exercise of the freedom of establishment.61 The Court found that the refusal of the Danish authorities to register the branch was not justified in the interests of the protection of corporate creditors. Usually, when the Court applies the principle of proportionality, it focuses on the necessity rather than the suitability of a restriction. But in Centros it questioned also its suitability. It held that the refusal of the Danish authorities was not such as to attain the objective of protecting creditors since, if the company concerned had conducted business in the United Kingdom, its branch would have been registered in Denmark, even though Danish creditors might have equally been exposed to risk. Also, since the company held itself out as one governed by the law of England and not as one governed by Danish law, its creditors were on notice that it was subject to laws different from those governing the formation of domestic   Centros (n 17 above).   Cadbury Schweppes (n 50 above). 57   TV 10 (n 18 above). 58  See the cases mentioned in n 16 above and note also Case C-200/02 Zhu and Chen v Secretary of State for the Home Department [2004] ECR I-9925. 59   Centros (n 17 above), para [26]. 60   Centros (n 17 above). 61   Centros (n 17 above), para [27]. 55 56

178 Takis Tridimas companies.62 Both these arguments are open to challenge. For one thing, it is reasonable for a regulator to assume that a company which has already conducted trade in its State of incorporation poses less of a commercial risk than a newly formed company which has transacted no business whatsoever there and is beyond the reach of its home regulator. For another, although it is true that Centros traded as a British company, creditors could not be assumed to be well-versed with the differences between Danish and English company law. By relying on caveat emptor to facilitate free movement, the judgment transfers a greater risk to creditors. This is not to say that Centros is wrongly decided but to put it in perspective. The Court accepted in effect that, in the circumstances of the case, it was worth upholding free movement at the expense of an increase in the commercial risks of creditors. The Court added that the fact that the host State may not refuse to register a branch of a company formed in accordance with the law of its home State does not preclude it from adopting any appropriate measure for preventing or penalising fraud, either in relation to the company itself or in relation to its members, where it has been established that they are in reality attempting to evade their obligations towards creditors. But combating fraud does not justify simpliciter a practice of refusing to register a branch of a company which has its registered office in another Member State.63 Centros clearly takes a pro-free market view and makes clear that harmonisation is not a prerequisite for inter-State corporate movements. The judgment establishes freedom of incorporation within the European Union and, together with Überseering BV and other subsequent judgments,64 may be seen as a triumph for the doctrine of incorporation and a corresponding defeat for the real seat doctrine. The Court accepted that Member States may take measures in order to prevent their nationals from circumventing their national legislation under cover of the rights guaranteed by the Treaty. The national court, however, must examine each case of alleged abuse on its facts. It may take account of abuse only on the basis of objective evidence and it must assess whether abuse has occurred in the light of the objectives of the provisions of EU law in issue. Centros begs a number of questions: in what kind of circumstances could the Court accept limitations imposed by the host State? What type of restriction would be permissible? What if, for example, the promoters of Centros had been subject to a disqualification order issued under Danish legislation or convicted for commercial fraud in Denmark? Could this justify restrictions on access to the Danish market? Also, could restrictions on the exercise of commercial activities be justified to the branch? What if, for example, Danish law imposed an obligation on companies to have all minimum capital as paid up in order to bid for certain contracts or engage in certain activities or have access to certain types of credit? The answer is that each restriction should be judged on its merits. As a general rule, artificial arrangements are less likely to be excused in the context of tax law. In Cadbury Schweppes, the Court emphasised that Member States may take measures to prevent ‘wholly artificial arrangements, which do not reflect economic reality’65   Centros (n 17 above), paras [35–36].   Centros (n 17 above), para [38].  Case C-208/00 Überseering BV v Nordic Construction Company Baumanagement GmbH [2002] ECR I-9919; Inspire Art (n 51 above). Cf the more recent judgment in Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641. 65   Cadbury Schweppes (n 50 above), para [55]. 62 63 64

Abuse of Rights in EU Law with a Focus on Financial Law  179 and which are aimed at circumventing national legislation. In particular, the right of establishment does not preclude Member States from being wary of ‘letter-box’ or ‘front’ companies.66 B.  Freedom of Establishment and Tax In Cadbury Schweppes,67 a company resident in the UK had set up two subsidiaries in Ireland solely for the purpose of benefiting from the lower tax regime available there. The UK Inland Revenue taxed their profits under legislation which subjected a resident company to taxation on the profits made by its controlled foreign subsidiaries. The legislation applied when the foreign subsidiary was subject to a level of taxation which was less than three-quarters of the amount of tax which would have been payable in the United Kingdom. The Court held that the UK law put at a disadvantage resident companies with foreign subsidiaries established in Member States with a lower level of taxation and then proceeded to determine whether the resultant restriction on freedom of establishment could be justified by overriding reasons of public interest. It reiterated the following general principles. First, the fact that a parent company derives a tax advantage by setting up a subsidiary in another Member State where there is low taxation does not by itself entitle the tax authorities to offset that advantage by imposing less favourable tax treatment on the parent company.68 Secondly, preventing the reduction of tax revenue is not included among the express grounds of derogation of Article 46(1) EC [now Article 52(1) TFEU], nor is it a matter of overriding general interest which would justify a restriction on a freedom introduced by the Treaty.69 Thirdly, the mere fact that a resident company establishes a subsidiary in another Member State cannot set up a general presumption of tax evasion and justify a measure which compromises the exercise of a fundamental freedom.70 The Court recalled, however, that a national measure restricting freedom of establish­ ment may be justified where it specifically relates to wholly artificial arrangements aimed at circumventing the application of the legislation of the Member State concerned.71 It held that, in order for a restriction on the freedom of establishment to be justified on the ground of prevention of abusive practices: ‘[T]he specific objective of such a restriction must be to prevent conduct involving the creation of wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory.’72 The Court held that the fact that the incorporation of the subsidiaries in Ireland was prompted by the intention to obtain tax relief did not suffice to conclude that there was a wholly artificial arrangement. In addition to subjective intention, there must be objective

66   Cadbury Schweppes (n 50 above), para [68]; see also Case C-341/04 Eurofood IFCS Ltd—Enrico Bondi v Bank of America et al [2006] ECR I-3813. 67   Cadbury Schweppes (n 50 above). 68   Cadbury Schweppes (n 50 above), para [49]. 69   Cadbury Schweppes (n 50 above), para [49]. 70   Cadbury Schweppes (n 50 above), para [50]. 71   Cadbury Schweppes (n 50 above), para [51]. 72   Cadbury Schweppes (n 50 above), para [55]. See also the discussion of abuse by AG Poiares Maduro in Cadbury Schweppes (n 50 above).

180 Takis Tridimas circumstances showing that, despite formal observance of the conditions laid down by EU law, the objective pursued by freedom of establishment has not been achieved.73 Thus, in Cadbury, the ECJ remained faithful to the philosophy of the freedom of establishment, stressing the objective element of abuse. It held that an artificial arrangement exists only where the elements of establishment are not present, ie only if there is no actual establishment in the host State or if the corporation does not intend to engage in the pursuit of genuine economic activities there. On that basis, it found the motives test provided by UK tax law too broad and, as such, incompatible with Article 43 EC [now Article 49 TFEU].74 Finally, the Court also pointed to the procedural aspects of abuse. It held that the undertaking in question must be given an opportunity to produce evidence that a subsidiary established in another Member State is not fictitious and its activities are genuine.75 Although it did not comment specifically on the allocation of the burden of proof, it pointed out that the competent national authorities may resort to the procedures for collaboration and exchange of information between national tax administrations provided by Community law to assess the evidence provided by the undertaking in question.76 C.  Freedom to Provide Services The Court took a less generous approach in TV 1077 which related to the freedom to provide services. The Dutch broadcasting authorities refused to TV 10, an undertaking based in Luxembourg, a licence to transmit television programmes in the Netherlands on the ground that it had been established in Luxembourg solely for the purpose of avoiding Dutch legislation applying to domestic broadcasters. The Court held that a Member State may not be denied the power to prevent service providers whose activities are entirely or principally directed towards its territory from exercising the freedom to provide services for the purpose of avoiding the rules which would be applicable to them if they were established within that State.78 That pronouncement had first been made in Van Binsbergen but there had the character of a general reserve in the scope of free movement without much direct impact on the proceedings.79

  Cadbury Schweppes (n 50 above), para [64].   The legislation provided for an exception from the obligation of the parent company to pay tax on profits of foreign subsidiaries when the so-called ‘motive test’ was satisfied. This involved two cumulative conditions. First, where the transactions which gave rise to the profits of the subsidiary for the accounting period in question produced a reduction in United Kingdom tax compared to that which would have been paid in the absence of those transactions, the resident company had to show that such a reduction was not the main purpose, or one of the main purposes, of those transactions. Secondly, the resident company had to show that it was not the main reason, or one of the main reasons, for the subsidiary’s existence in the accounting period concerned to achieve a reduction in United Kingdom tax by means of the diversion of profits. A diversion of profits was deemed to exist if it was reasonable to suppose that, if the foreign subsidiary did not exist, the receipts would have been received by, and been taxable in the hands of, a company resident in the UK. 75   Cadbury Schweppes (n 50 above), para [70]. 76   Cadbury Schweppes (n 50 above), para [71]. See especially, Council Directive 77/799 EEC concerning mutual assistance by the competent authorities of the Member States in the field of direct taxation [1977] OJ L336/15. 77   TV 10 (n 18 above). 78   TV 10 (n 18 above), para [20]. 79   Van Binsbergen (n 11 above), para [13]. 73 74

Abuse of Rights in EU Law with a Focus on Financial Law  181 Although in TV 10 its reasoning was brief, the ECJ was concerned, in effect, with defining the outer limits of the freedom to provide services. Dutch law imposed certain restrictions on domestic broadcasters which intended to maintain the pluralist and public service content of programmes and which, in previous cases, had been found to pursue legitimate cultural policy objectives.80 TV 10 was managed by Dutch nationals. Its target audience was the Dutch public, most of its employees were from the Netherlands, and it broadcasted advertisements made in the Netherlands. In light of those considerations, the Court saw the broadcasting activities of TV 10 as an attempt to circumvent the legitimate requirements of Dutch law. TV 10 suggests that the concept of abuse is wider in the freedom to provide services than in the right of establishment.81 This is a reflection of the difference between the freedoms. Whilst a Union national who exercises freedom of establishment becomes subject to the law of the host State, a person who provides services falls in principle outside the competence of the authorities of the State where the services are provided.82 The possibilities of regulatory arbitrage and the risk of using inter-State movement to circumvent national requirements is much higher in relation to the freedom to provide services, hence the Court’s greater vigilance. TV 10 remains, in fact, the only case where the Court made a conclusive finding that there was an abusive exercise of free movement. In all other cases, it has either rejected the claim or left it to the national court to decide. Although TV 10 was decided much earlier than Centros and Cadbury and at a time when the ECJ had not yet laid down the specific elements of the principle of abuse, it still appears to be good law. Subsequent judgments refer to it,83 thus suggesting that it is a reliable precedent. Furthermore, if one were to apply the objective and the subjective elements of abuse as articulated in Halifax and Cadbury to the facts of TV 10, it would appear that both elements were satisfied. V.  The principle of abuse in financial law

The doctrine of abuse, as a general principle of EU law, has had little direct impact in the field of financial law. Before entering into a more detailed examination of this area, the following preliminary points may be made. First, since the finding of abuse involves an inquiry to determine the outer limits of a right, our starting point has to be the Court’s general approach to the free movement of financial services. We will thus need to examine the application of the fundamental freedoms in financial law. Secondly, securities law provides a good example of the concretisation of the concept of abuse. Following the Financial Services Action Plan,84 securities law has been, in terms of legislative activity, the fastest growing area of EU law. Since 1998, the Union has adopted more than fifty

80  See Case C-288/89 Collectieve Antennevoorziening Gouda v Commissariaat voor de Media [1991] ECR I-4007, paras [22–23]; Case C-353/89 Commission v Netherlands [1991] ECR I-4069, paras [3, 29 and 30]; and Veronica (n 18 above), para [9]. 81   This is also the view of AG Stix-Hackl in Fidium Finanz (see the discussion at n 99 below). 82  See AG Mayras in Van Binsbergen (n 11 above) 1317. 83  See for example, Centros (n 17 above), para [24]. 84  See EC Commission, Financial Services: Building a Framework for Action COM (1998) 625 and Commission Communication of 11 May 1999 ‘Implementing the framework for financial markets: action plan’ COM (1999) 232.

182 Takis Tridimas measures, setting in motion a process towards the federalisation of securities regulation.85 Also, through the introduction of the Lamfalussy process, securities law has provided a fertile ground for regulatory experimentation. The end result has been the adoption of detailed measures and guidelines at various Lamfalussy levels, articulating specific regulatory requirements.86 In financial law, the term abuse has acquired specific statutory meaning following the adoption of Directive 2003/6 on Market Abuse.87 The Directive prohibits insider trading and market manipulation with a view to enhancing investor confidence in the securities markets. It is one of the Level 1 measures adopted under the Lamfalussy process,88 and makes for a great improvement over its predecessor.89 It provides for a higher degree of harmonisation, a more contemporary definition of insider dealing90 and, most importantly, in contrast to the previous Directive which only covered insider trading, it prohibits also market manipulation. Outside the scope of Directive 2003/6, the term abuse is used loosely in various contexts. Reference to the term is made, for example, in takeovers in relation to the board of directors of a target company who may seek to take action to frustrate a bid in abuse of their powers. Reference is also made to the ‘abuse’ of financial markets for the purposes of terrorist financing.91 Abuse has also been used with a specific meaning in recent post-financial crisis policy initiatives. Thus, to promote liquidity and credit supply, the Commission has allowed governments to provide guarantees to banks, invoking Article 107(3)(b) TFEU [ex Article 87(3)(b) EC] which allows the Commission to authorise state aid to remedy a serious disturbance in the economy of a Member State. To minimise, however, distortions in the conditions of competition and the ‘potential abuse’ of such guarantees by beneficiary banks, the Commission required Member States to impose behavioural constraints on beneficiaries, such as restrictions on advertising, share repurchases or issue of stock options to management.92 85  See for a summary of these measures, EC Commission, FSAP Evolution Chart: ec.europa.eu/internal_ market/finances/docs/actionplan/index/061003_measures_en.pdf. Note that the chart does not contain the Level 2 measures adopted under the Lamfalussy process. For an excellent discussion of the harmonisation programme, see N Moloney, EC Securities Regulation, 2nd edn (Oxford, Oxford University Press 2008). 86  See Final Report of the Committee of Wise Men on the Regulation of European Securities Markets, Brussels 2001 (Lamfalussy Report). The Report identified four levels of regulation: Level 1 consists of Framework principles adopted by the Council and the Parliament through the co-decision procedure. Level 2 consists of detailed rules adopted by the Commission under the comitology procedure with the participation of the EU Securities Regulators Committee. Level 3 consists of enhanced cooperation and networking among EU securities regulators to ensure consistent and equivalent transposition of Level 1 and 2 legislation through the Committee of European Securities Regulators (CESR) (common implementing standards). Level 4 consists of strengthened enforcement, underpinned by enhanced cooperation between the Member States. 87  Directive 2003/6 on insider dealing and market manipulation (market abuse) [2003] OJ L96/16. 88  See Lamfalussy Report, levels of regulation (n 86 above). 89  Directive 89/592/EEC of 13 November 1989 coordinating regulations on insider dealing [1989] OJ L334/30. 90  Notably, the Directive includes in the definition of insider a person who possesses inside information by virtue of his criminal activities: see Art 2(1)(d). This envisages cases where inside information originates from criminal activities, the preparation or execution of which could have a significant effect on the prices, and was included primarily to prohibit trading by those involved in terrorist activities. See Recitals 14 and 17 of the Directive. 91   This, which has received little attention by commentators, involves primarily the use of Informal Value Transfer Systems (IVTS) in what are otherwise legitimate transactions for the purposes of financing terrorist activities. See BJ Field and AN Guiora, ‘Using and Abusing the Financial Markets: Money Laundering as the Achilles Heel of Terror Financing’ (2007) 29 University of Pennsylvania Journal of International Economic Law 7. 92  See Commission Communication, The application of State aid rules to measures taken in relation to financial institutions in the context of the current global financial crisis [2008] OJ C270/8, para [27].

Abuse of Rights in EU Law with a Focus on Financial Law  183 Finally, some practices may be seen as abusive, and thus prohibited, on a temporary basis in the light of the prevailing circumstances in the financial markets. A good example is provided by short-selling. This consists in selling an asset, such as a security, currency or a commodities contract, which the seller does not possess in the hope that he will be able to acquire it at a lower market price before delivery is due. While under normal market conditions, short-selling is considered to be a legitimate market technique, in the extreme circumstances surrounding the financial crisis of September 2008, short-selling was seen as contributing to disorderly markets. As a result, it was temporarily prohibited by both the Financial Services Authority in the UK and the US Securities and Exchange Commission in relation to certain securities.93 Within the confines of this chapter, we will focus on the application of the Treaty freedoms on financial law, the free movement of capital, the single passport principle in relation to investment services, and the concept of market manipulation. The analysis is not intended to be exhaustive but rather to illustrate the various contexts in which the doctrine of abuse may be used. A.  The Treaty Freedoms and Financial Law In the context of financial services, as in other areas of EU law, the Court has interpreted free movement rights broadly. It has adopted an access to the market criterion and applied the principle of proportionality strictly. An example is provided by Caixa Bank.94 In that case, the Court held that French law which prohibited banks from offering interest in current accounts was in breach of the right of establishment even though it applied to all banks operating in France without discrimination. The Court reiterated that Article 43 EC [now Article 49 TFEU] encompasses all measures which prohibit, impede or render less attractive the exercise of the freedom of establishment.95 The French prohibition impeded access to the market for subsidiaries of foreign banks as it made it more difficult for them to raise capital from the public, and deprived them of the possibility to compete more effectively with French banks which have an extensive network of branches.96 The Court also rejected the submission that the restriction served overriding requirements relating to the public interest. The French Government argued that allowing the payment of interest in current accounts would substantially increase the operating costs of banks which, to recover them, would introduce charges for banking services hitherto offered free. This would have a detrimental effect, especially on poorer consumers. The ECJ found that, although that objective was legitimate, the prohibition was disproportionate. Even if allowing the payment of interest were to increase the cost of basic banking services, there were other ways to protect consumers. It was possible, for example, to provide for different kinds of current accounts, some of them offering basic services free of charge, and others making charges for services previously provided free, such as the issuing of cheques.97  See BBC online news, 19 September 2008: news.bbc.co.uk/2/hi/business/7624012.stm.  Case C-442/02 Caixa Bank v Ministère de l’Économie, des Finances et de l’Industrie [2004] ECR I-8961.   Caixa Bank (n 94 above), para [11] and, for previous case law, see for example, Case C-55/94 Reinhard Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano [1995] ECR I-4165, para [37]. 96   Caixa Bank (n 94 above), para [13]. 97   Caixa Bank (n 94 above), paras [21–22]. 93 94 95

184 Takis Tridimas Cases such as Caixa Bank suggest that, as a starting point, the Court opts for a broad definition of the fundamental freedoms in the field of financial services.98 The concept of abuse came to the fore in Fidium Finanz.99 The applicant in the main proceedings was a company based in Switzerland which offered sub-prime loans to customers resident abroad via its website and credit intermediaries. Customers resident in Germany accounted for ninety per cent of its business. The German authorities prohibited it from doing business in Germany on the ground that it had not obtained authorisation as a credit institution by the German authorities. The applicant argued that the prohibition was an unjustified restriction on the free movement of capital. The Court held that, although the situation fell both within the scope of the freedom to provide services and the free movement of capital, the requirement to obtain authorisation affected primarily the freedom to provide services, its restrictive effects on capital movements being merely an inevitable consequence. It then pointed out that, in contrast to Article 56 EC [now Article 63 TFEU], which liberalises both inter and intra Community capital movements, Article 49 EC [now Article 56 TFEU] only applies to Community nationals. It followed that, as a company established in a non-Member country, Fidium Finanz could not rely on the freedom to provide services. By this construction, the Court avoided entering into the merits of the authorisation requirement. The Advocate-General, by contrast, took the view that the free movement of capital was engaged, and proceeded to examine the argument of abuse. Advocate-General StixHackl considered that the subjective and objective elements which are required under the Halifax case law also apply in relation to the free movement of capital but left to the national court to decide whether there was abuse in the circumstances. In relation to the subjective element, she pointed out that the national court should determine whether Fidium Finanz intended to circumvent German law on banking supervision or obtain an advantage by creating artificially the conditions for the application of the free movement of capital.100 Notably, the Advocate-General considered that, if the standard of abuse articulated in TV 10 in relation to services were to be transposed to the freedom of capital, the requirement of authorisation would appear to be a justifiable restriction.101 She noted however that the concept of abuse is narrower in relation to the right of establishment than in relation to the freedom to provide services. Which concept should apply to the free movement of capital? Although the Advocate-General posed that question, she did not provide a clear-cut answer. She stated that, in contrast to the case law concerning the freedom to provide services, where the concept of abuse is invoked to prevent a person from undermining the requirements of the State of establishment under the guise of free movement of services,102 in the context of the free movement of capital there was no     98  For other cases in this area, see Case C-484/93 Svensson and Gustavsson v Minstre du Logement et de l’Urbanisme [1995] ECR I-3955; Case 205/84 Commission v Germany [1986] ECR 3755; Case C-222/95 Parodi v Banque H Albert de Bary [1997] ECR I-3899. Note also Case C-384/93 Alpine Investments BV v Minister van Financiën [1995] ECR I-1141 where the ECJ accepted as proportionate the Dutch prohibition on cold calling as a means of safeguarding the integrity and reputation of Dutch financial markets taking into account, inter alia, previous abuses committed by financial intermediaries. See para [53] of the judgment and the Opinion of AG Jacobs at 1164.     99  Case C-452/04 Fidium Finanz AG v Bundesanstalt für Finanzdienstleistungsaufsicht [2006] ECR I-9521. 100  See AG Stix-Hackl in Fidium Finanz (n 99 above), para [101]. 101  AG Stix-Hackl in Fidium Finanz (n 99 above), para [86]. 102  See the discussion above and Commission v Germany (n 98 above), para [22]; Van Binsbergen (n 11 above), para [13].

Abuse of Rights in EU Law with a Focus on Financial Law  185 danger that another freedom would be circumvented.103 This appeared to be an argument against the extensive application of the doctrine of abuse in the free movement of capital, although the Advocate-General did not pursue this analysis further. The dictum, however, that the freedom of capital cannot be used to undermine another freedom is not entirely correct. If one accepted, as the Advocate-General appeared to do, that the situation in Fidium Finanz came within the scope of Article 56 EC [now Article 63 TFEU] rather than Article 49 EC [now Article 56 TFEU] and that the standard of abuse under the first was narrower than the standard of abuse under the second, it would be possible to use capital movement as a way of achieving results unobtainable under the freedom to provide services. The approach taken by the Court in its judgment appears correct. The case was primarily concerned with the free movement of services and not the free movement of capital.104 It therefore followed that Fidium Finanz could not rely on the Treaty. In any event, it is submitted that, even if the German authorisation requirement were to be examined under the freedom of capital, the same standards of abuse should have applied as those applicable under the freedom of services. In the context of the case, there would be no justification for applying differential standards. It would thus be necessary to establish, first, whether there was an intention to obtain an advantage from Community law by creating artificially the conditions required for obtaining it, and, secondly, whether there was an objective element, namely, the existence of a situation where despite formal observance of the conditions laid down by the Community rules, their purpose was not achieved.105 In this context the findings in TV 10 would be important. As the Court pointed out in that case, a Member State may not be denied the power to prevent a person from relying on Article 49 EC [now Article 56 TFEU] if his activity is entirely or principally directed towards its territory for the purpose of avoiding the professional rules which would be applicable if he were established within that State.106 Thus, to make a determination of abuse, one would have to take account, among others, of the following considerations: the purpose of the requirement of authorisation, the reasons why Fidium Finanz decided to locate in a particular State, the proportion of business which it carried out in another State, whether it was subject to an equivalent authorisation requirement in its State of establishment, the types of credit services offered, and whether they were targeted to specialist professionals or the general public. In this respect, it is clear that the inquiry of abuse merges with the general inquiry of justification. In carrying out that inquiry, reliance is to be placed primarily on objective criteria. This is not to say that subjective criteria are not relevant but to suggest that the tenor of the case law is to focus first on the objectives of the rules and only secondarily on the intentions of the parties.  See AG Stix-Hackl in Fidium Finanz (n 99 above), para [90].   The freedom to provide services was triggered since the granting of credit on a commercial basis is a provision of services. The free movement of capital was also applicable since the provision of inter-State loans figures in the definition of capital movements under Directive 88/361. The ECJ was correct to suggest that, in the circumstances of the case, the predominant consideration was the freedom to provide services. The German rules prohibited companies established in non-Member countries from offering credit services to German residents and not, as such, the transfer of funds to and from Germany to third countries. Notably, the ECJ rejected the view that Arts 50 and 51(2) of the EC Treaty [Arts 57 and 58(2) TFEU respectively] establish priority of the freedom of capital vis-à-vis the freedom to provide services. See paras [32–33] of the judgment. The reasoning of the Court on this point, although it has been criticised, appears persuasive. 105  See AG Stix-Hackl in Halifax (n 22 above), paras [93–95]. 106  AG Stix-Hackl in Halifax (n 22 above), paras [93–95]. 103 104

186 Takis Tridimas In the specific circumstances of Fidium Finanz, it is notable that the applicant company was not subject to supervision by the Swiss authorities and, according to the referring court, there was substantial evidence to show that it had chosen to locate its registered office there with a view to avoiding supervision both in the State where it was established and the State where it carried out its commercial activities.107 B.  Single Passport in Investment Services Centros108 raises the fundamental issue of the division of powers between the home and the host State in circumstances where there are varying standards in national corporate laws. In the field of financial services, the Market in Financial Instruments Directive (MiFID),109 on which the single financial market edifice is based, carefully crafts a system of checks and balances in the division of competences between the home and the host State. The detailed provisions of the Directive make less necessary reliance on the general doctrine of abuse, although it continues to have a residual role both where statutory arrangements do not provide a solution and as an aid to their interpretation. Suffice it to outline here the main provisions of the Directive governing the principle of home State control. MiFID establishes the single passport principle in the field of investment firms.110 Under the scheme provided by the Directive, an investment firm must receive authorisation by its home Member State, ie the State where its registered office is located.111 The conditions for granting authorisation are specified in the Directive. Under Article 31, once authorisation has been granted by the home State, a firm is free to carry out the investment activities covered by its authorisation throughout the Union either by way of provision of services or through the establishment of a branch. The host State may not require the firm to be authorised by its own authorities nor may it impose any additional requirements in respect of the matters covered by the Directive.112 A firm who wishes to provide services in another Member State must communicate to the authorities of its home State the Member States in which it intends to operate and a programme of operations indicating the services it intends to offer. It may then commence its activities in the host State once the home State authorities notify that information to the competent authorities of the host State.113 A similar, albeit more detailed and elaborate, notification procedure is provided for the establishment of a branch.114  AG Stix-Hackl in Fidium Finanz (n 99 above), para [83].   Centros (n 17 above). 109  Directive 2004/39 on markets in financial instruments (MiFID) [2004] OJ L145/1. 110   The single passport principle is provided also, among others, in relation to banks by Directive 2006/48 relating to the taking up and pursuit of the business of credit institutions [2006] OJ L177/1, and in relation to life assurance institutions by Directive 2002/83 concerning life assurance [2002] OJ L345/1, as amended (see consolidated version of the Directive [2005] OJ L83/1). 111  MiFID (n 109 above), Art 5. 112  MiFID (n 109 above), Art 31(1), second sub-paragraph. Free movement may be exercised by the establishment of a branch or the provision of services. By contrast, if a firm wishes to establish a subsidiary in another Member State, the subsidiary is in its own right a corporate entity under the law of the host State and must receive authorisation by its competent authorities before being able to carry on any investment activities. The competent authorities of the State where the subsidiary is located must consult the competent authorities of the State of the parent firm before granting authorisation to the subsidiary: see Art 60 of MiFiD. 113  MiFID (n 109 above), Art 31(2) and (3). 114  MiFID (n 109 above), Art 32. 107 108

Abuse of Rights in EU Law with a Focus on Financial Law  187 The Directive does not provide any role for the competent authorities of the host State in the authorisation process.115 It does however provide a role in relation to the supervision of branches. Also, in certain specified cases, the authorities of the host State may take precautionary measures against firms carrying out activities in their territory.116 Although the detailed examination of those provisions is beyond the scope of this chapter, the following points may be made. Where a firm carries on activities in a host State via the free movement of services, it must comply with the conduct of business rules of the home State and supervision of its activities is entrusted to the authorities of that State. The host State may only take precautionary measures to the extent justified by the exception of Article 62.117 Where a firm establishes a branch in a host State, the authorities of that State have a number of powers. The most important are specified in Article 32(7). Under that Article, the host State authorities have responsibility for ensuring that the services provided by the branch within the territory of the host State comply with the obligations laid down in certain provisions of the Directive. These are Article 19 (conduct of business rules), Article 21 (the best execution principle), Article 22 (rules pertaining to the handling of client orders), Article 25 (obligation to uphold market integrity, report transactions, and maintain records), Article 27 (obligation to make public firm quotes), and Article 28 (obligations pertaining to post-trade disclosure). The authorities of the host State also have the power to examine branch arrangements and request such changes as are strictly necessary to enable them to enforce the above obligations.118 Although Article 32(7) does not make it clear, it appears that the exception of this article grants to the host State not only power of supervision but also power of regulation. In other words, in relation to the activities carried out by a branch, the investment firm must comply with the obligations stated above as they are implemented by the law of the State where the branch is located and not the law of the home Member State. This interpretation derives from Article 61(2) which refers to ‘the standards set by the host Member State’ in the areas covered by Article 32(7) and also by Article 62 which assumes that a branch must comply with the rules of the host State. It is also supported by the

115  See EC Commission, ‘Internal Market and Services DG Memorandum, Supervision of Branches under MiFID, MARKT/G/3/MV D(2007)’ (18 June 2007) Brussels, 2: ec.europa.eu/internal_market/securities/docs/ isd/mifid-branches_en.pdf. 116  See Art 62 (n 117 below). 117  Art 62 provides for enforcement action in the event where the authorities of the host State have ‘clear and demonstrable’ grounds for believing that an investment firm acts in breach of its obligations in their national territory. The article draws a distinction between freedom to provide services and establishment of a branch. If a firm which exercises the freedom to provide services in the host State acts in breach of its obligations, the host State authorities must refer the matter to the authorities of the home State. The host State authorities may take measures only if, despite the measures taken by the home State supervisor or because such measures prove inadequate, the firm persists in acting in a manner that is clearly prejudicial to the interests of host Member State investors or the orderly functioning of markets. By contrast, where the host State authorities establish a breach by a firm which has a branch in the host State, a distinction is drawn. If the breach relates to obligations which do not fall within the responsibility of the host State, the same arrangements as those described above in relation to services apply. But if the breach relates to obligations which, under the exceptions provided by the Directive, fall within the responsibility of the host State, the authorities of the latter may take direct enforcement action. 118  See Art 32(7), second sub-paragraph. Note that the host State authorities also have the powers provided by Art 61(1) (power to collect information for statistical purposes), Art 61(2) (power to require branches to provide the information necessary for the monitoring of their compliance of the standards provided for in Art 32(7)), and Art 62 (precautionary intervention powers).

188 Takis Tridimas Commission,119 and appears preferable since it would be less efficient to require the host State to monitor compliance with the law of the home State. Article 32(7) only grants supervisory power to the host State in relation to services provided by a branch within its territory. This begs the question of the allocation of supervisory functions in relation to services carried out by a branch in the territory of another Member State. In this context, the Commission has proposed a solution which is based on close cooperation between the authorities of the home and the host State.120 Finally, it should be noted that the supervisory power of the State where the branch is located does not replace that of the home State. The authorities of the latter may carry out on-the-spot investigations after informing the competent authorities of the host State.121 It is not clear however whether these investigations may relate also to observance by the firm of the rules of the host State applicable to it under Article 32(7). It appears that the answer is in the affirmative as the authorities of the home State retain residual responsibility to supervise all the activities of the firm covered by its authorisation wherever such activities are carried on. In effect, as the Commission points out, the dual supervision of branches is a reality and effective and efficient supervision may only be attained through cooperation between the national competent authorities.122 Notably, Article 5(4) of the Directive requires an investment firm to have its head office in the same Member State as its registered office. A similar provision is included in other directives which implement the single passport principle in the field of financial services law, including the banking and insurance directives.123 These provisions seem to require a substantive connection between, on the one hand, the place where the company is administered and, secondly, the territory where its registered office is based. Although the term ‘head office’ is not defined in any of the directives in common usage it denotes the location where most of the important functions of an organisation are concentrated, namely the place where the senior management and the executive officers of a company are located and the corporate policy-making functions are carried out. The purpose of Article 5(4) is to ensure that the competent authorities of the home State, ie the State where the registered office is located, can exercise effective regulation and supervision on an investment firm. The question which arises in this context is the following. Would it be permissible for an investment firm authorised in one State to establish a branch in another and exercise all its activities through that branch? It might be argued that this would be in breach of Article 5(4) although it might appear to be allowed under Centros. The answer to this question should be examined in the light of the following considerations. First, the fact that a firm exercises most of its activities through a branch located in another Member State does not necessarily mean that its head office has moved there. The location of  See EC Commission Memorandum (n 115 above) 2, para 6.  See Commission Memorandum (n 115 above). The Memorandum essentially provides that, if a branch carries on activities in the home State, then these will be supervised by the home State authorities. If it carries on activities in a third State, the authorities of the home and the host State are under a legal obligation to cooperate. It further suggests the drafting, under the auspices of the Committee of European Securities Regulators (CESR), of a Multilateral MoU for the supervision of branches under MiFID. See further CESR, The Passport under MiFID: www.cesr-eu.org/popup2.php?id=4603. 121  Art 32(8). 122  Commission Memorandum (n 115 above) 3–4. 123  See Directive 2006/48 on credit institutions (n 110 above), Art 11(2); Directive 2002/83 on life assurance (n 110 above), Art 6(3). 119 120

Abuse of Rights in EU Law with a Focus on Financial Law  189 the head office is an issue of fact to be determined in the circumstances of each case. Secondly, it is clear that MiFID does not oppose a firm’s expanding into other Member States nor generating most of its business from activities in one or more host States. The very purpose of the single passport regime is to enable firms to take advantage of business opportunities across the Internal Market. On the other hand, under the doctrine of abuse, it would not be permissible for a firm to avoid, through artificial arrangements, the supervisory powers of the ‘natural’ home State. The judgment in Centros facilitates free movement but, as it stands, the scope of the ruling is narrow. It requires the host State to recognise the existence of a branch in its territory and enter it in the register even where the company carries on no business in the State of incorporation. However, as already stated, Centros relates to the recognition of the existence of a branch, not to the exercise of any regulated activity. It is not a judicial carte blanche for unlimited forum shopping and therefore does not necessarily enable an investment firm to choose which Member State will serve as its home State without having a substantial connection with its territory. It is clear from the principles laid down in TV 10 that it is not possible for a firm which has its registered office, say, in the UK to move its central administration and control in France for the sole purpose of offering investment services in the UK whilst avoiding being subject to the regulatory and supervisory powers of the Financial Services Authority. Subject to the above, each case must be examined on its merits, taking account of all surrounding circumstances. C.  Concretisation of the Principle of Abuse: Market Manipulation Market manipulation provides a prime example of the concretisation of the notion of abuse through the articulation of specific statutory standards. It also shares some of the essential features of the general principle of abuse in that some forms of it consist in transactions which ostensibly come within the scope of legally permissible conduct but which, in fact, pursue improper purposes. In general, manipulation refers to practices which are intended to mislead investors by artificially affecting market activity, such as matched orders or wash sales.124 The key elements of the concept, as developed in common law and then in statute, are deception and artificiality.125 The distinction between a legitimate and an artificial transaction is difficult to draw and establishing whether there is a genuine market or a mirage may sometimes appear an impossible task. In EU law, the prohibition of market manipulation is concretised at three successive levels. A general definition is provided in the Market Abuse Directive. This is in turn exemplified by Level 2 measures which make the definition more specific and provide Member States with legislative guidelines. The definitions are further elaborated and concretised by non-legislative guidelines adopted by the Committee of European Securities Regulators (CESR). 124  See the definition given by the US Supreme Court under Rule 10(b)-5 in Schreiber v Burlington Northern, 472 US 1 (1985) 7; Santa Fe Industries, Inc v Green, 430 US 462 (1977). A matched order occurs where persons place contemporaneous orders for the sale and purchase of the same security at substantially the same price and quantity. The intention is to convey the appearance of an active market, thus inducing investors to trade. A wash sale involves trading in securities by persons acting in concert so that there is no change in the beneficial ownership of securities. 125  See Scott v Brown, Doering, McNab & Co [1892] 2 QB 724 (CA). This early case established that as long as deception was present, manipulation could occur without the dissemination of false statements.

190 Takis Tridimas Thus, Article 1(2)(a) of the Directive defines as manipulation, among others, transactions which give, or are likely to give, false or misleading signals as to the supply of, demand for, or price of financial instruments unless the person who entered into the transaction establishes that his reasons for so doing are legitimate and that these transactions ‘conform to accepted market practices on the regulated market concerned’. Commission Directive 2003/124, a Level 2 measure, provides for a detailed list of nonexhaustive signals, which although they should not necessarily be deemed in themselves to constitute market manipulation, must be taken into account by Member States when transactions or orders to trade are examined in order to decide whether they fall within the prohibition.126 Further, Commission Directive 2004/72, another Level 2 measure, provides a number of non-exhaustive factors which national authorities must take into account when deciding whether to accept a particular practice as an accepted market practice.127 Further detailed guidance is provided in the form of non-legislative guidelines issued at Level 3 by CESR.128 As stated above, securities regulation provides for an unusually detailed statutory framework at EU level. This has been the result of a policy choice on the part of the Union’s institutions and, through the Council, also of the national governments. Its origins lie in the Lamfalussy Report, which identified the drawbacks of the previous regime and sought to establish a close relationship between, on the one hand, more regulation at EU level and, on the other hand, the establishment of a genuinely pan-European securities market and the fulfilment of the objectives of the Lisbon agenda.129 One, however, may question the wisdom of the pursuit of complete harmonisation. Whilst it may be desirable in some areas, it appears to be less so in others, for example, the field of penalties and sanctions. VI. Conclusion

In European law, the principle of abuse serves as a residual anti-evasion doctrine to ensure fulfilment of the purposes of the law. Its recognition as a general principle in Kofoed is a sign of maturity in the development of the Union’s legal system. Its effect, however, has so far remained limited. One may distinguish two main categories of abuse. Cases where a person seeks to circumvent the requirements of national law by relying on a Union freedom, engaging, in effect, in forum shopping, and cases where a person seeks to derive an improper advantage from EU legislation. 126  Commission Directive 2003/124/EC of 22 December 2003 implementing Directive 2003/6/EC of the European Parliament and of the Council as regards the definition and public disclosure of inside information and the definition of market manipulation [2003] OJ L339/70, Art 4. See also Art 5 which concretises the form of manipulation specified in Art 1(2)(b) of the Market Abuse Directive. 127  See Commission Directive 2004/72/EC of 29 April 2004 implementing Directive 2003/6/EC of the European Parliament and of the Council as regards accepted market practices, the definition of inside information in relation to derivatives on commodities, the drawing up of lists of insiders, the notification of managers’ transactions and the notification of suspicious transactions [2004] OJ L162/70, Art 2. These include, for example, the level of transparency of the practice and the market in question, the structural characteristics of the market, the types of financial instruments traded, etc. 128  See CESR, Market Abuse Directive, Level 3—First set of CESR guidance and information on the common operation of the Directive, CESR/04-505b: www.cesr.eu/data/document/04_505b.pdf. 129  See Lamfalussy Report (n 86 above). The Report criticised the pre-existing Community directives, adopted largely in the 1980s, for giving rise to an inefficient regulatory system. The main criticisms were that they failed to provide clear Europe-wide regulation on a large number of issues, allowed too many choices to Member States, and were characterised by inconsistent implementation.

Abuse of Rights in EU Law with a Focus on Financial Law  191 It seems that the general conditions that must be satisfied are the same in both cases. There must be, first, a subjective element; namely, an intention to obtain an advantage from EU law by creating artificially the conditions required for obtaining it, and, secondly, an objective element; namely, the existence of a situation where, despite formal observance of the conditions laid down by the Union rules, their purpose has not been achieved. The second element is of particular importance. The finding of abuse involves an inquiry as to the objectives of the Treaty provision or legislation in issue so as to determine the scope of the right which it establishes and the range of interests that it is intended to protect. In the context of free movement, the Court of Justice has understood the principle narrowly in keeping with the expansive interpretation of the Treaty freedoms. The impact of the doctrine has remained limited. This is understandable: issues of abuse involve the determination of the scope of a right and should therefore be mostly internalised, ie, carried out as part of the general purpose-oriented process of interpretation. A difference between the objective and the subjective element of abuse is that whilst the first pertains to the interpretation of EU law and is therefore for the Court to ascertain, the second pertains to the subjective intentions of the parties and is therefore for the national court to ascertain. In financial law, the principle of abuse has been largely concretised through the articulation of detailed rules in various European directives. Since the adoption of Directive 2003/6, the term ‘market abuse’ has become a term of art referring specifically to insider trading and market manipulation. The definition of manipulation comes closer to the general principle of abuse since most of its forms involve artificial transactions which formally remain within the bounds of the law. As for the free movement of investment firms, the detailed provisions of MiFID make less necessary reliance on the doctrine of abuse as a general principle of EU law, although it continues to have a residual role both where statutory arrangements do not provide a solution and as an aid to their interpretation. One may assume that the greater the specificity of statutory provisions and guidelines articulating a prohibition, the brighter the lines which divide permissible from impermissible conduct and therefore the less the need to rely on the general principle of abuse. This assumption is not always correct. Sometimes, detailed regulation may lead to lack of transparency, contradictions, and uncertainty, making recourse to general principles essential. In such a context, the role of a general principle is not to fill gaps but make sense out of polynomy. Abuse of right is a principle of law which can only be invoked to create an exception. The case law seems to have struck the balance right. In the context of free movement, the fighting ground for drawing the elusive distinction between permissible and impermissible obstacles to trade remains the principle of proportionality.

13 Comments on Abuse of Rights in EU Law PIERRE SCHAMMO I. INTRODUCTION

A

s legal and market integration continue to progress at the hands of the Court of Justice and the European legislature, the concept of abuse of rights (or abuse of law) is assuming increasing significance in the EU legal order. As a concept or notion, it has long been recognised by the Court. The EU legislature too has sought to address abuse of rights by alternative legislative means. Professor Tridimas’ chapter1 not only provides a thorough analysis of the development of the concept of abuse by the Court, but also examines the broader questions that it raises in European legislation. It is especially with respect to EU financial law—ie in particular, the Directive on Markets in Financial Instruments, or MiFID2—that the chapter provides much food for thought and I will therefore concentrate on that aspect of his chapter. First, however, it is appropriate to start with a few brief observations about the concept of abuse. II.  THE CONCEPT OF ABUSE

Professor Tridimas’ chapter begins by looking back at the origins of the concept of abuse of rights and turns especially to French law where abuse of rights or abus de droit commonly refers to the exercise by a person of his or her rights in an unreasonable manner which causes harm to other persons.3 The chapter goes on to observe that in front of the Court of Justice claims of alleged abuse are typically raised in relation to two types of behaviour: that is, (i) attempts by a person to circumvent national laws by taking advantage of Treaty freedoms or (ii) attempts by a person to take ‘improper advantage’ of EU rights.4 Given the Court’s conceptual and terminological indeterminacy,5 a more useful starting point for examining its case law from a comparative law perspective is then often the French fraude à la loi doctrine.6 A fraude à la loi claim is typically raised 1   T Tridimas, ‘Abuse of Rights in EU Law: Some Reflections with Particular Reference to Financial Law’, ch 12 above. 2   Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments (MiFID) amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC [2004] OJ L145/1. 3   Tridimas (n 1 above), Part I. 4   Tridimas (n 1 above), Part I. 5   See also R de la Feria, ‘Introducing the Principle of Prohibition of Abuse of Law’ in the Introduction to this book above. 6   See P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351.

194  Pierre Schammo in a situation where a person performs a legal act or places himself or herself in a legal situation in order to escape mandatory provisions which would be applicable otherwise.7 Such a claim may be raised against a person who engages in regulatory arbitrage when attempting to exploit regulatory differences between jurisdictions. In the European context, it is well known that the free movement provisions tend to facilitate regulatory arbitrage.8 Economic actors who wish to profit from differences between the laws of Member States may seek to rely on primary rights, such as the right of establishment.9 Cases of alleged abuse in front of the Court of Justice hence often involve some form of regulatory arbitrage. For the Court, this type of behaviour is not, per se, illegitimate. It is perfectly valid to take advantage of differences between national laws by relying on the free movement provisions as long as the objectives of these provisions are being satisfied. But regulatory arbitrage can turn into illegitimate abuse for the Court if these objectives are not satisfied.10 III.  THE DIRECTIVE ON MARKETS IN FINANCIAL INSTRUMENTS

Some of the most thought-provoking points of Professor Tridimas’ chapter concern MiFID, and Article 5(4) in particular. The latter provides that an investment firm, which takes the form of a legal person, must have its head office (HO) in the same Member State as its registered office (RO). For investment firms which are not legal persons or which are legal persons but do not, under national law, have a RO, Article 5(4) states that the firm’s HO must be located in the Member State in which it conducts its affairs.11 The sanction for non-compliance with Article 5(4) is severe. The competent authority will refuse or withdraw the firm’s authorisation.12 The latter is a condition for the performance of investment services/activities at home and abroad.13 There is no definition of HO in the Directive, but it is obvious that the reason for requiring the HO and the RO to be in the same Member State is to make sure that the RO has some ‘substance’, the place of the HO being understood here as the place where a firm’s ‘mind and management’ is situated. Admittedly, the question of the location of the HO is an issue of fact.14 For our purposes, factual problems can be left aside. More interesting to consider, is the raison d’être of Article 5(4). Besides the immediate concern to improve the capacity of the home State authority to properly regulate and supervise an investment firm,15 Article 5(4) acts as a legislative means for containing undesirable regulatory arbitrage16 (or abuse) by firms seeking to circumvent potentially     7   For definitions, see for example, G Cornu, Vocabulaire Juridique (Paris, Presses Universitaires de France 2000); B Audit, ‘Fasc 50: Fraude à la Loi’ JurisClasseur Civil Code Art 3: global.lexisnexis.com/fr.     8   See generally, C Barnard and S Deakin, ‘Market Access and Regulatory Competition’ in C Barnard and J Scott (eds), The Law of the Single European Market: Unpacking the Premises (Oxford, Hart Publishing 2002) 197.     9   See generally, WG Ringe, ‘Sparking Regulatory Competition in European Company Law: The Impact of the Centros Line of Case Law and its Concept of “Abuse of Law”’, ch 8 above. 10   I have discussed the so-called intentional element of abuse elsewhere. See Schammo (n 6 above). 11   Hereafter, I will concentrate on the first scenario (ie the scenario of a legal person). 12   See MiFID Arts 7(1) and 8. See also Recital 22. 13   MiFID Art 5(1). 14   Tridimas (n 1 above), Part V. 15   Tridimas (n 1 above), Part V. 16   I use the term ‘regulation’ in its broadest sense here.

Comments on Abuse of Rights in EU Law  195 stricter standards in host Member States.17 In other words, by requiring the HO of a firm to be in the same Member State as the RO, one prevents an investment firm from setting up a letter-box company in one Member State—ie the home Member State—in order to take advantage of distinct (laxer?) standards of regulation, supervision or enforcement and subsequently (ab)using MiFID’s mutual recognition provisions in order to operate in other Member States—ie the host Member States—where the firm would, in fact, concentrate its ‘substance’.18 Article 5(4) can then be described as an ‘in-built’ defence against regulatory arbitrage. Indeed, if we are interested in abuse of rights, Article 5(4) appears to offer much food for thought. Professor Tridimas’ chapter does not examine Article 5(4) in a comprehensive manner and so it seems worth complementing his analysis by examining two points. The first concerns the distribution of competences between home and host Member States in relation to Article 5(4) and especially the question of whether a host Member State, which finds that a firm has transferred its HO into its territory (whilst maintaining its RO in the home Member State), is under the provisions of the Directive entitled to disregard the firm’s authorisation on the grounds that the firm has failed to comply with Article 5(4). While Tridimas’ chapter looks at the question of the distribution of competence, it does not examine the point in the context of Article 5(4). Yet, the point appears a natural corollary to the questions which it raises. Second, it is worth reflecting on the very need for a statutory provision such as Article 5(4). A.  Article 5(4) and the Authority of the Host State Among the questions which are likely to prove important for the functioning of the MiFID regime are questions that concern the competence of Member State authorities to take action under the provisions of the Directive if a firm has disregarded the Article 5(4) ‘legislative defence’ against regulatory arbitrage. It is especially the host Member State that is likely to have the most serious concerns. So the question arises whether a host Member State is authorised to sanction a firm by disregarding the firm’s authorisation if it finds that the firm no longer satisfies the requirement of Article 5(4). The question is examined next. i.  Home State Authorisation Prevails MiFID is based on the principle of home country control.19 As far as authorisation requirements are concerned, it is the firm’s home State competent authority which is competent to grant it authorisation to perform investment services or activities.20 Pursuant to Article 6(3), this authorisation is valid ‘for the entire Community’.21 Home State authorisation therefore allows a firm to perform investment services/activities ‘throughout   See also N Moloney, EC Securities Regulation, 2nd edn (Oxford, Oxford University Press 2008) 412.   Moloney (n 17 above) 412. 19   See also Moloney’s excellent analysis of MiFID in EC Securities Regulation (n 17 above). 20   MiFID Art 5(1). Moreover, Art 7(1) provides that the competent authority shall only grant authorisation if it is satisfied that the requirements of MiFID are met. Under Art 8, the competent authority is given the power to withdraw authorisation when the investment firm no longer satisfies the conditions under which it was authorised. 21   See also MiFID Recital 23. 17 18

196  Pierre Schammo the Community’,22 by setting up a branch or by way of the free provision of services pursuant to MiFID’s mutual recognition arrangements.23 These principles are essential for appreciating the margin of discretion which is left to host Member States. Under MiFID, the powers of host State authorities are limited.24 Certainly, as far as the authorisation process is concerned, there is no place for host State competence under the Directive.25 ii.  Article 62(1) As Professor Tridimas’ chapter mentions, MiFID lays down certain rules which allow host Member States to take precautionary measures in certain circumstances.26 For the present purposes, the relevant provision is Article 62(1) which envisages the situation in which a host Member State seeks to take action in relation to matters that fall under the competence of the home Member State. The relevant question is whether Article 62(1) may provide a legal basis for host State action in the case where a firm fails to comply with the requirements of Article 5(4). The first point to note is that Article 62(1) sets strict conditions on the use of precautionary measures by host Member States. MiFID thus requires the host State authority, in whose territory a firm is acting on the basis of the single passport, to have ‘clear and demonstrable grounds’ for thinking that the firm is infringing MiFID requirements. If the host State authority has such grounds, MiFID requires it to refer its findings to the home State authority first. Article 62(1) only empowers the host State authority to take measures ‘in order to protect investors and the proper functioning of the markets’ if, in spite of the measures adopted by the home State authority or because the home State measures are insufficient, the investment firm continues to act ‘in a manner that is clearly prejudicial to the interests of host Member State investors or the orderly functioning of the markets’. The home State authority will have to be informed first. The European Commission will also have to be informed. As for the measures which the host State authority is entitled to take, MiFID does not specify them in any exhaustive manner. As mentioned, Article 62(1) provides that the measures must be needed ‘in order to protect investors and the proper functioning of the markets’. It adds that among them are measures which prevent the firm ‘from initiating any further transactions’. That said, it is open to debate whether non-compliance with Article 5(4), which admittedly lays down requirements that a firm must meet in order to be authorised, would be considered severe enough to warrant reliance on the precautionary measures provision. More specifically, it is not readily apparent that a firm’s breach of Article 5(4) would be ‘clearly prejudicial’ to investors’ interests or the proper functioning of financial markets. Recall that a firm would fail to comply with Article 5(4) if it concentrated its ‘mind and management’ (or HO) in a host Member State whilst maintaining its RO in the home Member State. Certainly, seen from the perspective of host State investors (or for that matter, the proper functioning of markets), this type of action does not appear to raise   MiFID Art 6(3).   MiFID Arts 6(3), 31 and 32. 24   For exceptions to home State control, see, for instance, Arts 32(7), 61 or 62. See also Recital 32. 25   Cf European Commission, ‘Supervision of Branches Under MiFID’ (Brussels 18 June 2007) 2: ec.europa. eu/internal_market/securities/docs/isd/mifid-branches_en.pdf. The only situation where MiFID provides that authorities from other Member States shall be involved in the authorisation process concerns certain subsidiaries and certain controlled firms under MiFID Art 60, European Commission (this note above) 2, fn 1. 26   Tridimas (n 1 above), Part V. 22 23

Comments on Abuse of Rights in EU Law  197 the same level of concern as non-compliance with obligations which directly concern a firm’s dealings with its clients (eg best execution).27 This is especially so as long as a firm makes plain that it operates under the passport system in the host Member State and therefore is mainly regulated and supervised in another Member State. It should also be remembered at the outset that the level of harmonisation under MiFID is significant. Overall therefore, it is submitted that non-compliance with Article 5(4) would not be severe enough to justify unilateral host State action under Article 62(1). Moreover, given that the precautionary measures derogation has the potential to interfere significantly with the proper operation of home country control and ultimately with MiFID’s single passport, a strict interpretation of its conditions is justified.28 A host State authority that would ignore the distribution of competences would fail itself to comply with MiFID’s provisions. Moreover, its unilateral actions would constitute a failure to cooperate and a serious breach of trust between competent authorities which are ultimately necessary for the single passport to work effectively. B.  The Need for Article 5(4) MiFID The final point brings us back to the apparent assumptions that underlie Article 5(4) of MiFID. As a reminder, one of the underlying reasons for enacting Article 5(4) is the prevention of some form of regulatory arbitrage. Thus, the prevailing fear seems to be that, in the absence of Article 5(4), a firm might attempt to set up letter-box companies in a Member State—its home Member State—for the sole purpose of profiteering from (presumably) laxer standards whilst concentrating its ‘substance’ in a host Member State where it could operate under home country authorisation. If that is the raison d’être of Article 5(4), it raises a more fundamental question in view of the more comprehensive and prescriptive nature of the new MiFID regime: is there a need for Article 5(4) and the restriction which it imposes? MiFID replaced the Investment Services Directive (ISD) which was a minimum harmonisation Directive and famously incomplete.29 The ISD included a provision which was identical to Article 5(4) of MiFID. Yet, in comparison to the ISD, MiFID strikes one as a Directive of a different time and age. MiFID was adopted under the Lamfalussy process, a four-level approach to securities regulation. At Level 2 of the Lamfalussy process, MiFID’s provisions were fleshed out by implementing legislation. At Level 3, the Committee of European Securities Regulators (CESR) was put in charge of promoting cooperation and making sure that MiFID’s rules are consistently applied by national authorities. At Level 4, the European Commission committed itself to watch with new vigour over the proper implementation of the Directive. In comparison to the ISD, the level of harmonisation under the MiFID regime is thus far superior, not least because of MiFID itself (at Level 1).30 Having said this, the Lamfalussy approach, especially the 27   For branches, the responsibility to supervise best execution is, of course, entrusted to the Member State of the branch. 28   The host State authority could, however, bring the matter to the attention of the European Commission or the Committee of European Securities Regulators. 29   The Directive is examined in N Moloney, EC Securities Regulation, 1st edn (Oxford, Oxford University Press 2002). 30  Moloney (n 17 above) 412 and at 366 noting that ‘[a]lthough MiFID, at level 1, is like most of the FSAP, not formally expressed as a “maximum harmonisation” measure it is widely regarded as such’.

198  Pierre Schammo capacity of CESR at Level 3 to address differences in the interpretation or application of European rules between national authorities, has its limits.31 CESR’s members have thus on occasions failed to achieve the desired level of cooperation.32 Moreover, they have in the past simply defected at national level from agreed, albeit non-binding, CESR recommendations and standards.33 Admittedly, these are general points. But the problem with CESR is, in my view, that fundamentally it is a collective actor, a ‘club-like’ body.34 As such, it assesses problems through the lens of its members, which are distinct and autonomous national authorities which have their own interests and distinct normative goals. Therefore, CESR is not unambiguously driven by a shared set of interests and values or indeed by a common European interest. The point is important because regulatory arbitrage feeds on all types of differences, whether in terms of regulation, supervision or enforcement, as long as it offers a cost or other type of advantage. There is still space for regulatory arbitrage—and I use the term in its broadest sense—in European securities markets.35 The next months promise many new developments with respect to the institutional framework governing financial markets. The shock of the financial crisis and its effects on the real economy have created political momentum. The Lamfalussy Level 3 requirements are no longer seen as sufficient and important new proposals are being considered at European level following the report of the so-called de Larosière group.36 It remains to be seen whether the new reform proposals, which follow the de Larosière Report, and which, inter alia, suggest transforming CESR into a new European authority will be sufficient to address the problems which were identified above. For now, however, Article 5(4) may well continue having a raison d’être, although it remains arguably a rather crude legislative means for containing undesirable regulatory arbitrage. IV. CONCLUSION

Professor Tridimas provides an all-round view of abuse of law in the EU legal order. Rather than examining the Court’s case law on abuse of law,37 I looked mainly at financial law, especially MiFID and Article 5(4). As a final thought on Tridimas’ chapter, one 31  For example, European Commission, ‘Accompanying Document to the Communication From the Commission “European Financial Supervision”’ SEC(2009) 715 (May 2009) 24–25, Report of the high-level group on financial supervision in the EU (the ‘de Larosière Report’) (Brussels 25 February 2009) 75: ec.europa. eu/internal_market/finances/docs/de_larosiere_report_en.pdf. 32   de Larosière Report (n 31 above) 75. 33   See the testimony of E Wymeersch (chairman of CESR) in front of the House of Lords European Union Committee, ‘The Future of EU Financial Regulation and Supervision—Volume II: Evidence’ (14th Report of Session 2008–09): www.publications.parliament.uk/pa/ld200809/ldselect/ldeucom/106/106ii.pdf noting at 144: What I think is the main weakness in the present Lamfalussy organisation is the fact that we are doing excellent conveyance work, eg making recommendations which are very much supported by the markets, but we are not sure that the recommendations are followed up. Very often we find out that some of the Member States object, although when adopting these recommendations they agree with the text, but in practice the situation may be different.

And at 145: ‘The difficulty is that some of the members may have afterthoughts and then not apply the rule or change it.’ 34   See on the characteristic features of collective actors and clubs, F Scharpf, Games Real Actors Play: ActorCentered Institutionalism in Policy Research (Boulder, Westview 1997) 54–56. 35   With respect to MiFID, see also Moloney (n 17 above) 365. 36   de Larosière Report (n 31 above). 37   I have done this elsewhere in relation to the right of establishment. See Schammo (n 6 above).

Comments on Abuse of Rights in EU Law  199 could have looked at the broader constitutional questions which Article 5(4) raises and especially at how the requirement to have the HO in the same Member State as the RO squares with the Court of Justice’s case law on the right of establishment. Professor Tridimas addresses the question briefly when referring to the Court’s Centros decision.38 European legislation, just like national legislation, ought to respect EU primary law and so it also makes sense to examine whether a provision of a Directive—Article 5(4) MiFID in our case—falls foul of the interpretation of a primary freedom.39 He seems to suggest that the requirement of Article 5(4) to have the HO in the same Member State as the RO might contradict the Court’s Centros ruling.40 But he later notes that the scope of the Centros ruling is narrower.41 Presumably, therefore, Centros would not be applicable. But there are other reasons that suggest that Centros is not applicable. Thus, in Centros, the national measure, which was found to hinder the right of establishment, was imposed by the host Member State. Given that compliance with Article 5(4) is under home State control, it would have been worth considering cases such as Daily Mail and the extent to which the Court’s reasoning in this case, which has recently been clarified in Cartesio, can at all be transferred into the context of what is essentially a restriction to the right of establishment which was enacted by the European legislature.42 That said, Professor Tridimas’ chapter does neither pretend to examine Article 5(4) in any exhaustive manner, nor set out to draw a final conclusion on the question of the compatibility of Article 5(4) with the primary freedom.

  Tridimas (n 1 above), Part V.   In the company law context, see WG Ringe, ‘The European Company Statute in the Context of Freedom of Establishment’ (2007) 7 Journal of Corporate Law Studies 185; J Rickford, ‘Current Developments in European Law on the Restructuring of Companies: An Introduction’ [2004] European Business Law Review 1225. 40   Tridimas (n 1 above), Part V. 41   Tridimas (n 1 above), Part V. 42   Case 81/87 R v HM Treasury and Commissioners of Inland Revenue ex p Daily Mail and General Trust plc [1988] ECR 5483; Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-09641. 38 39

14 The Emsland-Stärke Abuse of Law Test in the Law of Agriculture and Free Movement of Goods Panos Koutrakos* I. Introduction

T

he law of agriculture and free movement of goods provides an interesting context within which to examine the application and development of the principle of abuse of law:1 on the one hand, the latter featured in the relevant case law as early as the 1970s; on the other hand, it was within this area that the Court responded to various calls about formalising the criteria for the application of the abuse of law principle and articulated the Emsland-Stärke test.2 The analysis in this chapter is structured as follows. First, the application of the principle of abuse in the period preceding the Emsland-Stärke test will be outlined. Second, the Emsland-Stärke test, its strands, and implications will be analysed. Third, the subsequent application of this test and, in particular, the role of national courts will be assessed. II.  Setting out the scene

It was in the area of agriculture where the Court dealt with the notion of abusive practices quite early on. In Cremer, it dealt with Community rules on refunds for exports of compound animal feeding-stuffs.3 Such refunds were determined on the basis of whether the exported compounds contained any ingredients covered by specific EC rules on the cereal market. The question arose whether an exporter would be entitled to a refund even if the compound feeding-stuff in question contained a very small proportion of one of these products.

  *  Many thanks to Fabian Amtenbrink, Agnes Behm, Achilles Skordas, Gabriele Springer, Ulrich Springer, and Antenor Hallo de Wolf for helping me with the judgments of national courts analysed in this chapter. The usual disclaimer applies. 1  For the genesis and development of the principle, see R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of A New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. 2  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 3  Case 125/76 Peter Cremer v Bundesanstalt für landwirtschaftliche Marktordnung [1977] ECR 1593.

204  Panos Koutrakos It is a relief to this author that the Court should have acknowledged that the secondary rules in question ‘were difficult to understand from the point of view of their wording and context’.4 Nonetheless, the Court felt confident to rule that: [T]he scope of [the secondary measures in question] must in no case be extended to cover abusive practices of an exporter in taking advantage of the flat-rate assessment in calculating the refunds especially as at the time it was not a question of adopting a comprehensive set of rules but only of creating a framework within which the national authorities were to regulate the market for the products in question at their own discretion.5

The above conclusion was based on the Court’s assessment of the objectives of the refund system as set out in secondary legislation, namely to compensate for the effect on the prices of the compound feeding-stuffs of the rules applicable to the ingredients used. It was for that reason that the refunds should be proportionate to the amount of the basic products covered by the EC cereal rules in the composition of the exported product. In practical terms, that meant that the grant of a refund presupposed the actual presence in the compound feedingstuff, in significant proportions, of the basic products. The Court, then, pointed out that, as the relevant EC rules did not provide any guidance as to which compound feeding-stuffs should give rise to an entitlement to the grant of refunds, ‘it is for the competent national authorities to judge the facts with a view to preventing undue payment of refunds as a result of manipulation by the producers of the proportion of the ingredients of compound animal feeding-stuffs’.6 However, it did offer some guidance: a trader would not be entitled to a refund for the export of a compound feeding-stuff which contained only one product covered by the EC cereal rules and that in insignificant proportions. In General Milk Products,7 the Court dealt with export refunds for cheddar cheese imported from New Zealand into Germany and then exported to other Member States. Whilst, under an arrangement between the EC and New Zealand, imports of cheddar from New Zealand into Germany did not make the trader entitled to compensatory arrangements, re-exportation to other Member States did on the basis of specific EC rules aiming to tackle the wide currency fluctuations between certain Member States. Following an amendment of the EC–New Zealand arrangement, the question arose whether re-exportation should still give rise to such an entitlement. The Court answered this question in the affirmative. However, it pointed out that ‘the position would be different only if it could be shown that the importation and re-exportation of that cheese were not realised as bona fide commercial transactions but only in order wrongfully to benefit from the grant of monetary compensatory amounts’ and noted that ‘[t]he bona fide nature of those transactions is a question of fact to be decided by the national courts’.8 The above judgments have the hallmarks of the abuse of law test which would follow in Emsland-Stärke: the artificial nature of transactions, the objectives of secondary rules not realised, the discretion of national courts to ascertain whether abuse has actually occurred, even the discretion of national authorities to make substantive judgments as to the application of secondary EU law rules. These will be analysed below. Another context in which the question of abuse of law arose was in regards to transactions carried out in order to deviate from the application of domestic rules. In   Cremer (n 3 above), para [13].   Cremer (n 3 above), para [21]. 6   Cremer (n 3 above), para [14]. 7  Case C-8/92 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas [1993] ECR I-779. 8   General Milk Products (n 7 above), para [21]. 4 5

The Emsland-Stärke Abuse of Law Test  205 Leclerc, the Court dealt with a reference about the compatibility with EC law of a French law requiring all retailers to abide by the selling prices for books fixed by the publisher or the importer.9 The relevant rules provided that, in cases of books published in France but imported from another State, the retail price should be no lower than that fixed by the publisher. The Court held that such a provision, whilst making no distinction between domestic and imported books, ‘discourages the marketing of re-imported books by preventing the importer from passing on in the retail price an advantage resulting from a lower price obtained in the exporting Member State’.10 However, the conclusion that it would violate Article 28 EC [now Article 34 TFEU] would not apply ‘where it is established that the books in question were exported for the sole purpose of re-importation in order to circumvent legislation of the type at issue’.11 The Court went on to rule that the French provision on the price of reimported books was not justifiable under Article 30 EC [now Article 36 TFEU], as the latter did not provide the grounds of justification put forward by the French government, namely consumer protection. This conclusion was repeated in subsequent judgments on book price fixing12 as well as national competition law.13 This seems to be the only context in relation to what was then Article 28 EC where the abuse of law principle has arisen. However, the conclusion reached by the Court is not unproblematic, as the abuse of law principle applies to a trader who seeks to escape from the application of a national rule which is not only inconsistent in principle with the rule of free movement, but also unjustified. In other words, rather than ensuring the application of EU law in a way which meets its objectives, the abuse of law principle appears to be used in order to sanction an illegal and unjustified restriction on free movement.14 On the other hand, the abusive behaviour would entail bringing within the scope of EU law an activity which would otherwise be subject to national law, as there would be no intraUnion dimension.15 In tackling trade in medicinal products in DocMorris, the Court ruled that no abuse of the free movement provision could be found if the exporter was not involved in their reimportation.16 This develops a thread which brings together the case law on Article 34 TFEU [ex Article 28 EC] with that on agriculture which is discussed in this chapter. There may appear to be a parallel between the process of examining the objectives of EU rules and ascertaining whether they are met by bona fide commercial transactions and that of examining whether the objectives of Article 36 TFEU are being manipulated     9  Case 229/83 Association des Centres distributeurs Édouard Leclerc et al v SARL ‘Au Blé Vert’ et al [1985] ECR 1. 10   Leclerc (n 9 above), para [26]. 11   Leclerc (n 9 above), para [27]. 12  Case 299/83 SA Saint-Herblain distribution, centre distributeur Leclerc et al v Syndicat des libraries de Loire-Océan [1985] ECR 2515 and Case 95/84 Boriello v Alain Darras and Dominique Tostain [1986] ECR 2253. In his very brief Opinion in the latter case, AG Slynn pointed out that the possibility of abuse did not seem to arise in the case before the referring court. 13  Case C-137/00 The Queen v The Competition Commission, Secretary of State for Trade and Industry and The Director General of Fair Trading ex p Milk Marque Ltd and National Farmers’ Union [2003] ECR I-7975, paras [114–16]. 14   In another context, AG Tesauro argued that ‘[c]ertainly, it seems difficult even to envisage the existence of a general rule of Community law capable of negating a right conferred by a Community provision, especially in a harmonised field such as the company law field involved here, in confrontation with a domestic provision infringing that right’ (Case C-367/96 A Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I-2843, para [18]). 15  Thanks to Cathryn Costello for discussing this point. 16  Case C-322/01 Deutscher Apothekerverband eV v 0800 DocMorris NV, Jacques Waterval [2003] ECR I-14887, para [130].

206  Panos Koutrakos by national authorities in order to introduce disguised restrictions on free movement. In Henn and Darby, the Court refers to ‘the proper purpose’ of the interests laid down in the first sentence of what was then Article 30 EC, the protection of which is the function of the second sentence of that provision, which is to prevent diversions from it in ‘such a way as either to create discrimination in respect of goods originating in other Member States or indirectly to protect certain national products’.17 This point was also made in Aragonesa18 and Gourmet,19 while in Commission v UK, the Court set out in quite some detail the grounds on which reliance upon Article 30 EC was, in fact, protectionist and constituted a disguised restriction on intra-Community trade.20 The above pronouncements raise questions as to the process of identifying the intention of national authorities21 which are similar to those raised by the subjective strand of the Emsland-Stärke test (and which will be analysed below). However, this process is really underpinned by an objective consideration; that is, whether national measures are protectionist, in which case reliance upon the public interests set out in Article 36 TFEU [ex Article 30 EC] would not make them justified. III. The Emsland-Stärke test

The main test was formulated by the Court in Emsland-Stärke.22 This was a reference from the Bundesfinanzhof on the application of EC legislation on export refunds on agricultural products.23 According to the relevant rules, a trader would be entitled to export refunds if the product has left the geographical territory of the Union unaltered within sixty days from the day of completion of the customs export formalities. In cases where there is serious doubt as to the true destination of the product, or where it is possible that the exported product may be reintroduced into the Union due to the difference between the rate of refund on it and the amount of the import duties applicable to an identical product on the day when customs export formalities are completed, additional requirements are imposed: the payment of refund would be conditional on the product’s having been imported into a non-Member State. Furthermore, additional proof may be required by the competent authorities of the Member States which would show, to their satisfaction, that the product on which an export refund is requested has actually been placed on the market in the non-Member State of import in an unaltered state. In Emsland-Stärke, the subject-matter of the reference from the Bundesfinanzhof was two export transactions carried out by the plaintiff. The first consisted of several consignments of a product based on potato starch to Switzerland which, however, their recipients arranged to have transported back to Germany unaltered and by the same  Case 34/79 R v Maurice Donald Henn and John Frederick Ernest Darby [1979] ECR 3795, para [21].   Joined Cases C-1/90 and C-176/90 Aragonesa de Publicidad Exterior SA and Publivía SAE v Departamento de Sanidad y Seguridad Social de la Generalitat de Cataluña [1991] ECR I-4151, para [20]. 19  Case C-405/98 Konsumentombudsmannen (KO) v Gourmet International Products AB (GIP) [2001] ECR I-1795, para [32]. 20  Case 40/82 Commission v United Kingdom [1982] ECR 2793. 21   See P Oliver, Free Movement of Goods in the European Community, 4th edn (London, Sweet and Maxwell 2003) 225–26. 22   Emsland-Stärke (n 2 above). 23  Commission Regulation (EEC) No 2730/79 of 29 November 1979 laying down common detailed rules for the application of the system of export refunds on agricultural products [1979] OJ L317/1, as amended by Commission Regulation (EEC) No 568/85 of 4 March 1985 [1985] OJ L65/5. 17 18

The Emsland-Stärke Abuse of Law Test  207 means of transport; once import duties were paid in Germany, they were released for home use. The second export transaction consisted of several consignments of a wheat starch-based product to Switzerland which their recipients arranged to have forwarded unaltered and by the same means of transport to Italy where, once the relevant import duties were paid, they were released for home use. The plaintiff in the main proceedings was the exporter to Switzerland who challenged before the German courts the decision of the German Customs Office to demand repayment of the export refunds. It is worth noting here that the requirements imposed on the circumstances under which export refunds are granted aim to prevent exporters from benefiting from differences between the amount of the export refund and that of the production refund. In EmslandStärke, for instance, the export refund in the starch sector was approximately twice the amount of the production refund. Considered against the very low import duties, this explains why exporters may be tempted to manipulate the system for financial benefit. A.  The Judgment In its judgment, having observed that all the formal conditions for the grant of export refunds as laid down in secondary law had been fulfilled, the Court pointed out that ‘the scope of Community regulations must in no case be extended to cover abuses on the part of a trader’.24 It then explained why, in that context, Community law objected to this specific practice: ‘the fact that importation and re-exportation operations were not realised as bona fide commercial transactions but only in order wrongfully to benefit from the grant of monetary compensatory amounts, may preclude the application of positive monetary compensatory amounts’.25 The Court went on to formulate the test of abuse of law:26 A finding of an abuse requires, first, a combination of objective circumstances in which, despite formal observance of the conditions laid down by the Community rules, the purpose of those rules has not been achieved. It requires, second, a subjective element consisting in the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it. The existence of that subjective element can be established, inter alia, by evidence of collusion between the Community exporter receiving the refunds and the importer of the goods in the non-member country.

It was for the national court to assess whether the above conditions were met on the basis of evidence adduced in accordance with the rules of national law. However, there was a qualification to this function: the effectiveness of Community law should not be undermined.27 The Court, then, dealt with two objections put forward by the exporter. The first was legal in nature and was about the alleged incompatibility between a requirement that he repay refunds and the principle of lawfulness. The Court dismissed this by pointing out that:   Emsland-Stärke (n 2 above), para [51].   Emsland-Stärke (n 2 above), para [51]. 26   Emsland-Stärke (n 2 above), paras [52–53]. 27   Emsland-Stärke (n 2 above), para [54]. 24 25

208  Panos Koutrakos [T]he obligation to repay is not a penalty for which a clear and unambiguous legal basis would be necessary, but simply the consequence of a finding that the conditions required to obtain the advantage derived from the Community rules were created artificially, thereby rendering the refunds granted undue payments and thus justifying the obligation to repay them.28

The second objection was pragmatic in nature: the exporter argued that it was not he who had reimported the goods. This was also dismissed by the Court which observed that it was he who ‘enjoyed the undue advantage of the grant of export refunds when he carried out an artificial operation in order to benefit from that advantage’.29 B.  Comment on the Emsland-Stärke Test The articulation of a test of what constitutes abuse in Emsland-Stärke should be viewed in its proper context. On the one hand, there had been a number of instances where the Court had referred to abuse of law in a variety of other contexts too, most notably in services30 and establishment.31 On the other hand, the Commission in its submissions put forward very elaborate suggestions, almost urging the Court not to deal with the specific reference as yet another specific case of abuse but to set out the parameters which would determine the conditions under which an abuse of law would be deemed to have occurred. There is a third factor which may explain why the Court chose to articulate the abuse of law test in this specific legal context. The reference in Emsland-Stärke, as well as those in Cremer and General Milk Products previously, touched upon the financial interests of the Community which ought to be protected. The significance of this dimension is illustrated by the express provision in secondary law: Article 4(3) of Council Regulation 2988/95 on the protection of the European Communities’ financial interests provides that ‘[a]cts which are established to have as their purpose the obtaining of an advantage contrary to the objectives of the Community law applicable in the case by artificially creating the conditions required for obtaining that advantage shall result, as the case shall be, either in failure to obtain the advantage or in its withdrawal’.32 This provision was referred to by the Commission in its submissions, but not the Court, in Emsland-Stärke.33   Emsland-Stärke (n 2 above), para [56].   Emsland-Stärke (n 2 above), para [57]. 30   See for instance, Case 33/74 Johannes Henricus Maria van Binsbergen v Besthuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299; Case C-211/91 Commission v Belgium [1992] ECR I-6773; Case C-148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487; Case C-23/93 TV 10 SA v Commissariaat voor de Media [1994] ECR I-4795. For an analysis of the relevant case law, see D Doukas, ‘Free Movement of Broadcasting Services and Abuse of Law’, ch 6 above. 31   See, for instance, Case C-212/97 Centros Ltd v Erhvervs-og Selskabsstyrelsen [1999] ECR I-1459, delivered approximately a year and a half previously. For an analysis of the relevant case law, see V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205; see also W-G Ringe, ‘Sparking Regulatory Competition in European Company Law: The Impact of the Centros Line of Case Law and its Concept of “Abuse of Law”’, ch 8 above. 32  Council Regulation (EC) No 2988/95 of 18 December 1995 on the protection of the European Communities’ financial interests [1995] OJ L312/1. Abuse is expressly prohibited in other areas of law, too: see, for instance, Art 11(1) Council Directive 1990/434/EEC of 23 July 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States [1990] OJ L225/1, as amended by Council Directive 2005/19/EC of 17 February 2005 [2005] OJ L58/19. 33  However, the Court referred to it in the judgments analysed below. 28 29

The Emsland-Stärke Abuse of Law Test  209 In this respect, a parallel has been drawn with the public finances of the Member States and the justification afforded by EU law to measures aiming to protect them.34 For instance, in the area of health care, the Court has consistently accepted that the right of the Member States to organise their social security systems covers their right to protect the financial stability of their health care insurance schemes.35 However, there is a distinction to be drawn: in the case of the latter case law, the Court protects the right of the Member States to organise and manage their social security system which would enable them to carry out their fundamental obligations to their citizens; in the case of agriculture law and free movement of goods, the question is whether an EU right granted to a trader under specific circumstances for specific reasons should have been granted at all. Viewed against prior case law, the Emsland-Stärke test may appear to provide clarity in the application of a principle which had been applied rather incrementally. It may also appear to assist national courts in their task of assessing whether an abuse of law is substantiated by the facts in the dispute before them. Viewed in the wider free movement of goods context, the Emsland-Stärke test might even appear to assume the function which the Keck formula was perceived to assume in the context of the definition of measures of equivalent effect to a quantitative restriction.36 One characteristic which it shares with the latter definition is its generality, as the criteria articulated by the Court are quite broad. In fact, they are so broad that their significance may only be assessed in relation to a specific set of factual circumstances. It is in the light of this that, in a subsequent case, Advocate-General Sharpston pointed out that the Court has developed ‘a broad, and pragmatic, concept of abusive practice’.37 However, this raises the question to what extent is this test more helpful than the approach previously articulated by the Court? In other words, to what extent can this test facilitate the role which national courts are called upon to assume in the area? This question becomes more pertinent in the light of the reference to the subjective part of the test. This is problematic not so much due to the specific characteristics of legal persons and their distinction from natural persons,38 but rather to the inherently indeterminate nature of any test to assess whether this criterion is met. In another context, Advocate-General La Pergola argued that ‘[s]o long as [the] right [of establishment] is 34   See D Triantafyllou, ‘L’interdiction des abus de droit en tant que principe général du droit communautaire’ (2002) 38 Cahiers de droit européen 611, 628–29. 35   See, for instance, Case 238/82 Duphar BV et al v The Netherlands State [1984] ECR 523, para [16]. For an analysis of the health care case law from that angle, see P Koutrakos, ‘Healthcare as an Economic Service under EC Law’ in M Dougan and E Spaventa (eds), Social Welfare and EU Law (Oxford, Hart Publishing 2005) 105. 36   Joined Cases C-267/91 and C-268/91 Criminal Proceedings against Bernard Keck and Daniel Mithouard [1993] ECR I-6097. Of the voluminous literature on Keck, see the analysis of the clarity issue in S Weatherill, ‘After Keck: Some Thoughts on How to Clarify the Clarification’ (1996) 33 Common Market Law Review 885; and P Koutrakos, ‘On Groceries, Alcohol and Olive Oil: More on Free Movement of Goods after Keck’ (2001) 26 European Law Review 391. On the more recent caselaw, see E Spaventa, ‘Leaving Keck Behind? The Free Movement of Goods after the Rulings in Commission v Italy and Micklesson and Roos’, (2009) 34 European Law Review 219. 37  AG Sharpston in Case C-279/05 Vonk Dairy Products BV v Productschap Zuivel [2007] ECR I-239, para [56]. 38  This had been relied upon by AG Lenz in his criticism of a subjective criterion for assessing abuse of law even prior to the articulation of the Emsland-Stärke test: in his Opinion in TV10 (n 30 above), he argued that ‘a legal person as such is not in a position to exhibit subjective attitudes’ and noticed the absence of a ‘uniform manner of imputing acts of natural persons to the sphere of responsibility of a legal person which was valid in Community law for all Member States’ (para [61]). Therefore, he deemed the reliance upon subjective criteria in order to assess the legally relevant conduct of a legal person as ‘problematic’ and suggested that ‘the avoidance of legal provisions by a legal person should be able to be determined using objective criteria’ (ibid).

210  Panos Koutrakos exercised in accordance with the Treaty, the motives, calculations and particular personal interests underlying the choice do not come into consideration and are consequently not open to judgment’.39 In fact, even the formulation of the subjective criterion in EmslandStärke itself suggests its relative role. The Court referred, by way of example, to evidence of collusion between the Community exporter receiving the refunds and the importer of the goods in the third State as a case where the subjective criterion of the test would be met. However, it would be more accurate to refer to this as objective circumstances which suggest that the purpose of the relevant EU rules has not been achieved. In other words, as Advocate-General Poiares Maduro pointed out, ‘it is not th[e] intention [of abusing Community law] that is decisive for the assessment of the abuse’ but ‘the activity itself, objectively considered’.40 Be that as it may, the broad test set out in Emsland-Stärke and the unclear manner in which it develops the previous strands of the Court’s case law on abuse suggest that at the very centre of its application lies the Court itself. This is further illustrated by the entire line of reasoning put forward in the judgment which is underpinned by a clear focus on the objective of the EC rules reliance upon which is alleged to amount to abuse of law. The Court refers to the need to establish that ‘the purpose of those rules has not been achieved’41 because a company ‘creat[es] artificially the conditions laid down for obtaining an advantage from them’.42 Therefore, the definition of the objective of the relevant EU rules is the starting point for any analysis of the abuse issue. As this definition is a matter entirely for the Court of Justice, the teleological interpretation upon which the assessment of abuse relies ensures the central role of the Union’s judiciary in the application of this principle. In effect, the approach of the Court stresses the duty of national courts to rely upon what is now Article 267 TFEU [ex Article 234 EC] in order to get an authoritative reading of the EU rules invoked before them. This is quite significant in the light of the role for national courts carved out by the Court in relation to the application of the abuse test. Whilst its role is quite pronounced, the Court of Justice leaves it to national courts to decide whether an abuse of law has actually occurred in accordance with the rules of national law. The only qualification which it imposes is that, in doing so, the effectiveness of EU law is not undermined. Therefore, the role of national courts in the application of the abuse test is central. It should be pointed out that the Court’s pronouncement should be examined within its proper context. In accordance with ‘the distinct separation of functions between national courts . . . and the Court of Justice’43 which underpins the   Centros (n 31 above), para [20].  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd v Commissioners of Customs and Excise [2006] ECR I-1609; Case C-419/02 BUPA Hospitals Ltd and Goldsborough Developments Ltd v Commissioners of Customs and Excise [2006] ECR I-1685; Case C-223/03 University of Huddersfield Higher Education Corporation v Commissioners of Customs and Excise [2006] ECR I-1751, para [70]. He went on, at para [71], to observe that: [I]t is not . . . a search for the elusive subjective intentions of the parties that ought to determine the existence of the subjective elements mentioned in Emsland. Instead, the intentions of the parties to improperly obtain an advantage from Community law are merely inferable from the artificial character of the situation to be assessed in the light of a set of objective circumstances . . . In such circumstances, the legal provision at issue must be interpreted, contrary to its literal meaning, as actually not conferring a right. It is consideration of the objective purpose of the Community rules and of the activities carried out, and not the subjective intentions of individuals, which, in my view, lies at the heart of the Community doctrine of abuse. 41   Emsland-Stärke (n 2 above), para [52]. 42   Emsland-Stärke (n 2 above), para [53]. 43  Case 35/76 Simmenthal v Ministero delle Finanze [1976] ECR 1871, para [4]. 39 40

The Emsland-Stärke Abuse of Law Test  211 preliminary reference procedure, it is for the former to apply the interpretation of EU rules given by the latter to the facts of the case before them. In effect, this is what the Court requires that they do in order to assess whether one of the parties in the dispute before them has created artificially the conditions laid down in EU rules in order to obtain an advantage from them. However, the limit imposed on national courts in assessing whether an abuse has occurred and the pivotal position that the definition of the objectives of the EU rules in question has for that assessment have led commentators to suggest that the power with which the national courts are endowed lacks substance.44 In other words, it has been argued that, in effect, the power which the Court appears to offer national courts in relation to the application of the abuse of law test is taken back because of the provisos attached to it. It is recalled that this argument had already been advanced in order to explain the approach of the Court to its relationship with national courts in the context of the preliminary reference procedure and, in particular, the latter’s Court-given power not to refer, even at last instance.45 This position may appear to suggest the following two propositions: that the approach underpinning the Emsland-Stärke test imposes too heavy a duty on national courts by circumscribing their function, and that it would be intellectually more honest if the Court applied the abuse of law test itself. It is suggested that both such propositions are misplaced. First, whilst tasked with the application of EU rules to the facts of a case before them, national courts carry out an EU law task—they give effect to the decentralised judicial system set out in European primary law and are required to act in accordance with the duty of cooperation as laid down in Article 5 TEU [ex Article 5 EC]. The articulation of the limits within which the national courts are required to act in carrying out this function is neither surprising nor suspect, as, in any case, their objective, namely to ensure that the effectiveness of EU law is not undermined, would be a conditio sine qua non of their task even if not articulated by the Court of Justice expressly. Second, the role of national courts becomes all the more central in the specific context of the application of the abuse of law principle, as it is mainly focused on the nature of the commercial activity which has given rise to a dispute: to ascertain whether that is abusive is a matter of fact and, as such, intrinsically linked to the function with which national courts are traditionally endowed in the context of the preliminary reference procedure. Third, whilst it is true that it has not shied away from making substantive judgments about the application of EU rules to the facts of a case pertaining to a reference by a national court,46 it is within the function of national courts as allocated in the symbiotic  A Kjellgren, ‘On the Border of Abuse’ (2000) 11 European Business Law Review 179, 191–92.   See the acte claire doctrine set out in Case 182/81 Sri CILFIT and Lanificio di Gavardo SpA v Ministry of Health [1982] ECR 3415 and the criticism of it in H Rasmussen, ‘The European Court’s Acte Claire Strategy in CILFIT’ (1984) 9 European Law Review 242. For the opposite view, see A Arnull, ‘The Use and Abuse of Article 177 EEC’ (1989) 52 Modern Law Review 622. 46   See, for instance, the application of the principle of proportionality in Joined Cases C-306/88, C-304/90 and C-169/91 Stoke-on-Trent City Council and Norwich City Council v B & Q plc [1992] ECR I-6457; Case C-312/89 Union départementale des syndicats CGT de l’Aisne v SIDEF Conforama, Société Arts et Meubles and Société Jima [1991] ECR I-997; Case C-332/89 André Marchandise, Jean-Marie Chapuis and SA Trafitex [1991] ECR I-1027; and the analysis in M Jarvis, The Application of EC Law by National Courts (Oxford, Oxford University Press 1998) 223–25, 269–72. On the other hand, the Court did not rule on this issue in Gourmet (n 19 above), ignoring the advice of AG Jacobs (for criticism of the Court’s approach in this case, see A Biondi, ‘Advertising Alcohol and the Free Movement Principle: The Gourmet Decision’ (2001) 26 European Law Review 44 45

212  Panos Koutrakos relationship established under Article 267 TFEU, to apply the authoritative interpretation of EU rules to the facts of the case. While criticism of the Emsland-Stärke test for effectively undermining the role of national courts was viewed above as misplaced, another issue arises, namely that of ensuring the effectiveness of European law.47 It would be a worrying phenomenon if national courts proved to be too keen to determine abuse of law—this would not only raise questions about compliance with the principle of legal certainty but it might also undermine the effectiveness of the relevant substantive EU law provisions insofar as EU rights bestowed in order to meet specific objectives would not be enforced. Is it not risky to endow national courts with a power which would enable them not to apply European law? Would this power not be too tempting for national judges to ignore? These questions need to be addressed against the context of specific examples of how national courts have handled the application of the abuse of law test. This is what the following section will examine. IV. The application of the test and the response by national courts

Four and a half years after the Court articulated the Emsland-Stärke test, it was asked to apply it in Eichsfelder Schlachtbetrieb.48 This was a reference from the Finanzgericht Hamburg about refunds regarding exports of beef to Poland. In Poland, the goods were used to make cooked meat roulades and then, under a contract concluded between the producer of these roulades and the buyer, they were exported back to Germany with the Polish customs duties being reimbursed. When confronted with an order to repay the export refunds, the plaintiff in the main proceedings argued that the goods for which the export refund had been paid had undergone substantial working in Poland. The question referred by the Finanzgericht Hamburg was whether, under secondary legislation, the fact that, following its release into free circulation in Poland, the product in question underwent substantial processing or working suggested that it had met the conditions set out in secondary law requiring that it be imported in the third country. The Court pointed out a number of factors which suggested that the goods had been genuinely imported into Poland: import duties had been paid, and a substantial working or processing of the goods within the meaning of the Customs Code had led to the creation of a new product, hence suggesting that those goods had been put to use in the third country and had actually been put on the market there, being released for consumption. All this ‘eliminates the risk—[which secondary legislation] seeks to obviate—of abusive reimportation of the initial goods into the Community, in breach of the aim pursued by the refund system’.49 The Court also dealt with the argument put forward by the Commission and the referring court that, given that import duties had been reimbursed, the exporter had no 616). See also the application of the principle of State liability in Case C-392/93 R v Her Majesty’s Treasury ex p British Telecommunications plc [1996] ECR I-1631. 47   See P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351, 361. 48  Case C-515/03 Eichsfelder Schlachtbetrieb GmbH v Hauptzollamt Hamburg-Jonas [2005] ECR I-7355. 49   Eichsfelder Schlachtbetrieb (n 48 above), para [33].

The Emsland-Stärke Abuse of Law Test  213 right to an export refund. The Court rejected it by making the following distinction: in cases where the customs formalities for release for consumption in the non-Member State have been completed and import duties paid, a subsequent reimbursement of those duties to an economic operator other than the exporter cannot retroactively render the export refunds unduly paid. If that was the case, ‘the exporter would be placed in a position of uncertainty, arguably in breach of the principle of legal certainty, and his right to a refund would depend on events or commercial conduct outside his control’.50 On the other hand, the above would not apply in cases where the exporter himself had participated in an abusive practice. In order to substantiate this conclusion, the Court refers to Regulation 2988/95 on the protection of the European Communities’ financial interests which states that ‘[a]cts which are established to have as their purpose the obtaining of an advantage contrary to the objectives of the Community law applicable in the case by artificially creating the conditions required for obtaining that advantage shall result, as the case shall be, either in failure to obtain the advantage or in its withdrawal’.51 Having repeated the Emsland-Stärke test, the Court went on to conclude that it was for the national court to determine whether these conditions had been met. In its judgment following the Court’s ruling, the referring court, the Finanzgericht Hamburg, applied the twofold test first articulated in Emsland-Stärke.52 In doing so, it showed considerable restraint. In relation to the objective part of the test, it pointed out that the purpose of the relevant EC rules would not have been achieved had the product not reached the market in Poland. However, it held that that was not the case as, after all, customs formalities had been met, and the product was marketed by being turned into roulades. In terms of the subjective part of the test, the national court pointed out that the crux of the matter was whether a normal transaction with economic objectives had been carried out or whether a transaction was carried out with the sole objective of achieving the grant of export refunds. The national court adds that the fact that what motivated the exporter was the grant of export refunds did not meet the subjective condition of the test; neither did the existence of artificial transactions between the importer and third parties. On the facts of the case, there was no evidence that the exporter was involved, participated in or had any influence on the importer’s conduct. The Court went on to rule that even knowledge on behalf of the exporter of the importer’s conduct would be irrelevant in legal terms. The reason for this is that knowledge of abusive behaviour does not constitute abuse in itself and, therefore, does not meet the subjective criterion of the Emsland-Stärke test. The Court added that, on the facts of the case, it was not clear whether the importer had engaged in artificial transactions. However, this was doubtful as its operations were clearly aiming to benefit from lower Polish costs in preparing roulades. The above suggests a degree of caution in the approach adopted by the national court. In ascertaining whether the conditions for the abuse of law test are met in general, and in setting a high threshold for the subjective element of abuse in particular, the national court appears as mindful of the possibility of abuse pursuant to the Emsland-Stärke test as it is of the requirement for legal certainty. This, in itself, is not a departure from the Court’s   Eichsfelder Schlachtbetrieb (n 48 above), para [36].  Council Regulation (EC) No 2988/95 (n 32 above), Art 4(3), mentioned verbatim in para [38] of the judgment. 52   IV 106/05, judgment of 26 January 2006. 50 51

214  Panos Koutrakos articulation of the principle. After all, the judgment in Eichsfelder Schlachtbetrieb did refer to ‘participation’ by the exporter himself.53 Another case where the abuse of law test was applied was Vonk Dairy.54 This was a reference from the College van Beroep voor het bedrijfsleven (Administrative High Court for Trade and Industry, The Netherlands) about the export of consignments of Italian pecorino cheese to the United States during a six-year period. Following investigations by the Netherlands authorities and US Customs, it was revealed that a number of the above consignments (seventy-five out of a total of 2,100) were almost immediately reexported to Canada by an intermediary of the exporter with the active involvement of the latter who was aware that the cheese was forwarded to Canada and was involved in the sale of those consignments in that country. The subject-matter of the reference was the request that the exporter, who was the applicant in the main proceedings, repay the export refunds it had received. Having pointed out that all the formal requirements for the grant of export refunds had been met, the Court ruled that the national courts would have to assess whether there was evidence of abuse on the part of the exporter pursuant to the Emsland-Stärke test. The referring court enquired whether, under the specific circumstances, a continuous or repeated irregularity had occurred. This question was made in the context of Regulation 2988/95 which set out a more liberal limitation period for proceedings against traders. The Court pointed out that the fact that the irregularity related to a relatively small proportion of all the transactions carried out in a given period and that the transactions in which the irregularity had been detected always concerned different consignments was immaterial. Instead, a continuous or repeated irregularity occurs ‘where it is committed by a Community operator who derives economic advantages from a body of similar transactions which infringe the same provision of Community law’.55 Again, it was for the national court to ascertain whether action constituting such an irregularity had taken place in the main proceedings. In doing so, it would have to act in accordance with the rule of evidence of national law, provided that the effectiveness of Community law was not undermined. When the case was sent back to the referring court, the latter, responding to the argument of the Dairy Products Board that the investigation reports included sufficient points of references substantiating abuse, pointed out that these were not mentioned in detail in the relevant decision. It also added that it was not for that court to go through the main dossiers, of which there were many, and which contained data about the dispute, in order to find out whether there has been an abuse in accordance with the test set out by the Court of Justice. Therefore, it concluded that there were no grounds on which the legal outcome of the contested decision should stand. Rather than indicating restraint or even caution, the approach of the national court may appear to be short on reasoning and suggests unwillingness to engage in a thorough examination of whether the conditions of the Emsland-Stärke test were met. However, the specific circumstances under which the reference was made suggests that the referring court had already made it clear that the order for repayment upheld in the contested decision was not based on abuse on the part of the appellant.56 In addition, Advocate  Eichsfelder Schlachtbetrieb (n 48 above), para [37].   Vonk Dairy (n 37 above). 55   Vonk Dairy (n 37 above), para [41]. 56  AG Sharpston in Vonk Dairy (n 37 above), para [62]. 53 54

The Emsland-Stärke Abuse of Law Test  215 General Sharpston pointed out that the contested decision was ‘at least to some extent, ambiguous’.57 Therefore, what seems to underpin the referring court’s final decision is an issue of national procedural law: is the assessment of whether the conditions for the abuse of law test are met a matter for the national authority responsible for examining refund papers or the national court reviewing the legality of that authority’s decision?58 As Advocate-General Sharpston observes, this is a matter for national law.59 As was the case in Eichsfelder Schlachtbetrieb, the application of the EmslandStärke test by the referring court in Vonk Dairy does not justify the alarm which the acknowledgment by the Court of their central role in determining abuse of law raised. In fact, a degree of caution seems to emerge from this sample of national reactions. Rather than applying the abuse of law test mechanically, national judges are prepared to engage in a detailed assessment of both their role as set out under national law and the facts of the case upon which they are asked to adjudicate. In relation to the latter, clear evidence of the abusive nature of the commercial activity in question is required. While the number of national judicial decisions examined in this section is, admittedly, very limited, it is in precisely the same area in which the Court of Justice articulated the abuse of law test. In any case, the role of national courts in the application of the abuse of law test ought to be viewed in the context of the preliminary reference procedure under which the abuse of law cases arise. Given the heavily factual dimension which underpins the test, as articulated in Emsland-Stärke, and the function of national courts in the preliminary reference procedure as defined by the Court of Justice—namely to apply the Court’s interpretation of EU law to the facts of the case before them—it would be difficult to see a task more suitable for national courts in this context than that articulated in Emsland-Stärke. As for the uncertainty to which the application of the abuse of law test may give rise, a degree of uncertainty is inherent in the preliminary reference procedure. In fact, it is the direct corollary of the central position which national courts enjoy in the constitutional architecture of the Union’s legal order in general and the decentralised system of enforcement set out in Article 267 TFEU. This is a given which has been acknowledged and built upon by the Court of Justice in its development and management of the Article 267 TFEU procedure. This is suggested by the role it has carved out for national courts in the application of the constitutionalising principles of EU law60 such as the duty of interpretation of national legislation in the light of a directive61 and State liability for a violation of EU law.62 It is also suggested by the central role it has granted them in the determination of substantive law issues such as the application of the principle of proportionality in free movement cases, as mentioned above in this chapter. To argue that the involvement of national courts in the above contexts is an essential part of the success of the multilayered system of the European legal order as developed over the years is to  AG Sharpston in Vonk Dairy (n 37 above), para [63].  AG Sharpston in Vonk Dairy (n 37 above), para [64]. 59  AG Sharpston in Vonk Dairy (n 37 above), para [65]. 60   See GF Mancini, ‘The Making of a Constitution for Europe’ in Democracy and Constitutionalism in the European Union: Collected Essays by Judge GF Mancini (Oxford, Hart Publishing 2000) 1; and C Timmermans, ‘The Constitutionalisation of the European Union’ (2002) 21 Yearbook of European Law 1. 61   See for instance, Case 14/83 Sabine von Colson and Elisabeth Kamann v Land Nordrhein-Westfalen [1984] ECR 1891 and Case C-106/89 Marleasing SA v La Comercial Internacional de Alimentacion SA [1990] ECR I-4135. 62   See for instance, Joined Cases C-46/93 and C-48/93 Brasserie du Pecheur SA v Germany and R v Secretary for Transport ex p Factortame Ltd et al [Factortame III] [1996] ECR I-1029. 57 58

216  Panos Koutrakos state the obvious. In the context of the present analysis, it is noteworthy that this system, with the vital position at its core granted to national courts and the inherent uncertainty which the latter’s performance may raise, has not produced such considerable problems which its decentralised character, the evolving interpretation of EU law and the exercise of the Court’s jurisdiction have not succeeded in addressing.63 This point is illustrated in the broader context of the law of free movement of goods in general, and the application of the principle of proportionality by national courts in particular, a parallel with which some authors have drawn:64 after all, in areas where national courts felt unclear as to how to apply it, their channel of communication with the Court of Justice proved efficient,65 whereas in cases where national courts were required to apply new principles of EC law, the Court of Justice assisted them of its own motion.66 Put differently, complete certainty in the application of European law principles is elusive in the constitutionally idiosyncratic EU legal order, a fact which has not necessarily been proved to be detrimental to the process of European integration. What this section suggests is that there is no evidence of abuse of the abuse of law test by national courts. V. Conclusion

The analysis of the precursor to the Emsland-Stärke test, as well as the latter’s application by national courts in the area of agriculture, suggests the existence of a number of common strands. These include the definition of the objective of the EU rules which a trader is alleged to contravene, the role of the national court to ascertain whether the transaction in question meets the criteria of the abuse of law test with due regard to the totality of circumstances, the extent to which the confines within which national courts ought to act are onerous, and, ultimately, the constitutional function of the Court in this process which is based, mainly, on its symbiotic relationship with the national judiciary. There is a question which seems to underpin these strands, namely whether the abuse of law test, as articulated in Emsland-Stärke, constitutes a useful step which clarifies the legal position of the individual, and facilitates the role which national courts are expected to assume, and, ultimately, their interaction with the Court of Justice. Whilst the Emsland-Stärke test clarifies the position of abuse of law as a general principle, this clarity is superficial, for its effect may only be determined by the content of its substantive provisions and the processes pursuant to which it is to be applied. Viewed from this angle, it is difficult to see what is the contribution of the Emsland-Stärke test other than to raise the profile of the abuse of law principle as a distinct, albeit limited,67 possibility in the 63  In relation to the application of the Francovich principle in particular, see M-P Granger, ‘National Applications of Francovich and the Construction of a European Administrative Jus Commune’ (2007) 32 European Law Review 157. 64   See for instance, LN Brown, ‘Is There a General Principle of Abuse of Rights in European Community Law?’ in D Curtin and T Heukels (eds), Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers, Vol II (The Hague, Martinus Nijhoff 1994) 511, 520. However, this parallel is not very helpful, insofar as the principle of proportionality is about the exercise of a right conferred by or acknowledged by EU law, whereas the abuse of law principle is about a right which is exercised in circumstances where it does not arise. 65   See the Sunday Trading line of cases. 66   See for instance, British Telecom (n 46 above). 67  In Kefalas (n 14 above), para [23], AG Tesauro pointed out that: ‘I consider that the risk of there being a gap

The Emsland-Stärke Abuse of Law Test  217 European legal order and, therefore, to make it even clearer that there are limits on the scope of rights granted under EU law. The ambiguous position of the subjective part of the test and its reliance upon objective considerations, the definition of the objective of the relevant EU rules by the Court of Justice, the central position of national courts within the context of the preliminary reference procedure, and the requirement that the latter courts rule in compliance with the effectiveness of the applicable EU rules all point towards the direction of applying standard principles of interpretation with which both the Court of Justice and national courts are familiar. Finally, the analysis in this chapter suggested that there is no evidence of abuse of the abuse of law test by the national courts in the law of agriculture and free movement of goods. Therefore, the role of the abuse of law test, as articulated in Emsland-Stärke, should not be overestimated, neither should its implications for national courts and, ultimately, the effectiveness of the integration process be exaggerated.

in the system—which is, after all, what the abuse of rights principle, like all other so-called catch-all provisions, seeks to avoid—is minor, or non-existent, in a legal order like that of the Community which, through judicial interpretation and case-law in general, is more promptly amenable to adaptation to the needs of society’. In this Opinion, as well as in the one in Case C-441/93 Panagis Pafitis et al v Trapeza Kentrikis Ellados AE et al [1996] ECR I-1347, he argued against the existence of abuse of law as a general principle of EC law. See also KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, who argues that the principle is of a narrow scope.

15 The Notion of and a General Test for Abuse of Rights: Some Normative Reflections JUKKA SNELL* I. INTRODUCTION

I

n Emsland1 the European Court of Justice, in a plenary composition and following the Opinion of Advocate-General Alber and the suggestion of the Commission, for the first time made clear that there is a general doctrine of abuse of rights2 in European Union law and offered a proper test for determining what amounts to such abuse. Although there had been earlier decisions that did refer to the notion, only in Emsland did the Court attempt to consolidate its approach.3 In so doing, the Court undertook a task of fearsome difficulty. As Joseph Voyame, Bertil Cottier and Bolívar Rocha write in a report for the Council of Europe: The European states have very few points in common as regards the definition of abuse of rights, conditions of implementation, areas of application and, finally, legal effects . . . [T]here are almost as many conceptions of the abuse of rights as there are member states of the Council of Europe . . . [T]his diversity is further accentuated by the fact that, within most countries, there is no unanimous agreement as to the scope of the prohibition of abuse of rights; doctrinal disputes and contradictory judgments are commonplace.4

 *  I wish to thank colleagues at Swansea University, in particular Dr Bebhinn Donnelly, Dr Helen Quane, and Professor Volker Röben, for stimulating discussions on the topic. 1   Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 2   The term ‘abuse of rights’ is used throughout this chapter for the sake of convenience, as it is the term that most frequently features in the case law, including Emsland itself, and in Union legislation, for example Art 35 of the European Parliament and Council Directive (EC) 2004/38 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States [2004] OJ L158/77. See however AG Poiares Maduro in Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, para [71] for an argument that the term ‘prohibition of abuse of [Union] Law’ is to be preferred. 3   Similarly, R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 408–11; and D Triantafyllou, ‘L’interdiction des abus de droit en tant que principe général du droit communautaire’ (2002) 38 Cahiers de droit européen 611, 632. 4   J Voyame, B Cottier and B Rocha, ‘Abuse of Rights in Comparative Law’ in Council of Europe, Abuse of Rights and Equivalent Concepts: The Principle and Its Present Day Application (Strasbourg, Council of Europe 1990) 23.

220  Jukka Snell The present chapter seeks to offer some normative remarks relating to the EU doctrine of abuse of rights, building on the detailed analysis of Emsland carried out by Professor Panos Koutrakos.5 It will proceed in two stages, first discussing the very need for the doctrine and then considering the actual test put forward by the Court. It will be argued that while the doctrine of abuse of rights is a necessary feature of Union law, it is questionable whether it was given the optimal form in the two-part Emsland test. II.  THE NEED FOR A GENERAL DOCTRINE OF ABUSE OF RIGHTS IN THE EU

It is well known that the doctrine of abuse of rights operates in some EU Member States, but not in all of them. In particular, comparativists have often drawn attention to the fact that many civil law systems employ the notion, while it is not known in common law jurisdictions. For example, in France in the celebrated Clément-Bayard case,6 the Cour de Cassation recognised the doctrine of abuse of rights in a situation involving property rights. Mr Coquerel had acquired a small piece of land near the airship hangar of the Clément-Bayard Company. He wished to sell the plot to the company at a high price, and proceeded to construct two tall wooden scaffoldings bristling with iron spikes on his land to encourage the sale, as in high winds they made it difficult for airships to take off and land from the hangar. The Court ordered Mr Coquerel to pay damages and to remove the iron spikes. The fact that the wording of the French Civil Code seemed to give him an absolute right to build structures on his property was of no assistance to him; it was held that he had abused his property rights, given that his intention was to cause damage. By contrast, in England in the Mayor of Bradford v Pickles judgment7 the House of Lords rejected entirely the notion of abuse of rights. Again, in issue was the use of property rights. Mr Pickles undertook various mining operations on his land. The effect was to pollute and reduce the springs and streams on a neighbouring property that supplied the town of Bradford with water. It was argued that the real purpose of Mr Pickles was not to work his minerals but to cause damage and induce the purchase of his land. The unanimous House of Lords held that the motives of Mr Pickles were irrelevant. As the owner of the property, he had the legal right to do what he was doing, and the fact that he may have acted maliciously or with the object of injuring another was of no significance.8 In their classic study, Professors Catala and Weir argue that the different approaches to abuse of rights in common and civil law jurisdictions can be explained by their different approaches to rights themselves. In common law, rights tend to be tightly circumscribed and contain their own limitations. If a right has been developed in case law, its remit is no wider than the ratio decidendi of the judgment, and it is typically in any event hedged by various qualifiers, such as reasonableness. Any statements by the judiciary going beyond the facts are obiter and therefore do not bind other courts. As a matter of legislative technique, statutes are often very detailed, broad rights are rarely granted, and in any 5   See P Koutrakos, ‘The Emsland-Stärke Abuse of Law Test in the Law of Agriculture and Free Movement of Goods’, ch 14 above. 6   Cour de Cassation, Civ Ch, 3 August 1915, Recueil Dalloz 1917.I.79. An English translation appears in P Catala and JA Weir, ‘Delicts and Torts: A Study in Parallel. Part II’ (1964) 38 Tulane Law Review 221, 224. 7   Mayor of Bradford v Pickles [1895] AC 587 (HL). 8   See generally on the judgment as the precedent ‘slamming the common law door shut on the civil law doctrine of abuse of rights’ M Taggart, Private Property and Abuse of Rights in Victorian England (Oxford, Oxford University Press 2002) 145–66.

The Notion of and a General Test for Abuse of Rights  221 event, courts are prepared to imply limits to them. By contrast, in civil law countries, rights are often conferred in a more sweeping fashion, with the result that the need for a doctrine curtailing their abuse is correspondingly greater.9 The EU legislative technique bears a closer resemblance to the civil law systems than to the common law ones. The TFEU itself is often described as a traité cadre that only lays down the key principles and leaves the details to secondary legislation,10 and it certainly offers extremely wide rights, in particular to cross-border movement. Ordinary legislation often takes the form of directives, which by their nature11 are apt to lack the detail of a common law statute.12 This is reinforced by Article 296 TFEU, which requires the EU institutions to respect the principle of proportionality when selecting the type of act to be adopted. As for the judicial method, the rulings of the Court of Justice do not lend themselves to a meaningful distinction between ratio and obiter,13 and in some judgments the Court has been prepared to make far-going pronouncements that are not strictly related to the facts of the case, but have nevertheless been treated as precedents.14 Thus, as a matter of legislative and judicial style, the Union legal system would seem to require a doctrine of abuse of rights more than a typical common law system. Incidentally, it has been argued that much the same is true for international law. Michael Byers writes that ‘international law is actually more like the civil law when it comes to rights, in that both systems tend to characterise some rights as general and primordial’.15 This leads him to argue that the principle has an important role to play in international law. Indeed, international tribunals, such as the WTO Appellate Body in the ShrimpTurtle case,16 have in practice referred to the notion of abuse of rights.17 Further, as the primary role of the doctrine of abuse of rights is to rectify shortcomings in legal rules,18 or ‘gap[s] in the system’ as Advocate-General Tesauro put it,19 it is clearly     9   Catala and Weir (n 6 above) 237–38. See however A di Robilant, ‘Abuse of Rights: The Continental Drug and the Common Law’ [2009] ExpressO, available at works.bepress.com/anna_di_robilant/1, who argues that the US doctrines of malice and reasonable use closely parallel abuse of rights. 10   See for example, the colourful description by Lord Denning MR in Bulmer Ltd v Bollinger SA [1974] 2 CMLR 91, paras [39–40]. 11   Art 288 TFEU [ex Art 249 EC] provides that ‘[a] directive shall be binding, as to the result to be achieved . . . but shall leave to the national authorities the choice of form and methods’. 12   See P Prescott QC in Oakley v Animal [2005] EWHC 210 (Ch), paras [39–40] for a discussion that is highly critical of the British style in the context of registered designs. See also A Arnull, A Dashwood, M Dougan, M Ross, E Spaventa and D Wyatt (eds), Wyatt & Dashwood’s European Union Law, 5th edn (London, Sweet & Maxwell 2006) 166–67. 13   A Arnull, The European Union and its Court of Justice, 2nd edn (Oxford, Oxford University Press 2006) 631. 14  See for example, the denial of horizontal direct effect for directives in Case 152/84 MH Marshall v Southampton and South-West Hampshire Area Health Authority (Teaching) [1986] ECR 723, para [48], which on the facts involved a vertical situation, or the statement in Case C-385/99 VG Müller-Fauré v Onderlinge Waarborgmaatschappij OZ Zorgverzekeringen UA, and between EEM van Riet and Onderlinge Waarborgmaatschappij ZAO Zorgverzekeringen [2003] ECR I-4509, para [103] on the applicability of what is now Art 56 TFEU [ex Art 49 EC] to a national health service in a case that concerned an insurance-based system. 15   M Byers, ‘Abuse of Rights: An Old Principle, A New Age’ (2002) 47 McGill Law Journal 389, 415. 16   United States—Import Prohibition of Certain Shrimp and Shrimp Products, Report of the Appellate Body, WT/DS58/AB/R, adopted 6 November 1998, in particular, para [158]. 17   See also LN Brown, ‘Is There a General Principle of Abuse of Rights in European Community Law?’ in D Curtin and T Heukels (eds), Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers, Vol II (Dordrecht, Martinus Nijhoff 1994) 516, who argues that the recognition of a general principle of abuse of rights in public international law lends support to its recognition by the EU. 18   See Voyame et al (n 4 above) 23, 48. 19   AG Tesauro in Case C-367/96 Alexandros Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I-2843, para [23].

222  Jukka Snell needed in the EU.20 The drafting of Union legislation presents particular challenges due to the multilingual nature of the EU and the way its legislative process operates.21 There is an acknowledged need for better lawmaking22 and some fairly gruesome examples of poorly drafted measures can be put forward, including the recent Services Directive,23 although it is of course likely that most EU Member States would also benefit from better lawmaking and examples of poor drafting can undoubtedly be found in every jurisdiction. More importantly, the correction of errors and the plugging of gaps are more difficult for the EU than for most of its Member States.24 The Treaties have proven difficult to change in recent years, in particular due to the need for referenda in some countries, and an intergovernmental conference followed by ratifications and possibly preceded by a Convention, or a unanimous decision of the European Council followed by approvals in accordance with national constitutional requirements25 are in any event particularly cumbersome corrective mechanisms. The amendment of secondary legislation is also notoriously difficult. Under a unanimity decision rule a single veto will stop any changes, creating the famous joint-decision trap.26 Even under qualified majority voting the requirements are formidable. Under the ordinary legislative procedure, after the expiry of transitional provisions, what is needed in favour is: the majority of Commissioners (to put forward a proposal), at least fifty-five per cent of the members of the Council, comprising at least fifteen of them and representing Member States comprising at least sixty-five per cent of the population of the Union, and a majority in the European Parliament.27 Further, the classical Union method and the principle of institutional balance mean that in most cases a number of institutions are involved in the overall decision-making process,28 at the very least in a consultative capacity, while the multilingual nature of the EU creates the need for translations. As a result, it is difficult to push through measures with any great speed, and in the case of directives the time needed for the transposition by the Member States will cause further delay. Consequently, it may well take years before the EU legislature can amend an unsatisfactory law. Thus, a doctrine of abuse of rights which allows the courts to step in seems desirable for the Union. Of course the idea of the judicial institutions correcting the shortcomings of the political ones brings us to familiar 20   Similarly in the context of international law H Lauterpacht, The Development of International Law by the International Court (London, Stevens & Sons 1958) 162. 21  See for example, J Flower, ‘Negotiating European Legislation: The Services Directive’ (2006–7) 9 Cambridge Yearbook of European Legal Studies 217, in particular 231–32 and 237–38; and J Welch, ‘Better Law-making’ in S Weatherill (ed), Better Regulation (Oxford, Hart Publishing 2007) 311, 314–16; and more generally H Xanthaki, ‘The Problem of Quality in EU Legislation: What on Earth is Really Wrong?’ (2001) 38 Common Market Law Review 651, 676 who argues that ‘the nature of EU law as droit diplomatique demands some sacrifices’ but that the EU legislature has also failed to focus sufficiently on the matter. 22   See for example, European Parliament, Council, and Commission Interinstitutional Agreement on better law-making [2003] OJ C321/1. For discussion, see the various essays in Weatherill (n 21 above). 23   European Parliament and Council Directive (EC) 2006/123 on services in the internal market [2006] OJ L376/36. For discussion, see for example, J Snell, ‘Freedom to Provide Services in the Case Law and in the Services Directive: Problems, Solutions, and Institutions’ in U Neergaard, R Nielsen and LM Roseberry (eds), The Services Directive: Consequences for the Welfare State and the European Social Model (Copenhagen, DJØF Publishing 2008) 171, 192–94. 24   See also P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351, 373–74. 25   See Art 48 TEU for the detailed rules. 26  FW Scharpf, ‘The Joint-Decision Trap: Lessons from German Federalism and European Integration’ (1988) 66 Public Administration 239, 257. 27   See in particular Art 16 TEU and 294 TFEU. 28   See G Majone, Dilemmas of European Integration: The Ambiguities and Pitfalls of Integration by Stealth (Oxford, Oxford University Press 2005) 44–51.

The Notion of and a General Test for Abuse of Rights  223 territory, as the European Court of Justice for a long time acted as the primus motor of integration when the EU legislature suffered from paralysis.29 However, Advocate-General Tesauro has presented a powerful counter-argument.30 He submits that there is little need for an abuse of rights doctrine ‘in a legal order like that of the [Union] which, through judicial interpretation and case-law in general, is more promptly amenable to adaptation to the needs of the society’.31 In other words, the powerful position of the courts in the system, reminiscent of the common law jurisdictions, ensures that provisions of EU law are interpreted sensibly in the first place, and therefore any formal doctrine of abuse of rights is superfluous. Instead of there being a right which is then found to be abused, the Union judicature can simply hold that the right does not exist in the first place, just like a common law court would do. There is no need for a two-stage right-abuse analysis; the analysis takes place at a single stage.32 There are at least three ways to counter this argument. First, it is not clear that the institutional capacity of the Court is sufficient for the task. Is it really able to fine-tune and adapt EU legislation to the needs of the entire ‘European society’ through detailed context-sensitive interpretations, given its case load and the changing nature of secondary legislation, which increasingly includes detailed norms in areas such as intellectual property, company law, insolvency and so on?33 Secondly, to be effective in combating abuse the Court would have to push interpretation beyond its proper limits. The activity under scrutiny typically complies with the letter of the law, but nevertheless needs to be censured by the judiciary.34 Yet a decision by a court that goes against the wording of the law no longer amounts to mere interpretation, as recognised by the Court itself.35 Such a ruling does not simply select among the various meanings allowed by the language, but creates a new legal text. A doctrinal tool that allows the Court to go beyond interpretation may therefore be needed. Finally, the Union legislature has rejected such an approach in Article 35 of the 2004 Citizenship Directive, where abuse is conceptualised as a derogation, not as a factor influencing the scope of the right.36 Thus, the Advocate-General has not fundamentally undermined the case for a doctrine of abuse of rights in Union law. III. THE EMSLAND TEST

While there seems to be a need for a doctrine of abuse of rights in the Union legal order, it still has to be asked whether the actual test laid down by the European Court of Justice 29   See JHH Weiler, The Constitution of Europe (Cambridge, Cambridge University Press 1999) 16–39 for a discussion of the interplay between the judicial and the political. 30   See also A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of the European Court of Justice on Circumvention, Fraud and Abuses of Community Law’ in M Andenas and W-H Roth (eds), Services and Free Movement in EU Law (Oxford, Oxford University Press 2002) 276. 31   AG Tesauro in Kefalas (n 19 above), para [23]. 32   AG Tesauro in Kefalas (n 19 above), para [25]. 33   See J Snell, ‘European Courts and Intellectual Property: A Tale of Hercules, Zeus, and Cyclops’ (2004) 29 European Law Review 178. 34   Emsland (n 1 above) offers a perfect example. 35   See in the context of the duty of consistent interpretation for example, Case C-334/92 Teodoro Wagner Miret v Fondo de Garantía Salarial [1993] ECR I-6911, para [22] and Case C-105/03 Pupino [2005] ECR I-5285, para [47], where the Court expressly stated that the duty cannot serve as a basis for contra legem decisions. See generally A Barak, Purposive Interpretation in Law (Princeton, Princeton University Press 2005) 16–25. Consider also the maxim in claris non fit interpretatio. 36   European Parliament and Council Directive 2004/38 (n 2 above). See C Costello, ‘Metock: Free Movement and “Normal Family Life” in the Union’ (2009) 46 Common Market Law Review 587, 609.

224  Jukka Snell in Emsland37 introduces it in an appropriate form for the EU. The principal problem with the doctrine in all jurisdictions is its scope. How can it be contained?38 This is important for a number of reasons. First, the doctrine enhances the institutional competence of the judiciary at the expense of the legislature.39 It creates a problem for the separation of powers. It allows the judges to decide a case contrary to the wording of the law, which obviously amounts to a rather extreme version of judicial discretion.40 A routine use of abuse of rights would tilt the balance between the legislature and the judiciary in the favour of the latter, something which would clearly be a matter for concern in the EU where the courts already occupy an unusually strong institutional position.41 Further, a doctrine of abuse of rights has the potential to curtail individual liberties,42 and more broadly undermine the rule of law. Behaviour that ex ante seems to comply with the law may nevertheless ex post be declared unlawful. This reduces predictability and is in conflict with the basic premise of liberal systems that everything which is not prohibited is permitted.43 In the Union, there is a final reason for ensuring that the doctrine of abuse of rights is strictly curtailed. The finding of abuse has to be made on a case-by-case basis, and is by nature a fact-specific exercise. As a result, the task will mostly fall on national courts. The danger is that different national courts may well come to very different conclusions on abuse, perhaps reflecting their domestic experience with the notion. This threatens the uniformity of the EU legal order, the importance of which has been emphasised by the Court of Justice time and again.44 It may also undermine the effectiveness of Union law.45 A national court faced with an exercise of a new and unfamiliar Union right that upsets the traditional ways of doing things may well be inclined to find abuse too readily.46 How can the doctrine then be contained? At first sight, the most natural method is to create a test that sets out the specific conditions that must be fulfilled before an abuse of rights can be established. This is what the Court sought to do in Emsland. It laid down a two-part test, with the first, objective, part focusing on the purpose of the right47 and the second, subjective, part on the intention of the party. Thus: A finding of an abuse requires, first, a combination of objective circumstances in which, despite formal observance of the conditions laid down by the [Union] rules, the purpose of those rules has not been achieved. It requires, second, a subjective element consisting in the intention to obtain an advantage from the [Union] rules by creating artificially the conditions laid down for obtaining it.48   Emsland (n 1 above).   See for example, Catala and Weir (n 6 above) 227. 39   See Schammo (n 24 above) 372–73. 40   Lauterpacht (n 20 above) refers to ‘judicial legislation’ at 162, but continues at 165: ‘The power to apply some such principle as that embodied in the prohibition of abuse of rights must exist in the background in any system of administration of justice in which courts are not purely mechanical agencies.’ 41   See for example, R Herzog and L Gerken, ‘Stop the European Court of Justice’ in euobserver.com (10 September 2008) for a rather extreme expression of the concerns. 42   See HC Gutteridge, ‘Abuse of Rights’ (1933–35) 5 Cambridge Law Journal 22, 42–45. 43   See generally, Taggart (n 8 above) 155–66. 44   See for example, Case 66/80 SpA International Chemical Corporation v Amministrazione delle finanze dello Stato [1981] ECR 1191, para [11]. For discussion, see Snell (n 33 above) 185–86. 45   The Court is clearly aware of this. See for example, Emsland (n 1 above), para [54]. 46   The chapter by Panos Koutrakos offers a more optimistic view (n 5 above). 47   A similar test was famously put forward in France by Professor Josserand and is discussed for example in Catala and Weir (n 6 above) 227–30. 48   Emsland (n 1 above), paras [52–53]. 37 38

The Notion of and a General Test for Abuse of Rights  225 The first part of the test has been uncontroversial. It fits well with the teleological method of interpretation familiar to EU lawyers,49 and reserves an important role to the Court, which at least initially ought to get the opportunity to pronounce on the purpose of a given rule.50 It is easier to apply in the Union than in many national systems, given that the objectives of the Treaties and secondary legislation are generally outlined in the first provisions or the preamble. Nevertheless, there are some instances where it is likely to lead to considerable difficulties. First, sometimes the purpose of a right may be fiercely contested, or it may have a number of different purposes.51 Just to give one example, it is by no means clear whether the aim of EU competition law is the promotion of consumer welfare or perhaps total welfare, the creation of the Internal Market, or the protection of economic freedom.52 This may render the test unpredictable and difficult to apply. Secondly, the purpose of some rights is the right itself.53 This is the case for example for some classical human rights.54 The idea is to provide a legally protected sphere of freedom, not to achieve some immediate social goal.55 As shown by anti-abuse provisions such as Article 17 of the European Convention on Human Rights (ECHR) and Article 54 of the Charter of Fundamental Rights, any abuse of such a right is likely to take place not because the right is used against its purpose, but because the use of the right excessively impinges on other legally recognised interests, given the circumstances.56 However, the main weakness of the objective element is that in isolation it would not contain the notion of abuse of rights but would allow it to be invoked in a large number of cases where rights are utilised in an unorthodox fashion or the purpose of a right is contested. On its own, it would appear incapable of reining in the doctrine.57 The second part of the test, focusing on the subjective intent, fits well with the Court’s referral in General Milk Products and Emsland itself to ‘bona fide commercial transactions’ 49   For an overview, see P Craig and G de Búrca, EU Law: Text, Cases, and Materials, 4th edn (Oxford, Oxford University Press 2007) 72–76. 50   See Koutrakos (n 5 above). 51   Majone (n 28 above) argues (for example, at 107) that EU policies in general pursue several objectives simultaneously. 52   See G Monti, EC Competition Law (Cambridge, Cambridge University Press 2007) in particular, 20–52 for an extended discussion. See Case T-168/01 GlaxoSmithKline Services v Commission [2006] ECR II-2969 and Joined Cases C-501, 513, 515, and 519/06 P GlaxoSmithKline Services Unlimited et al v Commission et al, [2009] ECR I-9291, for judicial disagreement. 53   See also D Anderson, ‘Abuse of Rights’ [2006] Judicial Review 348, 354, who argues that this is a more general feature of the English legal culture. 54   A Gambaro, ‘Abuse of Rights in Civil Law Tradition’ (1995) 4 European Review of Private Law 561, 570 makes this point in the context of property rights. 55   Of course it could be countered that even freedoms are granted for some ultimate purpose, such as human dignity or democracy. Whilst true, this may be unhelpful, as the search for the ultimate purpose is bound to lead to abstract discussion that is far removed from the facts of a case. For example, is the purpose of the right to property to protect human dignity or to offer a reward or an incentive for investment, and what are the implications of this? Further, such an ultimate purpose is more naturally used to define the scope of the freedom. For example, the purpose of the freedom of speech may well determine whether something can be classified as protected speech in the first place. By contrast, it would be more unusual for a court to say that something is covered by a freedom but goes against the purpose of the freedom. As a matter of logic, is a freedom that can only be used for certain defined purposes really a freedom? 56  Art 17 ECHR on prohibition of abuse of rights provides that ‘[n]othing in this Convention may be interpreted as implying for any State, group or person any right to engage in any activity or perform any act aimed at the destruction of any of the rights and freedoms set forth herein or at their limitation to a greater extent than is provided for in the Convention’. Art 54 of the Charter of Fundamental Rights is phrased virtually identically. 57   AG Poiares Maduro emphasised the need for restraint in Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641, para [29].

226  Jukka Snell as opposed to attempts ‘wrongfully to benefit’ from the financial support offered by the EU.58 A subjective element has appeared in many national abuse of rights doctrines and would certainly help to contain the notion also in the Union, directing it to situations of ‘real abuse’.59 Nevertheless, it has been much more controversial than the objective part. Advocate-General Lenz argued against such a subjective element in his Opinion in TV10,60 claiming that it could not be employed in the case of corporations, while in Akrich Advocate-General Geelhoed stated that ‘subjective criteria serve no purpose . . . since [they] and thus in particular the aim of those concerned may readily be subject to manipulation’.61 Neither objection is particularly convincing. The law is quite prepared to consider the intent of corporate bodies in a variety of circumstances. When construing agreements, courts in most jurisdictions examine the intent of the parties even if they are companies, just like the intent of the legislature is used in statutory interpretation, despite the fact that legislatures are not individuals. In the case of EU competition law, the intent of undertakings is taken into account inter alia when determining whether below-cost pricing constitutes an abuse of a dominant position in violation of Article 102 TFEU [ex Article 82 EC]62 and in setting the level of fines,63 and is a relevant factor also in many other jurisdictions.64 More generally, criminal liability is increasingly being used as a sanction against delinquent corporate behaviour both in the EU and elsewhere.65 Subjective intent is obviously difficult to determine, but the law is generally quite prepared to infer intent from objective evidence. Judges and juries everywhere have proven willing to convict in the absence of a confession on the basis of evidence that points to the requisite mens rea. Indeed, the focus on the artificial nature of the transaction can be understood as such objective evidence of intent. If the only possible reason for a transaction is to obtain an advantage from EU rules, this has to be the real reason for it. In fact, it is almost tautological to require an artificial arrangement and an intention to take advantage of EU rules. If the artificial arrangement is found, the intent is overwhelmingly likely to be there as well.66 Nevertheless, there are at least four issues relating to the subjective element that do need to be noted. First, the way this part of the test is formulated, namely the requirement 58   Case C-8/92 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas [1993] ECR I-779, para [21], which was quoted in Emsland (n 1 above), para [51]. 59   D Weber, ‘Casenote on Emsland’ (2004) 31 Legal Issues of European Economic Integration 43, 52. 60   Case C-23/93 TV10 v Commissariaat voor de Media [1994] ECR I-4795. 61   AG Geelhoed in Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607, paras [173–74]. 62   Case C-62/86 AKZO Chemie BV v Commission [1991] ECR I-3359, para [72]. See T Eilmansberger, ‘How to Distinguish Good from Bad Competition under Article 82 EC: In Search of Clearer and More Coherent Standards for Anti-Competitive Abuses’ (2005) 42 Common Market Law Review 129 in particular, 147–53 and 170–77 for a more general argument that the relevant intent is a decisive element for establishing ‘a market structure abuse’. 63   See Commission (EC) Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003 [2006] OJ C210/02, in particular paras [1] and [29]. Fines are only imposed on intentional or negligent infringements, and the level of fines may be reduced in the case of mere negligence. 64  International Competition Network, ‘Report on Predatory Pricing’ (2008) 24–26, available at www. internationalcompetitionnetwork.org/uploads/library/doc354.pdf. 65  See for example, Commission (EC) ‘Proposal for a Directive of the European Parliament and of the Council on the protection of the environment through criminal law’ COM (2007) 51 final, and in particular the accompanying ‘Impact Assessment’ SEC (2007) 160, 36. For a general discussion, see C Wells, Corporations and Criminal Responsibility, 2nd edn (Oxford, Oxford University Press 2001) in particular, 78–81. Within the last few years, such liability has been proposed, introduced, or strengthened—for example, in Luxembourg, Romania, Switzerland and the UK. 66   See also AG Poiares Maduro in Halifax (n 2 above), para [71].

The Notion of and a General Test for Abuse of Rights  227 of ‘the intention to obtain an advantage from [Union] rules by creating artificially the conditions laid down for obtaining it’, is not the appropriate subjective element for all abuse of rights cases. For example, if in issue is an abuse of a property right or the legal process,67 the intention to harm rather than the intention to gain an advantage might prove to be the appropriate criterion.68 Secondly, while it is clear that an intention to harm, which has featured in some national doctrines, is blameworthy in itself, the same cannot be said of an intention to obtain an advantage from rules. It is entirely legitimate to seek to plan one’s affairs in the light of the applicable law. This may change if the only reason for an activity is to obtain a regulatory benefit and, consequently, wholly artificial arrangements are employed, as in Emsland.69 However, actions often have more than one aim. For example, a transaction may be motivated predominantly by a desire to avoid taxes but also by other considerations. This has already created some difficulties for the Court. It seems that if the aim is ‘principally’ to obtain an advantage, an abuse may be found.70 This is a fairly low threshold and not an easy one to apply.71 Thirdly, the subjective element means that in preliminary reference proceedings the Court of Justice loses some of its control as the referring national court has the sole jurisdiction to make the relevant findings of fact, while the Court of Justice interprets matters such as the purpose of the relevant Union rule.72 In other words, in Article 267 TFEU [ex Article 234 EC] proceedings the Court cannot necessarily determine the actual result of a case if the finding of abuse depends on subjective elements.73 Finally, if the doctrine of abuse of rights is conceptualised as a principle of interpretation, the relevance of the subjective element becomes questionable. Accordingly, Advocate-General Poiares Maduro played down its significance in his Opinion in Halifax, where he suggested that the notion of abuse ought to be used to construe EU law.74 However, the problem with this conceptualisation is that 67   Vexatious litigation has been condemned as an abuse of a dominant position under Art 102 TFEU [ex Art 82 EC] if the proceedings ‘cannot reasonably be considered to be an attempt to assert the rights of the undertaking concerned and can therefore only serve to harass the opposing party’ in Case T-111/96 ITT Promedia NV v Commission [1998] ECR II-2937, paras [72–73]. 68   The intention to cause damage was the factor triggering the finding of abuse in Clément-Bayard (n 6 above) but might be considered too restrictive by some. See also the discussion of the ‘intention to escape tax normally due’ in M Lang and S Heidenbauer, ‘Wholly Artificial Arrangements’ in L Hinnekens and P Hinnekens (eds), A Vision of Taxes Within and Outside European Borders: Festschrift in Honour of Prof Dr Frans Vanistendael (The Hague, Kluwer Law International 2008) 609–10. 69   The Court referred to ‘sole purpose’ and ‘purely formal dispatch’ in Emsland (n 1 above), para [50]. Contrast the circumstances in the Opinion of AG Kokott in Case C-352/08 Modehuis A Zwijnenburg BV, Opinion of 16 July 2009, nyr, paras [40–47]. 70   See the discussion of ‘sole’, ‘essential’, and ‘principal’ aim in Case C-425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I-897, paras [40–45] and [62]. 71   Contrast the insistence on ‘wholly artificial arrangements’ in Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995, for example para [75] and AG Poiares Maduro in Halifax (n 2 above), para [89]. See for discussion the chapter of F Vanistendael, ‘Cadbury Schweppes and Abuse from an EU Tax Law Perspective’, ch 28 below. 72   See Snell (n 33 above) 187–93. 73  See for example AG Mengozzi in Case C-451/08 Helmut Müller GmbH v Bundesanstalt für Immobilienaufgaben, ECJ Opinion of 17 November 2009, nyr, para [106].This may not be a bad thing. See Koutrakos (n 5 above) and more generally G Davies, ‘Abstractness and Concreteness in the Preliminary Reference Procedure: Implications for the Division of Powers and Effective Market Regulation’ in N Nic Shuibhne (ed), Regulating the Internal Market (Cheltenham, Edward Elgar 2006) 210. 74   Halifax (n 2 above), paras [69–71]. See also V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law: Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205–06 and 221–24; M Lang, ‘Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective’, ch 30 below.

228  Jukka Snell it may either push interpretation beyond its proper limits,75 or alternatively deprive the doctrine of much of its utility by preventing contra legem decisions such as Emsland itself. An additional problem with the Emsland test is that it does not explain the Court’s case law very well. The ruling in Emsland was preceded by the judgment in Centros a year earlier and followed by Chen four years later.76 In neither case was abuse found, despite the fact that it could and has been argued that both parts of the Emsland test were fulfilled. In Centros two Danish citizens living in Denmark acquired an off-the-shelf company in the UK. The company had a share capital of £100 and did not conduct any business in the UK. The owners sought to register a branch in Denmark through which the actual trading would be carried out. The reason for the scheme was to avoid the strict Danish rules on minimum capital, which required a paid-up capital of approximately £20,000. As to the objective element of the Emsland test, it could be argued that the scheme ran against the purpose of the right of establishment, which according to the Court’s earlier and subsequent case law ‘involves the actual pursuit of an economic activity through a fixed establishment in another Member State for an indefinite period’.77 Here Centros Ltd was not pursuing any economic activities whatsoever in the UK. Equally, given the aim of circumventing the Danish rules by acquiring a UK letter-box company, the subjective part of the test was met. There was an intention to obtain an advantage of the EU rules on the right of establishment by creating artificially the relevant conditions.78 Indeed, had the sequence of events been reversed and the Emsland case decided first, it is conceivable that the Danish court would not have sent a reference to Luxembourg at all, but would simply have declared that an abuse of the right of establishment had taken place. In actual fact, the reference was of course made and the European Court of Justice, sitting as a full Court and following Advocate-General La Pergola, found that the right had not been abused. Chen involved the EU free movement rules as well. Mrs Chen, a Chinese national seeking to avoid the consequences of China’s ‘one child policy’, had travelled to Northern Ireland to give birth to her second child. In accordance with the Irish nationality law, her daughter was automatically given Irish citizenship by virtue of being born on the island. The parties agreed that Mrs Chen had gone to Northern Ireland to obtain the Irish citizenship for her child and to create the consequent Union law right to reside in the UK for herself. Again, the Emsland test could have been employed.79 The subjective element seemed to be there, as Mrs Chen had intended to take advantage of the EU citizenship rules by manufacturing an Irish nationality for her daughter. As to the objective element, the purpose of Article 21 TFEU [ex Article 18 EC] and the relevant directive did not appear to be achieved, not least because Mrs Chen and her child never actually moved between the Member States but stayed in the UK territory the whole time. Indeed, as   See text at n 35 above.   Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459; Case C-200/02 Kunqian Catherine Zhu, Man Lavette Chen v Secretary of State for the Home Department [2004] ECR I-9925. 77   Case C-221/89 The Queen v Secretary of State for Transport ex p Factortame Ltd et al [1991] ECR I-3905, para [20]; Cadbury Schweppes (n 71 above), para [54]. 78   See also Edwards and Farmer (n 74 above) 218–20; Schammo (n 24 above) 375. 79   See also Arnull (n 13 above) 531; de la Feria (n 3 above) 417; FG Jacobs, ‘Citizenship of the European Union: A Legal Analysis’ (2007) 13 European Law Journal 591, 596; B Kunoy, ‘A Union of National Citizens: The Origins of the Court’s Lack of Avant-Gardisme in the Chen Case’ (2006) 43 Common Market Law Review 179, 185. 75 76

The Notion of and a General Test for Abuse of Rights  229 pointed out by Professor Arnull,80 the behaviour seemed to fall within the anti-abuse provision of Directive 2004/38,81 which admittedly was not in force at the material time. According to Recital 28 of the Preamble, Member States should have the possibility to act against marriages of convenience or ‘any other form of relationships contracted for the sole purpose of enjoying the right of free movement and residence’. Nevertheless, the Full Court, following Advocate-General Tizzano in result, ruled that Mrs Chen and her daughter had the right to reside in the UK by virtue of the Union citizenship rules. Despite the fact that both elements of the abuse test seemed to be fulfilled in both Centros and Chen, the Court can be defended for not finding abuse. If there was a problem, it was not due to the EU free movement law, but rather the consequence of the permissive nature of the British rules on company formation and the Irish nationality law.82 These rules were operating in the normal way in both the cases.83 Any artificial arrangements were the result of the national laws that allowed or even encouraged them.84 Had the Court found an abuse of rights, it would in effect have condemned the permissive policies chosen by the national legislatures.85 Further, in neither case was the principal right holder behaving abusively. Neither Centros Ltd nor the infant daughter of Mrs Chen were acting improperly; the problematic activity took place earlier in the chain, when the Danish owners obtained the company without any economic activities in the UK to circumvent the Danish rules, and when the parents contrived to acquire the Irish nationality for their child, despite a lack of a genuine link to the country.86 It should also be noted that the Court has always been hesitant to interfere with rules relating to nationality, whether of companies or natural persons. The Daily Mail judgment, which denied companies the right to transfer their central management to another Member State while retaining their legal personality, is a significant break from the trend in the field of establishment.87 Equally, the Court has refrained from second-guessing Member State decisions on nationality,88 despite the fact that it has been adamant to maintain control of the other constitutive concepts of EU rights, such as ‘worker’.89 Unfortunately, the two-part Emsland test has insufficient room for these nuances, and therefore appears over-inclusive.

  Arnull (n 13 above) 531.   Art 35 of European Parliament and Council Directive 2004/38 (n 2 above). 82   See AG Tizzano in Chen (n 76 above), paras [124–25]. 83   The British government strongly supported Centros Ltd in its observations, and nobody challenged the grant of Irish nationality in Chen. 84   In any event, legal personality and nationality are formal concepts that can always be described, in a sense, as artificial. 85   See also KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 448. 86   See D Edward and N Nic Shuibhne, ‘Continuity and Change in the Law Relating to Services’ in Arnull et al (n 74 above) 243, 246–47. I am grateful to Miguel Poiares Maduro for pointing this out in respect of Chen. 87   Case 81/87 The Queen v Her Majesty’s Treasury & Commissioners of Inland Revenue ex p Daily Mail and General Trust plc [1988] ECR 5483. For criticism, see AG Poiares Maduro in Cartesio (n 57 above), paras [26–31]. The Court upheld Daily Mail, stating at para [109] that the nationality of companies and citizens is a matter of national law. 88   Case C-192/99 The Queen v Secretary of State for the Home Department ex p Manjit Kaur [2001] ECR I-1237. See Case C-135/08 Janko Rottmann v Freistaat Bayern, ECJ Opinion of 30 September 2009, nyr, for in-depth discussion. 89  See for example, Case 75/63 Mrs MKH Hoekstra (née Unger) v Bestuur der Bedrijfsvereniging voor Detailhandel en Ambachten (Administration of the Industrial Board for Retail Trades and Businesses) [1964] ECR 177. 80 81

230  Jukka Snell Altogether, the setting out of a precise test in Emsland is not without its dangers. First, it is very difficult to craft a test that is satisfactory and covers all the myriad ways that rights can be abused, as the discussion above has sought to demonstrate. Indeed there is something illogical in the law seeking to set out the criteria for contra legem decisions and to establish a formal test to counteract excessive formality.90 Secondly, a clearly defined test may encourage national courts to make decisions without referring cases to the Court of Justice under Article 267 TFEU [ex Article 234 EC].91 As long as the concept of abuse remained undefined, national courts could not claim that the issue was acte claire.92 By contrast, provided that the purpose of the Union rule is clear or has been clarified in earlier case law, national courts may now come to feel that the main problems are of factual nature and therefore a matter for them rather than for the Court. As a result, the test that should enhance the uniformity and effectiveness of EU law by ensuring that national courts follow a consistent approach might paradoxically undermine these values by depriving the Court of the opportunity to offer case-by-case guidance.93 Finally, and most importantly, creating a test may normalise the doctrine of abuse of rights and make its application seem routine as yet another two-stage test to be employed in the appropriate circumstances. By contrast, when faced with an amorphous concept of abuse that lacked a precise definition courts may have been more hesitant in its application, reserving it for the most exceptional of circumstances. This might have been preferable, given that the contra legem decision that a finding of an abuse of right entails should never be made as a matter of routine. Such a decision involves the violation of the fundamental principles and values underpinning the legal order and the rule of law, including the separation of powers, legal certainty and so on.94 IV. CONCLUSION

To sum up, a doctrine of abuse of rights may be cautiously welcomed in the Union legal order. It may be of use, given that the EU system is in many respects akin to the civil law, in particular as both grant wide-ranging rights, and given the difficulty and delay any amendment of Union law entails. Whether the two-stage Emsland test is the appropriate incarnation for the doctrine is more questionable. Both the objective and the subjective parts of the test are open to criticism and may not be able to deal with all the permutations that abuse of rights may take.95 The test may also be over-inclusive, as it would seem to cover cases where the Court has not found abuse. Most fundamentally, the desirability 90   Lauterpacht (n 20 above) argues at 162 that ‘[t]he determination of the point at which the exercise of a legal right has degenerated into abuse of a right is a question which cannot be decided by an abstract legislative rule, but only by the activity of courts drawing the line in each particular case’. 91   It should be noted that considerations relating to the case load of the Court of Justice do not seem decisive in the present context, as abuse of rights cases ought to be relatively few and far between. 92   Case 283/81 Srl CILFIT and Lanificio di Gavardo SpA v Ministry of Health [1982] ECR 3415. See also Kjellgren (n 30 above) 272–73. 93   See also de la Feria (n 3 above) 441; A Arnull, ‘What is a General Principle of EU Law?’, ch 2 above. Koutrakos is perhaps less concerned. 94   See G Palombella, ‘The Abuse of Rights and the Rule of Law’ in A Sajó (ed), Abuse: The Dark Side of Fundamental Rights (Utrecht, Eleven International Publishing 2006) 5 for a discussion from the jurisprudential perspective. 95   See also the suggestion by AG Kokott in Case C-168/08 Laszlo Hadadi (Hadady) v Csilla Marta Mesko, married name Hadadi (Hadady), ECJ Opinion of 12 March 2008, nyr, para [71] that the recourse to the doctrine should be more difficult in some areas of law than in others.

The Notion of and a General Test for Abuse of Rights  231 of the two-part test may be questioned. The crystallisation of the prohibition of abuse of rights to a two-stage test may normalise the doctrine and make its application seem routine. Yet the notion should be employed only in the most unusual of circumstances, given the extraordinary character of contra legem decisions. Ideally of course the doctrine of abuse of rights should be utilised extremely sparingly. Instead, the legislature and the courts should craft carefully circumscribed rights that contain their own limitations. Abusive attempts to invoke a right should generally be met with the finding that the situation is outside the scope of the right altogether, as in the case of a person seeking to rely on the free movement of workers but engaged in a purely marginal and ancillary activity.96 The less frequently the notion of abuse is employed the more successful has the Union legal system been, and under no circumstances should the doctrine be allowed to become a substitute for detailed legislation and case law.

96   Case 53/81 Levin v Staatssecretaris van Justitie [1982] ECR 1035, para [17]. See Sørensen (n 85 above) 427–30 for discussion.

16 Abuse of Law in EU Private Law: A (Re-)Construction from Fragments Axel Metzger*

A

s with most other principles of EU law, the principle of abuse of law did not enter the scene in a private law setting. Rather, it was cases concerning fundamental freedoms, agricultural levies, and later, corporate and tax law that introduced the concept to European law.1 A similar narrative could be recounted for good faith, unjustified enrichment, interest for late payments, damages and other fundamental concepts of primary relevance for private law. They were all brought up for the first time before the ECJ in administrative law, in staff cases or agricultural policy cases, or in the framework of Article 288 of the EC Treaty [now Article 340 TFEU].2 But even though introduced in areas which, from a continental perspective, would be classified as ‘public law’, some of them returned to their private law roots as principles of the later emerging European private law.3 This chapter tries to ascertain whether this is also the case for the doctrine of abuse of law. It starts by defining what is meant by ‘European private law’, ‘general principle of law’ and ‘abuse of law’ (Part I) before considering three areas of EU private law: intellectual property, civil procedure and contracts (Part II). The final section then summarises the results (Part III). I.  Preliminary Definitions of the Concepts Used

A.  What Is ‘European Private Law’? ‘European private law’ for many lawyers still seems to be, in large part, terra incognita. The conceptual uncertainties start with the notion of ‘European’ private law. For some,   *  I would like to thank Malek Barudi for research assistance, Michael Friedman for intense language editing and Professor Simon Whittaker for his rich and thoughtful comments. 1  See R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. 2  See A Metzger, Extra legem, intra ius: Allgemeine Rechtsgrundsätze im Europäischen Privatrecht (Tübingen, Mohr Siebeck 2009) 325. 3  See for example, the principle of ‘good faith’ which was introduced in an early public administration staff case (Joined Cases 43/59, 45/59 and 48/59 Eva von Lachmüller, Bernard Peuvrier, Roger Ehrhardt v Commission of the European Economic Community [1960] ECR 463) but was later also used for the assessment of private agreements on jurisdiction under Art 17 of the Brussels Convention on jurisdiction and the enforcement of judgments in civil and commercial matters of 27 September 1968 [1972] OJ L299/32 (Case 25/76 Galeries Segoura v Société Rahim Bonakdarian [1976] ECR 1851).

236  Axel Metzger the true European private law is vested in the common traditions4 or the common concepts to be found in today’s national private law systems.5 For others, only the EU private law (consisting of directives and regulations in the fields of consumer contracts, company law, insurance law, intellectual property, jurisdiction and enforcement, private international law, etc) and the case law handed down by the Court of Justice and the General Court [formerly the Court of First Instance] should be considered as the ‘existing’ European private law.6 A third group combines the two approaches,7 whereas again others conceptualise European private law as a multi-level system comprising EU law as well as national systems in their congruent and divergent parts.8 Although it is true that the multi-level approach seems best suited to draw a comprehensive picture of today’s European private law, the approach used here will be restricted to the EU private law in order to match the concept of the conference.9 B.  What Is a ‘General Principle of Law’? An even more difficult exercise is to define what is meant when talking about a ‘general principle of law’. It cannot be the task of this chapter to present a full-blown theory of the general principle of law.10 But it seems nevertheless necessary to raise the curtain for a moment and to reveal what conception of ‘general principle’ has been used when drafting this chapter. i.  A Procedural Definition of ‘General Principle of Law’ For the purpose of this chapter a very basic definition must suffice: a general principle of law is a legal standard that is derived from legal rules by way of inductive generalisation.11 The sources for this process of induction may be taken from the same legal system. This is the case when the European courts infer general principles from specific rules of the acquis communautaire.12 The same method is applied by the courts of civil law countries when using multiple rules of the civil codes to establish general principles.13 It can also be  See for example, H Coing, Europäisches Privatrecht (Munich, CH Beck 1985/1989).  This has been the approach of the influential ‘Commission on European Contract Law’, see O Lando and H Beale (eds), The Principles of European Contract Law, Parts I and II (The Hague, Kluwer Law International 1999) xxv.     6  This is the approach followed by the ‘Acquis-Group’, see: www.acquis-group.org.     7  See for example, J Basedow, U Blaurock, A Flessner, R Schulze and R Zimmermann, ‘Editorial’ [1993] Zeitschrift für Europäisches Privatrecht 1; E Hondius and M Storme, ‘Editorial’ [1993] European Private Law Review 1.     8  See for example, C Joerges, ‘The Impact of European Integration on Private Law: Reductionist Perceptions, True Conflicts and a New Constitutional Perspective’ (1997) 3 European Law Journal 378, 386.     9   For a comparative law perspective see the contribution of J Gordley, ‘The Abuse of Rights in the Civil Law Tradition’, ch 4 above. 10  See for more details on the following theoretical framework, Metzger (n 2 above) 11–108. 11  The definition is ‘procedural’ in the sense that it defines the general principle of law as the outcome of an inference from certain sources and not by any substantial criterion. See Metzger (n 2 above) 25–32. 12  See for example, the line of arguments in Case 17/74 Transocean Marine Paint v Commission [1974] ECR 1063, para [15] on the right to be heard. 13  See for example, the argument used by the Austrian Oberster Gerichtshof for the justification of the principle that long-term contracts may be terminated under extraordinary circumstances, OGH, 31 March 1966 [1967] Juristische Blätter 209; OGH, 22 May 1962, Handelsrechtliche Entscheidungen no 3178; OGH, 25 January 1968, Handelsrechtliche Entscheidungen no 6474; OGH, 10 October 1974, Handelsrechtliche Entscheidungen no 9323.     4     5

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  237 found in common law courts when general principles are inferred from single instances in the case law.14 But the sources for general principles may also be taken from other legal systems. The generalisation is then one over different legal systems. ‘General’ in this case refers to ‘internationally accepted’.15 This concept is at the core of the Court’s method of deriving general principles from the Member States’ legal systems.16 On the basis of this definition, two types of principles of EU law can be distinguished: those inferred from EU legislation and those inferred from the laws of the Member States.17 Recognising the inductive process of inferring principles from rules as the main characteristic of general principles of law does not mean that they can be derived from the underlying sources without making any policy choices. Legal theory has long since learned from epistemology that general principles can never be inferred from single instances without explaining why a certain class of cases should be treated equally according to a principle and why the instances should not be seen as mere exceptions.18 But accepting this inherent weakness of inductive inferences does not mean that it is superfluous to ask for the support of a principle in the sources. The recognition of a principle is different from a ‘free’ policy choice exactly because it is supported—at least to a certain extent—by the sources. ii.  ‘Legal Principles’ As a matter of fact, not every general principle one might derive from the rules has to be qualified automatically as ‘law’. Otherwise, every speculation made by scholarship with some support in the rules could claim to be recognised as binding law, a vision close to a nightmare, especially for common lawyers. Therefore one should accept that a principle derived from rules can only be considered as a truly ‘legal’ principle if it is recognised in legal practice—such recognition being shown either by parties who conform their behaviour to the norm or courts who enforce the principle through their judgments.19 Principles lacking any actual recognition may be thoughtful and reasonable suggestions by scholarship, but they have not (yet) entered the realm of law. Taken together with the requirements from the definition, a standard should therefore be considered as a ‘legal principle’ of EU law under two conditions: firstly, if it can be derived from the rules of positive law by way of induction which implies that it finds sufficient ‘institutional support’20 in the acquis communautaire or in the law of the 14  See for example, the speech of Lord Atkin in Donoghue v Stevenson [1932] AC 562, 580 using several precedents to establish a general tort of negligence. 15  This is the concept followed by Art 38 of the Statute of the International Court of Justice of 26 June 1945, 33 UNTS 993 (‘general principles of law recognized by civilized nations’). 16  See for example, the line of arguments in Case 155/79 AM & S v Commission [1982] ECR 1575 on the legal privilege. 17  See also TC Hartley, The Foundations of European Community Law, 5th edn (Oxford, Oxford University Press 2003) 133; JA Usher, General Principles of EC Law (London, Longman 1998) 7. 18  See Metzger (n 2 above) 52. 19  See RM Dworkin, Taking Rights Seriously (Cambridge, Cambridge University Press 1977) 40 (‘sense of appropriateness developed in the profession and the public over time’). See also J Basedow, ‘Die UNIDROITPrinzipien der internationalen Handelsverträge und das deutsche Recht’ in H Schack (ed), Gedächtnisschrift fürAlexander Lüderitz (Munich, CH Beck 2000) 1, 5. 20   It is the dominant position in legal theory that only principles with some support in the sources can be qualified as ‘legal’ principles. See Dworkin (n 19 above) 40; N MacCormick, Legal Reasoning and Legal Theory (Oxford, Clarendon Press 1978) 238; J Raz, ‘Legal Principles and the Limits of Law’ (1972) 81 Yale Law Journal 823, 848; R Sartorius, ‘Social Policy and Judicial Legislation’ (1971) 8 American Philosophical Quarterly 151,

238  Axel Metzger Member States and, secondly, if it is recognised as a European principle by legal practice, especially by the Court of Justice and the General Court but also by the national courts applying EU law. iii.  Functions of General Principles of Law General principles of law typically fulfil auxiliary functions with regard to legislative rules or case law.21 They serve secundum legem as criteria for interpreting the existing rules, they may be applied for gap-filling or other praeter legem adjudication and, most controversially, they may serve as a yardstick for the correction of existing rules by overruling or by contra legem decisions.22 All three functions can be found in EU legal practice, especially in the Court’s case law. The Court refers to principles to interpret Treaties, Regulations and Directives.23 They are invoked to fill the numerous gaps in the still very ‘open texture’ of the acquis communautaire.24 And they serve as tools for the correction of EU and Member State legislation.25 Irrespective of whether they are used for interpretation, gap-filling or correction, general principles of law cannot be applied in an ‘all-or-nothing’ fashion.26 Principles always have to be balanced with other principles or arguments. They have a ‘dimension of weight’ as Dworkin has put it in his famous work on principles.27 One can explain this feature of principles by the inductive method applied when deriving principles from rules.28 The validity of rules for special cases cannot ensure that a generalised form of this standard has to be applied in all unprovided-for cases. This is another lesson the law should learn from epistemology when dealing with inductive inferences: one might observe a high number of swans all being white but one still cannot predict with certainty whether the next swan to come along will also be white.29

154. But see J Esser, Grundsatz und Norm in der richterlichen Fortbildung des Privatrechts (Tübingen, Mohr Siebeck 1956) 132 and 201. 21   In the classical language of J Boulanger: ‘Dans une législation codifiée, les principes sont l’apanage du législateur.’ J Boulanger, ‘Principes généraux du droit’ in Le droit privé français au milieu du XXe siècle—Études offertes à Georges Ripert (Paris, LGDJ 1950) 51, 63. 22  The distinction of ‘secundum, praeter, contra legem’ is often used in French legal theory. See for example, B Oppetit, ‘Les “principes généraux” dans la jurisprudence de cassation’ (1989) 5 JCP Cahiers de droit de l’entreprise 14, 15; F Terré, Introduction générale au droit, 6th edn (Paris, Dalloz 2003) 272. 23  See for example, Case 61/81 Commission v United Kingdom [1982] ECR 2601. See also J Bengoetxea, The Legal Reasoning of the European Court of Justice (Oxford, Clarendon Press 1993) 226, 255; A Bredimas, Methods of Interpretation and Community Law (Amsterdam, North-Holland Publishing Company 1978) 126; T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press 2006) 29, 51. 24  See for example, Case 17/74 Transocean Marine Paint v Commission [1974] ECR 1063. See also Bengoetxea (n 23 above) 227; U Everling, ‘Richterliche Rechtsfortbildung in der Europäischen Gemeinschaft’ [2000] Juristenzeitung 217; M Herdegen, ‘The Origins and Development of the General Principles of Law’ in U Bernitz and J Nergelius (eds), General Principles of Community Law (The Hague, Kluwer Law International 2000) 3, 17; Tridimas (n 23 above) 17. 25  See for example, Case C-473/00 Cofidis v Fredout [2002] ECR I-10875. See also RE Papadopoulou, Principes généraux du droit et droit communautaire (Brussels, Bruylant 1996) 17; Tridimas (n 23 above) 31; Usher (n 17 above) 123. 26   Dworkin (n 19 above) 24. 27   Dworkin (n 19 above) 26. 28   Metzger (n 2 above) 52. 29  See Metzger (n 2 above) 36.

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  239 C.  What Constitutes ‘Abuse of Law’? It remains to be defined what conception of ‘abuse of law’ is used here. For the purpose of this chapter it seems most appropriate to take the Court’s case law as a starting point, not only because this will help to uncover common strands between EU private law and other areas of EU law, but also because the Court frequently uses general principles recognised in one area of law at a later point in other areas.30 Hence, one should expect that future references to abuse of law in EU private law will be justified by reference to the older cases on fundamental freedoms, agricultural policy, corporate or tax law. The most elaborate and influential definition of the abuse of law principle has been established in the Emsland-Stärke decision of 2000, an agricultural policy case in which the ECJ required two conditions for a finding of abuse of law: A finding of an abuse requires, first, a combination of objective circumstances in which, despite formal observance of the conditions laid down by the Community rules, the purpose of those rules has not been achieved. It requires, second, a subjective element consisting in the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it. The existence of that subjective element can be established, inter alia, by evidence of collusion between the Community exporter receiving the refunds and the importer of the goods in the non-member country.31

The definition takes up in its first prong the central element of the continental doctrine, according to which abuse of law requires the use of a legal position for a purpose contrary to the objective of the rule granting it. This objective concept was introduced by Josserand in modern French theory and has been influential in France and other continental countries.32 But the Court does not stop here. It combines this objective test with a more subjective element which is the ‘intention to obtain an advantage . . . by creating artificially the conditions laid down’. The allusion to the French idea of ‘intention de nuire’ is obvious33 but the Court indicates that objective criteria may suffice as evidence for the required ‘intention’, especially the artificial nature of the parties’ conduct. This pragmatic approach to the subjective element has only recently been underlined by the Court’s ruling in Halifax in 2006 in the field of tax law:34 In view of the foregoing considerations, it would appear that, in the sphere of VAT, an abusive practice can be found to exist only if, first, the transactions concerned, notwithstanding formal 30  See for example, the line of arguments in Joined Cases C-295/04 to C-298/04 Manfredi v Lloyd Adriatico Assicurazioni SpA (C-295/04), Antonio Cannito v Fondiaria Sai SpA (C-296/04) and Nicolò Tricarico (C297/04) and Pasqualina Murgolo (C-298/04) v Assitalia SpA [2006] ECR I-6619, paras [95–97] where the Court allowed the claim for compensation for loss of profits and for interest in a competition law case with reference to older administrative law cases. 31   Case 110/99 Emsland-Stärke v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, para [52]. 32  See L Josserand, De l’esprit des droits et de leur rélativité (Paris, Dalloz 1927) 368. A comprehensive survey of the current French practice and theory is provided by L Cadiet and P Tourneau, ‘Abus de droit’ in Encyclopédie Dalloz, Droit civil (Paris, Dalloz 2002). For Germany see the influential study by W Siebert, Verwirkung und Unzulässigkeit der Rechtsausübung (Marburg, Elwert 1934) 68, and more recently HP Haferkamp, Die heutige Rechtsmißbrauchslehre—Ergebnis nationalsozialistischen Rechtsdenkens? (Berlin, Nomos 1995) 152. 33  See Cadiet and Tourneau (n 32 above) 24; Terré (n 22 above) 395. 34  On the subjective element of the test see also, de la Feria (n 1 above) 395, 410, 423; KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 451, 456.

240  Axel Metzger application of the conditions laid down by the relevant provisions of the Sixth Directive and the national legislation transposing it, result in the accrual of a tax advantage the grant of which would be contrary to the purpose of those provisions. Second, it must also be apparent from a number of objective factors that the essential aim of the transactions concerned is to obtain a tax advantage. As the Advocate General observed in point 89 of his Opinion, the prohibition of abuse is not relevant where the economic activity carried out may have some explanation other than the mere attainment of tax advantages. (...) As regards the second element, whereby the transactions concerned must essentially seek to obtain a tax advantage, it must be borne in mind that it is the responsibility of the national court to determine the real substance and significance of the transactions concerned. In so doing, it may take account of the purely artificial nature of those transactions and the links of a legal, economic and/or personal nature between the operators involved in the scheme for reduction of the tax burden (see, to that effect, Emsland-Stärke, paragraph 58).35

According to Halifax, it is not a specific intention that has to be proved but the aim of the behaviour, and this can be done by taking into account objective factors such as the purely artificial nature of the actions taken. Admittedly, Emsland-Stärke and Halifax concerned cases from agricultural and tax law and not EC private law. Nevertheless, they established an abuse of law test of an overarching nature that may also be useful as a yardstick for other areas of law if isolated from its agricultural and tax law setting. II.  Abuse of Law and EU Private Law: Three Sketches

Having defined the essential concepts used, the stage appears to be set for consideration of the chapter’s main subject. This will be done by offering three sketches from specific fields of European private law, those being intellectual property, international civil procedure and contract law. For each field it will be analysed whether there is institutional support in the EU legislation for the principle of abuse of law and whether the principle has been recognised by European and national courts. It will not be the task of this chapter to analyse systemically whether such a principle could also be founded on the common traditions of most or at least some Member States, since other contributions to the present volume are devoted to this subject. Therefore, the method used should neither be seen as exhaustive nor should it be understood as a plea for pure ‘acquis principles’ nor as being against a comparative approach. A.  Abuse of Law in EU Intellectual Property Law The classical area of application for the principle of abuse of law in civil law countries is property law.36 Looking to European law, one might doubt at first glance whether there 35   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, paras [74, 75 and 81]. 36  See for a French perspective, Cadiet and Tourneau (n 32 above) 45; G Marty and P Raynaud, Droit civil: Introduction générale à l’étude du droit, 2nd edn (Paris, Sirey 1972) 170. For Germany see, H Fleischer, ‘Der Rechtsmißbrauch zwischen Gemeineuropäischem Privatrecht und Gemeinschaftsprivatrecht’ [2003] Juristenzeitung 865.

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  241 are any cases or rules in the acquis communautaire concerning property rights because harmonisation of national private laws has, at least so far, ignored the differences in the national property regimes of the Union’s Member States. But this is only true with regard to property in movables and real property.37 For intellectual property there is a remarkable body of EU legislation and case law.38 And it is of little surprise that both the EU legislator and the courts—although so far only national courts—have used the principle of abuse of law to prevent abusive practices with regard to the harmonised aspects of intellectual property. i.  Abusive Registration of Trade Marks The first reference to abuse of intellectual property rights in Community law appears to have been Article 3(2)(d) of Directive 2008/95/EC of the European Parliament (formerly Directive 89/104/EEC) and the Council of 22 October 2008 to approximate the laws of the Member States relating to trade marks (codified version), OJ L 299/25,39 which allows Member States to refuse the registration of a trade mark if ‘the application for registration of the trade mark was made in bad faith by the applicant’. The Community legislator asserted the concept of bad faith again in Article 51(1)(b) of Council Regulation (EC) No 207/2009 of 26 February 2009 on the Community trade mark (codified version), OJ L 78/1 (formerly Regulation 40/94).40 Although bad faith should not be equated with abuse of law,41 some of the cases classified under the concept of bad faith in the Trade Mark Directive could be qualified as abuse of law under the Court’s definition of the principle. Hence, as is often the case with general principles of law, the deviation in terminology should not prevent recognition of the common concept. Unfortunately, neither the Directive nor the Regulation specifies what exactly bad faith means. In addition, there is no case law of the Court of Justice or the General Court on the notion of ‘bad faith’ in European trade mark law. However, a considerable number of cases of the Member States’ courts have explored the concept and have developed a list of arguable grounds of ‘bad faith’, some of which bear strong resemblance with typical abuse of law cases. Two examples may highlight this special area of application of the principle.42 According to German legal practice, the ‘blocking’ of trade marks is held to be abusive and can be prevented under the bad faith provisions of the Trade Mark Directive. Blocking means the practice of registering a trade mark which the applicant does not intend to use himself in his business but which is supposed to prevent another from filing a similar trade mark or using the sign in the marketplace. The German Bundesgerichtshof has 37   But see Art 4 of the Directive 2000/35/EC of the European Parliament and of the Council of 29 June 2000 on combating late payment in commercial transactions [2000] OJ L200/35. 38  See for example, the survey by A Metzger and W Wurmnest, ‘Auf dem Weg zu einem Europäischen Sanktionenrecht des geistigen Eigentums?’ [2003] Zeitschrift für Urheber- und Medienrecht 922. 39   First Council Directive 2008/95/EC of the European Parliament (formerly Directive 89/104/EEC) and the Council of 22 October 2008 to approximate the laws of the Member States relating to trade marks (codified version), OJ L 299/25. 40  Council Regulation (EC) No 207/2009 of 26 February 2009 on the Community trade mark (codified version), OJ L 78/1 (formerly Regulation 40/94). 41  See from a comparative perspective, R Zimmermann and S Whittaker, Good Faith in European Contract Law (Cambridge, Cambridge University Press 2000) 694. 42  See also D Füllkrug, ‘Gedanken zur markenrechtlichen Einordnung der Spekulations- oder Sperrmarke’ [2006] Wettbewerb in Recht und Praxis 664; P Lange, Marken- und Kennzeichenrecht (Munich, CH Beck 2006) 223; S Middlemiss and J Phillips, ‘Bad Faith in European Trade Mark Law and Practice’ (2003) 25 European Intellectual Property Review 397.

242  Axel Metzger explicitly endorsed the principle of abuse of law prior to the implementation of the Trade Mark Directive43 and has upheld the doctrine after its implementation in the Classe E case as a specific category of bad faith registration.44 The second example is provided by the practice of the UK Trade Marks Registry, which holds a trade mark registration to be in bad faith if evidence shows that the applicant makes use of the registered sign in a manner other than as registered. In the case Betty’s Kitchen Coronation Street it was clear from the facts that the applicant had no intention to use the sign as registered but that his plan was only to use the words ‘Betty’s Kitchen’ and ‘Coronation Street’ separately. As concerns ‘Coronation Street’, the registry would never have granted a trade mark, since ‘Coronation Street’ is a well-known soap opera. Therefore the application was found to be a bad faith application.45 In both cases the formal requirements for a trade mark registration were met but the grant of the trade mark would have been contrary to the purpose of the Trade Mark Directive. In addition, one could argue that evidence had shown that the conditions for the registration of a trade mark had been fulfilled ‘artificially’ since the applicant did not have the intention to use the sign as registered. Thus both requirements from the abuse of law tests of Emsland-Stärke and Halifax had been met.46 Seen from the continental doctrine of abuse of law, the two cases could be classified as cases of abusive acquisition of property rights.47 Thus, one could argue that there is some ‘institutional support’ for this subdivision of the abuse of law principle in the current EU legislation on trade marks. One could also argue that the principle is applied by the national authorities in European trade mark law settings and therefore recognised in legal practice, even though the Court of justice and the General Court have not yet referred to the principle in trade mark cases. ii.  Abusive Enforcement of Intellectual Property Rights Having had a specific doctrine in trade mark law since the late 1980s, it was 2004 before the European legislator decided to expand the area of application of the abuse of law principle to all intellectual property rights through Directive 2004/48/EC on the enforcement of intellectual property rights.48 According to Article 2(1), the Enforcement Directive applies to ‘any infringement of intellectual property rights as provided for by Community law and/or by the national law of the Member State concerned’. Hence, the Directive is not only applicable to unitary EU rights (such as the Union’s trade mark) and to harmonised national rights but also to non-harmonised national intellectual property rights such as patents.49 It provides rules on remedies for infringement of intellectual property rights, especially on damages and injunctions. In addition, it offers a comprehensive set of 43   BGH, 24 February 1961 [1961] Gewerblicher Rechtsschutz und Urheberrecht 413; BGH, 23 March 1966 [1967] Gewerblicher Rechtsschutz und Urheberrecht 298; BGHZ 46, 130. 44  BGH, 23 November 2000 [2001] Gewerblicher Rechtsschutz und Urheberrecht 242. See also BGH, 9 October 1997 [1998] Gewerblicher Rechtsschutz und Urheberrecht 412. The concept also seems to be recog­ nised in the United Kingdom, see Middlemiss and Phillips (n 42 above) 399. 45  Trade Marks Registry, 14 October 1999, Reports of Patent, Designs and Trade Mark Cases 117 (2000) 825. 46  See Emsland-Stärke (n 31 above) and Halifax (n 35 above). 47  See for example, D Looschelders and D Olzen (eds), Staudinger Kommentar zum Bürgerlichen Gesetzbuch (Berlin, Sellier-de Gruyter 2005) § 242 para 240. 48   Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights [2004] OJ L195/16. 49   Patent law has been mostly spared by EU law with the exception of Directive 98/44/EC of the European Parliament and of the Council of 6 July 1998 on the legal protection of biotechnological inventions [1998] OJ L213/13.

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  243 procedural instruments, eg on the taking and preserving of evidence, and on provisional and protective measures. Article 3 stipulates a ‘general obligation’ for all kinds of remedies and procedural measures: Article 3: General Obligation 1.  Member States shall provide for the measures, procedures and remedies necessary to ensure the enforcement of the intellectual property rights covered by this Directive. ( . . . ) 2. Those measures, procedures and remedies shall also be effective, proportionate and dissuasive and shall be applied in such a manner as to avoid the creation of barriers to legitimate trade and to provide for safeguards against their abuse.

The clause at the end of Article 3(2) is taken from Article 41 of the TRIPS Agreement.50 Due to the broad area of application of the Directive, the abuse of law principle codified at the end of paragraph (2) is applicable to almost all remedies and procedural measures in European intellectual property law. It will be interesting to see in the coming years how the national and European courts will interpret the very open wording of the rule. Since most Member States have just recently implemented the Directive, there is currently no reported case law on Article 3. Nevertheless, one should consider abuse of law to be established as a general principle of EU intellectual property law as far as remedies are concerned. B.  Abuse of Law in EU Civil Procedure In civil law countries the area of civil procedure has, in addition to property law, always been one of the typical fields of application for the abuse of law principle.51 Therefore, it is by no means surprising that the European legislator and the Court rely on the principle when dealing with procedural law. In recent decades, the centre of interest of European civil procedure has been the Brussels Convention of 1968 on jurisdiction and the enforcement of judgments in civil and commercial matters52 which was transformed into the ‘Brussels I Regulation’ in 2001.53 The texts of the Convention and the Regulation and the interpretation given by the Court make several references to the abuse of law principle.54

50   Art 41(1) phrase 2 of the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS): ‘These procedures shall be applied in such a manner as to avoid the creation of barriers to legitimate trade and to provide for safeguards against their abuse.’ 51  See Cadiet and Tourneau (n 32 above) 113; Fleischer (n 36 above) 865, 866. See also Sørensen (n 34 above) 423, 435 (referring to misuse of the procedure of Art 230 EC [now Art 263 TFEU] according to Case 104/79 Foglia v Novello [1980] ECR 745). 52   Brussels Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters of 27 September 1968 [1972] OJ L299/32. 53  Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2001] OJ L12/1. 54  See also the contributions of A Briggs, ‘The Rejection of Abuse in International Civil Procedure’, ch 18 below; G Cuniberti, ‘The Discreet Influence of Abuse of Law in International Civil Procedure’, ch 19 below; and A Nuyts, ‘The Enforcement of Jurisdiction Agreements Further to Gasser and the Community Principle of Abuse of Rights’ in P de Vareilles-Sommières (ed), Forum Shopping in the European Judicial Area (Oxford, Hart Publishing 2007) 55.

244  Axel Metzger i. Abuse of Derived Jurisdiction under Article 6(2) Brussels I Regulation Article 6(2) of the Brussels Convention and of the Brussels I Regulation provides the clearest legislative affirmation of the abuse of law principle in the Brussels system. According to Article 6(2), initial proceedings against one party may be extended to join a third party in case of actions on warranty or guarantee. The typical case for the provision is the case of a defendant in an action for warranty seeking redress from his supplier and joining the claim against his supplier with the claim of his customer for breach of warranty. In this case, jurisdiction over the third party is granted at the forum of the initial proceedings even if there would be no jurisdiction over the third party under the general rules. However, Article 6(2) provides a reservation from this approach for cases in which the initial proceedings ‘were instituted solely with the object of removing him [the third party] from the jurisdiction of the court which would be competent in the case’. Hence, using the specific jurisdiction rule of Article 6(2) to bypass the general jurisdiction rules of the Brussels I Regulation is deemed abuse of law and is prohibited by the Regulation.55 In this scenario, the two elements of the abuse of law test from Emsland-Stärke and Halifax are met.56 The specific jurisdiction rule is being used against the purpose of the provision; the court for the initial proceedings is chosen artificially to remove the third party from the forum of his general jurisdiction. Seen from this perspective, Article 6(2) may be read as providing clear legislative support for the abuse of law principle in EU civil procedure. ii.  Artificial Place of Performance and Article 5(1) Brussels I Regulation A second area of application for the abuse of law principle may be found in the field of special jurisdiction for contractual obligations under Article 5(1)(a) of the Brussels I Regulation. Under this provision, a person may be sued in matters relating to a contract ‘in the courts for the place of performance of the obligation in question’. Although it is generally admitted that the parties to a contract may specify the place of performance in their contract and that this choice must also be respected with regard to jurisdiction, the ECJ decided in 1997 in the case Mainschiffahrts-Genossenschaft that the parties may not choose a wholly artificial place without actual connection with the real subject-matter of the contract which was never meant to be the place of performance but which was only intended to determine the place of jurisdiction.57 In such case, the agreement has to comply with the specific requirements of choice of court agreements of Article 23 of the Regulation: Thus, where there is such an agreement, there is not only no direct connection between the dispute and the courts called upon to determine it, but there is also circumvention of Article 17 [Article 23 of the Regulation], which, whilst providing for exclusive jurisdiction by dispensing with any objective connection between the relationship in dispute and the court designated . . . 55  See U Magnus, P Mankowski and H Muir Watt (eds), Brussels I Regulation (Munich, Sellier European Law 2007) Art 6 para 37; T Rauscher and S Leible, Europäisches Zivilprozessrecht, 2nd edn (Munich, Recht und Wirtschaft 2006) Art 6 para 21. According to a recent decision of the ECJ, Case C-98/06 Freeport v Arnoldsson [2007] ECR I-8319, paras [51–54], the abuse-proviso of Art 6(2) is not applicable under Art 6(1). This could have been understood differently under the rulings of the ECJ in Case 189/87 Kalfelis v Schröder [1988] ECR 5565, paras [8, 9]; and Case C-103/05 Reisch v Kiesel [2006] ECR I-6827, para [32]. In this sense Nuyts (n 54 above) 65. 56  See Emsland-Stärke (n 31 above) and Halifax (n 35 above). 57   Case C-106/95 Mainschiffahrts-Genossenschaft (MSG) v Les Gravières Rhénanes [1997] ECR I-911.

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  245 requires, for that very reason, compliance with the strict requirements as to form which it sets out.58

Although the Court emphasised that allowing such agreements would amount to a circumvention of Article 23 of the Regulation, the case could also be construed as an abuse of Article 5(1)(a). The right to determine the place of performance is being abused to evade the specific requirements of Article 23.59 Again, both elements of the EmslandStärke and Halifax test60 are fulfilled so that Mainschiffahrts-Genossenschaft may be construed as recognising the abuse of law principle in EU civil procedure. iii.  Abuse of Exclusive Jurisdiction under Article 22(5) Brussels I Regulation Another example for the recognition of the abuse of law principle in EU civil procedure is provided by the decision of the ECJ in AS Autoteile v Malhé.61 In that case, the question arose whether the defendant in an enforcement procedure under Article 22(5) Brussels I Regulation may plead set-off with a claim over which the Court could not independently assert jurisdiction. The Court denied such a defence, relying explicitly on the abuse of law principle: It follows from the specificity of the connection required by Article 16 [Article 22 of the Regulation] that a party cannot make use of the jurisdiction conferred by Article 16(5) [Article 22(5) of the Regulation] on the courts of the place of enforcement in order to bring before those courts a dispute which falls within the jurisdiction of the courts of another contracting state under Article 2. The use for such a purpose of the application to oppose enforcement is contrary to the division of jurisdiction which the Convention intended to establish between the court of the defendant’s domicile and the court of the place of enforcement. In this case, since the German courts have already held that they have no jurisdiction over the claim relied on as a set-off, the use of that claim in order to oppose the enforcement of an order for the costs incurred in the same proceedings amounts to a clear abuse of the process of the part of the plaintiff for the purpose of obtaining indirectly from the German courts a decision regarding a claim over which those courts have no jurisdiction under the Convention.62

Once again the doctrine of abuse of law was used to restrict the exercise of a procedural right granted under the Convention that would contradict the purpose of this right. iv.  Abuse of Procedure and the Principle of ‘Mutual Trust’ However, it is not in every instance that opportunistic behaviour amounts to abuse of law under the Brussels I Regulation. Countervailing principles and arguments may justify a different assessment of behaviour that might at first blush appear to be abusive. This has been demonstrated by the ECJ judgments in Gasser and Turner v Grovit, which both invoked the principle of ‘mutual trust’ underlying the Brussels Convention and declined to find an abuse of law.   Mainschiffahrts-Genossenschaft (n 57 above), para [34].  See Rauscher and S Leible (n 55 above) Art 5 para 44a; Nuyts (n 54 above) 66. But see A Briggs and P Rees, Civil Jurisdiction and Judgments, 4th edn (London, LLP Professional Publishing 2005) 171, fn 804. 60  See Emsland-Stärke (n 31 above) and Halifax (n 35 above). 61   Case 220/84 AS Autoteile Service GmbH v Pierre Malhé [1985] ECR 2267. See also Nuyts (n 54 above) 66. 62   AS Autoteile (n 61 above), para [17]. 58 59

246  Axel Metzger In Gasser,63 one party filed suit in a court of a Member State (Italy) which it knew to proceed more slowly than the courts of the Member State the other side intended to select for the same cause of action (Austria). The question arose whether an exception from the lis pendens rule of Article 27 of the Regulation should be made in cases where the sole intention of the first proceedings was to block proceedings before the faster-working, second seised court. The United Kingdom had submitted exactly that position to the ECJ,64 but the Court did not follow the argument, holding that the principle of mutual trust did not allow any exceptions from the lis pendens rule even if the proceedings before the court first seised took excessively long.65 A similar argument was used with regard to anti-suit injunctions in the case of Turner v Grovit, decided by the ECJ in 2004.66 In that case, a British employee had filed suit against his former employer, a Spanish company, before an English court. After this suit was brought, the employer started proceedings on the same cause of action before a Spanish court. The employee asked the English courts to issue an injunction against the employer, restraining him from pursuing the proceedings commenced in Spain. Such an ‘anti-suit injunction’ was granted. The employer appealed to the House of Lords, claiming that the English courts did not have the power to issue restraining orders preventing the pursuit of proceedings in other EC Member States. The House of Lords referred the case to the ECJ where the employee and the United Kingdom submitted that such an injunction would not interfere with the Brussels I Regulation since it would only prevent an abuse of procedure. The ECJ again took the opposite view, holding that the principle of mutual trust would not allow the courts of one Member State to review the jurisdiction of another Member State. In both Gasser and Turner v Grovit, opportunistic behaviour was not assessed as abusive because of the countervailing argument that courts should in no case decide on the jurisdiction of other courts under the Brussels I Convention. This makes it very clear that abuse of law in procedural matters is not an ‘all-or-nothing’ rule but has to be balanced with other legal principles.67 C.  Abuse of Law in EU Contract Law (and EU Law of Obligations) Having some support for the abuse of law principle in the fields of intellectual property and procedural law, one might expect a similar picture in EU contract law. The European legislator has been very active in this area since the 1980s, especially in the field of consumer contracts, and one could well imagine cases in which the use of contractual remedies such as the consumer’s right to withdraw from the contract could be seen as abusive. But the principle has until now found little support in EU legislation and has only rarely been invoked in European and national courts with regard to contractual rights and remedies rooted in the acquis communautaire. Moreover, the rare examples do, for   Case C-116/02 Gasser v MISAT [2003] ECR I-14693.   Gasser (n 63 above), paras [61–64]. 65   Gasser (n 63 above), paras [70–73]. 66   Case C-159/02 Turner v Grovit [2004] ECR I-3565. See also, A Dutta and C Heinze, ‘Prozessführungsverbote im englischen und europäischen Zivilverfahrensrecht’ [2005] Zeitschrift für Europäisches Privatrecht 428; C Hare, ‘A Lack of Restraint in Europe’ (2004) 63 Cambridge Law Journal 570 . 67   For a similar line of argument in the Court’s corporate and tax law cases see, de la Feria (n 1 above) 395, 405 and 423 (‘legitimate circumvention’ or ‘planning without abuse’) and Sørensen (n 34 above) 423, 436. 63 64

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  247 most part, fail to match the concept of abuse of law as it has emerged in the Court’s case law in other areas of law. i.  Abuse of Law and the Unfair Terms Directive When talking about abuse of law in EU contract law, one misconception should be avoided from the outset. ‘Unfair terms’ under the Directive 93/13/EEC on unfair terms in consumer contracts68 do not present a case of abuse of law even though the French language version reads ‘clause abusive’ and the German version ‘missbräuchliche Klauseln’. The aim of the Directive is to protect consumers from the unfair behaviour of professionals using standard terms. These professionals are not abusing a legal position— apart from their contractual freedom—but rather the factual setting of non-negotiated and lengthy standard terms used against consumers who typically do not read a single word of what they accept as the terms of the contract. The behaviour of the professional using standard terms may indeed conflict with the provisions of the Directive and the more general concept of good faith in contractual dealings.69 It does not, however, fall under the concept of abuse of law stricto sensu as it has been established in the ECJ’s decisions of Emsland-Stärke and Halifax.70 ii.  Consumers Disguised as Businessmen: Abuse of EU Consumer Protection But what about abuse of consumer protection rights guaranteed by EU legislation? Although there seems to be no case law of the Court applying the abuse of law principle to consumer contract rules, the German Bundesgerichtshof has nonetheless entered the arena and restricted consumer protection in case of abusive practices of the consumer. In the landmark decision from December 2004,71 the court had to rule whether a consumer may rely on the specific remedies of the Directive 1999/44/EC on the sale of consumer goods72 if he has pretended to be a professional when negotiating the sales contract. In that case, the seller had made clear that he would only sell the used car, a Fiat Barchetta, to a professional because he wanted to exclude any warranty in his standard terms. The buyer pretended to be a businessman and bought the car under the terms of the seller. Later, he discovered that the car had a technical defect and asked for rescission of the contract, pleading for invalidity of the exclusion of warranty under the consumer protection rules of section 475 of the German Civil Code which is the transposition of Article 7 of the Directive.

68  Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts (Unfair Terms Directive) [1993] OJ L95/29. 69   In France, unfair terms are seen as a sub-category of abuse of law, see for example Cadiet and Tourneau (n 32 above) 9, 37. The reason for this broad concept seems to be that good faith as a general principle is traditionally less developed in France than in other countries, eg Germany. Therefore French law is using a broader concept of abuse of law to cover cases that would fall under the principle of good faith elsewhere. See Zimmermann and Whittaker (n 41 above) 695. 70  See Emsland-Stärke (n 31 above) and Halifax (n 35 above). In Germany, unfair terms are not put under the conceptual umbrella of abuse of law, see for example, Fleischer (n 36 above) 865, 871; FJ Säcker, R Rixecker and GH Roth (eds), Rebmann Münchener Kommentar zum BGB, 5th edn (Munich, CH Beck 2007) § 242 para 211. 71   BGH, 22 December 2004 [2005] Neue Juristische Wochenschrift 1045. 72   Directive 1999/44/EC of the European Parliament and of the Council of 25 May 1999 on certain aspects of the sale of consumer goods and associated guarantees [1999] OJ L171/12.

248  Axel Metzger The Bundesgerichtshof denied rescission. Without addressing the general question of whether the ‘consumer’ had to be qualified according to an objective or a subjective test, the court stated that it would be against the principle of good faith to grant consumer protection to a buyer who pretends to be professional during the negotiations. Protecting the consumer would be against the doctrine of venire contra factum proprium, a doctrine which is seen in Germany as a subdivision of the abuse of law principle.73 But the court did not content itself with the application of the German principle of good faith or abuse of law. It went on to conclude that the principle of good faith is also part of EC law and that under the European standard no different result could be reached.74 Although the Bundesgerichtshof explicitly emphasised the abusive character of the consumer’s behaviour, the case is not squarely covered by the concept of abuse presented by Emsland-Stärke and Halifax.75 One might consider the consumer’s conduct as against the purpose of consumer protection law. Yet one could hardly say that the consumer had artificially fulfilled the statutory requirements to obtain the advantages of EC consumer law since the case was not about the abusive acquisition of a legal status but about the abusive exercise of it. In this respect, the case has more resemblance to ECJ abuse of law cases like Van Binsbergen76 or the Broadcasting cases.77 iii.  Abusive Assignment of an Insolvent Debtor Even though the Court has not yet applied the abuse of law principle to EU consumer protection legislation, there are cases in which the Court has made use of it in the law of obligations at large. One example is provided by the DEKA case decided in 1983.78 In DEKA, the Community was sued for damages under Article 215 of the EEC Treaty [later Article 288 EC and now Article 340 TFEU] and wanted to set off a claim for recovery of unlawfully paid subsidies which it had against the plaintiff. Before the suit was filed, the insolvent plaintiff had assigned the claim against the Community to another company which, after informing the Community as to the assignment, pleaded that the Community could not set off its claim for recovery against the assignee. The Court dismissed the argument by applying the principle of abuse of law: In the case of an insolvent trader, the assignment of the trader’s claim against the Community authorities to a third party may, depending on the circumstances, amount to an abusive transaction of such a nature that it must be regarded as invalid as against those authorities. According to a general principle of law common to the laws of the Member States, certain acts of a debtor to the detriment of the interests of creditors and, in particular, those which are of a fraudulent nature vis-à-vis creditors, either cannot be pleaded against the creditors or may be set aside under procedures specifically prescribed for that purpose.79  See for example, Looschelders et al (n 47 above) 286; Säcker et al (n 70 above) § 242, para 255.  The Bundesgerichtshof was on the right track with regard to EC consumer law as shown by the decision of the ECJ in Case C-464/01 Gruber v Bay Wa AG [2005] ECR I-439, which was decided only four weeks later and which used the same line of arguments for the denial of consumer protection under the rules of Arts 13 to 15 of the Brussels Convention on Jurisdiction and Enforcement (n 52 above). 75  See Emsland-Stärke (n 31 above) and Halifax (n 35 above). 76  See Case 33/74 Johannes Henricus Maria van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299. 77  See for example, Case C-148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487. 78  See Case 250/78 DEKA Getreideprodukte GmbH & Co KG, iL (formerly Firma Contifex Getreideprodukte GmbH & Co KG) v European Economic Community [1983] ECR 421. 79   DEKA (n 78 above) Summary, para [2]. 73 74

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  249 As a result, the right to assign claims—which was governed by Community law—was restricted by reference to the principle of abuse of law, according to which the assignment of a claim may not be pleaded against the creditors if it was executed to the detriment of the creditors. But the concept of abuse of law referred to in DEKA, once again, does not match the concept established by Emsland-Stärke and Halifax. It is hard to see any statutory or contractual right acquired or exercised by formal observance of the conditions laid down by the rules of EU law but which is against the purpose of those rules. The only ‘right’ that may have been abused in DEKA seems to be the (unwritten) right to assign claims under EU law. But such an explanation would appear to be rather far-fetched. It is more likely that the Court wanted to refer to a broader concept of good faith and wanted to prevent the debtor from taking advantage of his assignment in bad faith. Therefore DEKA does not provide very compelling support for the concept of abuse of law in EU contract law. iv.  Abuse of Law and Fixed-Term Employment Contracts Another admittedly marginal anchor for the abuse of law principle in EU contract law can be found in Directive 1999/70/EC concerning the framework agreement on fixed-term work. In Article 5(1) of the annexed ‘framework agreement on fixed-term work’, the Directive provides rules on the prevention of abuse of fixed-term employment contracts: To prevent abuse arising from the use of successive fixed-term employment contracts or relationships, Member States, after consultation with social partners in accordance with national law, collective agreements or practice, and/or the social partners, shall, where there are no equivalent legal measures to prevent abuse, introduce in a manner which takes account of the needs of specific sectors and/or categories of workers, one or more of the following measures: reasons justifying the renewal of such contracts or relationships; ·   objective maximum total duration of successive fixed-term employment contracts or relationships; ·  the · the number of renewals of such contracts or relationships. 80

The abuse of fixed-term employment contracts is at the crossroads of the abuse of a mere factual position (ie the dominance of the employer enabling him to keep the employee in successive fixed-term contracts) and a legal position (ie the right of the employer to proceed with fixed-term contracts without having the duties of a contract of indefinite duration). Only if seen from the second perspective can the abuse of fixedterm employment contracts be conceptualised as a specific affirmation of the abuse of law principle.81 v.  Abuse of Law in the ‘Acquis Principles’ In view of the weak support of the abuse of law principle in EU legislation and case law in respect of contracts, it is somewhat surprising to find a fairly clear affirmation of the principle in Article 7:102 of the ‘Principles of the Existing EC Contract Law’ or ‘Acquis Principles’: 80   Council Directive 1999/70/EC of 28 June 1999 concerning the framework agreement on fixed-term work concluded by ETUC, UNICE and CEEP [1999] OJ L175/43. 81  On the interpretation of Art 5, see Case C-212/04 Konstantinos Adeneler et al v Ellinikos Organismos Galaktos (ELOG) [2006] ECR I-6057.

250  Axel Metzger Good faith in the exercise of rights The creditor must exercise its rights to performance and remedies for non-performance in accordance with good faith.82

The principle in Article 7:102 is not on good faith in contract formation. Good faith in contract formation is dealt with in Article 2:101 of the Acquis Principles and has to be distinguished from abuse of law. In the former case, the question is about fairness and reasonableness of the parties when negotiating contracts, whereas in the case of abuse of law, the question is whether somebody makes use of a right or remedy in a way that contradicts the purpose of that right or remedy. Article 7:102 points to such a bad faith exercise of rights and remedies and the comments on the Acquis Principles state explicitly that Article 7:102 should be understood as a general rule preventing any abuse of law.83 However, the sources cited by the comments do not provide sufficient support for the existence of such a principle in EU contract law. Article 3(1) of the Unfair Terms Directive 93/13 deals—as has been mentioned earlier—with unfairness in contract formation,84 and Article 4 of the Commercial Agents Directive 86/653 concerns the duty of the principal to provide information to the agent.85 The cited ECJ case law deals with EC administrative law cases in which the Court invoked the principle of good faith in order to avoid arbitrary decisions of the Community authorities against its staff,86 for the recovery of unlawfully paid State aids in cases in which the national authority was responsible for the illegality of the aid decision87 or the recipient was acting in good faith.88 Admittedly, all sources and cases deal with good faith in a broader sense, but they do not support the existence of an abuse of law principle for the exercise of contractual rights and remedies as suggested by Article 7:102 of the Acquis Principles.89 III. Conclusion: The Current Status of the Abuse of Law Principle in EU Private Law

What then is the current status of the abuse of law principle in EU private law? This chapter has examined three areas of EU private law and has yielded mixed findings. In EU intellectual property law there is strong legislative support for the principle of abuse of law in Article 3 of the Enforcement Directive and in trade mark law. It has been applied by national courts when interpreting the European trade mark rules; however, the Court has not yet recognised the principle explicitly in this area. In EU civil procedure, the 82  See the Aquis Research Group on the Existing EC Private Law, Principles of the Existing EC Private Law— Contract I (Munich, Aquis Group 2007) 261. 83   Aquis Research Group (n 82 above) 264. 84   Unfair Terms Directive (n 68 above). 85   Council Directive 86/653/EEC of 18 December 1986 on the coordination of the laws of the Member States relating to self-employed commercial agents (Commercial Agent Directive) [1986] OJ L382/17. 86  See Joined Cases 43/59, 45/59 and 48/59 Eva von Lachmüller, Bernard Peuvrier, Roger Ehrhardt v Commission of the European Economic Community [1960] ECR 463. 87  See Case C-24/95 Rheinland-Pfalz v Alcan [1997] ECR I-1591. 88  See Case C-336/00 Republik Österreich v Martin Huber [2002] ECR I-7699; Case C-209/03 The Queen v London Borough of Ealing [2005] ECR I-2119. 89  See the detailed critique of the ‘Acquis Principles’ by N Jansen and R Zimmermann, ‘Grundregeln des bestehenden Gemeinschaftsprivatrechts?’ [2007] Juristenzeitung 1113. It is of special interest, that the comments on Art III – 1(103) of the Draft Common Frame of Reference on ‘Good faith and fair dealing’ do not refer explicitly to the concept of abuse of law and do not cite any EU sources.

Abuse of Law in EU Private Law: A (Re-)Construction from Fragments  251 legislative support provided by the Brussels Convention and the Brussels I Regulation— albeit Article 6(2)—is not as significant as in EU intellectual property, but here the Court has referred to abuse of law more frequently. Viewing the picture as a whole, one can convincingly argue that abuse of law has been established as a general principle of law in both areas, although not with the same clarity and insistence as in corporate or tax law or in the application of fundamental freedoms. And the finding would be even stronger if the comparative law materials from the Member States were taken into account. But this second basis of justifying legal principles of EU law has been deliberately omitted from this chapter and left to another contribution to this volume.90 Abuse of law is used in EU intellectual property and civil procedure as a correction tool for counteracting the use of rights and remedies in a manner that would contradict the purpose of the legal rules granting those rights and remedies. The principle is not merely used as a tool of interpretation but as a self-standing principle that can lead to results which run counter to the literal meaning of the rules at hand. From this perspective, abuse of law in EU intellectual property and procedural law has strong similarities to the principle as established by the Court in other areas of EU law. But the establishment as a general principle of law does not mean that abuse of law can be applied as a hard and fast rule. Rather, the Court’s case law indicates that countervailing principles and arguments may justify opportunistic behaviour that would otherwise appear abusive. For EU contract law the result seems to be different. Abuse of law is neither supported by EU legislation nor is it recognised by the Court’s case law. Although it is true that concept appears in some minor statements in specific directives and in single instances within the Court’s case law, these fragments do not amount to the recognition of a general principle of law. Therefore Article 7:102 of the Acquis Principles is more of a prophecy than an accurate statement of the ‘existing’ EU contract law. But there are arguments suggesting this prophecy will come true in the future. One argument might be that in the continental civil law systems as well, the principle of abuse of law was first established in property and procedure before entering the less obvious area of contracts.91 Another more policy-oriented argument might be that it is hardly conceivable for any legal order to provide parties with rights and remedies without providing at the same time the limits of those rights and remedies.92 This may be done by the legislator if a detailed style of legislation is preferred. If the written law does not provide such safeguards, this task falls to the courts. And here the principle of abuse of law is a natural candidate for EU private law to fulfil this limiting function.

 See Gordley (n 9 above).  See Fleischer (n 36 above) 865. 92  See S Grundmann, Europäisches Schuldvertragsrecht (Berlin, de Gruyter 1998) 135. 90 91

17 Comments on ‘Abuse of Law’ in European Private Law Simon Whittaker I. Introduction

L

et me start by saying that I found Professor Metzger’s chapter very interesting indeed, with a helpfully wide range of material reflecting his broad view of what can be called ‘European private law’.1 His insights are telling and repay careful consideration and reflection. In these comments, I shall concentrate my attention on those elements of his paper which relate to substantive private law rather than private international law, given that this was the subject of a separate discussion at this conference.2 My comments are basically three. II.  ‘General Legal Principles’

The question as to which legal propositions we should include within the category of ‘general legal principle’ and the significance which we then give them as a result is, in my view, increasingly difficult in the European legal context, given that there are a number of very different understandings of the nature and function of ‘legal principle’ in national laws and in European law itself. As regards ‘European private law’, let me say that I think that what is meant by ‘principle’ and which ‘principles’ should be included within the projected ‘Common Frame of Reference’ for European contract law will be one of its most controversial features.3  This contribution to this book of essays retains the oral character of the original conference observations.   G Cuniberti, ‘The Discreet Influence of Abuse of Law in International Civil Procedure’, ch 19 below. 3  The Commission has announced that a ‘Common Frame of Reference’ (CFR) to be used as a ‘tool-box’ for European contract law and, possibly in the future, as an ‘optional instrument’ will be constructed drawing on, in particular, the Draft Common Frame of Reference (DCFR): Communication from the Commission to the Council and the European Parliament on European Contract Law COM (2001) 398 final; Communication from the Commission to the European Parliament and the Council, A More Coherent European Contract Law, An Action Plan COM (2003) 68 final; European Contract Law and the Revision of the Acquis: The Way Forward (‘The Way Forward’) COM (2004) 651 final. The DCFR is published by the Study Group on a European Civil Code and the European Research Group on Existing EC Private Law (Acquis Group): C von Bar, E Clive and H Schulte-Nölke (eds), Principles, Definitions and Model Rules of European Private Law, Draft Common 1 2

254  Simon Whittaker I appreciate that this is not the occasion to develop a general theory of legal principles (or ‘general legal principles’), nor to comment on the nature of Ronald Dworkin’s contribution to such a discussion, but it is important to consider Professor Metzger’s ‘procedural’ definition of ‘principle’ which strikes me as very narrow. He states that: [A] general principle of law is a legal standard that is derived from legal rules by way of inductive generalisation.4

Professor Metzger notes that the sources of these legal principles may be derived either from a single legal system or from a range of systems,5 and then continues by explaining that a general principle (though discernable from individual examples by a particular jurist) cannot count as being a ‘legal principle’ unless it is recognised as such by the courts. Moreover, in his view, the functions of a general legal principle are quite particular: they may act as an aid to the interpretation of existing legal rules; they may be applied for ‘gap-filling’ (that is, they can be allowed to create new legal rules or, at least, different decisions on facts); and most controversially, they may (sometimes) be used as a means of correcting existing rules.6 ‘General principles’ in the sense in which Professor Metzger identifies them are not to be applied in an ‘all-or-nothing fashion’ but are to be balanced by other principles or arguments: ‘the validity of rules for special cases cannot ensure that a generalised form of this standard has to be applied in all unprovided-for cases’.7 This view therefore appears to put forward ‘legal principles’ as being general proposi­ tions of law which are not given overt expression by the law or, perhaps, by legislation. But, with respect, this is surely not the case—or rather, this surely uses ‘principle’ in a very particular sense, as something important to be ‘discovered’ by juristic or judicial construction (in both the sense of interpreting something which is there and of building something which is not there yet). For there are very important overt legislative statements of principle. So, for example, a French lawyer would see Article 1134 of the Civil Code as setting out a principle of good faith in the performance of contracts8 and Articles 1382 and 1383 as setting out the general principle of personal (extra-contractual) liability for fault.9 Where facts fall within the terms of these legislative general principles, they apply without more unless a particular exception is allowed: these are not principles to be balanced against some other principle or principles. Moreover, while the formal recognition of a general legal principle is most obvious in a legislative codification, it is not restricted to it. So, the English common law recognises some legal principles which are not merely generalisations from particular instances: in Frame of Reference (DCFR), Outline Edition (Munich, Sellier 2009). As can be seen, the inclusion of ‘principles’ is announced by the title of this work, but its authors themselves express some hesitation as to what should be included under this heading: see 9, 13. See further S Whittaker, ‘A Framework of Principle for European Contract Law?’ (2009) 125 Law Quarterly Review 616. 4  A Metzger, ‘Abuse of Law in EU Private Law: A (Re-)Construction from Fragments’, ch 16 above. 5  Metzger (n 4 above), Part I.B. 6  Metzger (n 4 above), Part I.B.i. 7  Metzger (n 4 above), Part I.B.iii. 8  Indeed, the recent French reform proposals for contract law include a section headed ‘principes directeurs’ which includes a statement of freedom of contract, the binding force of contract and of good faith in contract: Ministère de la justice, Projet de réforme du droit des contrats (July 2008) Arts 15–18, referred to in the rapport de presentation (p 3) as ‘les trois piliers’ of ‘la liberté contractuelle, la sécurité contractuelle et la loyauté entre les parties’. 9   See notably, G Viney and P Jourdain (eds), J Ghestin (gen ed), Traité de droit civil: Les conditions de la responsabilité, 2nd edn (Paris, LGDJ 1998) 315, no 439, whose first sentence discussing Article 1382 refers to it as setting out a ‘principe’.

Comments on ‘Abuse of Law’ in European Private Law  255 the context of contract, I would include here the principles of freedom of contract and the binding force of contract, and the principle of privity of contract. These ‘general principles’ argue in favour of a series of particular results unless there is some qualification on them or exception to them in a particular context. They may or may not be susceptible to being outweighed by ‘counter-principle’, depending on the context.10 But the existence of a formally recognised general principle—and in the EU context, a legislatively recognised general principle—remains significant as it casts a different light on the principles which Professor Metzger seeks to identify by means of ‘inductive generalisation’. I can see, however, why Professor Metzger has adopted the definition of ‘general principle’ which he has in the case of ‘abuse of law’: there clearly is no legislative principle of the ‘abuse of law’ in European private law (at least if we stick to the EU acquis) and no overt recognition of such a principle in the case law of the Court of Justice as regards ‘private law’. The question therefore arises as to whether such a principle should be seen in (I would say, should be constructed from) the particular instances where either EU legislation or case law use this notion. Here, though, I think that three questions must be addressed. First, do the justifications of policy behind each particular instance of recourse to ‘abuse of law’ still apply beyond their own scope to other situations, and, if so, to which? Secondly, if they do so apply, should the courts use the notion of ‘abuse of law’ as the legal technique or mechanism which best gives effect to these policies in the new context or contexts? And, thirdly, even if these two questions are answered positively, should it be the courts (and in this case, the European courts) rather than the legislature which should take the step of extending the use of the notion to the new situation, or, more broadly, to a set of new situations by recognition of a principle? Recourse to the language of principle and inductive reasoning should not disguise the reality of these three choices: choices of policy, technique, and appropriate (European) lawmaker. In the case of ‘abuse of law’ in the context of what is known as European private law, I think that the question of whether or not one should recognise such a ‘principle’ is acute. For, in particular, are the instances of the use of ‘abuse of law’ in EU legislation or case law to be seen either as examples of a wider general principle or as exceptions to other principles which are recognised as such by EU legislation or the Court of Justice? Let me explain this last point by taking one of the examples given by Professor Metzger illustrating the recognition of a principle of abuse of law in the context of EU civil procedure, that is, private international law.11 Article 6(2) of the Brussels Regulation12 provides a special ground of jurisdiction where a third party is joined to original proceedings, ‘unless these were instituted solely with the object of removing [the third party] from the jurisdiction of the court which would be competent in his case’. From this provision, Professor Metzger suggests that the two elements of the abuse of law test which he derives from the decisions in Emsland-Stärke and Halifax are met.13 For ‘the 10   For further discussion of the significance and content of some of these principles in the English context see H Beale (gen ed), Chitty on Contracts, 30th edn (London, Sweet & Maxwell 2008) paras [1-010–1-034] (S Whittaker) which also explains the significance of English law’s rejection of a legal principle of good faith in contracts. 11  Metzger (n 4 above), Part II.B.i. 12  Council Regulation 44/2001/EC of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (Brussels Regulation) [2001] OJ L12/1. 13  Metzger (n 4 above), Part I.C, Part II.B.i. Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt HamburgJonas [2000] ECR I-11569. Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609.

256  Simon Whittaker specific jurisdiction rule is being used against the purpose of the provision; the court for the initial proceedings is chosen artificially to remove the third party from the forum of his general jurisdiction. Seen from this perspective, Article 6(2) may be read as providing clear legislative support for the abuse of law principle in EC civil procedure.’14 However, and with respect, I find this a surprising reading of Article 6 and, indeed, of the Court’s decision in Freeport plc v Arnoldsson to which Professor Metzger refers in a footnote.15 Indeed, I am led to exactly the opposite conclusion. First, the Brussels Regulation does use the language of ‘general principle’16 but to refer to the ground of jurisdiction based on a defendant’s domicile found in Article 2 of the Regulation, from which the grounds of special jurisdiction found in Articles 5 to 22 derogate. This therefore provides a good example of a true legislative principle in that it will determine an outcome in the absence of an exception being established. Secondly, in Freeport17 the Court was urged by the EC Commission to recognise that the prohibition on misuse of the rules on jurisdiction found in Article 6(2) ‘is a general principle which must also be observed in the application of Article 6(1) of the Regulation’, the latter providing for special jurisdiction where the person sued ‘is one of a number of defendants, in the courts for the place where any one of them is domiciled, provided the claims are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings’. However, the Court explicitly rejected the Commission’s argument: first, on the basis that Article 6(1) (unlike Article 6(2)) does not say that it is so qualified; and, secondly, on the basis that there is no need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the Member State where one of the defendants is domiciled, given that (as earlier held by the Court) Article 6(1) applies only where ‘it is expedient to hear and determine claims together to avoid the risk of irreconcilable judgments resulting from separate proceedings’.18 So, far from being a ‘general principle’, Article 6(2)’s qualification by reference to the (improper) object of including the third party in the proceedings is needed in the context owing to the way in which the substantive provision is drafted; whereas in the case of Article 6(1), such a qualification is redundant. The decision in Freeport therefore appears to emphasise the special nature of the instance of ‘abuse of law’ in Article 6(2) (assuming that this is a proper identification, of course) rather than suggesting that it can be used as an example of a wider general principle. Moreover, in the case of the substantive provisions of ‘European private law’ (substantive as opposed to the ‘procedural’ provisions of private international law), there is a further reason why they do not form a suitable body of rules from which a court (European or national) should construct a principle of the nature of the ‘abuse of law’ in the way which has been suggested. For even if the construction of an ‘inductive principle’ of the sort proposed is an entirely appropriate way of reasoning in the context of a system of rules which is intended to be substantively coherent and complete—as is certainly the case as regards the continental civil codes and is, I would add, also the case as regards the English common law—I do not think it appropriate where the body of rules from which it is intended to be drawn is neither complete nor coherent in the same sense. If  Metzger (n 4 above), Part II.B.i.  Case C-98/06 Freeport plc v Olle Arnoldsson [2007] ECR I-8319; Metzger (n 4 above), Part II.B.i. 16   Brussels Regulation (n 12 above) Recital 11. 17   Freeport (n 15 above), para [50]. 18   Freeport (n 15 above), para [54]. 14 15

Comments on ‘Abuse of Law’ in European Private Law  257 for present purposes we restrict ‘European private law’ to the legislative acquis, it is clear that this collection of rules (itself somewhat indeterminate, given the lack of technical definition of this expression) does not form a complete ‘coherent and consistent’ body of law in the manner of a continental civil code or of the common law. Rather, the acquis is a collection of particular legislative interventions targeted at particular economic or social problems: indeed, this must be the case given the limited competences on which the legislation is based.19 And it is this ‘sectoral’ or ‘pointillist’ character which lies behind the desire of some continental private lawyers to establish a European civil law ‘codification’ aimed (at the very least) at the production of a greater coherence and consistency in EU legislation.20 But until such a ‘codification’ or ‘Common Frame of Reference’ comes into being (and it is not yet clear what form it will take),21 we are left with what is properly to be viewed as a series of exceptional treatments: exceptional, either in terms of the character of the contract (timeshare,22 package holidays23); the manner in which the contract is made (distance contracts,24 electronic commerce25), and in particular the character of the parties (notably, contracts between businesses and consumers26). As can be seen, this is true of the consumer contract acquis where EU contract law is most developed, and applies even more to the legislation outside this area, such as the Commercial Agents Directive27 or the Late Commercial Payments Directive.28 Moreover, the vast majority of directives belonging to the ‘private law acquis’ require the introduction of qualifications on ‘general principles’ which are accepted as such by the Court of Justice—that is, freedom of contract and the binding force of contract.29 The difficulty, therefore, in constructing ‘general principles’ from the provisions of the ‘private law acquis’ is that one runs the serious risk of generalising exceptions at the expense of the accepted fundamental principles from which they derogate.

19  This is notably the case as regards the Directives made under what are now Arts 115 and 114 TFEU [initially Arts 100 and 100A EEC and later Arts 94 and 95 EC respectively]. 20   See in particular, WH Roth, ‘Transposing “Pointillist” EC Guidelines into Systematic National Codes— Problems and Consequences’ (2002) 10 European Review of Private Law 761. 21   See notes to ‘Common Frame of Reference’ (n 3 above). 22   Parliament and Council Directive 2008/122 EC on the protection of consumers in respect of certain aspects of timeshare, long-term holiday product, resale and exchange contracts (Timeshare Directive) [2009] OJ L158/59. 23  Council Directive 90/314/EEC of 13 June 1990 on package travel, package holidays and package tours (Package Travel Directive) [1990] OJ L158/59. 24   Parliament and Council Directive 97/7/EC of 20 May 1997 on the protection of consumers in respect of distance contracts (Distance Selling Directive) [1997] OJ L144/19. 25   Parliament and Council Directive 2000/31/EC of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market (E-Commerce Directive) [2000] OJ L178/1. 26  Notably, Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts (Unfair Contract Terms Directive) [1993] OJ L95/29. 27  Council Directive 86/653/EEC of 18 December 1986 on the coordination of the laws of the Member States relating to self-employed commercial agents (Commercial Agents Directive) [1986] OJ L382/17. 28  Directive 2000/35/EC of the European Parliament and of the Council of 29 June 2000 on combating late payment in commercial transactions (Late Commercial Payments Directive) [2000] OJ L200/35. 29  Case C-240/97 Spain v European Commission [1999] ECR I-6571, para [99] (‘general principle of civil law’ in the context of the common agricultural policy); Case C-277/05 Société thermale d’Eugénie-les-Bains v Ministère de l’Economie, des Finances et de l’Industrie [2007] ECR I-6415, paras [21, 24, 28 and 49] (VAT); Case T-170 Alrosa Co Ltd v European Commission [2007] ECR II-2601, para [108] implicitly referring to ‘the fundamental economic freedoms guaranteed by the [EC] Treaty’ (competition law).

258  Simon Whittaker

III.  ‘Abuse of Law’ and Other ‘Qualifying’ Concepts

My second comment relates to the notion of ‘abuse of law’ itself in the context of private law and the way in which we should identify its expressions or examples. Professor Metzger (in my view rightly) discerns the influence of Josserand’s famous doctrine of the ‘abus de droit’ and, more generally, French juristic work on this notion in the case law of the ECJ in Emsland-Stärke in 2000 and Halifax in 2006.30 The notion of ‘abus de droit’ is indeed recognised by French lawyers (including the courts) as a very general ‘legal doctrine’ (théorie du droit), with expressions throughout French private law and French civil procedure,31 and possessing clear relations (and historical connections) with French public law theories of the ‘abuse of power’. There is indeed in the French materials a tension of the sort which Professor Metzger identifies between more ‘subjective’ approaches to the doctrine (which concentrate on a person’s ‘intention to harm’) and on more ‘objective’ approaches, which ask whether a person is using a right in accordance with or contrary to its social purpose. Having said that, two points arise. First, the English expression ‘abuse of law’ (which has, I think, no real resonance to most English lawyers) appears to mix up two related but nevertheless distinct elements, these being expressed in French as abus de droit and fraude à la loi. Briefly put, the purpose of the doctrine of ‘abus de droit’ (to be translated as ‘abuse of rights’) is to control the exercise of a right held by an individual (un droit subjectif  ). So, for example, a right to sue, a right not to contract, a right to terminate a contract for non-performance under an express termination clause. It can easily be seen that a control on the exercise of an individual’s right which goes beyond denying its effect where it is malicious (that is, intended to harm) and invalidates it where it is ‘contrary to its social purpose’ appears to strike at the very core of the liberal conception of rights. It is for this reason, indeed, that the doctrine of the ‘abuse of rights’ has remained theoretically controversial in France; and for this reason also, that French courts have declined to accept any one approach to the doctrine, sometimes requiring an intention to harm, sometimes taking a broader, more teleological view of the rights in question. The case law here varies considerably depending on the legal context in question. Fraude à la loi consists of the intentional attempt to avoid the application of a mandatory rule of law without a legitimate ground. So, the ‘fraudulent’ element has a sense of illegitimacy and deliberateness, but it is related here to a mandatory rule of law. So, it can be seen that the doctrines of ‘abuse of rights’ and of ‘fraud on the law’ have in common that they include elements of ‘intention to harm’ (though the latter is not necessary for all cases of the ‘abuse of rights’)32 and, according to a leading work, ‘the fundamental idea is the same: the law cannot allow its legal doctrines or other mechanisms to be diverted from their purposes’.33 However, there are differences between the two doctrines. In particular, the doctrine of the ‘abuse of rights’ is used to control the exercise of ascertained and delineated individual rights, whereas most examples of ‘fraud on the law’ do not concern the exercise of a right, but rather the ‘diversion of a legal rule  Metzger (n 4 above), Part I.C.   J Ghestin, G Goubeaux and M Fabre-Magnan (eds), J Ghestin (gen ed) Traité de droit civil: Introduction générale, 4th edn (Paris, LGDJ 1994) 747. 32   Ghestin et al (n 31 above) 722. 33   Ghestin et al (n 31 above) 822 (author’s translation). 30 31

Comments on ‘Abuse of Law’ in European Private Law  259 so as to acquire a right which one would not otherwise do’.34 The expression of the ‘abuse of law’ appears to merge these two distinct ideas. This then leads me to my second point. Even if we agree that general liberties (for example, a right to choose with whom to contract or to break off from contract, or to convey one’s property when and to whom one likes) or particular rights (for example, a right to performance or to termination of a contract) require qualification in some way, the question still arises as to the legal mechanism or mechanisms by which such a qualification is to be achieved. In this respect, recourse to a ‘general legal principle’ of the sort identified by Professor Metzger is only one type of legal mechanism, using only one type of language. First, such a qualification can be achieved by restricting the particular conditions under which a particular right (or legal result) arises. We have already seen an example of this in Article 6(1) of the Brussels Regulation.35 Secondly, a person’s European rights may be made subject to an evaluative concept without this being seen as an expression of a principle of the ‘abuse of rights’/‘abuse of law’. So, notably, I do not see references to ‘good faith’ in either the legislative acquis (as in the case of the control of unfair terms in consumer contracts or the duties owed by parties to contracts of commercial agency36) or the case law of the Court of Justice (notably, in Gruber v Bay Wa AG37) as being illustrative of a principle of the ‘abuse of rights’/‘abuse of law’. Similarly, I do not see the qualifications on a consumer/buyer’s rights to repair or replacement of the goods by reference to the notion of ‘disproportionality’ in the Consumer Sales Directive of 1999 as so illustrative.38 These are indeed qualifications on the unrestricted behaviour of the contracting parties—and may also be qualifications on their rights. But their orientation and content differs from the orientation and content of the ‘abuse of law’ doctrine adopted by the Court in the context of common agricultural policy or tax. Instead of seeing these various types of qualifying concepts as examples of such a broad mega-principle, they should rather be seen in their own contexts, each having their own proper function in the context of the rules and purposes pursued by the Directive in question. In my view, the use by the legislative ‘European private law’ acquis of good faith, proportionality and, sometimes, reasonableness demonstrates the variety of techniques available instead of the notion of the abuse of law. IV.  ‘General Principles’ and Lawmaking Power

This leads me to my third comment. This relates to the constitutional appropriateness of recognition of a ‘general principle’ of ‘abuse of law’ in the context of the ‘European private law’ or EU contractual acquis given the nature of the legislation in this area. For, each directive attempts to balance the needs of freedom of contract and the binding force   Ghestin et al (n 31 above) 823 (author’s translation).   Brussels Regulation (n 12 above). 36  Unfair Contract Terms Directive (n 26 above) Art 3(1); Commercial Agents Directive (n 27 above) Art 3(1); similarly Directive 2002/65/EC of the European Parliament and of the Council of 23 September 2002 concerning the distance marketing of consumer financial services and amending Council Directive 90/619/EEC and Directives 97/7/EC and 98/27/EC (Distance Marketing Directive) [2002] OJ L271/16, Art 3(2). 37  Case C-464/01 Johann Gruber v Bay Wa AG [2005] ECR 1-439, para [53]. 38  Directive 1999/44/EC of the European Parliament and of the Council of 25 May 1999 on certain aspects of the sale of consumer goods and associated guarantees (Consumer Sales Directive) [1999] OJ L171/12, Art 3(3). 34 35

260  Simon Whittaker of contract, on the one hand, against its particular social and economic concerns (for example, the protection of consumers or commercial agents and the encouragement of the Internal Market through the harmonisation of laws). This balance is expressed through the rules which are set out in the directives and should not be disturbed by reference to a qualifying principle of the breadth of ‘abuse of law’ (or, I would add, good faith) as this would give too much power to the European Court in Luxembourg (to which it would fall to decide where this new principle would bite) and/or to national courts (if they were given by the Court of Justice the task of assessing whether a person had indeed ‘abused’ the law). The recognition of such a general principle would mark a very significant shift in lawmaking power from the European legislator to the European and/or national courts. V. Conclusion

We should not allow the naturally seductive character of general principles, once formulated, to obscure the real choices which remain: choices of policy, technique and appropriate lawmaker. Moreover, the gathering together of ‘qualifying concepts’ (such as the abuse of law/rights, good faith, disproportionality or reasonableness) under a general heading may result in the distortion of individual components, each of which possesses its own significance or range of significances and its own utility in different contexts.

18 The Rejection of Abuse in International Civil Procedure Adrian Briggs* i. Introduction

I

t is sometimes thought to be good practice to start with a confession. To someone who deals only with private international law, and almost never with substantive European law, it is not precisely clear what ‘abuse of law’ actually, really, is. By way of acquiring an education, one could do a lot worse than to read and concentrate upon two particular papers. The first is the most interesting publication by Rita de la Feria,1 the second is the paper by Professor Nuyts, given by him in Oxford in 2006.2 The first of these describes the way in which the European Court has developed, if not always evenly and openly, a principle by reference to which national law may prevent the exercise of rights seemingly granted by European legislation where that exercise would be abusive, or which allows a national court to conclude that the exercise of a right seemingly conferred by European legislation would be an abuse of that legislation: it is not clear to me that these are congruent, but they certainly overlap, whatever the precise form of it. There may come a point at which, even though a person can bring himself or his activity within the wording of a provision of European law, he or it still falls so far outside the spirit of the legislation that it will not protect him after all. Such a technique is, one imagines, found in all legal systems. An English lawyer will most easily recall the principle that equity will not permit a statute to cloak a fraud. Others will express similar ideas in their own ways. The second paper, though written earlier in time, examined whether the notion of abuse of rights had any part to play in the operation of the Brussels Convention. We can, for these purposes, take the Brussels Convention,3 and the Lugano Convention,4 and

 * Circumstance made it impossible to speak to the paper in person at the conference for which this chapter was prepared, and as a result there was no opportunity to debate the differences which showed themselves to exist between the view advanced here and in the analysis of Professor Cuniberti. See G Cuniberti, ‘The Discreet Influence of Abuse of Law in International Civil Procedure’, ch 19 below. 1  R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. 2   A Nuyts, ‘The Enforcement of Jurisdiction Agreements Further to Gasser and the Community Principle of Abuse of Right’ in P de Vareilles-Sommières (ed), Forum Shopping in the European Judicial Area (Oxford, Hart Publishing 2007) 55. 3   Brussels, 27 September 1968, given effect in the United Kingdom by Civil Jurisdiction and Judgments Act 1982. In effect now superseded by the Brussels I Regulation (n 5 below). 4   Lugano, 16 September 1988, given effect in the United Kingdom by Civil Jurisdiction and Judgments Act 1991, but being superseded by a revised Convention, published in [2007] OJ L339/3, but not yet fully in force.

262  Adrian Briggs the Brussels I Regulation,5 to be functionally interchangeable. The conclusion to which the writer came was that there were one or two places in which the principle of abuse of rights had made an appearance, and that it was still possible, if in a constricted way, for it to operate alongside the Convention. The trouble for the argument was that in its most recent utterances, the Court of Justice has shown itself to be unmistakeably and unrelentingly hostile to the application of any such principle in the particular context of jurisdiction and the recognition of judgments in civil and commercial matters. The most obvious illustration of the point came in Erich Gasser GmbH v MISAT srl.6 In that case, a long-term sales relationship between an Austrian seller and an Italian buyer had broken down. The case was one in which the contractual relationship between the parties was not in dispute: though there were mutual accusations of its breach, there was none as to its intrinsic validity. Neither was there any dispute that the contract contained a jurisdiction agreement for the Austrian courts. Nevertheless, when it apprehended that the weather was turning litigious, the Italian buyer commenced proceedings in Rome, claiming various forms of relief. It was, perhaps, never found as a fact or determined as a certainty, but little intelligence is required to deduce the motivation for the behaviour of the Italian buyer. The so-called ‘Italian torpedo’ tends to be launched by the party who fears being found to be in the legal wrong. He will therefore start proceedings in Italy, the court which time forgot. He will confidently expect that the proceedings will make glacial, or even slower, progress and will, while they are pending, prevent the party who thinks of itself as being in the right from being able to sue in the court which would have otherwise been his right and preference. In common or garden cases, the opportunity for wrongful tactical behaviour is obvious; the invitation and intention to pervert the course of justice is plain enough. In intellectual property cases, which was what provoked Mr Franzosi7 to coin this term of art, where every day for which the patent-holder is prevented from enforcing his property rights is a day of unwarranted gain for the infringer, the menace is all the more acute. It seemed so to the Higher Regional Court at Innsbruck when it asked, in effect, whether the Austrian court, nominated by a written agreement on choice of court, could be entitled to exercise jurisdiction even though the proceedings brought before the court in Rome had priority in terms of time. The Innsbruck court was given a resoundingly negative answer: the chosen Austrian court was not entitled to undertake any review of the jurisdiction of the Italian court, which was the court first seised, and given the terms of Article 21 of the Convention,8 that was that. The European Court of Justice knew perfectly well that its answer gave the green light to chicanery of the kind which MISAT srl had practised, and responded to the fact by jamming its fingers in its ears. At paragraph [53] it said: Finally, the difficulties of the kind referred to by the United Kingdom Government, stemming from delaying tactics by parties who, with the intention of delaying settlement of the substantive dispute, commence proceedings before a court which they know to lack jurisdiction by reason of the existence of a jurisdiction clause are not such as to call into question the interpretation of any provision of the Brussels Convention, as deduced from its wording and its purpose. 5  Council Regulation (EC) 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2002] OJ L12/1. 6  Case C-116/02 Erich Gasser Gmbh v MISAT srl [2003] ECR I-14692. 7  M Franzosi, ‘World-wide Patent Litigation and the Italian Torpedo’ (1997) 7 European Intellectual Property Review 382. 8  Now Art 27 of the Regulation.

The Rejection of Abuse in International Civil Procedure  263 It could hardly have been put more clearly than that; there is no ambiguity to be found. Even where a person started proceedings in a court which he knew9—not, be it noted, which he suspected or believed, but knew—to have no jurisdiction, those objectionablybrought proceedings maintained the status of temporal priority over those brought in the agreed court. If there was ever going to be a clearer refutation of the idea that the invocation of a rule of European jurisdictional law might be found to be an abuse, and disregarded as a result, this was it. As far as the European Court of Justice was concerned, the principle was that even when there is misbehaviour by a contract-breaker, or other abuser, the duty on the national court is to see no evil, hear no evil, and do no good. It is therefore from this point which any analysis of abuse of rights in international civil procedure must begin, and to which it must eventually return. II.  Abuse of process, right, or law

The context in which the possibility of abusive exercise of rights arises may, one supposes, shape the manner in which the principle operates. The cases examined in the paper of Dr de la Feria, to which reference has been made,10 appear to show some variation in the degree of seeming self-justification which can be tolerated before it crosses the line and becomes abusive. It seems to me that this need not be so: if we were able to agree on a precise and durable meaning of ‘abuse’, there would be no reason in principle for it to differ in effect from one legal area to another. The definition proposed in the EmslandStärke GmbH v Hauptzollamt Hamburg-Jonas11 case seems to meet the standard, in requiring: First, a combination of objective circumstances in which, despite formal observance of the conditions laid down by the Community rules, the purpose of those rules has not been achieved . . . Second, a subjective element consisting in the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it.

So let us consider the behaviour of the proprietors of MISAT srl, on the hypothesis that it conforms to the pattern of behaviour which the United Kingdom proposed as significant. Taking the second first: did they have ‘the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it’ when they ‘with the intention of delaying settlement of the substantive dispute, commence[d] proceedings before a court which they know to lack jurisdiction by reason of the existence of a jurisdiction clause’? It seems undeniable that they did. They took advantage of the rule now in Article 27 of the Brussels I Regulation to assert the priority of the Italian proceedings brought, on this hypothesis, in a court known to have no jurisdiction. The second condition is well satisfied. As to the former condition, can it then be said that there was ‘a combination of objective circumstances in which, despite formal observance of the conditions laid down by the Community rules, the purpose of those rules has not been achieved’ when there     9  It is not said that the particular Italian buyer knew that the Italian court had no jurisdiction, even if this is the most plausible explanation for his behaviour. The point is that the Court, in making its ruling, explained that it would be undeflected even by proof that the party in question did have this guilty knowledge. 10  See de la Feria (n 1 above). 11  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, paras [53] and [54].

264  Adrian Briggs were ‘delaying tactics by parties who, with the intention of delaying settlement of the substantive dispute, commence proceedings before a court which they know to lack jurisdiction’? The answer can only be yes. The purpose of the scheme of rules put in place by the Brussels I Regulation may be gathered in part from the preamble to the Regulation itself. The tenth paragraph of the preamble says that: The rules of jurisdiction must be highly predictable and founded on the principle that jurisdiction is generally based on the defendant’s domicile and jurisdiction must always be available on this ground save in a few well-defined situations in which the subject-matter of the litigation or the autonomy of the parties warrants a different linking factor.

The counterpart to this is to be found in the judgment in Benincasa v Dentalkit srl,12 where the Court said: Next, as the Court has consistently held, the objectives of the Convention include unification of the rules on jurisdiction of the Contracting States’s courts, so as to avoid as far as possible the multiplication of the bases of jurisdiction in relation to one and the same legal relationship and to reinforce the legal protection available to persons established in the Community by, at the same time, allowing the plaintiff easily to identify the court before which he may bring an action and the defendant reasonably to foresee the court before which he may be sued (Case 38/81 Effer v Kantner [1982] ECR 825, paragraph 6, and Case C-125/92 Mulox IBC [1993] ECR I-4075, paragraph 11).

Now if proceedings are instituted in a court which has no jurisdiction over the defendant in relation to them, for the purpose of delaying settlement of the substantive dispute, the aims of the EU rules do not seem to be met; if this is done outside the contractuallychosen court, legal certainty and predictability is set at naught. So far as it is a matter to be determined by the Emsland case, there would therefore be no objection to the finding, by a national court in the position of the Austrian court in Erich Gasser, that the bringing of proceedings before the Italian court was abusive: a manipulation of the rules for an advantage so illegitimate that it defeats the very purpose for which those rules were adopted. But the Court of Justice has told us that we are wrong, and we need to ask what this teaching actually means for us. Despite what the European Court of Justice said in its judgment in Erich Gasser, not everyone appears to accept that the principle of abuse of law has been excluded from the realm of the Brussels Convention. In his paper in 2006, Professor Nuyts looked at part of the material which will be mentioned below before coming to the conclusion that: [I]t is equally true that the abuse of right correction is a particularly narrow one, which applies only in very particular circumstances, when there is clear evidence that a litigant is seeking to derive, to the detriment of another party, an improper advantage manifestly contrary to the objective of the Brussels regime. But though this tool should be used sparingly, it is still available under current European law and should indeed be relied upon to sanction13 the most blatant abusive tactics used by unscrupulous litigants within the European judicial area.14

It is necessary to ask in this short chapter—short for several reasons, not excluding the paucity of available material—to consider why it is that English lawyers look at the  Case C-269/95 Francesco Benincasa v Dentalkit srl [1997] ECR I-3767, para [26].  It seems that ‘sanction’ is used in the sense of ‘to penalise or discourage’ rather than ‘to approve or to encourage’. 14  Nuyts (n 2 above) 55, 73. 12 13

The Rejection of Abuse in International Civil Procedure  265 judgments of the European Court and draw from them the conclusion that ‘abuse of law’ has been rejected in international civil procedure. It is possible that the title allocated to this session will be seen by some as controversial. Indeed, it sometimes appears that almost anything a common lawyer ventures to say about the operation of the Brussels Convention and the instruments descended from it is assumed to be controversial: even if it appears to be utterly orthodox, the suspicion is that this is in fact a clever application of veneer to conceal something which needs to be covered up. The analysis is made more complicated by the fact that some judgments and Opinions in the Court can be read as supporting the idea that a national court may, perhaps must, control abusive behaviour by those who appear before it. Professor Nuyts gave two examples which, however, do not seem to be material; the one which he should have cited went unmentioned by him. We should first dispose of the two irrelevant authorities. First in Kalfelis v Bankhaus Schröder Münchmeyer Hengst & Co,15 it was held that a claimant was not entitled to invoke the special jurisdiction of a court under Article 6(1) of the Brussels Convention16 where his doing so had as its sole object the ousting of the jurisdiction of the courts of a particular defendant’s domicile. In other words—or the Court considered it to be in other words, which is perhaps not the same thing—there had been enough of a link between the claims made against the first defendant and the second defendant, domiciled elsewhere, to justify removing the second defendant from the court which had or would have had general jurisdiction over claims against him. But this has little or nothing to do with a principle of abuse of rights: the Court interpreted Article 6(1), as it was then worded, in a manner which restricted the circumstances in which it would be available to claimants in preference to the general jurisdictional rule of Article 2, and in so doing, no doubt, made it more difficult for it to be used. The Court has a record of giving restrictive interpretations to various jurisdictional rules in the Convention and now in the Regulation. When it does so, it is driven by a primitive urge to restrict jurisdictions other than that of the particular defendant’s domicile. It is not motivated by a concern to prevent an assertion of jurisdiction which is abusive, but by the need to give a meaning to statutory words which is conservative, which is not the same thing at all. By its judgment in Kalfelis, the Court said ‘On a true construction of its words, Article 6(1) does not apply on facts such as these, so Article 2 does’. It did not say ‘Article 6(1) could apply in a case like this, but it is an abuse for you, the claimant, to rely on it, so you may not’. Kalfelis did not really assist the argument that there is a role for a doctrine of abuse of rights. And more recently, there is more. In its most recent judgment in Freeport plc v Arnoldsson,17 the Court appears to have gone back even on this limited acceptance of the possibility that the right to take advantage of the jurisdictional rule in Article 6(1), as it is now worded, may be withdrawn where it is used for such an improper purpose. The point was explicitly raised by an English company which had been sued in Sweden by a claimant who had joined it in proceedings brought also against a Swedish company. The English defendant was of the opinion that the claimant was bringing proceedings against the Swedish company for the sole purpose of ousting the English company from the jurisdiction of the courts of its domicile. The answer given by the Court, in the second paragraph of its ruling, was that even if this were so it was nihil ad rem:  Case 189/87 Athanasios Kalfelis v Bankhaus Schröder, Münchmeyer, Hengst and Co et al [1988] ECR 5565.   Which gives special jurisdiction over a defendant sued in another Member State where the claim is brought against several defendants of whom one is sued in the State of his or her domicile. 17  Case C-98/06 Freeport plc v Olle Arnoldsson [2007] ECR I-8319. 15 16

266  Adrian Briggs Article 6(1) of Regulation No 44/2001 applies where claims brought against different defendants are connected when the proceedings are instituted, that is to say, where it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings, without there being any further need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the Member State where one of the defendants is domiciled.

It seems fair to read this as saying and meaning that the issue of abusive behaviour is simply not relevant, whether it is present or not. To my mind, this development, of which he obviously could have had no inkling, makes the argument advanced by Professor Nuyts even more difficult to defend than it was at the time. The second example was the decision in MSG v Les Gravières Rhénanes sarl,18 in which the Court examined a contractual stipulation that the place of performance of a contract for carriage by river was Würzburg, a German city, even though the contractual obligations, and all of them, fell to be performed at river ports in France. It held, entirely rationally, that the provision spelled out in the contract was ineffective to give the courts at Würzburg special jurisdiction over matters relating to the contract. This decision, perfectly sensibly reached, was unconcerned with any notion of abuse. Article 5(1) gave jurisdiction to the place for performance of the obligation in question and not, if it were different, to the place which the parties had stated as the place for performance. In that case, Würzburg was not the place for performance of the obligation in question, for the good and sufficient reason that it was not the place where the contractual duties were to be performed. The parties were simply wrong about it, and ignorance does not make the law. Even the dullest English undergraduate—and there is no shortage of competition for that title—can usually remember that: The manufacture of a five-pronged implement for manual digging results in a fork even if the manufacturer, unfamiliar with the English language, insists that he intended to make, and has made, a spade.

Lord Templeman’s homespun wisdom in Street v Mountford19 makes the simple point that the law does not always take the parties’ word for it. This principle, and not anything about abuse of rights, explains the result in MSG. It is fair to say that the Court made rather heavy weather of the point, but it got to the right answer mostly for the right reason: the place for the performance of the obligation in question was not Würzburg, and nothing the parties did could alter that fact. This decision, therefore, fails to assist Professor Nuyts in his submission. The case which represents the best, but weak, authority for the view that abuses in the assertion of jurisdiction should be prevented on the ground that they are abuses, is The Tatry.20 When a cargo of soya bean oil arrived damaged and corrupt at the ports of Rotterdam and Hamburg, the various cargo owners were displeased. The shipowners knew perfectly well that they could expect to be sued, so seized the initiative to bring proceedings before the Dutch courts for a declaration that they owed no liability to the cargo interests in respect of the damage to the cargo. Unfortunately, they listed as 18  Case C-106/95 Mainschiffahrts-Genossenschaft eG (MSG) v Les Gravières Rhénanes SARL [1997] ECR I-911. 19   Street v Mountford [1985] AC 809, 819 (HL). 20  Case C-406/92 The owners of the cargo lately laden on board the ship ‘Tatry’ v The owners of the ship ‘Maciej Rataj’ (The Tatry) [1994] ECR I-5439.

The Rejection of Abuse in International Civil Procedure  267 defendants only some of the cargo interests; and in due course proceedings were begun in England (by the arrest of a sister ship as security for the claim, followed by service of the writ on the vessel before it is released, as admiralty law allows) by two groups of cargo owners: those who had not been brought into the Dutch proceedings as defendants, and those who had. The English Court of Appeal referred five questions to the European Court of Justice. For our purposes it is sufficient to observe that the Court stated that in applying the lis pendens rule in Article 21 of the Brussels Convention, which gives jurisdiction to the court seised first, where the same cause of action between the same parties is brought for a second court, the latter court has no jurisdiction. The court seised second is obliged to apply the rule of temporal priority simply and straightforwardly. No exception is allowed for the case in which the court first seised is seised of proceedings which look contrived, as claims for a declaration of non-liability frequently21 do: the Court saw no need to say so; it simply observed that where the two sets of proceedings involved the same parties and were founded on the same cause of action, Article 21 of the Convention applied without modification or further ado. Advocate-General Tesauro, however, was not completely deaf to the suggestion that there was something artificial, self-interested even, in the way the shipowners had brought proceedings in which they claimed no substantive relief. He said:22 Where, conversely, the rules of private international law or the case law, or both, adopt connecting factors which better correspond to the nature and characteristics of the relationship, and to the expectations of the parties who originally created it and ‘devised it’, the possibilities of biased or even abusive use of procedural and private international law, as a whole, are also reduced. In any event, it will be incumbent upon the court seised to ensure that any abuse is thwarted. (Emphasis added.)

It is not clear whether the Court’s non-mention of this aspect of the Opinion has any silent significance, but the approach of the Advocate-General is plain enough: the court seised has a general duty to ensure that the abusive use of procedural and private international law is thwarted. It seems that this should be the point of departure for an argument, such as that which Professor Nuyts might have made. There will be those cases in which the bringing of proceedings may be said to be abusive, but maybe the Brussels Convention itself is not the source of any legal rule which allows a court to measure or assess whether the bringing of proceedings is abusive or not. III. Cases from the English courts

The European Court of Justice was recently asked by the English Court of Appeal whether, when proceedings arising out of a tort committed in Jamaica, to which five Jamaican and one English party were sued as defendants, might (as against the English defendant) be stayed on the ground that the natural forum for the trial of the claim was the court of Jamaica. The answer given by the Court in Owusu v Jackson23 was a predictable 21  Not inevitably, though. In certain circumstances (such as where an insurer or reinsurer needs to know whether the policy is valid and binding, so that a decision can be taken whether to defend the claim in the name of the insured or reinsured) it is an entirely proper and prudent course to take. The same could not so obviously be said of the shipowners’ claim in The Tatry. 22   AG Tesauro in The Tatry (n 20 above), para [22]. 23  Case C-281/02 Andrew Owusu v NB Jackson, trading as ‘Villa Holidays Bal-Inn Villas’ et al [2005] ECR I-1383.

268  Adrian Briggs negative: negative, because the judgment makes it plain that the Court considers the jurisdictional rules stated in the Convention to be absolute and unconditioned by any rule of national law, and predictable, because the continental hostility to what it believes24 to be the doctrine of forum non conveniens is inveterate. But the legal doctrine of forum non conveniens in England—in Scots law it is otherwise—has its origin in the principle that the English court has a duty, when asked, to prevent abusive recourse to its procedure. In St Pierre v South American Stores (Gath & Chaves) Ltd25 the Court of Appeal had been asked to decline to exercise jurisdiction over the defendant in respect of the claim; having reviewed the law it refused to grant the relief applied for. The test was that a stay of proceedings would not be ordered unless the bringing of the proceedings in England was ‘oppressive or vexatious’, and even then it would not be ordered unless the defendant, who had to apply for the relief, was able to convince the court that it was not unjust to order the relief. The relationship between this standard of ‘oppression or vexation’ and the current English doctrine of forum non conveniens is debatable, but nowadays not much debated in England. Everyone knows that the line of authority which began with The Atlantic Star26 in 1974 was designed to water down the stringency of these terms.27 The High Court of Australia, which generally disapproves of the direction which English law has taken, in this area in particular, has adhered28 to the law as laid down in St Pierre, and authorises judges to grant a stay of proceedings if the bringing of proceedings in Australia is29 ‘vexatious or oppressive in the Voth30 sense of those terms, namely, that they are “productive of serious and unjustified trouble and harassment” or “seriously and unfairly burdensome, prejudicial or damaging”’. If the defendant to a case before the English courts, over whom and which the court has jurisdiction by reason of Article 2 of the Brussels Regulation, is able to show that the bringing of the claim against him in England is oppressive or vexatious in the sense in which the High Court of Australia continues to use those terms, it is easy enough to see the bringing of the action as something which abuses the law, or which abuses the processes of the court. If there remains a right or a duty on the English judge to ‘ensure that any abuse is thwarted’, as Advocate-General Tesauro said, the proper conclusion would be that the proceedings may be stayed after all. On this view of the matter, the answer given by the European Court of Justice in Owusu, which is plainly binding in its entirety on the English courts, would not extend to the case in which the court was persuaded by the defendant that the bringing of the proceedings was ‘productive of serious and unjustified trouble and harassment’ or ‘seriously and unfairly burdensome, prejudicial or damaging’. In other words, English law would be permitted to revert, on this point, to the position which it held before the standard of ‘vexation or oppression’ was first watered down. 24   Writings which have explained that the manner in which the doctrine is misunderstood outside the common law are many, but the objectors sometimes appear to be deaf to reason. Certainly, if the doctrine worked in the way some writers have represented it—an unpredictable, unreviewable, maverick discretion to refuse to hear a case even though both parties want it heard by the judge—most English lawyers would object to it as well. 25   St Pierre v South American Stores (Gath & Chaves) Ltd [1936] 1 KB 382 (CA). 26   The Atlantic Star v Bona Spes [1974] AC 436 (HL). 27  See MacShannon v Rockware Glass Ltd [1978] AC 795 (HL). The reference to ‘watering down’ is in the speech of Lord Russell of Killowen, 823. 28  See in particular, Régie National des Usines Renault SA v Zhang (2003) 210 CLR 491. 29  The quotation is taken from the judgment in CSR Ltd v Cigna Insurance Australia Ltd (1997) 189 CLR 345, 401. 30  The reference is to Voth v Manildra Flour Mills Pty Ltd (1990) 171 CLR 538.

The Rejection of Abuse in International Civil Procedure  269 Would this be accepted by the Court of Justice as justification for an English court to decline to exercise jurisdiction conferred by the plain words of (say) Article 2? It is not clear that it would. The judgment in Owusu can be read as denying the possibility that an action may be dismissed, in any circumstances and for any reason, in order that it be tried in the court of another State unless the rule by which the decision is taken is expressly stated in the Convention. There being no textual reference to vexation or oppression, or to abusive proceedings, in the Convention, it may be deduced that a claim may not be disposed of by finding that its being brought is oppressive or vexatious. On the other hand, the Opinion of Advocate-General Tesauro ought to count for something. And even though the gist of the judgment in Owusu is that a claimant who has brought proceedings in a court which has statutory jurisdiction to hear them should not be told to go and sue somewhere else by reference to a rule which is not written in the Convention, the objection to the doctrine of forum non conveniens is, according to the Court, its width and its capacity to upset legal certainty.31 The material question is therefore this: if a claimant brings proceedings in circumstances where the English court is persuaded by the defendant that it is ‘productive of serious and unjustified trouble and harassment’ or ‘seriously and unfairly burdensome, prejudicial or damaging’ that they be brought in England rather than elsewhere, may it grant the appropriate relief? If the answer is negative, the conclusion will be that abuse of rights means abuse of substantive rights. One enters the realm of prediction, but a pessimistic guess would be that a court may not grant relief on this ground.32 One explanation for this would be an apprehension that it would represent the thin end of a wedge, the thick end of which is the doctrine of forum non conveniens. But another would be that the notion of abuse of rights does not apply to procedural law, because procedural law is not founded on the notion of litigants having rights. If there are no rights, there is nothing which can be used to generate their abuse. Though in English law we are completely accustomed to see rules relating to jurisdiction in terms of rights (to invoke, to oppose invocation) and obligations (to attend court, to not summon another to attend court), it is now apparent that jurisdiction and civil procedure is not so understood, or is not understood in quite the same way, in other European systems. It is necessary to be clear about the one case which appears to decide that, even if there is blatant abuse by a party who seeks to take advantage of a jurisdictional rule of the Brussels Convention or Regulation, there is no remedy which may be ordered to prevent this wrongdoing, even if national law would have made such a remedy available. In Turner v Grovit,33 the English courts were seised of proceedings brought by an impecunious former employee, Turner, when the defendant employer, Grovit, caused proceedings to be brought against Turner before the Spanish courts. It is far from clear that the Spanish court was made aware of the pending English proceedings,34 but it was pretty clear to the English court that the institution of the Spanish proceedings was abusive in the Voth v Manildra sense and, indeed, in any sense of the word. The Court of Appeal ordered Grovit to cease his wrongful behaviour; the European Court of Justice held that it had 31  See Owusu (n 23 above), paras [41] and [42]. The error in para [42], where the Court supposes that an English court may stay proceedings contrary to the wishes of the defendant, is crass but will, no doubt, not be allowed to invalidate the reasoning which followed from it. 32  This conclusion may be supported by reference to Turner v Grovit (n 33 below). 33  Case C-159/02 Turner v Grovit [2004] ECR I-3565. 34  Turner, defendant in Madrid, did not have the funds to allow him to make an appearance.

270  Adrian Briggs no right to do so. There can be no intelligent doubt that Grovit was using the Brussels Convention for an illegitimate, abusive purpose, but this was not permitted to lead the English court to the conclusion that his misbehaviour was something which it could deal with. There may have been an abuse, but it was not an abuse which it was within the power of the English court to restrain: see no evil, hear no evil, speak no evil, and do no good. IV. Reasons for the silence of the law on abuse

So why is it that those of us whose interest in European law extends no further than the Brussels Convention, and the instruments descended from it, do not appear to have the benefit of a principle that rights may be not asserted when it would be abusive for them to be exercised? One may identify four possible explanations. The view taken here, derived from the limited material at our disposal and from a rather sour view of the quality of the reasoning deployed in the judgments of the European Court of Justice, is that all of these are justifiable as explanations, but that the last one is the most telling. A.  The Doctrine Exists to Prevent a Rule of European Law Excluding a Rule of National Law which Would Otherwise Be Applicable, but Does Not Apply to Settle the Inter-Relationship between Rules of European Law It would not be a complete surprise to be told that there is a geometrical expression for this: that it has diagonal effect, or something similar, but not horizontal or vertical effect. It is hard to know what these expressions mean. As far as one may tell, viewing it all from the outside, the principle that there may not be an abusive exercise of rights conferred by European law comes into consideration when a national court is invited to apply a rule of national law, only to be told that this is incompatible with a rule of European law, only for this to provoke the retort that for the rule of European law to be applied in these circumstances would amount to an abuse of that European right. Even if there is a case which does not conform to that basic pattern, it will be the exception rather than the rule. But in the context of the Brussels Convention, there is not usually—Turner and Owusu are the exceptions—a conflict between a rule of national law and a provision of the Convention which, on the face of it, conflicts with it. Rather, the conflict is internal, and confined to the rules of the Convention itself. In Erich Gasser, for example, the question was whether the lis alibi pendens rule in what was then Article 2135 of the Convention prevented the assertion of jurisdiction in accordance with Article 17;36 in MSG, with whether the rule of jurisdiction in Article 5(1) or Article 17 applied to assess the validity and effectiveness of a contractual specification of which court was to have jurisdiction. In The Tatry the question was again whether Article 17 served to prevent the jurisdiction which would otherwise be available by virtue of Article 5737 of the Convention, and so on. As far as these cases are concerned, the conflict of legal rules  Now Art 27 of the Brussels I Regulation.  Now Art 23 of the Brussels I Regulation. 37   Which preserves the integrity of jurisdictional provisions of another international Convention; now Art 71 of the Brussels I Regulation. 35 36

The Rejection of Abuse in International Civil Procedure  271 is not one which arises between a jurisdictional rule of national law and a jurisdictional rule of the Regulation. When seen in that light, it is not surprising that the resolution of jurisdictional issues is seen as a matter of ordinary statutory construction, and not as a question which requires one to be able to characterise one of the rival arguments as an abuse of process. This is correct as far as it goes. But this by itself cannot explain the outcome in Turner and Owusu. There was no question in those cases of needing to reconcile the material scope of two provisions of the Convention; and as an explanation it would be incomplete. We need say no more about it. B.  The Doctrine of Abuse Exists To Curb the Abusive Assertion of Rights, but the Brussels Convention Does Not Confer Rights on Individuals, but Only Imposes Duties on Courts May it be said that the doctrine of abuse exists to control the abusive assertion of rights, but that the Brussels Convention does not confer rights, and the doctrine of abuse does not and cannot apply to it? The paper by Rita de la Feria, to which reference has been made,38 is interesting at several levels, but what appears to be taken for granted is that the doctrine of abuse operates, where it operates at all, where one is dealing with rights and with the claim by a party to assert them. These may be sometimes referred to as freedoms,39 but the freedom lies in the individual right to exercise whichever freedom it is with which one is concerned. The exercise of a right to establishment, or to receive or provide services, et cetera, may not be exercised where there is something abusive about it. The principle which seems to have its origin in the judgment in Van Binsbergen40 was concerned with restrictions placed on the right of free movement in the context of the right to provide services. The case, and those which came after it, appear to be concerned with restrictions imposed by national law to restrict the exercise of substantive European law rights (or freedoms). It may be granted that the doctrines of oppression or vexation, or even forum non conveniens, are restrictions imposed by national law. But they exist to control what in English law are procedural rights, and what in some other systems do not appear to be thought of as rights at all. It would be no surprise to discover that a doctrine which had been admitted by the Court of Justice, if erratically and occasionally grudgingly, to control the exercise of substantive rights simply had no application to the exercise of procedural non-rights. In England, it has been common for a long time to theorise our rules on jurisdiction and its exercise—not that we theorise very much—in terms of claimants having rights, defendants having rights, and of the court, when deciding whether to give effect to the rules which govern jurisdiction, as having much the same discretion as it does when called upon to specifically enforce a contract. It is not that jurisdiction has been privatised, but that a very simple jurisdictional foundation—anyone may sue anyone who is present in England when the proceedings are commenced—is modified in its operation by the  See de la Feria (n 1 above).   Presumably because there is no correlative duty on an individual, as distinct from on a state. ‘Rights’ would more naturally suggest that there was a correlative duty. 40  Case 33/74 Van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299. 38 39

272  Adrian Briggs principle of forum non conveniens and the legal fact that the parties to litigation have procedural as well as substantive rights, about which they may make agreements which are contractually binding, and for the enforcement of which they may look to a judge. If the law is in essence written in terms of rights, it is clear that those rights may sometimes be exercised in a wrongful, abusive fashion; and courts have mechanisms for dealing with abuses.41 But if one imagines rules which define judicial jurisdiction as being written as legislative instructions to judges, directing them when they should adjudicate and when they should not adjudicate, it is less obvious that they confer rights on individuals.42 To be sure, individuals may benefit or derive advantage from the way these rules operate. The idea that a stranger to a relationship derives an advantage from the observance of the duties of that relationship is well known. Here the relationship is between legislator and judge, and the private litigant is no more than an interested bystander. The idea that rules of jurisdiction can be exercised abusively is, from this point of view, practically inadmissible. If there is no right, there is no abuse of right. Such an explanation is at least consistent with the result, and with some of the alleged reasoning, in Turner. But it cannot explain Owusu. C.  The Doctrine Is Overshadowed by and Yields to a Principle Which Is Even More Important: That There Is Such a High Degree of Mutual Trust between the Courts of Member States That a Submission Which Even Suggests That a Foreign Court Has Failed to Prevent Wrongful Recourse by a Party Is Inadmissible, Whether or Not It Is True Be it now supposed that there is a complaint which can be made, that the assertion of a jurisdiction conferred by the Convention is a perversion of the scheme which the Convention sought to establish. This, after all, is not a hard conclusion to draw from the facts of Erich Gasser and it is, as said above, even more of a menace in intellectual property disputes, where an action which is to be based on the infringement of a patent may be torpedoed by proceedings brought in the Italian or Belgian courts, and which are founded on a pleading which claims a declaration that there was no wrongful infringement. These torpedoes still work; and even though the Italian Supreme Court is said to have made some effort to restrict their ultimate effectiveness,43 none of this overcomes the immense and hair-tearing delay which may be made to descend upon jurisdictional challenges in the Italian legal system. Proceedings brought, as the United Kingdom Government evidently suggested, for such wrongful purposes should not be brought and should not be permitted to have the effect which they are wrongfully designed to have. The reason that the European Court of Justice in Erich Gasser refused to countenance the idea that the court which had been prorogated with jurisdiction might exercise it despite the proceedings launched and then marooned in the Italian courts, was that for 41   A full working out of this view of the common law may be found in A Briggs, Agreements on Jurisdiction and Choice of Law (Oxford, Oxford University Press 2008). 42  To take an English analogy, even though the law may impose a statutory duty, it does not necessarily mean that it generates a private personal right, such as would allow a complaint to be made if the duty is not discharged by the person upon whom it is laid. 43  In BL Macchine Automatische v Windmöller & Hölscher Decision no 19550 of 19 December 2003, Corriere Giuridico 2/2004, 162.

The Rejection of Abuse in International Civil Procedure  273 the Austrian court to do any such thing was to assess and to make a judgment about the jurisdiction of the Italian court. Even though the Austrian court would have been applying the very same legal rule as the Italian court (one day) would, it amounted to an interference with a matter which was for the Italian court, and not for anyone else, to adjudicate. Exactly the same objection was made to the conclusion, drawn by the English Court of Appeal in Turner v Grovit, that the antics of Grovit in Madrid were abusive in character: it was not for the English court to ask this question, still less to answer it, and least of all to grant a remedy to deal with it: hear no evil, et cetera. The principle of mutual trust which the courts of Member States are expected to have for each other’s adjudications would be damaged if the Austrian court in Erich Gasser were to have been permitted to do as it was asked by the claimant to do; likewise if the English court were to have granted relief in Turner v Grovit. It is hard to avoid the conclusion that there was more than a trace of suspicious hostility to the habits and thinking of the common law in this, a conclusion not challenged by the manner in which the Court set out the arguments of the United Kingdom in order to dismiss them: we return to this point below. There is, of course, a perfectly sensible basis for the decision in Erich Gasser. Even if one were to take the view that the proceedings in Italy should never have been brought, and would inevitably be dismissed, the fact still was that they were in existence when the Austrian court was asked to take jurisdiction. Their existence was an indisputable fact; they were not legally non-existent. It may be that the Italian proceedings would, in due course, be terminated. It might have been declared that they were unfounded, flawed from the outset, and liable to be set aside as though they had been rescinded. But they cannot have been a nullity. It is a perfectly familiar principle of English law that, where a court has made an order, it requires observance, even though one party will contend that it should never have been made. English law forbids a party to say, in effect: ‘I will succeed in having this order set aside or annulled, and in the meantime I am entitled to ignore it, as it is an order that should never have been made’. The law does not work like that.44 Yet that was, in effect, what Erich Gasser GmbH was asking the Austrian court to do: to proceed as though the Italian judicial summons had never been made. It is not hard to accept that it was quite right for the European Court of Justice to reject an argument which would have allowed the Austrian court, in effect, to act in response to the Italian proceedings as (one supposes) it would not have allowed a party to act in relation to one of its own judicial orders. There was no room for the argument that there was abusive behaviour where the conclusion which the court would then be invited to draw was that the Italian judicial proceedings were a legal nullity. If the correct legal principle is that only a court which has made an order can lift it, it is obvious why the argument of the claimant in Erich Gasser was irrelevant: there was nothing the Austrian court could do, for only the Italian court could retract the judicial summons which it had issued. It may be regrettable that this be so: it is all the more regrettable where one may fear that the processes of the Italian court may be perverted for illicit advantage; but the existence of the Italian proceedings was a legal fact, and there was no argument which could justify an Austrian court in proceeding to adjudicate as though something which was a fact was not a fact. This sensible answer would follow from sensible law. It has little or nothing to do with a principle of mutual trust. It has been argued—convincingly, to my way of thinking—that   For authority: Isaacs v Robinson [1985] AC 97 (PC).

44

274  Adrian Briggs trust is something to be earned but not something to be imposed.45 But mutual respect for the judicial orders made by judges in the various Member States seems perfectly sensible. Article IV of the Constitution of the United States contains a provision that ‘full faith and credit’ is to be given to the ‘public acts, records, and judicial proceedings’ of the states of the union. It simplifies matters to point to this as the reason why one court may not question what another has done, but there is no difficulty in drawing the conclusion that a court in one state may not find that, as far as it is concerned, a judicial act made by a court in another jurisdiction should be treated as non-existent. Though all sorts of people will have all sorts of objection to the idea that a provision of the Constitution of the United States might translate into the context of the European Union, it seems to fit like a glove. Erich Gasser was right, if not for the reasons given. The reasoning in the case is unsophisticated and unconvincing, but the conclusion is correct. The proper analysis of Erich Gasser leaves no room for one court to make the judgment that the invocation of proceedings before another court, even if these proceedings are jurisdictionally unfounded, may be treated as a nullity. But Turner v Grovit is quite different. The proposition that a court may not order a person, over whom it already has jurisdiction, to act in accordance with English law, is altogether surprising. The European Court of Justice ruled that whether or not the behaviour of Grovit was liable to be analysed as wrongful, the English court was precluded from ordering Grovit to show self-restraint and to bring an end to his Spanish adventure. It did not make any reference to the contention that Grovit may have been invoking the jurisdiction of the Spanish court in an abusive fashion for, as it does rather seem to many, it was so determined to disparage the anti-suit injunction that it was unprepared to descend to technical analysis of the arguments, perhaps for fear that if it had done so, it would have invited debate and disagreement. However, it would seem to follow that the decision whether a person—usually the claimant—has invoked the jurisdiction wrongly is something which may be determined only by the court whose jurisdiction has been so invoked. It seems further to follow that if there is such a thing as the invocation of jurisdiction in a way which is an abuse, then that also is a conclusion which may be reached only by the judge whose jurisdiction has been invoked. That limited proposition is consistent with the statement of Advocate-General Tesauro in The Tatry. That does not mean that there is no room for the principle of abuse in this area, but if a court is precluded from looking at what has gone on elsewhere, and finding it to be an abuse, there is really no role for it at all. And if (when) a court comes to the conclusion that its own jurisdiction has been invoked in accordance with the letter of the Convention, but otherwise wrongfully, Owusu makes it very difficult for any consequential order to be made; even then, all roads seem to be blocked. Even the prospect of a limited scope for what Advocate-General Tesauro said seems to vanish. D.  There Is a Suspicion That If the Argument Were to Find a Foothold, Common Lawyers Would Run Amok English lawyers are traditionally unsurprised by the picture of one court making orders which have the effect of interfering with proceedings before another court. For several 45   F Blobel and P Späth, ‘The Tale of Multilateral Trust and the European Law of Civil Procedure’ (2005) 30 European Law Review 528.

The Rejection of Abuse in International Civil Procedure  275 centuries it has been part of our legal tradition that the exercise of legal rights in one court (common law) could be prevented by an order made by another court (Chancery), administering a separate system of rules (equity), requiring that the party addressed exercise self-restraint and act in accordance with what the restraining court determines that his conscience should have told him to do. It would not be surprising, therefore, that an English court was prepared—not keen, perhaps, but prepared where the law required it—to make such orders where the proceedings to which legal objection could be made were taken overseas. The analysis proceeds according to three steps: (i) not denying that proceedings in another court are lawful so far as the law applied in and by that court is concerned, (ii) then finding it to be sufficiently wrongful to take advantage of the law applied in and by that court (iii) and therefore making an order directing the party who is so conducting himself to exercise self-restraint. This is as obviously sensible to a common lawyer as it evidently is not to a civilian. The opposition to this is distinctly peculiar. The idea that the only court which is entitled to order a wrongdoer to cease doing his wrong should be the place where he is doing it (and where it may not be wrong) is not one generally found in our national laws: the courts of the locus delicti may have jurisdiction over wrongs, but it is rare to be told that they have exclusive jurisdiction. Yet if we see the commencing of proceedings which should not be commenced as a wrong, it leads inevitably to the twin conclusions that the Brussels Convention or Regulation does not or should not be understood to permit it, and that the English courts may grant the remedies which follow from any judicial finding that a defendant has committed a civil wrong. One rather suspects that the aim of the recent cases is to nip this line of argument in the bud by establishing that it is not a wrong to institute civil proceedings; it is not an abuse of the Convention to start civil proceedings in a court which will, in due course, have to decide for itself whether it has jurisdiction; and as it is neither a wrong nor an abuse, there is no conclusion which an English court may draw. In other words, the legal principle which allows a national court to reason that reliance on provisions of European law is an abuse of those provisions, and to do something about it in consequence of that abuse, does not apply where the rules of European law in question are rules about the jurisdiction of courts. If Owusu is taken as laying down a general rule, it makes no difference whether the court in question is in England or in any other Member State. The principle that the rules of jurisdiction may not be relied on in a manner which may be characterised as abusive and dealt with as such is, on this view, not part of the law of jurisdiction in civil and commercial matters. V. Conclusions

The principle that rights conferred by European legislation may not be exercised in a manner which would constitute an abuse of their fundamental purpose is not a complex idea. The principle appears to be a member of a family of similar and related doctrines which are essential in any legal system, that allow a litigant to say and a court to agree that the legal rule in question was never meant to extend to such a case, or to licence such an outcome. It would be extraordinary to think that such a principle was incompatible with the legislative rules put in place by the organs of the European Union. But it will be equally true that every court which accepts that there is such a principle will in its turn see the risk that the principle will be abused. In England, the Statute of Frauds 1677 and

276  Adrian Briggs the legislation which has re-enacted it, and the case law which has had to grapple with the operation of it, shows brilliantly the opposing forces at work. European law is no different. It is just law. It is made by legislators. It does not always work the way, and only in the way, the legislators had in mind; and one also suspects that, not infrequently, they had nothing precise in mind to begin with. But the European Court seems to have set its face against admitting that the principle, whatever it is, has a role in the operation of the Brussels Convention and instruments derived from it. Theoretical explanations for this can be essayed, but they are pretty unconvincing. Even though the Convention and Regulation made procedural law rather than substantive law, procedural legal rules may be utilised with the same wrongful intent as can be brought to bear on substantive rules. If abuse involves the wrongful use of legal rules, there is no self-evident justification for confining this to rules of substantive law. Another related possibility is that disputes as to jurisdiction are supposed to be quick and easy to resolve, and are supposed to keep the court clear of any decision on matters of substance. This, after all, has tended to be said to justify the judicial decision that the question whether the parties have agreed on the jurisdiction of a court is to be answered without regard to whether the contract which is said to underpin the claim was and remained valid as a source of legal obligations: one sees this most clearly in Benincasa v Dentalkit srl.46 Yet this cannot be the reason either. It is childish to believe that disputes as to jurisdiction can be insulated from the intellectual and practical processes of dispute, legal reasoning, and appeals against decisions. It may be an inconvenient truth, but if parties elect to join issue on a question of judicial jurisdiction, this was their choice. No one forces them to; and for lawmakers to suggest that this is not the proper place for thorough legal analysis of legal rules is unworthy. It may also come as a surprise to those who find themselves promoted or shunted sideways to the bench of the Court of Justice, but in the adult world, commercial litigation is like that. The result is that there is really only one conclusion left. Everyone knows that the rules of the Brussels Convention and Brussels Regulation may be utilised by a litigant for a purpose which is abusive. The Court knows it too. Be it repeated: [T]he difficulties . . . stemming from delaying tactics by parties who, with the intention of delaying settlement of the substantive dispute, commence proceedings before a court which they know to lack jurisdiction by reason of the existence of a jurisdiction clause are not such as to call into question the interpretation of any provision of the Brussels Convention, as deduced from its wording and its purpose.47

It is improbable that the Brussels Convention was made, and the Brussels Regulation adopted, to allow for the delaying of settlement of legal disputes. This cannot be an aim; it must therefore be a tolerated side effect. The consequence of Erich Gasser and of Turner v Grovit is that a court other than that before which the particular proceedings are brought may be asked to find that the jurisdiction of that court has been invoked abusively. The consequence of Owusu v Jackson is that the power of the court before which the proceedings have been brought to find that it has been seised abusively is, at its highest, of uncertain but very restricted scope; and it may actually be non-existent. If this results from a fear that courts in common law jurisdictions will make use of their traditional jurisdictional principles, it is regrettable, not least for the manner in which it   Dentalkit (n 12 above).   Erich Gasser (n 6 above), para [53]. (Emphasis added.)

46 47

The Rejection of Abuse in International Civil Procedure  277 shows that the mutual trust which has been said to underpin the law and its operation in this area is something from which the Court of Justice is, uniquely, exempt. For the present, it means that if it is said to a court that a jurisdictional provision of the Brussels Regulation has been invoked in a manner which abuses that right, the authorised judicial response is ‘so what?’ Perhaps it is utopian to believe that when a litigant can bring his opponent before a judge and demonstrate to that judge that a wrong is being committed, then the judge should be able to order a remedy which will put it right. But then, Utopia is not a Member State.

19 The Discreet Influence of Abuse of Law in International Civil Procedure Gilles Cuniberti I. INTRODUCTION

T

he purpose of this chapter is to react to the chapter by Professor Briggs1 on abuse of law in international civil procedure. I understand that the organisers of the conference for which these chapters were prepared chose the main speakers and their discussants with the hope that they would have different points of view on the topic. As a discussant, I always fear that I might agree with everything the speaker has said and thus have a hard time presenting after him. This is especially so when the speaker not only was once the supervisor of your doctoral thesis, but is also someone you regard as one of the very best specialists of the field in Europe. Remarkably, however, I have found after reading Adrian Briggs’ chapter that we have a very different perception of the law on this topic. This certainly puts me in a much easier situation in respect of the organisers of the conference, and it also shows that they did a great job at identifying potential reasons for different appreciation of the law on this topic. I should underline, however, that I do not think that Adrian Briggs and I are in full disagreement. I think that our understanding of the state of the law is quite different. He thinks that the Court of Justice has consistently rejected the doctrine of abuse of law in our field, while I think the doctrine has had some influence on the development of the law, and has certainly not been rejected. Yet, we have a similar analysis of the reason why abuse of law has been either rejected, or has at least been less influential in our field. So, in a way, we probably agree on the most important reason: what the peculiarity of our field is in respect of other parts of European law, and why this has led to a much more limited operation of the doctrine in international civil procedure. It seems to me that one of the reasons we have different opinions on the matter is that there is no decision of the Court which expressly has applied the concept in the context of international civil procedure, let alone a decision which has provided a general definition of abuse of law in this context, or at least has contributed to the making of a new general principle of EU law. This is not to say, however, that because a concept was not expressly applied by courts it was not applied at all, or it was wholly inoperative in the relevant field. Courts could have applied the concept without feeling the need to insist on it and to rationalise its operation.  See A Briggs, ‘The Rejection of Abuse in International Civil Procedure’, ch 18 above.

1

280  Gilles Cuniberti But such operation is of course harder to notice for scholars and outside observers. And if the result it produces could have been reached through the application of other concepts or tools, reasonable people may eventually disagree on whether one concept or the other was actually applied in a given decision. The risks of such disagreement will increase dramatically if the following conditions are met. First, the relevant concept is still in the making, and its definition and legal regime have not yet been fully worked out. Second, the decisions can be read by observers in different languages. I submit that this is what is happening in the field of international civil procedure as far as a possible application or rejection of abuse of law is concerned. II.  WAS ABUSE OF LAW REJECTED IN INTERNATIONAL CIVIL PROCEDURE?

The first disagreement I have with Professor Briggs is about the Gasser case.2 In the opinion of Professor Briggs, this case has clearly rejected the operation of abuse of law in international civil procedure. I am not sure that is right. In Gasser, the dispute had arisen between an Austrian seller, Gasser, and an Italian buyer, MISAT. Gasser had sent invoices to the buyer which had all included a jurisdiction agreement for the Austrian courts. MISAT initiated proceedings in Italy. Gasser later sued in Austria, arguing that the jurisdiction clause had conferred jurisdiction to the Austrian courts. MISAT, however, challenged the jurisdiction of Austrian courts, on the ground that the jurisdiction agreement was non-existent. It also argued that the lis pendens rule of Article 21 of the Brussels Convention imposed on the Austrian court, which was seised second, to stay proceedings until the court first seised ruled on its jurisdiction. The case has been presented as a clear example of abuse of law.3 Article 21 of the Convention and the doctrine of lis pendens would have been used abusively for the purpose of circumventing the operation of Article 17 of the Convention, which provides for the enforcement of jurisdiction agreements. Proceedings in Italy would have been initiated with clear knowledge that there was no chance that the Italian court could ever retain jurisdiction in the case. The Italian party would have earned some time by launching the so-called ‘Italian torpedo’, owing to the fact that a lot of cases take an exceedingly long time to be tried in Italy. This would clearly be abusive. However, I am not absolutely convinced that the case is such a perfect example of abusive behaviour. After all, the actual acceptance by the Italian party of the jurisdiction agreement appearing on invoices was unclear,4 and it may well be that MISAT genuinely thought that it could convince the Italian court that there had not been an actual agreement on jurisdiction. The case would certainly have been much more conclusive if the agreement had been part of the main contract and thus clearly accepted by both parties. In any case, the reason why Gasser is presented as a clear rejection of abuse of law is not so much the result that the ECJ reached in the case, but rather a dictum in the decision which rejected an argument relying on the idea of abuse:   Case C-116/02 Erich Gasser Gmbh v MISAT srl [2003] ECR I-14692.  Briggs (n 1 above). 4   A Nuyts, ‘The Enforcement of Jurisdiction Agreements Further to Gasser and the Community Principle of Abuse of Right’ in P de Vareilles-Sommières (ed), Forum Shopping in the European Judicial Area (Oxford, Hart Publishing 2007) 55, 62. 2 3

The Discreet Influence of Abuse of Law in International Civil Procedure  281 Finally, the difficulties of the kind referred to by the United Kingdom Government stemming from delaying tactics by parties who, with the intention of delaying settlement of the substantive dispute, commence proceedings before the court which they know to lack jurisdiction by reason of the existence of a jurisdiction clause are not such as to call into question the interpretation of any provision of the Brussels Convention, as deduced from its wording and its purpose.5

At first sight, this dictum seems to clearly reveal the unwillingness of the Court to accept the operation of a doctrine of abuse of law in the context of the Brussels Convention. As already explained, what could be perceived as abusive in this case was the commencement of proceedings in Italy for the purpose of delaying proceedings in Austria. The ECJ refused to take this into account in its decision. I am not sure, however, that what the ECJ meant was that the doctrine of abuse of law would not apply in international civil procedure. Actually, I am not even sure that it meant that the very behaviour that it discusses in this dictum would necessarily stay unsanctioned. I think that it is important to read the dictum in context. The Court was addressing an issue which was not whether there should be remedies in case of abuse of law. The ECJ was answering the second question referred to it by the Austrian court. This question was understood by the ECJ as being whether the lis pendens rule of Article 21 of the Convention also applies when the national court seised second has exclusive jurisdiction by virtue of a jurisdiction agreement.6 So, unsurprisingly, what the ECJ did was to answer this question. The judgment discusses the scope of Article 21 and whether there is a good reason to exclude Article 17 from it. This was the first time this important issue was addressed by the Court. So I would submit that the judgment means to decide this question. The Court did not have to address the issue of abuse of law. It was concerned with a more general question, which was the question which had been asked by the referring court. It happens to be that, in the course of the proceedings, the UK government argued that in order to answer this general question, it was desirable to take into account the potential for abuse of the possible answers. That was perhaps a sensible position. But this did not convince the Court. The ECJ answered that these are two different issues which should not be confused. The general question regarding the lis pendens rule of Article 21 of the Convention is the one which should be answered. Abuse of law is a different issue. Thus, in the words of the Court, this different issue does not ‘call into question the interpretation of any provision of the Brussels Convention, as deduced from its wording and its purpose’. In essence, the ECJ ruled that the two issues are unconnected, and that one does not inform the other—if only because the question of the relationship between Article 21 and Article 17 can occur even in the absence of any abuse. Indeed, in Gasser, it was arguably not abusive to sue first in Italy, as the existence of the jurisdiction agreement was not obvious. I think that this dictum should therefore be read narrowly. It did not reject abuse of law; it only refused to address the issue. In my mind, this is far from ruling out that the Court of Justice will one day be asked expressly to discuss the operation of the doctrine in the context of international civil procedure in a case where there is evidence of an actual abuse. This interpretation of Gasser is consistent with an understanding of abuse of law as an autonomous corrective device. The principle of abuse of law would not modify rules or   Gasser (n 2 above), para [53].   Gasser (n 2 above), para [28].

5 6

282  Gilles Cuniberti doctrines of EU law, but rather operate to correct abusive use of such rules or doctrines. As an autonomous doctrine, it ought to be discussed autonomously, and should not be confused with the interpretation of other rules or doctrines of EU law. To put it another way, it would not be a principle of construction of European law, but a truly autonomous principle.7 III. THE DISCREET INFLUENCE OF ABUSE OF LAW

My disagreement with Professor Briggs goes further. I am of the opinion that not only was the doctrine of abuse of law not rejected in our field, but also that it has actually been influential and has contributed to the development of the law. In a couple of cases, the ECJ gave reasons for its decision which very much look like the application of a principle of abuse of law. Professor Briggs thinks that they were ‘unconcerned with any notion of abuse’.8 I think that they show at the very least that the doctrine of abuse of law exercises some influence in the field. A.  Gravières Rhénanes The first of these cases was MSG v Gravières Rhénanes.9 In this case, the ECJ was concerned with the interpretation of Article 5(1) of the Brussels Convention, which provided for the jurisdiction of courts in contractual matters. Article 5(1) granted jurisdiction to the court of the place of performance of the contractual obligation. The question referred to the ECJ was whether such place of performance could be defined abstractly by the parties; that is, without aiming at defining the place where performance ought to actually occur, but solely at establishing which court would have jurisdiction under Article 5(1). The answer to this question could have been given without considering any other provision of the Convention. The Court could merely have answered that the reference to the place of performance in Article 5(1) was to be interpreted as a reference to the actual place of performance of the obligation. This is indeed what the Court answered.10 This was a very sensible interpretation of the provision. As Adrian Briggs rightly pointed out, this was a good and sufficient reason for the decision. However, this is not the sole reason that the ECJ gave. The Court also insisted that the result of admitting that the parties define abstractly the place of performance of the obligation would be to circumvent another provision of the Convention, namely Article 17. Article 17 provides for the enforcement of agreements on jurisdiction, but under certain conditions only. In effect, parties defining the place of performance abstractly solely to establish jurisdiction conclude an agreement on jurisdiction. The Court held: Secondly, it should be considered that to specify a place of performance which has no actual connection with the real subject-matter of the contract becomes fictitious and has as its sole     7   On this distinction, see R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 438.     8  Briggs (n 1 above).     9   Case C-106/95 Mainschiffahrts-Genossenschaft eG (MSG) v Les Gravières Rhénanes SARL [1997] ECR I-911. 10   Gravière Rhénanes (n 9 above), para [32].

The Discreet Influence of Abuse of Law in International Civil Procedure  283 purpose the determination of the place of the courts having jurisdiction. Such agreements conferring jurisdiction are governed by Article 17 of the Convention and are therefore subject to specific requirements as to form. Thus, where there is such an agreement, there is not only no direct connection between the dispute and the courts called upon to determine it, but there is also circumvention of Article 17, which, whilst providing for exclusive jurisdiction by dispensing with any objective connection between the relationship in dispute and the court designated (Zelger v Salinitri, paragraph 4), requires, for that very reason, compliance with the strict requirements as to form which it sets out.11

The second reason for the decision of the ECJ was thus to avoid any ‘circumvention of Article 17’ by allowing the parties to use Article 5(1) of the Convention ‘fictitiously’ and with ‘the sole purpose’ of designating the competent court. Although the Court does not expressly say so, this very much looks like an application of a doctrine of abuse of law.12 Interestingly enough, the term was used by the referring court in the question it asked to the Court: [T]he Bundesgerichtshof has doubts about whether such ‘abstract’ agreements are valid under the Convention in so far as they involve a risk of abuse, that is to say, circumvention of the rules as to form set out in Article 17 of the Convention.13

B.  Other Cases Continental scholars believe that, likewise, in several other cases, the doctrine of abuse of law was influential in shaping the European law of international civil procedure.14 I cannot discuss all of them in this short essay.15 I will thus only present the most interesting of them, which are those which influenced the interpretation, and indeed the reform of Article 6(1) of the Brussels Convention. IV. THE PUZZLE OF ARTICLE 6(1) OF THE CONVENTION

Article 6(1) of the Brussels Convention enables plaintiffs to sue all defendants in the courts for the place where any one of them is domiciled. In the late 1990s, the issue arose as to whether this provision should be interpreted as meaning that there must be a connection between the actions against the various defendants. In Kalfelis v Bankhaus Schröder Münchmeyer Hengst & Co,16 the ECJ answered that it should. The judgment could be interpreted in many ways, but I would argue that the rationale for the rule in Kalfelis is indeed to prevent abuse of law.   Gravière Rhénanes (n 9 above), paras [33–34].  Nuyts (n 4 above) 55, 66; H Gaudemet-Tallon, ‘Case Comment on Gravières Rhénanes’ [1997] Revue critique de droit international privé (Fr) 576; H Muir Watt and D Bureau, Droit international privé (Tome 1) (Paris, Presses Universitaires France 2007) 223. 13   Gravière Rhénanes (n 9 above), para [10]. (Emphasis added.) 14  See Muir Watt and Bureau (n 12 above) 223; E Cornut, Théorie critique de la fraude à la loi (Paris, Répertoire Defrénois 2006) 351; A Metzger, ‘Abuse of Law in EU Private Law: A (Re-)Construction from Fragments’, ch 16 above. 15  They are the Autoteile case (Case C-220/84 AS-Autoteile Service GmbH v Pierre Malhé [1985] ECR 2267) and the Kalfelis case (n 16 below). 16   Case 189/87 Athanasios Kalfelis v Bankhaus Schröder, Münchmeyer, Hengst & Co et al [1998] ECR 5565. 11 12

284  Gilles Cuniberti When the Brussels Convention was replaced by the Brussels I Regulation, Article 6(1) was amended along the lines of Kalfelis. The new provision did not mention abuse of law, but arguably was amended for the purpose of addressing the issue. Recently, the ECJ was asked in Freeport plc v Olle Arnoldsson17 whether, as abuse of law ought to be considered as a general principle of EC law, it should be deemed generally applicable in the context of Article 6(1). The ECJ held that it should not. A.  Kalfelis In Kalfelis,18 the ECJ was asked to interpret a provision which did not make any reference, direct or indirect, to the concept of abuse. The issue was whether Article 6(1) of the Brussels Convention should be interpreted as meaning that there must be a connection between the actions against the various defendants. The ECJ answered positively, but explained that the reason was to exclude the possibility to call into question the principle of the Convention vesting jurisdiction in the courts of the domicile of the defendant. It then added that ‘that possibility might arise if a plaintiff were at liberty to make a claim against a number of defendants with the sole object of ousting the jurisdiction of the courts of the State where one of the defendants is domiciled’.19 In other words, the ECJ laid down an interpretation of Article 6(1), observing that the rationale of this interpretation was to avoid use of such a provision to circumvent the one which provides for the jurisdiction of the court of the defendant’s domicile, namely Article 2. Nowhere in its decision did the ECJ mention abuse of law. It is thus fair to interpret Kalfelis as a mere effort of interpretation of the relevant provision, as Professor Briggs did. At the same time, for any civil lawyer, especially one from a francophone background such as Professor Nuyts20 or me, the reference of the decision to the concept of fraude à la loi seems obvious.21 Fraude à la loi is defined as the use of one rule with the sole object of ousting the application of another. This seems to be clearly the issue that the ECJ addressed and the rationale of the decision. Of course, fraude à la loi is a concept of French and Belgian domestic law. I am intentionally using the French term to emphasise the point that the legal culture of the lawyer reading a decision of the Court of Justice is important, and that it can easily change the perception of a concept’s meaning. As lawyers trained in the francophone tradition, both Professor Nuyts and I immediately saw our own doctrine of abuse of law in action in Kalfelis. There are several reasons for this. First, in substance, the reasoning very much looked like an application of the doctrine. Second, as a good example of continental abstract and general thinking, fraude à la loi is meant to apply generally to all rules and not to some special cases. Third, francophone Cours de cassation deliver judgments which typically do not say clearly what they mean. As a consequence, a francophone lawyer could easily ignore the fact that the ECJ did not say expressly that it was applying the doctrine of abuse of law.

  Case C-98/06 Freeport plc v Olle Arnoldsson [2007] ECR I-8319.   Kalfelis (n 16 above). 19   Kalfelis (n 16 above), paras [8–9]. 20  Nuyts (n 4 above) 55, 65. 21  See also Muir Watt and Bureau (n 12 above) 215. 17 18

The Discreet Influence of Abuse of Law in International Civil Procedure  285 Now, I do not want to say that Adrian Briggs is wrong, and that my francophone perspective is right. Kalfelis is a judgment from the ECJ, not from a Cour de cassation. There is also something puzzling about Kalfelis from my perspective. The doctrine of fraude à la loi is traditionally used directly as an autonomous corrective device. In Kalfelis, it was used indirectly as a rationale for a rule which was laid down by the ECJ. The reasoning of the Court refers to some doctrine of abuse, but the language of the rule does not.22 It covers one specific case of abuse, and neutralises it by adding a requirement which makes the abuse impossible. In a way, this is surprising, because the next paragraph of the same provision, by contrast, seems to call explicitly for the application of a doctrine of abuse.23 Article 6(2) provides that: A person domiciled in a Member State may also be sued: [...] 2. as a third party in an action on a warranty or guarantee or in any other third party proceedings, in the court seised of the original proceedings, unless these were instituted solely with the object of removing him from the jurisdiction of the court which would be competent in his case. (Emphasis added.)

Again, from the perspective of a francophone lawyer, this provision, which appeared in the 1968 Brussels Convention and was never amended since then, looks like an obvious application of the doctrine of fraude à la loi.24 As such, French lawyers could easily ignore it as stating the obvious: all rules are subject to the application of the general doctrine of fraude à la loi. And Article 6(2) shows that it is fortunately the same in EU law. B.  Freeport In Freeport,25 the ECJ was asked again whether the new Article 6(1) of the Regulation could be used ‘solely in order to have a claim against another defendant heard by a court other than that which would otherwise have had jurisdiction to hear the case’.26 This was much the same question as had been asked in the Kalfelis case. There were, however, two differences between the two cases. First, one of the parties in the Freeport litigation expressly referred to the danger of ‘abuse’27 of jurisdictional rules, and presented its prohibition as a ‘general principle’.28 Second and most importantly, Article 6(1) had been amended in the meantime to incorporate the rule in Kalfelis. In effect, what the ECJ was asked was whether more needed to be done, and indeed should be done. New Article 6(1) already incorporated a tool meant to fight against abuse. Was it necessary to do more? 22   Article 6(1), as amended after Kalfelis when the Brussels I Regulation replaced the Brussels Convention, reads: ‘A person domiciled in a Member State may also be sued: 1. where he is one of a number of defendants, in the courts for the place where any one of them is domiciled, provided the claims are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings.’ (Emphasis added.) 23  Nuyts (n 4 above) 55, 65. 24  See for example, B Audit, Droit international privé, 5th edn (Paris, Economica 2008), para 543; M-L Niboyet and G de Geouffre de la Pradelle, Droit international privé (Paris, LGDJ 2007) 438. See also Metzger (n 14 above). 25   Freeport (n 17 above). 26   Freeport (n 17 above), para [21]. 27   Freeport (n 17 above), para [50]. The reference, however, is only in the French text (‘abuser’), not in the English text (‘misuse’). 28   Freeport (n 17 above), para [50].

286  Gilles Cuniberti The Freeport court refused to do more. It held that the new Article 6(1) applies when the statutory condition (as initially laid down in Kalfelis) is met ‘without there being any further need to establish separately that the claims were not brought with the sole object of ousting the jurisdiction of the courts of the Member State where one of the defendants is domiciled’.29 In other words, it did not follow the party which was asking for a general application of the doctrine of abuse of law in this context. One could therefore think that the unwillingness of the Court to import the doctrine of abuse of law in international civil procedure is clear. Certainly, the case does not reveal the enthusiasm of the Court for such an import. But I would not say that one can conclude from this case that the doctrine has been rejected in the field. To begin with, as already mentioned, the present state of the law (that is, the language of Article 6(1)) is arguably the result of the discreet operation of the doctrine. Furthermore, the reasons given by the ECJ rely heavily on the legislative history of Article 6. The Court underlines that when Article 6(1) was amended, the possibility of adopting a general test, much like the one in Article 6(2), was rejected by the Member States. The Commission put forward that an objective test such as the one laid down by the Kalfelis court had eventually been preferred.30 So, it seems to me that the real rationale behind the Freeport decision is judicial self-restraint. This issue had been addressed by the European legislator. The ECJ saw no reason to contradict the will of the Member States. I believe that Freeport does not show that the Court of Justice rejects the application of the doctrine of abuse of law in the context of international civil procedure. However, Freeport may show that, if the doctrine is influential, it is not a general principle of EU law, and as such, hierarchically superior to European legislation, since the ECJ accepted that European institutions had discretion to mould it. C. Conclusion What do these cases teach us? Contrary to Professor Briggs, I would submit that they show that abuse of law, or at least some form of abuse of law, can have, and has had, a role in international civil procedure. I am of course not submitting that this role has been major. It has been modest, but I would not regard it as insignificant. My conclusion on the present state of the European law of international civil procedure is therefore that there has not been a clear rejection of the principle of abuse of law. Indeed, I think that there are signs that the concept has already been influential and has contributed, albeit modestly, to the crafting of some of the rules in the field.31 This is not to say, however, that I am optimistic about the role that abuse of law could play in the future in this field. Even if I believe that it has not been rejected by the Court of Justice, it is clear that abuse of law has not played as important a role in international civil procedure as it has in other fields of European law. The question I therefore want to ask is why this role has been so limited.

  Freeport (n 17 above), para [54].   Freeport (n 17 above), para [51]. 31  Nuyts (n 4 above) 55, 65; Muir Watt and Bureau (n 12 above ) 223. 29 30

The Discreet Influence of Abuse of Law in International Civil Procedure  287

V. EXISTENCE OF A JURISDICTIONAL OBSTACLE

My sense is that there is indeed an obstacle for the development of abuse of law in the European law of international civil procedure. This obstacle is one of the four reasons that Professor Briggs pointed out as explanations of the silence of the law on the issue under consideration, and it is the one I find the most convincing. This obstacle is jurisdictional.32 Some courts only have jurisdiction to address the issue of whether an abuse occurred. This would not be a problem if such courts were perceived as disinterested adjudicators able to address the issue fairly. But that is not the case. It is felt that they are not the appropriate courts for that matter. Parties would therefore want to raise the issue of abuse of law before another court. But other courts do not have jurisdiction, so it cannot be done. The roots of the problem are to be found in the peculiarity of rules of international civil procedure. Such rules are essentially about the jurisdiction of courts. They grant or deny jurisdiction to the national courts of the European Union. If an issue of abuse of law is raised, it therefore aims at challenging the ruling of a court on its jurisdiction. It is widely admitted that only the relevant court could entertain such a challenge. Comity prohibits courts of one national state from ruling on the jurisdiction of courts of another state, let alone from ordering a foreign court to retain or decline jurisdiction. In the context of European integration, it was added that this prohibition is also grounded in the mutual trust that national courts of the EU ought to owe to each other. I agree with Adrian Briggs33 that we do not need to go that far for our present purposes, although it is how the Court of Justice has consistently put it. It is only necessary to say that courts have exclusive jurisdiction to rule on their own jurisdiction.34 As courts have exclusive jurisdiction to rule on their jurisdiction, they also have exclusive jurisdiction to rule on any allegation of abusive taking or denial of their jurisdiction. The problem is that such courts might not always be in a position to address the issue appropriately, either because they would not be able to do so fairly, or because they would not be able to do so in a timely fashion. An example of a court unable to address the issue fairly can be derived from the facts of the Turner case.35 If a court has retained jurisdiction, it might not want to hear that it did so abusively. It is not fully disinterested. If a court seised second nevertheless retained jurisdiction, it will not be easy to convince it that this ruling was abusive. But the defendant cannot go before the courts of another Member State to seek a remedy for the abuse, because that would amount to a ruling on the jurisdiction of a foreign court, and this would not be acceptable. An example of a court unable to address the issue in a timely fashion can be derived from the facts of the Gasser case.36 If a court is known to take an exceedingly long time to rule on its jurisdiction, this creates incentives for parties to seise it abusively for the sole reason of delaying the proceedings. If such court is seised first, the court which has jurisdiction will have to stay its proceedings. This is the ‘Italian torpedo’. A remedy should be available   Muir Watt and Bureau (n 12 above) 223.  Briggs (n 1 above). 34  It is the consequence of a principle of international law, according to which states are free to organise themselves as they wish, and that they therefore have exclusive jurisdiction to establish administrative bodies (including courts) and decide on the power of such bodies. 35   Case C-159/02 Turner v Grovit [2004] ECR I-3565. 36   Gasser (n 2 above). 32 33

288  Gilles Cuniberti against this abusive practice. The problem is that the only court which has jurisdiction to grant any remedy is the Italian court, as the central element of the alleged abuse is its lack of jurisdiction. And the Italian court will not make such a decision for a very long time. There is, therefore, a severe limitation to the operation of the doctrine of abuse of law in the context of international civil procedure. Because of this limitation, the Court would have had great difficulties ruling on the doctrine in Gasser, as it was asked whether the court seised second could deem abusive the proceedings developing before another court. This is not to say, however, that there is no room for abuse of law in the field. First, the court which has jurisdiction (often the court seised first) to rule on the abuse could do so at some point. Second, there are rules of international civil procedure which do not concern the jurisdiction of courts. For instance, there are rules on service or on recognition of judgments. Those rules may not raise the same issue.

20 ‘Prohibition of Abuse of Law’: A New General Principle of EU Law Jonathan Faull* I.  INTRODUCTORY REMARKS

I

thought initially that imprecise law gave rise to manipulation and abuse, but I understand from what has already been said that very detailed tax rules also seem to create incentives to test up to, and beyond, the limits of interpretation. That said, the problem of European1 law is more often a lack of precision than an excess of detail. The way in which it is made, indeed its very nature, makes it particularly hard to interpret. It is conceived, debated, negotiated, and enacted by a variety of actors, most of whom are working in a foreign language. A key participant in the legislative process, the Permanent Representatives Committee (Coreper), is a committee of ambassadors, skilful international negotiators but not necessarily experts in legal drafting. The same can be said of Parliamentary Committees. The end product of this process is translated from one into many other languages. The efforts of Parliament, Council and Commission legal services are considerable and their role could be enhanced in order to improve the quality of legislation.2 However, the duty of clarity which the legislative institutions of the Union must observe is imperfectly discharged. Too many regulations and directives are enacted thanks to ‘constructive ambiguity’: compromise language chosen in English or French and then translated with a loss of ambiguous nuance into the other languages. The Court of Justice also has a duty of clarity. The Metock judgment3 explicitly overruled a previous judgment in Akrich,4 which was not yet five years old, but failed to

  * This chapter has been written in a personal capacity. I am grateful to Freja Sine Thorsboe for sharing her views on several of these issues. 1   Meaning the European Union [and formerly the European Community]. 2  In another area, asylum law, AG Poiares Maduro has recently commented on Directive 2004/38 (n 6 below) that: ‘the person seeking an interpretation is condemned to try to reconcile what seems prima facie irreconcilable’. AG Poiares Maduro in Case C-465/07 Meki Elgafaji and Noor Elgafaji v Staatssecretaris van Justitie [2009] ECR I-921. 3   Case C-127/08 Blaise Baheten Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I-6241. 4   Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607.

292  Jonathan Faull provide clear reasoning for this reversal.5 Was it a new reading of the Directive?6 In fact, Akrich can be said not to be a reading of the Directive at all. The judgment was handed down on 23 September 2003, months before the adoption of the Directive on 29 April 2004. However, the Council had reached political agreement on the text of the Directive on 23 September 2003, the same day as the Akrich judgment.7 It is not surprising that several Member States relied on Akrich when transposing the Directive into national law. To complicate things further, the Directive is to a considerable degree a codification and simplification of previous legislation and case law. The Court held that: In order to be able to benefit in a situation such as that at issue in the main proceedings from the rights provided for in Article 10 of Regulation No 1612/68, a national of a non-Member State married to a citizen of the Union must be lawfully resident in a Member State when he moves to another Member State to which the citizen of the Union is migrating or has migrated.8

Article 10(1) had provided that: The following shall, irrespective of their nationality, have the right to install themselves with a worker who is a national of one Member State and who is employed in the territory of another Member State: (a) his spouse and their descendants who are under the age of 21 years or are dependants; (b) dependent relatives in the ascending line of the worker and his spouse.

This provision was repealed by Directive 2004/38. Its wording was modernised, but its principle was certainly not abandoned. So can Akrich be said after all to constitute a reading of the current Directive’s essentially unchanged rules? What is Metock if not an interpretation of the Directive? A new application of the Treaty? A new application of fundamental rights law? In the presence of a succinct judgment and the absence of possibly helpful analysis by an Advocate-General,9 the reader is left to speculate on the source of the Court’s new position. This is not only an academic problem. In the world of politics, Metock caused a storm in several Member States and has led to calls for the amendment of Directive 2004/38. In Denmark, in particular, politicians have demanded legislation to ‘cancel Metock’ and ‘return to Akrich’, ie restore the rule that prior lawful residence is required of a person seeking to take advantage of his/her partner’s EU free movement rights to settle in a Member State. But would an amendment of the Directive ‘cancel’ Metock? Perhaps only if Metock is based on an interpretation of the Directive alone, and we do not know if that is the case. I say ‘perhaps’, because the  The Court held in Metock (n 3 above), para [58]: It is true that the Court held in paragraphs 50 and 51 of Akrich that, in order to benefit from the rights provided for in Article 10 of Regulation No 1612/68, the national of a non-member country who is the spouse of a Union citizen must be lawfully resident in a Member State when he moves to another Member State to which the citizen of the Union is migrating or has migrated. However, that conclusion must be reconsidered. The benefit of such rights cannot depend on the prior lawful residence of such a spouse in another Member State (see, to that effect, MRAX, paragraph 59, and Case C-157/03 Commission v Spain, paragraph 28). 6   Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States amending Regulation (EEC) No 1612/68 and repealing Directives 64/221/EEC, 68/360/EEC, 72/194/ EEC, 73/148/EEC, 75/34/EEC, 75/35/EEC, 90/364/EEC, 90/365/EEC and 93/96/EEC [2004] OJ L158/77, and corrigenda [2004] OJ L229/35, [2005] OJ L30/27, [2005] OJ L197/34, and [2007] OJ L204/28. 7   See: ec.europa.eu/justice_home/news/intro/news_230903_1_en.htm. 8   Akrich (n 4 above), para [61]. 9  AG Poiares Maduro’s ‘View’ of 11 June 2008 was published later and is now to be found with the judgment. Metock (n 3 above) 6247. 5

‘Prohibition of Abuse of Law’: A New General Principle of EU Law  293 Directive itself is a codification of legislation and case law rules stretching back to the 1960s, to which the modern label of ‘citizenship’ has been added. When law is imprecise and it is unclear whence it comes among a complex hierarchy of norms, justice is poorly served and political reaction to change is likely to be toxic and directionless. Furthermore, the Metock controversy occurs in a specific, but not unique, legal and political environment, mixing a fundamental internal EU freedom with immigration policy characterised by political controversy, opt-outs, and limited Union competence. Purely domestic or international situations may be regulated more severely than cross-border EU cases, creating incentives to manipulate or abuse and much political consternation to boot. Directive 2004/38 provides in Article 35 that: Member States may adopt the necessary measures to refuse, terminate or withdraw any right conferred by this Directive in the case of abuse of rights or fraud, such as marriages of convenience. Any such measure shall be proportionate and subject to the procedural safeguards provided for in Articles 30 and 31.

It seems surprising that the ECJ in Metock did not consider the notion of ‘abuse’. However, the referring national court had not made the issue central to its case,10 which perhaps explains the ECJ’s silence on the point. To conclude, the embryonic principle of abuse of rights demands more clarity on the part of all EU institutions. More widely, the European legal system’s effectiveness and legitimacy require greater effort to provide clarity and quality in the institutions’ lawmaking functions.

10  The Irish High Court’s judgment seeking a preliminary ruling from the Court of Justice is at [2008] IEHC 77: www.bailii.org/ie/cases/IEHC/2008/H77.html. Ms Justice Finlay Geoghegan held that: 70. The single question upon which I need the preliminary ruling of the Court of Justice to enable me to determine the applications for certiorari and underlying challenge to the validity of article 3(2) of the 2006 Regulations is: Does Directive 2004/38/EC permit a Member State to have a general requirement that a non-EU national spouse of a Union citizen must have been lawfully resident in another Member State prior to coming to the host Member State in order that he or she be entitled to benefit from the provisions of Directive 2004/38/EC? 71. I must respectfully emphasise that, for the reasons set out at paragraphs 43 to 48 above, the question must be answered independently of the immigration history of the applicants except of the fact that the non-EU spouse was not lawfully resident in another Member State prior to arriving in Ireland if it is to be of assistance to me in determining the claims for certiorari in these proceedings. In refusing the three relevant applicants, the Minister only relied on their failure to provide evidence of prior lawful residence in another Member State and not on their personal immigration history in Ireland. Further, the prior lawful residence requirement in article 3(2) of the 2006 Regulations applies, irrespective of the immigration status of the nonEU applicant spouse. Other applications in which leave to seek Judicial Review has been granted by the High Court (some of which have been settled and others not yet heard) demonstrate reliance upon article 3(2) to refuse applications for residence from non-EU national spouses of Union citizens who had been permitted to enter Ireland on a ‘join EU spouse visa’ and other types of visa. 72. In the context of the submissions made on behalf of the applicants during the hearing before this Court, the position of non-EU spouses of Irish citizens is not relevant to this Court’s decision. However, as the Court of Justice took into account in Jia the analogous position under Swedish law it may be relevant for the Court of Justice to be aware that the Minister, through his counsel, informed this Court that applications for residence for non-EU spouses of Irish citizens are considered and determined by the Minister as part of his executive functions. There are no statutory provisions applicable. The Minister has indicated that each application is decided on its own facts. There is no automatic entitlement to residence and there is no automatic exclusion even of non-EU nationals who may have entered or be in Ireland illegally.

21 ‘Abuse of Law’ in the Context of the Free Movement of Workers KATJA S ZIEGLER I.  The Codification of the Prohibition of Abuse in the Free Movement of Workers

A

rticle 35 of Directive 2004/38 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States recognises the concept of abuse of rights explicitly.1 The Directive codifies pre-existing case law, the evolution of which will be considered below. It has also, in turn, been referred to recently by the Grand Chamber of the European Court of Justice in the Metock case of 25 July 2008, where the Court stated that there are limitations of the rights of the Directive on the grounds of public policy, security, health and abuse of rights or fraud, mentioning the classic derogation clauses and abuse of rights or fraud side by side.2 The Directive is applicable to two broad categories of people: economically active persons (workers and the self-employed) and economically independent citizens. The latter category is discussed in the chapter by my colleague, Cathryn Costello.3 Article 35 provides: Abuse of rights Member States may adopt the necessary measures to refuse, terminate or withdraw any right conferred by this Directive in the case of abuse of rights or fraud, such as marriages of convenience. Any such measure shall be proportionate and subject to the procedural safeguards provided for in Articles 30 and 31.

It can be concluded from this provision, firstly, that the principle is recognised in EU secondary legislation in the context of free movement of workers and EU citizens. As will be shown, as is the case in many respects with the Directive, Article 35 is just a codification of case law.4 Secondly, dogmatically, Article 35 of the Directive equates the concept of abuse of rights with a justification for Member States to except or derogate from the application of EU rights in limited circumstances, in much the same way as the 1  Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States amending Regulation (EEC) No 1612/68 and repealing Directives 64/221/EEC, 68/360/EEC, 72/194/EEC, 73/148/EEC, 75/34/EEC, 75/35/EEC, 90/364/EEC, 90/365/EEC and 93/96/EEC (2004) OJ L158/77–123. 2   Case C-127/08 Blaise Baheten Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I-6241, paras [74–75]. 3   See C Costello, ‘Citizenship of the Union: Above Abuse?’, ch 23 below. 4   Recital 3 explicitly states this.

296  Katja S Ziegler exceptions of public policy, public security and public health can justify restrictions of free movement rights.5 This can be inferred from the wording and the fact that abuse of rights is subject to the limitations of proportionality.6 Several observations can be made in this context: Firstly, the concept of abuse in this construction does not modify the content of the right; it is not inherently part of the definition of the right, nor does it remove aspects of a right from the scope of the Treaty.7 Member States are only entitled to ‘refuse, terminate or withdraw’ any right, ie restrict in one form or another the scope of a right that otherwise exists. Secondly, abuse of rights does not rule out the existence of a right per se.8 This reflects a division between the right and its exercise which is not uncommon in a comparative perspective.9 Thirdly, this also means that abuse of rights is not identical with lack of rights because certain conditions of a right are not present. Fourthly, the relationship between the concept of abuse providing an exception for the exercise of a right and the restriction of a right under ordinary derogation clauses (public policy, public security, public health) is uncertain. Exception clauses might severely limit the application of a doctrine of abuse of rights and make the application of an abuse of rights principle superfluous in the individual case or even as a separate category. Fifthly, the relationship between abuse and fraud in the wording of the Directive is not clear. It would appear that fraud refers to a deception about a condition of a right being met, where the condition is not formally fulfilled, such as the forgery of a marriage certificate where there is no marriage and where marriage of two people is the condition for a right.10 Fraud as a category will mainly be relevant where formalities such as certificates or documents are the immediate source of rights. Where they are just pretending to be declaratory of a condition (and in effect lying about a condition of a right), it is possible to deny the right for lack of one of its conditions itself. In contrast, abuse might be taken to refer to the situation where pro forma all requirements are met, giving otherwise rise to a right, but there is a flaw in the motive or purpose. ‘Marriages of convenience’ can hence be considered as an example of abuse of rights (and only of abuse), as a specific example for a motive which is considered illegitimate by the Directive. Finally, abuse of rights in EU law is used here as a concept in a hierarchical relationship, or one that might be called one of subordination: it is not used to restrict one individual right vis-à-vis another individual right at the same level, which is its main application in national legal systems and in international law. Instead, it operates in a vertical dimension between individuals, the EU and the Member States.   Arts 45(3), 36, 52, 52 with 62 TFEU [ex Arts 39(3), 30, 46, 46 with 55 EC].   The ECJ seems to assume this in Metock (n 2 above), paras [74–75].     7   See for the discussion of derogation versus removal from the scope of the Treaty, Case 33/74 Van Binsbergen v Bestuur van de Bedrijfvereniging voor de Metaalnijverheid [1974] ECR 1299, paras [12–13]. Cf Case C-285/95 Kol v Land Berlin [1997] ECR I-3069, paras [24–25] (fraud rules out a right under the Treaty); A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of the ECJ on Circumvention, Fraud and Other Misuses of Community Law’ (2000) 11 European Business Law Review 179, 192.     8   In this vein however, KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 432.     9   See also AG Slynn in Case 39/86 Sylvie Lair v Universität Hannover [1988] ECR 3161, [1989] 3 CMLR 545, 553. 10   Kol (n 7 above), para [24].     5     6

‘Abuse of Law’ in the Context of the Free Movement of Workers  297

II.  The Relevance of the Scope of The Freedom of Movement of Workers

The notion and scope for a finding of abuse of the free movement of workers is directly dependent on the scope of the freedom of movement of workers.11 The correlation has two dimensions: on the one hand, a wide scope of the freedom may reduce the legal scope of abuse, because potential abusive practices squarely fall into the scope of the right and therefore are permitted, although quantitatively the potential for abuse might increase if more actions come under the scope of the right. A narrow scope of the right, on the other hand, may mean that certain practices that might be considered abusive do not come into the ambit of the right in the first place and therefore are not permitted. At the same time the quantitative likelihood of behaviour that falls into a separate category of abuse is reduced. The crucial point therefore seems to be not so much what amounts to abuse and whether abuse is an independent general principle, but which kinds of behaviour are approved or disapproved as a matter of legislative policy decision. These decisions need to be made when drafting (and interpreting) a right and its boundaries.12 They are not qualitatively different from defining what amounts to an abuse. The difference is that the definition of a right is more specific. It does not claim to have such a general effect as a general principle of abuse would have. In the following, some of the cases within the free movement of workers will be considered from the perspective of the concept of abuse of rights. This involves looking at cases that mention abuse as well as cases where the Court of Justice might have referred to abuse but has not done so—often in spite of the fact that the abuse argument had been raised by governments of Member States. Before turning to this, the scope of the concept of a worker and the relevance of being thus classified will be recapitulated.

A.  Scope of ‘Worker’ A ‘worker’ is defined as someone who pursues or desires to pursue for a certain period of time services for and under the direction of another for remuneration (Lawrie-Blum).13 Each of these conditions has been interpreted very broadly. All ‘effective and genuine activities,14 to the exclusion of activities on such a small scale as to be regarded as purely marginal and ancillary’15 qualify as services; this includes part-time work,16 work rendered

  Cf also Sørensen (n 8 above) 427–30.   In the same vein, AG Tesauro in Case C-367/96 Alexandros Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I-2843, para [25]; Kjellgren (n 7 above) 191–92. 13   Case 66/85 Deborah Lawrie-Blum v Land Baden-Württemberg [1986] ECR 2121, para [17]. 14   It has not decided for immoral or illegal activity in the context of workers, but in the affirmative for the freedom of establishment, Case C-268/99 Jany v Staatssecretaris van Justitie [2001] ECR I-8615. 15   Case 53/81 Levin v Staatssecretaris van Justitie [1982] ECR 1035, para [17]. The main objective of following a rehabilitation programme, however, was found to make the activity merely ancillary, see Case 344/89 Bettray v Staatssecretaris van Justitie [1989] ECR 1621, paras [13], [17]; see also Case C-456/02 Trojani v Centre public de l’aide sociale de Bruxelles (CPAS) [2004] ECR I-7573, paras [15], [18]. 16   Levin (n 15 above), para [17]. 11 12

298  Katja S Ziegler in a family context,17 in a preparatory or training relationship18 or employment that is financed by public funds, such as a programme that facilitates the employment of persons who otherwise would be recipients of social assistance;19 an activity with a non-economic aim, such as in the fields of culture or sport is sufficient.20 Furthermore, the worker cannot be required to reside in the host State.21 Neither the short duration of the economic activity in relation to the total duration of the stay,22 nor the fact that the worker is employed and paid on an on-call basis,23 rule out the status as a worker in themselves, but the duration and regular or irregular nature may be taken into account when considering whether the activity is effective and genuine.24 Further, the fact that the remuneration obtained from the employment falls below the minimum wage or needs supplementing by social assistance is irrelevant, as long as the activity is not purely marginal and ancillary.25 B. Relevance of Being a Worker (Worker versus EU Citizen) Until the introduction of Union citizenship with the Treaty of Maastricht in 1992, the question whether an EU national qualified as a worker was decisive, both for the right to stay and for other rights derived automatically from the status of a worker, such as rights for third country national (TCN) family members and rights to social benefits. It is in these two factual scenarios that the case law around abuse of rights in the context of the free movement of workers arises. After the introduction of Union citizenship and through secondary legislation, most notably Directive 2004/38, the residence rights of family members no longer depend on the status as a worker, but merely on the status as EU citizen of the person from whom the rights are derived (although generally a cross-border movement is necessary). With the expansion of rights to non-economically active EU citizens, more cases that might have been discussed as abuse of rights scenarios clearly become ‘legitimate’ rights where raising a possible abuse argument is out of the question. Hence the notion of abuse is reduced even further to a mere vanishing point. The recent case of Metock expanded the constituency of eligible family members of EU citizens and hence of rights holders under EU law even further.26 17   Case C-337/97 CPM Meeusen v Hoofddirectie van de Informatie Beheer Groep [1999] ECR I-3289, paras [14–15]. 18   Lawrie-Blum (n 13 above), para [19]; Case C-3/90 Bernini v Minister van Onderwijs en Wetenschappen [1992] ECR I-1071, paras [14–15]. 19   Case C-1/97 Birden v Stadtgemeinde Bremen [1998] ECR I-7747, paras [25–31], esp [28]. 20   Case C-415/93 Union Royale Belge des Sociétés de Football Association ASBL et al v Jean-Marc Bosman et al [1995] ECR I-4921. 21   Case C-350/96 Clean Car Autoservice GmbH v Landeshauptmann von Wien [1998 ] ECR I-2521; see also Case C-212/05 Hartmann v Freistaat Bayern [2007] ECR I-6303, paras [15–19]—even if it means relying on worker status against the home State. 22   Sylvie Lair (n 9 above), paras [40–44]; Case C-413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I-13187, para [25]; cf also Case C-209/03 The Queen, on the application of Dany Bidar v London Borough of Ealing and Secretary of State for Education and Skills [2005] ECR I-2119, paras [49–63], esp [61] (regarding length of stay of Union citizens). 23   Case C-357/89 VJM Raulin v Minister van Onderwijs en Wetenschapen [1992] ECR I-1027, paras [9–11]. 24   Raulin (n 23 above). 25   Levin (n 15 above), para [17]; Case 139/85 RH Kempf v Staatssecretaris van Justitie [1986] ECR 1741; Bettray (n 15 above), paras [13], [17]; Raulin (n 23 above), para [15]; Ninni-Orasche (n 22 above), para [32]. 26   Metock (n 2 above), paras [48–80], esp [54], [58], [67], [80].

‘Abuse of Law’ in the Context of the Free Movement of Workers  299 Furthermore, when the ECJ in 2001 in the case of Grzelczyk characterised citizenship as ‘destined to be the fundamental status of nationals of the Member States’27 and started to derive rights other than the right to merely stay in the territory of another Member State from that status through the right of non-discrimination,28 non-workers (‘mere citizens’) could also under certain circumstances be entitled to certain social benefits. Hence the categorisation as a worker became somewhat less important. However, in order to derive access to a benefit from the principle of non-discrimination under the status of citizen, the refusal of the benefit must amount to a disproportionate restriction of the residence right flowing from Union citizenship. Relying merely on the status as EU citizen therefore has disadvantages related to the substantive conditions and possibly the procedural consequences: substantively, the lack of proportionality will normally require a certain manifestation of attachment to the host State—time and degree of integration are considered to be relevant factors by the Court. Procedurally, this hinging of (aspects of ) a substantive right on proportionality, which by its very nature requires a balancing exercise, necessitates an extremely individualised decision-making process. This makes outcomes unforeseeable and leads to corresponding transaction costs—added time required for an individualised assessment and a greater likelihood of litigation over the ‘right’ balancing of the relevant factors.29 For these reasons, it is still much easier to invoke the status of a worker, which automatically entails equal treatment, and especially the same social and tax advantages as nationals according to Article 7 of Regulation 1216/68. Hence there is extensive case law around social benefits, much of it concerning student maintenance grants or loans, and around residence rights of TCN family members of a person claiming to be a worker. III.  Evolution of Case Law of Abuse in the Context of Free Movement of Workers

Where the criteria of a worker are met, two factual scenarios give rise to the claim of abuse. The first relates to an objective ‘time-or-scale element’: if someone has been a worker, but only for a short time or only in an employment on a small scale, it may appear to be ‘disproportionate’ to grant further rights. The time element may apply to ongoing employment situations. More frequently it will be problematic in situations where the employment lies in the past, but a link to the current activity—or inactivity in case of involuntary unemployment—can be made.30   A formulation picked up in Recital 3 of Directive 2004/38 (n 1 above).   Art 18 TFEU [ex 12 EC]; Case C-184/99 Rudy Grzelczyk v Centre public d’aide sociale d’OttigniesLouvain-la-Neuve [2001] ECR I-6193, para [31]. 29   See in regard to the conflict between the generality of a rule (with its practical consequence of reduced individualised decisions) and a ‘just’ decision in an individual case: AG Mazák in Case C-388/07 The Incorporated Trustees of the National Council on Ageing (Age Concern England) v Secretary of State for Business, Enterprise and Regulatory Reform [2009] ECR I-1569, para [54], referring to the fact that lack of the possibility to create general and typified grounds of exception would ‘unduly restrict the scope of justification’, and paras [68–83]; see now the ECJ’s turn in Case C-158/07 Jacqueline Förster v Hoofddirectie van de Informatie Beher Groep [2008] ECR I-8507, paras [53–60]; see also Sørensen (n 8 above) 453–54; cf in a similar vein the criticism voiced against the individualised proportionality test regarding patients’ rights following Case C-372/04 The Queen ex p Yvonne Watts v Bedford Primary Care Trust and Secretary of State for Health [2006] ECR I-4325. 30   See today Art 7(3) Directive 2004/38 (n 1 above); Case C-85/96 María Martínez Sala v Freistaat Bayern [1998] ECR I-2691, para [32]; AG Geelhoed in Ninni-Orasche (n 22 above), para [45]; Case 197/86 Steven Malcolm Brown v The Secretary of State for Scotland [1988] ECR 3205. 27 28

300  Katja S Ziegler The second scenario relates to a subjective ‘motive element’: where a worker applies for a social benefit after having become a worker, but where it might be suggested that he or she entered the Member State already with the purpose of deriving a further right from worker status, be it the claim of a benefit or the right of a family member, the question is raised whether this amounts to abuse of the right of free movement. As will be shown, the Court does not reject the possibility of abuse outright, but acknowledges that an abuse of the free movement right is possible in principle. It relies on the objective ‘time’ or ‘scale’ element in this respect. However, the objective ‘time’ or ‘scale’ element which gives rise to the concern of an abuse of rights, is only relevant for the question whether there is a genuine and effective economic activity which is already a condition for being a worker, ie the test is one of effectiveness. In contrast, the Court rejects the relevance of motive outright today, although not all cases in the past are consistent in this respect. From a legal perspective, the notion of abuse of rights in the free movement of workers, is, therefore, construed very narrowly, leaving virtually no space for its operation (with the only exception being marriages of convenience codified in Article 35 of Directive 2004/38). Nevertheless, the Court has held that the assessment whether specific facts amount to abuse is one for the national courts to make. It also has regularly given very specific guidance as to what does not amount to abuse, but the use of a right. Further, due to a wide interpretation of who is a worker and the automatism of consequential rights derived from that status, it seems impossible to affirm abuse without recourse to motive.31 A family member is not a ‘true’ family member if the marriage is one of ‘convenience’. The marriage is validly concluded, but it is the motive that is considered objectionable by the law. In contrast, fraudulent behaviour implies manipulating the factual pre-conditions of a right, through deception: a worker would not be a ‘true’ worker if the parties conspire to conclude an employment contract only on paper to trigger immigration rights and no services are exchanged for remuneration. However, even without resort to the notion of fraud, worker status could already be ruled out for lack of a substantive condition. In the following, some of the cases concerning the claim of benefits derived from the status of a worker and concerning derived residence rights of TCN family members will be considered in order to sketch out the approach of the Court. A.  Benefits Cases i.  Levin 1982: ‘effective and genuine’ activity A case that would probably be obsolete today under the concept of Union citizenship was Levin.32 A couple consisting of a British wife and a South African husband with sufficient own resources tried to derive their resident right in the Netherlands through the wife’s (EU national) part-time employment as a chambermaid which only earned her a low income at what was thought to be below subsistence level. 31   See also text at n 10 above. In the same vein Kjellgren (n 7 above); D Triantafyllou, ‘Abuse of Rights versus Primacy?’ (1999) 36 Common Market Law Review 157, 190. For examples in the case law see: Kol (n 7 above), paras [24–25] (residence permit based on false statements regarding marriage to a German woman does not give rise to independent residence right after the relevant period of residence); see also the pending case Case C-303/08 Metin Bozkurt v Land Baden-Württemberg (Reference for a preliminary ruling from the Bundesverwaltungsgericht (Germany) lodged on 8 July 2008). See also text at n 10 above. 32   Levin (n 15 above).

‘Abuse of Law’ in the Context of the Free Movement of Workers  301 It is significant that the Court rendered its judgment on the basis of the conditions of the right to free movement of workers and did not even discuss the case in terms of abuse of a right. The beneficiary of the free movement right and the connected rights must be a worker. At issue then was the question whether the level of income thus generated was decisive or sufficient for meeting these criteria. The Court adopted its much cited functional approach, stating that ‘effective and genuine activities, to the exclusion of activities on such a small scale as to be regarded as purely marginal and ancillary’33 qualify as activities of a worker. What is ‘effective and genuine’ goes to the scope of the right under what is now Article 45 TFEU [ex Article 39 EC]. Actions that might be considered abusive would in a binary way be either (still) a right or not (anymore), thus leaving no scope for a separate concept of abuse. Importantly, the ECJ stated further that the motive was irrelevant for any of the conditions of what is now Article 45 TFEU [ex Article 39 EC] where a genuine activity is pursued.34 It thus does not even mention the concept of abuse. In Bettray the Court held that the condition of ‘effective and genuine economic activity’ was not fulfilled by an employment that constituted merely a means of rehabilitation or reintegration for the persons concerned and the purpose of paid employment, which is adapted to the physical and mental possibilities of each person, is to enable those persons, sooner or later to recover their capacity to take up ordinary employment.35

Although the Court regarded the purpose of the activity (rehabilitation) in this case, it is not the purpose alone which is decisive. The decision is based on an overall assessment of the nature of the activity which was shaped by adjustments made to the activity to achieve rehabilitation. Likewise, the Court has held in Brown that where the status of worker is exclusively the result of having been accepted to a university course, ie he would not have been employed otherwise and the activity thus is in fact part of the course, this amounts to an activity merely ancillary to the studies.36 ii.  Lair 1988: ‘sole purpose’ The Court referred obiter dictum in Lair to the possibility that an abuse of rights may limit the free movement of workers in order: [T]o prevent certain abuses, for example where it may be established on the basis of objective evidence that a worker has entered a Member State for the sole purpose of enjoying, after a very short period of occupational activity, the benefit of the student assistance system in that State, it should be observed that such abuses are not covered by the Community provisions in question.37

This was, however, not held to be the case on the facts. Ms Lair had been employed continuously for two-and-a-half years in Germany, followed by a further period of more than three years of alternate periods of unemployment, retraining and employment. The period in employment was sufficiently long to be qualified as genuine and hence for her to   Levin (n 15 above), para [17].   Levin (n 15 above), paras [21–22].   Bettray (n 15 above), para [17]. In Trojani the Court referred the assessment of the precise nature of his work in a reintegration programme with the Salvation Army back to the national court, Trojani (n 15 above). 36   Brown (n 30 above), paras [27–28]. 37   Lair (n 22 above), para [43]. 33 34 35

302  Katja S Ziegler acquire and retain the status of a worker.38 A German statute requiring her to have been employed for at least five years prior to being entitled to a student maintenance grant therefore imposed an unjustified restriction of the free movement right. She therefore had a right to the social benefit. It may be observed that the reference to the purpose of entering the Member State in Lair does not fit comfortably with the supposed irrelevance of the motive in Levin.39 Or rather, there may be two types of motive, as reflected by Advocate-General Slynn in his Opinion in Lair.40 The first type of motive is ‘collateral’ and irrelevant. These are motives which are behind exercising a ‘genuine and effective’ job (Levin) which satisfies the necessary characteristics of an employment relationship (Lawrie-Blum): If he is there as such a worker the collateral intentions behind his going (e.g. that he wants his wife and children to be in a particularly agreeable area or near to a particular educational institution) are irrelevant.41

The second type of motive is relevant because it rules out the genuineness of the status of worker: But if he goes there not genuinely in the capacity of a worker, but, e.g. in order to become a student or to gain a short, useful experience before his studies begin, then it does not seem to me that he is to be regarded as a worker for the purposes of Article 7(2) and (3) of the regulation, even if during that period he is doing genuine and effective work which satisfies the test in LawrieBlum. Rights under those provisions are given only to persons in a Member State genuinely in the capacity of a worker.

Advocate-General Slynn considers a period that ‘cannot reasonably exceed one year’ as the practical test period within which motives might be considered as doubtful, but that this period may be shorter where there are no doubts about the genuine nature of the activity. The case is illustrative in that it reveals both the explicit reference to abuse of the free movement of workers, but also the accommodation or absorption of a potential abuse situation in the definition of the concept of a worker. In doing so, the Court does not support a freestanding exception of abuse of rights in the free movement of workers. Furthermore, in construing the concept of a worker broadly, the scope for scenarios that may potentially be considered an abuse is reduced. The cases Raulin,42 Ninni-Orasche and Förster43 resulted from factually similar situations, namely short-term employments followed by a claim of social benefits. iii.  Raulin 1992 Raulin concerned an employment of a French national in the Netherlands on an on-call contract where the overall number of hours of work was not determined beforehand. There was no obligation to follow each call of the employer, and in the end Ms Raulin had 38   The Court considered the former employment sufficiently linked with the ‘occupational retraining’. Lair (n 22 above), para [37]. 39   Likewise AG Geelhoed in Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607, para [103]. 40   AG Slynn in Lair (n 9 above) 553–54 (CMLR). 41   See also AG Slynn in Levin (n 15 above) 1061. 42   Raulin (n 23 above). 43   Förster (n 29 above).

‘Abuse of Law’ in the Context of the Free Movement of Workers  303 only worked for sixty hours as a waitress in the first two weeks of the eight-month on-call contract. The Court, following the Advocate-General, refers the case back to the national court with the guidance that the on-call contract does not a priori exclude that someone is a worker. While on-call availability was one factor in the assessment whether the activities were effective and genuine or purely marginal and ancillary, the duration of the activities was also a relevant aspect. The Court does not even mention abuse. Advocate-General van Gerven, while acknowledging the legal concept of abuse in his Opinion, dismisses it without further reasoning as the case clearly did not involve an abuse.44 iv.  Ninni-Orasche 2003 In Ninni-Orasche the referring national court picked up on the obiter dictum in Lair. The Austrian administrative court had to decide on a claim of an Italian national to a student benefit which it considered to be abusive. Mrs Ninni-Orasche was married to an Austrian and had resided in Austria for three years, but had only been employed for the short period of two-and-a-half months before applying for the benefit. The Austrian court subsumed these facts under the ECJ’s ‘sole purpose’ test, namely that where a national from another Member State enters ‘for the sole purpose of enjoying, after a very short period of occupational activity, the benefit of the student assistance system in that State’, this would amount to abuse not covered by EU law.45 The ECJ, referring to the effective and genuine nature of the activity,46 held that Mrs Ninni-Orasche was a worker and retained that status, and that ‘factors relating to the conduct of the person concerned before and after the period of employment are not relevant for establishing the status of worker’.47 The ECJ held that it was for the national court to assess on the facts whether Mrs Ninni-Orasche only became a worker in order to claim the benefit, but gave strong ‘guidance’ that there was no abuse, pointing to several factors, especially that she entered the country to live with her husband and that she had unsuccessfully searched for a new job when her temporary employment ended.48 It may be noted that the Court shifted in its reply to the national court from the motive for taking up employment to the motive for entering the country, which widened the scope of the relevant alternative motives—including conjugal life. It thus was able to ‘guide’ the national court in clearer terms as to the absence of an exclusive motive to gain a right to a benefit. v.  Förster 2008 Similarly, Advocate-General Mazák discussed the concept of abuse in his Opinion of 10 July 2008 in the case of Förster. Jacqueline Förster, a German from the Dutch border region, moved to the Netherlands to commence her studies there. She also worked alongside her studies for a period of more than three years, amongst others as a school teacher, during which she qualified as a worker. She then ceased her employment and made an application for a student benefit when she needed to focus on finishing her   AG van Gerven in Raulin (n 23 above), para [9].   Ninni-Orasche (n 22 above), para [36]. 46   Ninni-Orasche (n 22 above), para [26]. 47   Ninni-Orasche (n 22 above), para [28]. 48   Ninni-Orasche (n 22 above), paras [45–47]. 44 45

304  Katja S Ziegler studies.49 The benefit was refused by the Dutch authority on the grounds that she was a) no longer a worker and b) had not fulfilled the residence requirement of the relevant Dutch statute of five years to become eligible for the benefit. Advocate-General Mazák examines the continuing effect of her status as a worker, relying on Lair, Brown, Raulin and Ninni-Orasche and stresses that the continuity between work and studies should ‘not be interpreted too strictly in order to avoid excluding a substantial part of working students from the benefits of their rights as Community workers despite the fact that they have already been economically active and have entered the employment market of the host Member State’.50 The Advocate-General acknowledges that to determine the link and continuity of the studies with the previous employment and hence the continuing effects of having been a worker is a question of fact, and therefore for the national court to decide. However, he clearly states, followed insofar by the ECJ,51 that even if the motive of coming to the Netherlands was ‘principally to study’,52 this does not preclude meeting the objective criteria of a worker. He does not see evidence of abuse in the light of the facts that there was a substantial employment relationship for three years and a personal relationship with a Dutch citizen. So it could not be said that entry to the country (nor taking up the employment in the first place for that matter) occurred with the sole purpose of enjoying the student benefit.53 Although the Court does not follow the result of the AdvocateGeneral’s Opinion neither with regard to the continuing effect of the status of worker,54 nor with regard to the disproportionality of a five-year residence requirement,55 it follows him in dismissing the relevance of the motive56 and hence abuse. B.  Family Members Cases In the context of residence rights of family members derived from a worker, three cases deserve mention: Singh, Akrich and Metock. i.  Singh 1992 Singh was the first case in which the ECJ ruled on the question whether a residence right of a family member of a worker under EU law can be claimed against the state of nationality of the worker upon return. This case implied that a TCN family member who did not have a right of residence in the home State of the EU worker (but who stayed legally in another Member State) could return with the worker to the latter’s State of nationality. Otherwise the worker would be deterred from and hindered in the exercise   Cf Grzelczyk (n 28 above) which was, however, decided under the citizenship rules of the EC Treaty.   AG Mazák in Bettray (n 15 above), paras [85] and [121]. 51   Förster (n 29 above), para [44]—albeit in regard to free movement rights under the citizenship rules. 52   AG Mazák in Förster (n 29 above), paras [80–81]. 53   AG Mazák in Förster (n 29 above), paras [86–87]. 54   The ECJ does not follow the Advocate-General’s argument that there was a sufficient link between the employment and the course of studies, denying Mrs Förster the possibility to rely on the continuing effect of having been a worker. 55   AG Mazák at paras [128–30]. The ECJ departs from Bidar (n 22 above), paras [61], [63]) in finding that a generalised five-year residence requirement is restricting the EU citizens’ residence proportionately, Förster (n 29 above), paras [53–58]. 56   Förster (n 29 above), para [44]—albeit in regard to free movement rights under the citizenship rules. 49 50

‘Abuse of Law’ in the Context of the Free Movement of Workers  305 of his or her free movement right. Advocate-General Tesauro acknowledged the need to prevent evasion of national immigration rules.57 The ECJ made only a passing reference to abuse, although it did not find it to be present on the facts of the case.58 ii.  Akrich 2003 Akrich concerned a British-Moroccan couple who had explicitly declared to have gone and worked in Ireland in order to trigger immigration rights under EU law when they eventually would return to the UK. UK immigration Regulations explicitly excluded a residence right for family members in the return scenario where the UK national left the UK ‘in order to acquire rights under those regulations and thereby to evade the application of United Kingdom immigration law’. This rule clearly seems to be intended to tackle ‘abuse’.59 Recalling the formulation in the student benefits cases, it could be argued that the couple went to Ireland ‘for the sole purpose of enjoying, after a very short period of occupational activity, the benefit of’ a residence right in the home State.60 The scenario differs from the student benefits cases only in that the benefits are claimed in the host State, whereas the residence right is claimed against the home State of the worker. The ‘return’ constellation was already ruled to give rise to rights under EU law in principle in Singh. The Court therefore needed to address the relevance of the express motive. It rejected outright the abuse in a few paragraphs, because the motives which may have prompted a worker . . . to seek employment in another Member State are of no account as regards his right to enter and reside in the territory . . . provided that he there pursues or wishes to pursue an effective and genuine activity.61

Following Advocate-General Geelhoed, whose discussion of abuse in the context of free movement of workers counts among the more detailed discussions of the question,62 the ECJ considered the scenario one of use, not abuse of EU Law, confirming that the motives for the exercise of a free movement right are irrelevant.63 It explicitly describes, however, marriages of convenience as an example of abuse and hence as an exception to the irrelevance of the motive.64 The dual test of a possible general prohibition of abuse requiring a) objective circumstances and b) the intention to obtain an advantage from the EU rules by artificially creating the conditions laid down for obtaining it, which was developed in Emsland-Stärke in the context of export refunds,65 seems to be abandoned66— 57   AG Tesauro in Case C-370/90 The Queen v Immigration Appeal Tribunal and Surinder Singh ex p Secretary of State for the Home Department [1992] ECR I-4265, para [14]. 58   Singh (n 57 above), para [24]. 59   Regulation 11(1) of the EEA Regulations 2000, cited in Akrich (n 39 above), para [28]. 60   Cf representations of the UK in Akrich, AG Geelhoed in Akrich (n 39 above), para [170]. 61   Akrich (n 39 above), para [55]. 62   AG Geelhoed in Akrich (n 39 above), paras [96–105], [172–85]. 63   Akrich (n 39 above), para [56]. It may be pointed out that AG Geelhoed in the case of Case C-1/05 Jia v Migrationsverket [2007] ECR I-1, para [75] refers to the Akrich scenario in a factual sense as one of abuse. AG Poiares Maduro seems to adopt a more far-reaching approach to abuse in Metock (n 2 above), para [14] where the Akrich case is described as one of abuse of national immigration legislation. 64   Akrich (n 39 above), para [57]. 65   Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, paras [52– 53]. 66   Cf AG Geelhoed in Akrich (n 39 above), paras [99–105].

306  Katja S Ziegler as previously in Centros.67 If there is a genuine exercise of an economic activity as defined by the Court of Justice, its preconditions cannot at the same time be created artificially. However, the absence of abuse of the free movement of workers still remained only a second leg of the Court’s argument, if not an obiter dictum. Although the motive of working in Ireland did not rule out that Mrs Akrich pursued an effective and genuine activity (test from Levin), Mr Akrich as the spouse could not derive a right to first immigration into the Union from the freedom of movement of workers which then presumed prior lawful residence in a Member State in the EU.68 iii.  Metock 2008 The requirement of prior lawful residence of a TCN family member was explicitly overturned by the Court in Metock.69 The Court70 confines itself to saying that there are limits to the now even wider scope of application of the free movement of workers and citizens rules, limits in the form of the normal grounds of justification and abuse of rights or fraud.71 However, on the facts of the case, the marriages were not sham marriages. It remains to be seen whether this broadening of the residence rights of TCN family members leads to the obsolescence of the principle of abuse of rights in the area or, conversely, to a greater use of it. IV.  Reasons for the Court’s Reluctance to Find an Abuse

The case law discussed above suggests that the principle of abuse does play a role in the free movement of workers, but that its role is reduced to merely verbal acceptance as a legal principle. Although some decisions refer the factual side of the question whether there is an abuse back to the national court, the Court generally has not applied or given scope for the prohibition of abuse—with the exception of codified expressions of the principle of abuse of rights. This is the result of two features of the Court’s case law on free movement of workers. On the one hand, the Court of Justice merges abuse with the conditions of the free movement right. On the other hand, that very right is given such importance that it outweighs any allegations of abuse which are made by the Member States. The scope for an abuse principle seems to be limited to the specified, defined (and doubtful) case of marriages of convenience. This eliminating approach to the concept of abuse may not be a bad thing. In the light of suggestions that the Court is—supposedly inappropriately—reluctant to apply a general principle72 of abuse in the area of free   Case C-212/97 Centros v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459, para [27].   Although in effect, as is well known, the immigration right was in the end derived from human rights as general principles of EU law, in this case the right to family life as expressed in Art 8 ECHR, Akrich (n 39 above), para [58], thus in contrast to the overall result of the Advocate-General’s Opinion. 69   Metock (n 2 above), para [58]. 70  Decided under the new fast-track procedure. 71   Metock (n 2 above), para [75]. 72   It might be added that in light of the willing finding of a general principle of the prohibition of age discrimination in the Mangold case (Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9981) that the reluctance is not due to general reservations in this respect. For the difficulties in determining general principles see M Herdegen, ‘General Principles of EU Law: The Methodological Challenge’ in U Bernitz, J Nergelius and C Cardner (eds), General Principles of EC Law in a Process of Development (Alphen aan den Rijn, Wolters Kluwer 2008) 343. 67 68

‘Abuse of Law’ in the Context of the Free Movement of Workers  307 movement of workers and that this is out of line with other areas of EU law,73 the reasons for this reluctance deserve further reflection. Firstly, there may be doubts about the existence or at least application of the notion of abuse of rights as a general legal principle. The principle of abuse would be merely a descriptive umbrella term for certain factual scenarios without independent legal significance.74 Or, more modestly, the difficulty of defining the vague notion leads to a reluctance of the Court to apply it where people’s movement rights are at stake. Secondly, an assumed general principle of abuse of law would require concretion. Legal significance would result only from area-specific codification or casuistic concretion of the principle, yet such concretion is difficult and runs counter to the rationales of integration and effectiveness of EU law where free movement of persons is concerned. Thirdly, there may be policy considerations of the Court not to apply the concept of abuse of law in the context of free movement of persons in order to further integration through free movement, citizenship and the constitutionalisation of the EU legal order. Constitutionalisation implies the need for a greater concern for the rule of law and protection of human rights. Finally, as a general principle, abuse of rights is at most used to fill gaps in the law. Hence its application would be exceptional in any case. Where other limitations exist or can be construed by interpretation, it is not necessary to resort to the notion of abuse. A.  Dogmatic Rationale and Definitional Problems: What Does ‘Abuse of Law’ Really Mean? Advocate-General Geelhoed rightly identifies three difficulties with applying the concept of abuse of EU law which can be classified as problems related to the delimitation of the concept. Firstly, abuse of law is inherently linked to subjective intentions, which, in his view, serve no purpose in and therefore are not relevant to the free movement of persons.75 People act for a variety of purposes and the exercise of free movement will hardly ever be an aim in itself or for purely economic motives. The Member States themselves acknowledged this by including the citizenship provisions in the Treaty. In addition, there would be significant evidentiary problems in establishing motives and intentions. The explicit ‘confession’ of motives in Akrich is not the ‘ordinary’ scenario. Motives are usually difficult to determine—and subject to manipulation.76 The inconsistency of the case law with regard to subjective motives seems to confirm this. Secondly, a test for abuse based on objective criteria, such as minimum residence periods, can be opposed on the basis that such criteria are subject to circumvention.77 Where the line between use and misuse is a fine one, this may be a practically convincing argument. Furthermore, dogmatically, the need to define conditions and limits of rights is not a 73   R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 411. 74   LN Brown, ‘Is There a General Principle of Abuse of Rights in European Community Law?’ in D Curtin and T Heukels (eds) Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers, Vol II (Dordrecht, Martinus Nijhoff 1994) 511; Kjellgren (n 7 above) 192. 75   AG Geelhoed in Akrich (n 39 above), para [173]. 76   AG Geelhoed in Akrich (n 39 above), para [174]. 77   AG Geelhoed in Akrich (n 39 above), paras [175–76].

308  Katja S Ziegler special feature of a concept of abuse. But it is difficult to distinguish objective criteria for abuse from the conditions of a right in the first place. Therefore the relevance of an independent legal concept of abuse may be questioned. It appears as a ‘contradiction in terms’ in a binary decision-making situation in which someone either acts lawfully because he or she uses a right, or exceeds the right and therefore acts unlawfully.78 The third of Advocate-General Geelhoed’s reasons is related: it is nigh impossible to draw the dividing line between use of a right (or EU law) for a purpose for which it was not foreseen and misuse of EU law79 where the unforeseen purpose directly results from the exercise of a freedom, or in the English wording of Centros, where it is ‘inherent in the exercise, in a single market, of that freedom’:80 The installation of Mr and Mrs Akrich in Ireland must be viewed as a use of EC law for a purpose not contemplated by the EC legislature but which is inherent in EC law. The EC legislature did not intend to create a right that can be used in order to evade national immigration laws but did create a right in favour of a national of a Member State to install himself in another Member State together with his spouse. Installation in that other Member State constitutes the key element of the freedom given by Community law to nationals of the Union.81 (Emphasis added.)

The Advocate-General raises the question whether in other cases the concept of abuse has been applied more successfully.82 In the case of TV10 the Court has restricted the exercise of a right directly following from the freedom of establishment to prevent precisely those national regulations, aiming at preserving cultural pluralism in the media, from being circumvented. However, the Court achieves this by finding that the activity falls within the scope of a right and by applying the justification of a restriction83 ‘designed to ensure the pluralist and non-commercial content of programmes’.84 It thus follows the ordinary operation of market freedoms in the Treaty, justifying a restriction through the exception clauses or mandatory requirements without needing to rely on a separate concept of abuse.85 The term ‘abuse’ then is reduced to a merely factual description of the behaviour rather than a legal principle in this case. The development of area-specific derogations based on specified instances of circumvention or abusive practices, rather than a vague reference to abuse, seems to be in line with not applying ‘abuse’ as a general principle. In the area of free movement law, the codified abuse in the form of marriages of convenience in the context of TCN family members in Article 35 of Directive 2004/3886 provides such an example. Even though this specific example is questionable from a substantive perspective, considering the need to enquire into the motives of a marriage and the difficult or even arbitrary lines that have to be drawn, the underlying principle of specifying in positive law what amounts to an illicit 78   M Planiol, Traité élémentaire de droit civil, Vol 2 § 978, 2nd edn (Paris, LGDJ 1947) cited by R O’Sullivan, ‘Abuse of Rights’ (1955) 8 Current Legal Problems 61. 79   AG Geelhoed in Akrich (n 39 above), para [178]. 80   Centros (n 67 above), para [27] (in German: ‘unmittelbar’ aus der Niederlassungsfreiheit folgend); AG Geelhoed in Akrich (n 39 above), paras [179–81], esp [180]. 81   AG Geelhoed in Akrich (n 39 above), para [180]. 82   AG Geelhoed in Akrich (n 39 above), para [98]. 83   Cf G Dahm, J Delbrück and R Wolfrum, Völkerrecht, Vol I/3 § 171, 2nd edn (Berlin, Walter de Gruyter 1989) 849–50 for the restriction approach to the principle of abuse of rights. 84   Case C-23/93 TV10 v Commissariaat voor de Media [1994] ECR I-4795, paras [13], [16] and [21]. 85   See also Case C-148/91 Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487, para [13]. 86   The ‘marriages of convenience’ rule allows exceptions from the right under EU law where a specific motive is present, cf also Sørensen (n 8 above) 431.

‘Abuse of Law’ in the Context of the Free Movement of Workers  309 motive which slips through the net of the conditions of the free movement right, points in the right direction. In the end, deciding what amounts to an abuse of a right is a value judgment. What are relevant and what are irrelevant motives behind free movement? The principle is therefore dependent on concretion and development for specified groups of cases. Whether this happens also depends as much on the willingness of the Court as a matter of policy as on the doctrinal questions. This might explain the variances between different areas of EU law. B.  The Integrationist or Effectiveness Rationale The general approach to interpreting free movement rights of individuals widely and exceptions narrowly, coupled with the teleological interpretation guided by considerations of ‘effectiveness’, seems to be one reason why the Court has rejected the relevance of the motive outright in its more recent case law. It appears to refer to the principle of abuse only in passing, as if to placate Member States.87 To ‘strengthen the right of free movement and residence’ and hence an integrationist rationale, is also the declared aim of Directive 2004/38.88 The fact that far-reaching rights of family members are derived from a free movement rationale in Metock,89 in which the Court emphasised the ‘importance to eliminate obstacles to the exercise of fundamental freedoms’90 demonstrates this. The integrationist rationale is well summarised by Advocate-General Geelhoed in his Opinion in Akrich: ‘It is precisely the objective of Community Law to promote mobility’.91 The ‘prevention of misuse cannot give rise to limitations of fundamental freedoms which are widely interpreted by the Court’; it must not ‘prejudice the full effect and uniform application of Community Law in the Member States’ and ‘the restriction may not relate to a matter inherent in the exercise of a freedom guaranteed by the EC Treaty’.92 The case of Paletta93 of 1996, a ‘benefits case’ in the wider sense, is an example of how strongly the Court emphasises the effectiveness of the freedom of movement of workers also with regard to procedural formalities and the burden of proof.94 Paletta was an Italian national employed in Germany who reported ill from Italy while on annual leave, presenting a medical certificate from an Italian doctor. This attested illness would trigger the employer’s obligation to continue to pay his wages for up to six weeks. Mr Paletta’s employer doubted the veracity of the medical certificate and hence his incapacity to work and required further evidence, relying on case law in German labour courts. According to this case law, employers could require further evidence if there were serious doubts about the 87   The question of abuse of EU law/rights by an individual is, in a different dimension of the effectiveness rationale, also a question of who is competent: the Member States or the EU. Hence the question of an abuse of rights might become a question of abuse of competence (of the EU) if the right is stretched too far. 88   Recital 3; this has been stressed by the ECJ recently in Metock (n 2 above), para [59]. 89   Metock (n 2 above), paras [67–68]. 90   Metock (n 2 above), para [56], quoting amongst others Case C-60/00 Mary Carpenter v Secretary of State for the Home Department [2002] ECR I-6279, para [38]; Case C-459/99 Mouvement contre le racisme, l’antisémitisme et la xénophobie ASBL (MRAX) v Belgian State [2002] ECR I-6591, para [53]; Case C-291/05 Minister voor Vreemdelingenzaken en Integratie v RNG Eind [2007] ECR I-10719, para [44]. 91   AG Geelhoed in Akrich (n 39 above), para [179]. 92   AG Geelhoed in Akrich (n 39 above), para [99]. See also Kefalas (n 12 above), para [22] and AG Tesauro, para [25]. 93   Case C-206/94 Brennet AG v Vittorio Paletta [1996] ECR I-2356. 94  For a general discussion see Sørensen (n 8 above) 454–56.

310  Katja S Ziegler truthfulness of a medical certificate and substantiated evidence. The Court acknowledges explicitly the notion of abuse or fraudulent conduct as a possible limit to a right conferred by EU law.95 But it held that the employer was bound by this medical certificate under EU secondary legislation and that he was not allowed to impose further burdens or formalities on the worker. Contrary to the German courts, the ECJ did not reverse the burden of proof: it would be for the employer to prove abuse.96 This makes it de facto very difficult to prove abuse and the case therefore provides another example of the limited scope of application of the principle due to attempts to make EU free movement law effective. C.  The People/Citizenship Rationale In the context of free movement rights, it would not be surprising if the underlying policy reason besides a wide scope of the concept of a worker (and hence EU law) was furthered not only by considerations of economic but also of social integration and advancement, strengthening the case for non-discriminatory treatment. This transpired already in the discussion of Advocate-General Mazák’s Opinion in the Förster case above.97 This rationale is explicitly laid down in the recitals to Regulation 1612/68:98 Whereas freedom of movement constitutes a fundamental right of workers and their families; whereas mobility of labour within the Community must be one of the means by which the worker is guaranteed the possibility of improving his living and working conditions and promoting his social advancement, while helping to satisfy the requirements of the economies of the Member States; whereas the right of all workers in the Member States to pursue the activity of their choice within the Community should be affirmed.

Or, as Advocate-General Jacobs has expressed it: The recital makes it clear that labour is not, in Community law, to be regarded as a commodity and notably gives precedence to the fundamental rights of workers over satisfying the requirements of the economies of the Member States.99

In other words: the parallels in the paradigm of market integration for goods and people have limits—people aren’t goods. Nowadays, Directive 2004/38 is very explicit regarding the relevance of social integration,100 and there is a widespread reflection of this in the case law on citizenship. In the context of the workers cases101 considered above, the Court does not expressly refer to the integration in the host society. But the facts, for example of Lair, reveal a long period of residence in the host State, even if only a shorter one of employment. The AdvocateGeneral refers explicitly to the fact that Ms Lair went to Germany ‘in the capacity of a genuine worker economically integrated into the host State’ and that she subsequently was ‘involuntarily unemployed during a large part of eight years during which she has   Paletta (n 93 above), paras [23–24].   Paletta (n 93 above), para [26].     97  See Bettray quote with n 50 above.     98   Recitals, para [7].     99   AG Jacobs in Bettray (n 15 above), paras [28–29]. 100   Recitals 18, 23, 24; Art 28; see also Joined Cases C-482/01 & 493/01 Orfanopoulos and Oliveri v Land Baden-Württemberg [2004] ECR I-5257. 101   In contrast to the citizenship cases it does not need to rely on integration in the context of rights of workers.     95     96

‘Abuse of Law’ in the Context of the Free Movement of Workers  311 resided in Germany’.102 This sounds like a harbinger of the rationale adopted later in the context of citizenship cases where the degree of integration (or genuine link) was a direct consideration for deriving further rights from citizenship. The Court rejected typified approaches in regard to standard residence requirements as suggested by Advocate-General Slynn in his Opinion in Lair,103 and imposed financial solidarity between Member States through the requirement of an individualised proportionality test.104 Advocate-General Geelhoed in the (citizenship) case of Bidar exemplifies this point: Finally, it was submitted by various intervening Governments that the Member States have a legitimate interest in preventing abuse of their student support schemes and in preventing ‘benefit tourism’. I do consider that this is indeed a legitimate concern of the Member States, but the manner in which this should be ensured should not be such as to undermine the fundamental rights of EU citizens residing lawfully within their territory. A simple residence requirement is too non-selective for achieving this aim. In my view it can be achieved adequately in the context of establishing whether or not an applicant has a genuine link with the national education system or society as set out above.105 (Emphasis added.)

The people/citizenship rationale might receive added momentum through its actual or perceived link with the democratic deficit debate with the perhaps optimistic logic that affording more rights to EU citizens might lead to more realisation of relevance of, and engagement in, the political process of the EU, and hence decrease the democratic deficit, insofar as it results from the lack of input legitimacy through voter apathy in European Parliamentary elections. D. Human/Individual Rights Rationale Related but going beyond the citizenship rationale is a human rights rationale which has many facets. There are two types of human rights consideration in the case law on free movement of workers. The first one refers to the substance of the right at issue. A sympathetic approach to the right to family life is visible both in Akrich106 and Metock.107 It is also an aim and one rationale of Directive 2004/38 to ‘maintain the unity of the family in a broader sense’108 and to facilitate ‘the free movement of family members who are not nationals’.109 The ability to ‘lead a normal family life in the host Member State’110 was also a decisive factor both for the substantive decision in Metock and for the procedure under which the case was decided. Because of the urgency of a decision, the fast-track procedure was chosen—to limit the time of uncertainty about the family situation.111   AG Slynn in Lair (n 9 above) 553–54 (CMLR).   See text at n 40 above. 104   Case C-85/96 María Martínez Sala v Freistaat Bayern [1998] ECR I-2691; Grzelczyk (n 28 above), para [44]; Bidar (n 22 above), para [56]; AG Geelhoed in Akrich (n 39 above), paras [32], [67]; and AG Mazák in Förster (n 29 above), paras [55–56], [124–25]. 105   AG Geelhoed in Bidar (n 22 above), para [66]; see also AG Mazák in Förster (n 29 above), paras [128–33], esp [130]. See however the more restrictive approach of the Court now in Förster (n 29 above), para [54]. 106   Akrich (n 39 above), para [58]. 107   Cf Metock (n 2 above), paras [69], [79]. 108   Recital 6. 109   Recital 8. 110   Metock (n 2 above), para [62]. 111   Metock (n 2 above), Order of 17 April 2008, paras [15–16]: ‘A reply from the Court within a very short period could, therefore, bring a swifter end to that uncertainty, which is preventing the persons concerned from leading a normal family life.’ 102 103

312  Katja S Ziegler Further, the difficulty of distinguishing between use and abuse of a right under EU law raises at least indirectly human rights considerations on a different level: those of foreseeability, legal certainty and uniform application of rights under EU law. That national exception rules are in conformity with human rights is a direct concern of EU law and the Court,112 even if it is prima facie more concerned about the effectiveness of EU law. Divergence in the application of individual rights under EU law across Member States may also cause a loss of coherence of EU law as applied in different Member States and hence have repercussions for the legitimacy of the EU. It may be added that the notion of abuse of rights or similar concepts such as the principle of good faith in national legal systems as well as in public international law,113 primarily apply in horizontal situations and not in situations of subordination of the individual to the state. The ‘horizontal’ concept of abuse can be found in international law. Here, the question of abuse of rights arises between states as sovereign equals— concerning the exercise of sovereign rights of states114 which structurally is more akin to claims under private law. It is reflected by the Roman law principle sic utere jure tuo ut alterum non laedas or in the principle of interpretation of contracts in good faith115 (or indeed international treaties116) or the performance of obligations117 in good faith. These are structurally different situations from those where the individual finds himself in a hierarchical relationship with a state, as is the case with rights concerning the free movement of persons. Where individual rights against the state are concerned, it is therefore problematic to create vague and indeterminate grounds which allow the Member State to derogate further from rights than is permitted by the limitation clauses which are already intended to strike the balance between individual and state interests. The notion of abuse of rights in a vertical dimension (individual versus state) in the context of individual rights has a very different meaning and thrust. From a comparative perspective, the notion of abuse appears in special derogations clauses which can be invoked by the state against the individual. They are ultima ratio provisions. They are not intended to resolve the ordinary balancing between state and individual interests, but to defend the existence or constitutional order of a state in the context of defensive democracy. Hence they apply in the situation where rights are used to abolish the very existence of rights or the existence or democratic structure of a state, a totally different situation from those in which the notion of ‘abuse of rights’ is discussed in the context 112   See however Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-9609, paras [33–39] and Case C-112/00 Eugen Schmidberger, Internationale Transporte und Planzüge v Republik Österreich [2003] ECR I-5659, paras [73–80]. 113   The concept is highly contentious there as well, see the overview by M Beyers, ‘Abuse of Rights: An Old Principle, A New Age’ (2002) 47 McGill Law Journal 389, 397–404, 410–19; J Kokott, ‘Mißbrauch und Verwirkung von Souveränitätsrechten bei gravierenden Völkerrechtsverstößen’ in U Beyerlin, M Bothe, R Hofmann, E-U Petersmann (eds), Recht zwischen Umbruch und Bewahrung: Völkerrecht—Europarecht— Staatsrecht: Festschrift für Rudolf Bernhardt (Berlin, Springer 1995) 135, 139 fn 20; GDS Taylor, ‘The Content of the Rule against Abuse of Rights in International Law’ (1972–73) 46 British Yearbook of International Law 323. The seminal work is still that of A Kiss, L’abus de droit en droit international (Paris, LGDJ 1952); see also A Kiss, ‘Abuse of Rights’ in R Wolfrum (ed), Max Planck Encyclopedia of Public International Law (currently available only as online database) (Oxford, Oxford University Press 2008). 114   See the categories of abuse in international law, developed by Kiss in L’abus (n 113 above) 184, 186–87. 115   See for example § 157 of the German Civil Code (BGB). 116   Art 31 of the Vienna Convention on the Law of Treaties (1969) (VLCT). See also Appellate Body of the WTO, United States-Import Prohibition of Certain Shrimp and Shrimp Products, para [158]. 117   Such as in § 242 BGB or Art 26 VCLT.

‘Abuse of Law’ in the Context of the Free Movement of Workers  313 of free movement rights. In this situation the extraordinary limitation of forfeiture of (human) rights may apply, as exemplified by Article 17 of the European Convention on Human Rights,118 which—in contrast to the concept of abuse—leads to the a priori exclusion of an activity from the scope of protection of the right.119 EU free movement rights, at least in their origin, are of course different from human rights. Their common characteristic, however, is that they are individual rights, and they have even been equated by some to human rights because they overlap with human rights in the economic sphere but also because of doctrinal parallels, especially in the area of free movement of persons.120 It may therefore be suggested that the same principles about restricting individual rights apply to both categories, free movement rights and human rights, and that therefore prime importance should be attached to clarity about scope and limitations of an individual right. V.  Conclusion: Last Resort, Gap-filling and Progressive Development of the Law as Raison d’être?

It might be asked why the Court of Justice refers to the abuse of rights at all in the context of free movement of persons, and whether appeasing Member States might be the only reason.121 By referring to abuse of rights as a principle, the Court seems to keep at least the options open. It may be used as an interpretative tool122 and to help decide cases in grey areas where the law is not (yet) sufficiently clear and precise.123 Here, the principle might provide a balancing tool,124 adding coherence to or ‘rationalising’ an argument without independent normative content of its own,125 adding ‘subtlety and depth to the relatively crude black and white quality of primary norms’126 until clearer rules emerge.127 However, 118   The possibility to forfeit human rights seems to address what the abuse of a right vis-à-vis a state (vertical relationship) might encompass. It is a common feature of many human rights codifications which carries a high threshold. Cf Art 54 of the EU Charter of Fundamental Rights, stating that nothing in the Charter shall be interpreted as implying any right to engage in any activity or to perform any act aimed at the destruction of any of the rights and freedoms recognised in the Charter; likewise Art 17 ECHR; Art 3 of the Optional Protocol to the ICCPR; see also national constitutions—eg Art 18 of the German Constitution (Grundgesetz). 119   See, for example Pavel Ivanov v Russia (admissibility decision) (App no 35222/04) 20 February 2007, para [1]; Case of the United Communist Party of Turkey and others v Turkey (App no 19392/92) Reports 1998-I, para [60]; Case of the Socialist Party of Turkey and others v Turkey (App no 21237/93) Reports 1998-III, para [53]. See in contrast, the effect of the abuse principle as a limitation in Art 35 of Directive 2004/38 (n 1 above), text at (n 8 above). 120   T Oppermann, Europarecht, 2nd edn (Munich, CH Beck Verlag 1991) para [490]; see also C Walter, ‘Geschichte und Entwicklung der Europäischen Grundrechte und Grundfreiheiten’ in D Ehlers (ed), Europäische Grundrechte und Grundfreiheiten (Berlin, De Gruyter 2002) § 1, paras [42–44]. 121   Cf Kjellgren (n 7 above) 192. 122  See AG Poiares Maduro in Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, para [68]; AG Pergola in Centros (n 67 above), para [20]; AG Tesauro in Kefalas (n 12 above), para [25]. See also A Arnull, ‘What is a General Principle of EU Law?’, ch 2 above. 123   Regarding the effects of a general principle, see Arnull (n 122 above). 124   Cf Sørensen (n 8 above) 434, who compares the operation of abuse with the principle of proportionality, however, without defining the differences of the two principles. 125  V Lowe, ‘The Politics of Law-making: Are the Method and Character of Norm Creation Changing?’ in M Byers (ed), The Role of Law in International Politics: Essays in International Relations and International Law (Oxford, Oxford University Press 2000) 207. 126   Lowe (n 125 above) 218. 127   This would lead to further harmonisation—as happened in Art 35 of Directive 2004/38 (n 1 above); see in this respect also de la Feria (n 73 above) 407–08.

314  Katja S Ziegler as already discussed,128 it does ultimately depend on an underlying value judgment whether an activity is considered illicit and therefore excepted from the exercise of a right. Classifying an action as abuse is therefore not much more than a label that suggests rationality. It has to be acknowledged nevertheless that there is a need in legal systems to decide the undecided or unregulated and to give reasons for such decisions.129 In that sense reference to abuse is serves the function of keeping open the possibility to use a blanket concept in an individual case which was not foreseen by the legislator.130 The Court may therefore refer to abuse to keep this flexibility, even though it is not really applied, at least not in the area of free movement of workers and citizens. The reference to the principle by Advocate-General Kokott in her Opinion in the Hosse case could be explained in this way. Hosse concerned the claim of an allowance for the care of a disabled child of a German frontier worker, resident in Austria and employed in Germany, and therefore benefiting from the free movement of workers. The AdvocateGeneral considered the mere possibility of claiming a care allowance twice, from the State of residence and from the State of nationality, not to be a sufficient reason to discriminate and to refuse the payment of the benefit, absent harmonisation of the benefit system. However, the Advocate-General considered that double-claiming would amount to an abuse of rights and suggested that it should be countered by appropriate declarations with and cooperation of the relevant authorities,131 ie by regulation of the individual phenomenon of abuse. A principle of abuse of rights thus may have a residual gap-filling and interpretive function. ‘Abuse’ as a factual phenomenon and term, however, is better accommodated in the teleological construction of the scope of the free movement of persons provisions and their exceptions. If restrictions beyond this prove necessary, area-specific codifications of further exceptions at European level are preferable because of, on the one hand, the need to make a value judgement about tolerable and intolerable motives, and, on the other hand, of the need for clarity and legal certainty when individual rights are restricted.

  See above (after n 86).   This is normally done through the methods of interpretation, cf AG Poiares Maduro in Case C-465/07 Elgafaji v Staatssecretaris van Justitie [2009] ECR I-921, paras [31–35]. 130   Cf the less far-reaching and in a sense diametrically opposed methodological instrument of analogy. 131   AG Kokott in Case C-286/03 Hosse v Land Salzburg [2006] ECR I-1771, para [110]. 128 129

22 Comments on Abuse of Law and the Free Movement of Workers Eleanor Spaventa* I. Introduction

T

he excellent chapter by Katja Ziegler1 examines the relevance of the ‘abuse of law’ doctrine in the field of the free movement of workers. In this respect, the main claim put forward by Dr Ziegler, a claim that finds me in total agreement, is that the category of abuse of law is factually irrelevant in this field. Thus, despite the fact that the Court of Justice has indicated that in theory the provisions on the free movement of workers, and those on the free movement of persons more generally,2 might not help an abusive claimant, it has never applied this category to a specific case.3 According to the case law of the Court, the notion of abuse implies an ‘improper use’ of the Treaty freedoms to circumvent national law. Thus, the notion of abuse seems to imply the relevance of a psychological element, so that the reason why the right holder exercises her right becomes relevant to the assessment of whether there is an abuse capable of projecting the factual situation outside the protection afforded by EU law. However, in the case of the free movement provisions, the Court has repeatedly, and consistently, declared the irrelevance of the reasons—the psychological element—that might have led the individual to exercise her rights. Thus, for instance, the fact that the right to move and work in another Member State is exercised with the sole aim of triggering the TFEU so as to benefit from the rights for family members of EU nationals, which is to say with the intention of circumventing national migration law, is immaterial to the enjoyment of such rights.4 In her chapter, Ziegler identifies four possible reasons that might explain the irrelevance of the abuse category in the field of free movement of persons. First of all, the nonexistence of a principle of abuse of law, which ‘would be merely a descriptive umbrella

  *  I am grateful to the organisers and participants to the conference for an interesting and lively discussion. 1   K Ziegler, ‘“Abuse of Law” in the Context of the Free Movement of Workers’, ch 21 above. 2  Neither Dr Ziegler nor I are concerned with the free movement of corporate entities; therefore the expression ‘free movement of persons’ will be used to indicate the free movement of individuals. 3   See for example, Case C-200/02 Zhu and Chen v Secretary of State for the Home Department [2004] ECR I-9925; Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607. The issue of abuse might be relevant in the case of the instrumental use of the free movement provisions with the sole purpose of eluding national tax legislation; on this point see AG Geelhoed in Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107. 4  This was to a certain extent established in Case 53/81 Levin v Staatssecretaris van Justitie [1982] ECR 1035; and confirmed in Akrich (n 3 above).

316  Eleanor Spaventa term for certain factual scenarios’.5 A principle of abuse of law might be particularly relevant in relation to corporate tax law, where the concept of abuse might usefully be employed to limit the possibility for corporate entities to exploit gaps in the legislation by simply being a step ahead of the legislative process.6 In this respect, abuse of law would be a useful tool to avoid the age-old problem of passing legislation through a cumbersome and long legislative process only to find that companies having access to good legal advice find a way around it. In this respect, the abuse of law category might well indicate the application of a sui generis mandatory requirement, in that the imposition by the Member State of its rules is a legitimate and proportionate way to protect an aim compatible with European law (eg, the need to combat fiscal elusion). Should one not adhere to the claim that ‘abuse of law’ is a non-existent legal category, Ziegler puts forward alternative reasons why the abuse of law doctrine has never been applied by the Court in relation to the free movement of workers. Thus, the Court’s case law in the field of the free movement of persons is driven by a very strong integrationist rationale; in interpreting the rights that individuals derive from the free movement provisions, the Court is also driven by a citizenship rationale; and finally, the fact that the case law in this field is driven by fundamental rights considerations. II.  Some Remarks on Abuse of Individual Rights

And yet, as pointed out by Ziegler in the last part of her chapter, the real issue revolves around a more fundamental question: is the notion of abuse of workers’ rights a legally defined (and legally helpful) category? And in many respects, this is exactly the broader question proposed by Rita de la Feria as the theme of the conference which gave rise to the chapters collected in this volume: to what extent can a right really be abused? And what does such an abuse consist of? Generally, it is not crystal clear that the notion of abuse is legally persuasive: after all, the doctrine of abuse serves the purpose of transforming something that would be otherwise legal into unlawful behaviour. And, as pointed out elsewhere in this volume, the transformation from legality to illegality happens via means of judicial interpretation in the case at issue and therefore might raise problems of legal certainty as well as democratic accountability. In any event, the reasons which might justify this metamorphosis are also not well defined. In this respect, the Court refers to the possibility for a Member State to ‘prevent certain of its nationals from attempting, under cover of the rights created by the [EC] Treaty, improperly to circumvent their national legislation or to prevent individuals from improperly or fraudulently taking advantage of provisions of Community law’ (emphasis added).7 Thus, it is unclear whether the category of abuse of law would ever be relevant,   Ziegler (n 1 above), Part IV.   For an interesting comment about abuse of law in tax cases, particularly useful for non-tax lawyers, see Confédération Fiscale Européenne, Opinion Statement of the CFE ECJ Task Force on the Concept of Abuse in European Law, Based on the Judgments of the European Court of Justice Delivered in the Field of Tax Law (November 2007): www.cfe-eutax.org/sites/default/files/cfe-opinion-statement-on-abuse-in-european-law_nov2007.pdf. The paper also points at a possible distinction between abuse of rights (mainly to the detriment of a private party) and abuse of law. 7  Joined Cases C-151/04 and 152/04 Claude Nadin, Nadin-Lux SA and Jean-Pascal Durré [2005] ECR I-11203, para [45]. 5 6

Comments on Abuse of Law and the Free Movement of Workers  317 at least in relation to non-corporate entities, beyond the instances of fraud, which maybe would be best kept distinct in order to avoid confusion. Fraudulent behaviour might well exclude the individual from the scope ratione personae or ratione materiae of the Treaties. For instance, marriages of convenience are excluded, and were excluded before Directive 2004/38,8 from the scope of the Treaty simply because, even though they might be otherwise valid, they are not relevant marriages for EU law purposes. Similarly, the principle according to which individuals cannot exploit the Treaty so as to benefit twice from welfare provisions simply translates in to the fact that there is no right in European law to double an entirely overlapping welfare provision.9 For this reason, there is no possibility to raise an EU law defence in relation to fraudulent behaviour. In this respect, the fact that Article 35 of Directive 2004/38 mentions both abuse and fraud is not, in my opinion, a sufficient reason to hold that the two are conceptually distinct. More fundamentally, it should be queried whether the concept of abuse of law can ever be relevant in relation to the exercise of individual rights. Thus, for instance, consider the situation at issue in Akrich.10 Mr and Mrs Akrich moved from the UK to Ireland, and then back to the United Kingdom, with the sole purpose of triggering the EC Treaty so that Mr Akrich would gain a residence permit pursuant to the Singh case law.11 Leaving aside the fact that since Mr Akrich was unlawfully resident in the UK, the Court excluded the applicability of Article 10 of Regulation 1612/68,12 it was made clear in that case that the motives that led Mrs Akrich to Ireland and back were irrelevant in assessing the legal situation of the couple.13 Indeed, it would have been untenable to hold that since Mrs Akrich wanted to take advantage of one of the ancillary rights expressly conferred by Community law, and used the primary rights to achieve that end, then she should be excluded from the scope of Community law. In other words, it is unlikely that the fact that an individual exercises her rights in order to be able to enjoy rights conferred upon her by EU law should matter in defining the scope of such rights. Or else, it would mean that intention to enjoy rights might disqualify the claimant from enjoyment of such rights. Furthermore, it would be illogical to take motivation into account: there are endless variables that can influence individual choices, from mere economic ones to more individual ones such as personal preferences, the location of family and friends, etc. It would be impossible to find a common rationale capable of determining how and when those personal preferences are relevant in defining the legitimate or abusive exercise of the rights at issue; and when such personal motivations are irrelevant. And in any event,     8  Art 35, Directive 2004/38 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States, amending Regulation (EEC) No 1612/68 and repealing Directives 64/221/EEC, 68/360/EEC, 72/194/EEC, 73/148/EEC, 75/34/EEC, 75/35/EEC, 90/364/EEC, 90/365/ EEC and 93/96/EEC [2004] OJ L229/35; see Akrich (n 3 above).     9   See for example, Regulation 883/2004 on the coordination of social security systems [2004] OJ L 166/1 as implemented by Regulation 987/2009 [2009] OJ L 284/1. 10   Akrich (n 3 above). 11   In Case C-370/90 R v Immigration Appeal Tribunal and Surinder Singh ex p Secretary of State for the Home Department [1992] ECR I-4265, the Court held that a British national returning from Germany to the UK after having worked there was protected by the EEC Treaty and the relevant secondary legislation so that her husband gained residency rights in the UK. 12  Regulation 1612/68 on freedom of movement for workers within the Community OJ Sp Ed 1968 L257/2, 475. This case law has now been reversed; see Case C-127/08 Blaise Baheten Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I-6241. 13   Akrich (n 3 above), paras [55–56].

318  Eleanor Spaventa it would be difficult to reconcile the notion of abuse with the hermeneutic principles established by the Court. Thus, the free movement provisions have always been given a teleological interpretation, which is aimed not only at facilitating movement, but also at not deterring the use of both primary Treaty rights and rights contained in secondary legislation (which, in any event, usually derive directly from the former rights).14 It would be therefore difficult to justify the imposition of limits, which might well appear arbitrary, to the use of those rights by focusing on a pre-defined (by whom?) notion of the manner in which those rights should be used. This is all the more the case given that the interpretation of the free movement provisions espoused by the Court has gone far beyond the (supposed) intention of the EEC Treaty drafters. Thus, in a dynamic and evolving system such as the European Union, it would be very difficult to limit the rights granted by the TFEU because of the fact that the claimant is not exercising the right in a manner which is compatible with the reasons for which the right was originally granted. After all, the Court’s interpretation has been developing according to an ever changing telos (market integration first, constitutional citizenship now) which has, in some instances, openly disregarded the political will expressed in primary and secondary legislation. Thus for instance, consider not only the move from discrimination to barrier to movement,15 with all that that implies in terms of repartition of regulatory competences,16 but also those cases in which the Court pushed the scope of Community law beyond its letter. One could usefully recall the health care case law,17 where the Court accepted that Article 56 TFEU could be interpreted so as to confer a right to seek health care abroad (subject to limits imposed by the Court itself) beyond what was allowed by Regulation 1408/71, or the citizenship case law on welfare benefits for non-economic migrants.18 It is difficult to maintain that the limitations in Regulation 1408/71, the limitation of the scope of the EC Treaty to cover only services provided for remuneration in Article 50 EC, or the limitations to the principle of equal treatment provided for in Article 24(2) of Directive 2004/38, were not intended to curtail the possibility of using the Treaty freedoms to gain access to health care and benefits respectively.

14   See the case law on barriers that make movement less attractive and the general definition of barrier given in Case C-55/94 Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano [1995] ECR I-4165, para [37]. 15   Gebhard (n 14 above); Case C-415/93 Union Royal Belge des Sociétés de Football Association ASBL v JeanMarc Bosman [1995] ECR I-4921; Case C-60/00 M Carpenter v Secretary of State for the Home Department [2002] ECR I-6279. 16  See generally, N Bernard, Multilevel Governance in the European Union (The Hague, Kluwer Law International 2002) esp ch 2; and more specifically, E Spaventa, ‘From Gebhard to Carpenter: Towards a (NonEconomic) European Constitution’ (2004) 41 Common Market Law Review 743. 17   Cf Art 22, Council Regulation (EEC) 1408/71 of the Council of 14 June 1971 on the application of social security schemes to employed persons and their families moving within the Community. See for example, the health care cases: Case C-157/99 BSM Geraets-Smits v Stichting Ziekenfonds VGZ and Peerbooms v Stichting CZ Groep Zorgverzekeringen [2001] ECR I-5473; Case C-385/99 Müller-Fauré v Onderlinge Waarborgmaatschappij OZ Zorgverzekeringen UA, and van Riet v Onderlinge Waarborgmaatschappi ZAO Zorgverzekeringen [2003] ECR I-4509; Case C-372/04 The Queen (on the application of Yvonne Watts) v Bedford Primary Care Trust and Secretary of State for Health [2006] ECR I-4325. 18   For example, Case C-224/98 MN D’Hoop v Office national d’emploi [2002] ECR I-6191; Case C-138/02 Brian Francis Collins v Secretary of State for Work and Pensions [2004] ECR I-2703; Case C-209/03 The Queen (on the application of Dany Bidar) v London Borough of Ealing and Secretary of State for Education and Skills [2005] ECR I-2119. The ruling in Bidar might have been partially overruled by the ruling in Case C-158/07 Jacqueline Förster v Hoofddirectie van de Informatie Beheer Groep [2008] ECR I-8507.

Comments on Abuse of Law and the Free Movement of Workers  319 Given the ever evolving scope of the free movement provisions, it would be near impossible to draw a line between abusive use of the free movement provisions and non-abusive use. And if such a line were to be drawn it might appear rather arbitrary. Thus, the fact that we find references to the concept of abuse both in the case law and in Directive 2004/38, might indicate simply the fact that, as said above, in instances in which the individual is trying to rely on EU law to shield herself from the consequences of fraudulent, and therefore either illegal or borderline illegal, behaviour, EU law cannot be used as a defence. III. Abuse of Law in Horizontal Situations

A more interesting question might arise in relation to the possibility of invoking the abuse of law doctrine in relation to horizontal situations. As pointed out by Dr Ziegler, it is not unusual for fundamental rights documents to contain an ‘anti-abuse’ clause according to which the exercise of fundamental rights cannot be aimed at the destruction of the rights recognised in those documents.19 Again, it is a matter of debate whether the anti-abuse provisions delimit the scope of the rights granted by such documents, or rather imply a self-standing legal category.20 Usually the doctrine, or inherent limitation as the case might be, applies both to vertical and horizontal situations. In particular, it might be used to prevent a party from using a right, for instance freedom of expression, in ways which go against the very purpose of the rights granted; or in order to allow a State to regulate such rights to avoid ‘abuses’. For instance, the European Court of Human Rights has accepted that Member States might legitimately prevent the expression of ideas, such as racist or fascist ideas, that would undermine the very values which the Convention seeks to protect.21 In relation to the free movement of workers provisions, it is not clear whether the doctrine of abuse might be relevant in relation to horizontal situations. This question seems particularly relevant following the horizontal effect of the free movement of persons provisions. In Angonese,22 the Court clarified that the prohibition of non-discrimination on grounds of nationality is a mandatory principle of Community law which also binds private parties. In the more recent cases of Viking and Laval,23 the Court accepted that the EC Treaty’s free movement provisions might impose a positive duty on trade unions to refrain from taking collective action against trans-frontier employers, or action with the aim of preventing an employer from exercising its Treaty free movement rights. The horizontal boundaries of the free movement provisions have yet to be properly defined. However, it is not impossible to foresee situations in which claimants might 19   For example, Art 54 of the EU Charter of Fundamental Rights states: ‘Nothing in this Charter shall be interpreted as implying any right to engage in any activity or to perform any act aimed at the destruction of any of the rights and freedoms recognised in this Charter or at their limitation to a greater extent than is provided for herein’. Similar provisions are found in Art 17 of the European Convention for the protection of Human Rights and Fundamental Freedoms, and in Art 18 of the Universal Declaration of Human Rights. 20   For a comparative discussion see E Reid, ‘The Doctrine of Abuse of Rights: Perspectives From a Mixed Jurisdiction’ (2004) 8.3 Electronic Journal of Comparative Law: www.ejcl.org/83/art83-2.html. 21   For example, Garaudy v France (App no 65831/01) Decision of 24 June 2003, ECHR 2003-IX (extracts). 22  Case C-281/98 Roman Angonese v Cassa di Risparmio di Bolzano SpA [2000] ECR I-4139. 23  Case C-438/05 The International Workers’ Federation and the Finnish Seamen’s Union v Viking Line [2007] ECR I-10779; Case C-341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet et al [2007] ECR I-11767.

320  Eleanor Spaventa attempt to artificially bring themselves within the scope of the TFEU in order to gain an undue advantage against a private party.24 Take for instance the case of Evans: that case concerned litigation between an ex-husband and wife as to the use of fertilised eggs, when the ex-husband withdrew consent for insemination. In that case, Community law had not been triggered and therefore the balancing act as to the mutual rights and obligations of the two parties rested with national law, and with the European Court of Human Rights.25 However, one could ask oneself whether, if Ms Evans had sought implantation of the fertilised eggs in France with the sole purpose of triggering the intra-Community element and the jurisdiction of the Court, that would qualify as an abusive use of the free movement provisions. In the present writer’s opinion, even in such an extreme case, it would not be a matter of abusive exercise but rather of defining in a more precise way the extent to which individuals can rely on their right to move during the course of proceedings against another person, when their right to move is only tangentially relevant. Thus, in the hypothetical case of Ms Evans in France, the real issue is not that it is an abuse of rights but rather a constitutional issue as to the respective scope of application of European and national law. In other words, the right to have infertility treatment in France would not be abusively exercised, but rather the Court of Justice would have to decide whether to impose its own fundamental rights standards or leave it to national law to balance the conflicting fundamental rights.26 IV. Concluding remarks

As argued by Katja Ziegler, it seems doubtful that the doctrine of abuse of law is a legally useful concept in relation to the free movement of workers. The cases in which abuse of law might be invoked can be grouped into two categories: first, those cases in which the claimant has engaged in fraudulent behaviour which excludes the application of the TFEU ratione personae (eg marriages of convenience, where the spouse is not a spouse for Community law purposes), or ratione materiae (eg cases involving double benefit claims). Secondly, the expression ‘abuse of law’ might serve as an all-encompassing safety net that can be relied upon in exceptional circumstances to justify a limitation of the Treaty rights. Thus, it could be seen as a sui generis mandatory requirement, indicating nothing more than the legitimate and proportionate application of a rule which is itself compatible with EU law. Overall then, at least in relation to the free movement of workers and physical persons more generally, the concept of abuse—far from being a self-standing legal category—seems to be a tool which might be used to delimit the rights at issue.

24  This is common in competition law as well as in vertical cases in relation to free movement of persons; see for example, the rather artificial intra-Community link in Carpenter (n 15 above). 25   Evans v UK (App no 6339/05) Judgment of 10 April 2007. 26  On this point see E Spaventa, ‘Federalization versus Centralization: Tensions in Fundamental Rights Discourse in the European Union’ in M Dougan and S Currie (eds), 50 Years of the European Treaties (Oxford, Hart Publishing 2009) 343.

23 Citizenship of the Union: Above Abuse? Cathryn Costello*

T

his chapter explains and defends the limited role for ‘abuse of rights’ claims in the case law on citizenship of the Union. By engaging with the multifarious contexts in which pleas of ‘abuse’ arise, the chapter provides a constructive reading of the cases. Part I of this chapter contains a primer on abuse, setting out some general observations and concerns about the doctrine. Ultimately while rejecting the existence of an autonomous general principle of abuse, I acknowledge the concept to the extent that it embodies an inherent aspect of teleological reasoning under certain conditions. Part II examines the first relevant feature of EU citizenship, namely that access thereto depends on holding the nationality of a Member State, at present a matter of national law. There should be no room for abuse of rights claims in the case of nationality designation. The Court of Justice rightly insists on a strict duty of mutual recognition. Part III also deals with questions of personal scope, focusing on the residence rights of third-country national (TCN) family members of EU citizens. The case law here, culminating in the recent ruling of the Grand Chamber of the Court in Metock1 illustrates that EU rights entail licit circumvention of domestic immigration law. This section also explores the incorporation into EU law of the concept of ‘marriage of convenience’, a specific instantiation of abuse developed in national legal systems. Part IV then examines the interplay between EU citizenship and free movement, revealing that abuse arises only in limited instances of problematic or illicit circumvention. First, the cases illustrate that reliance by migrant EU citizens on the EU equal treatment guarantee to claim benefits available in their host State does not generally unsettle the allocation of regulatory power in the Union. The doctrine of abuse plays a limited role, mainly in policing the boundaries between various economic statuses, where these entail different entitlements. Secondly, I examine u-turn situations, where individuals rely on EU rules against their home State. These situations are potentially troubling to the allocation of regulatory competence to the home State, so the doctrine of abuse may have a role to play. Part IV also illustrates the impact of EU legislative intervention. Where there is a framework of minimum legislative standards to underpin mutual recognition, general claims of ‘abuse’ are likely to fail, but fraudulent behaviour may nonetheless relieve States of their duties of mutual recognition.   * The author thanks Professor Gareth Davies, Dr Catherine Donnelly and Professor Michael Dougan for comments on a much earlier draft. The usual disclaimer applies. 1   Case C-127/08 Blaise Baheten Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I-6241; reaffirmed by way of reasoned order in Case C-551/07 Deniz Sahin v Bundesminister für Inneres [2008] ECR I-10453. See further, C Costello, ‘Metock: Free Movement and “Normal Family Life” in the Union’ (2009) 46 Common Market Law Review 587.

322  Cathryn Costello I. A PRIMER ON ABUSE

The phrase ‘abuse of rights’ is question-begging. To suggest that a right is being ‘abused’ seems to suggest behaviour beyond the scope of the right, in which case there is no right to speak of. In this way, ‘abuse’ seems self-contradictory. However, we may ease the apparent contradiction if we distinguish between the coverage of a right and the protection it grants, and conceive of a claim of abuse as part of an incremental judicial process of refining the scope of rights case-by-case.2 Alternatively, the notion of abuse may be treated as determining the permissible limitations on rights. However, treating abuse as an aspect of limitation of rights seems to render ‘abuse’ redundant. After all, limitation of rights and proportionality assessment is omnipresent in EU law. To gain some purchase on the distinctiveness of the concept of abuse, some mapping may be helpful. At the outset, if we speak of ‘abuse of rights’, we need to identify the right holders, the nature of the rights at stake, and the criteria for distinguishing use from abuse. A. Abuse of Individual Rights Abuse of rights in the civilian legal tradition serves as a specific prohibition on the use of private rights in an offensive manner or with a purpose deemed improper, often due to its detrimental effects on the general good. Similarly in EU law, abuse should be considered in the context of individual behaviour, mainly as improper reliance on EU rights against other individuals, Member States, or the EU. Its private law roots mean that it is less inapt for recognition as a general principle, as the notion of a general principle is rooted in EU public law. Although the Court in Emsland-Stärke3 held that there was a ‘general principle’ of abuse of rights, that statement must not lead us to equate ‘abuse’ with the other general principles of European law.4 The autonomous normative force of those general principles is well established. They are principles of public law, binding on both the EU institutions and the Member States when acting within the scope of EU law. Moreover, they are constitutional and general in character. Thus, they determine the validity of EU acts, and consequently dictate how those acts ought to be interpreted. The ruling in Audiolux5 provides helpful delineation of the notion of a general principle. Advocate-General Trstenjak stated that ‘general principles are distinguished primarily by their constitutional status within the Community legal order. As a rule, general principles embody fundamental legal concepts and values inherent in a legal order’.6 While the Opinion treats constitutional status and specificity as distinct features, the Court seems to regards them as inextricably linked. The principle proposed does not have ‘constitutional status’ in that it is ‘characterised by

 See further, F Schauer, ‘Can Rights Be Abused?’ (1981) 31 Philosophical Quarterly 225.   Case C-110/09 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. See further, P Koutrakos, ‘The Emsland-Stärke Abuse of Law Test in the Law of Agriculture and Free Movement of Goods’, ch 14 this volume. 4  See further, M Dougan, ‘Some Comments on the Idea of a General Principle of Union Law Prohibiting Abuses of Law in the Field of Free Movement for Union Citizens’, ch 24 this volume. 5   Case C‑101/08 Audiolux SA et al v Groupe Bruxelles Lambert SA (GBL) et al and Bertelsmann AG et al, [2009] ECR I-9823. 6  AG Trstenjak in Audiolux (n 5 above), para [87]. 2 3

Citizen of the Union: Above Abuse?  323 a degree of detail requiring legislation to be drafted and enacted at Community level by a measure of secondary Community law’.7 In addition, it is inappropriate to speak of Member States abusing their rights under EU law. In contrast to the position under public international law, where the doctrine of abuse is invoked to temper the excesses of States’ ‘primordial rights’,8 EU law heavily constrains the Member States, frequently treating them as mere administrative agents of the EU. The Member States have extensive legal duties under the Treaties, the text of which is complemented by extensive duties developed under Article 4(3) TEU [ex Article 10 EC], which incorporates the type of good faith considerations that under public international law may be covered under the abuse of rights doctrine. We have no need for a doctrine of ‘abuse’ further to constrain the Member States. The closest public law analogue to ‘abuse of rights’ is the ground of ‘misuse of powers’ under Article 263 TFEU [ex Article 230 EC], a limited ground of review only arising if it can be established that an EU institution has used a power solely or mainly for an improper purpose.9 It is difficult to make out as it usually requires proof of a specific illicit purpose,10 whereas other grounds going to lack of competence or illegality11 offer litigants greater prospects of success, reflecting the maturation of the rule of law at the EU level. B.  Indicia for Distinguishing Use from Abuse i.  Abuse Contrasted with Fraud Article 35 of the 2004 Citizenship Directive12 refers to both abuse and fraud. It states that: Member States may adopt the necessary measures to refuse, terminate or withdraw any right conferred by this Directive in the case of abuse of rights or fraud, such as marriages of convenience. Any such measure shall be proportionate and subject to the procedural safeguards provided for in Articles 30 and 31.

‘Abuse’ is a distinct concept from fraud, the latter connoting an objective falsehood and dishonesty in claiming entitlement to a benefit. Fraud seems straightforward. We would not expect the law to accept that the holder of a forged EU passport is an EU citizen. The   Audiolux (n 5 above), para [63].   G Schwarzenberger, ‘Uses and Abuses of the “Abuse of Rights” in International Law’ (1956) 42 Transactions of the Grotius Society 147; M Byers, ‘Abuse of Rights: An Old Principle, A New Age’ (2002) 47 McGill Law Journal 389. Byers emphasises the potential role of abuse of rights ‘in those few areas where the rights of states are still conceived of as general or primordial, by mediating between or otherwise limiting the exercise of rights’.     9  Earlier accounts of abuse included ‘misuse of powers’ within their ambit. See for example, N Brown, ‘Is There a General Principle of Abuse of Rights in European Community Law?’ in T Heukel and D Curtin (eds), Institutional Dynamics of European Integration, Vol II (Dordrecht, Martinus Nijhoff Publishers 1994) 511, 513–15. 10  See further, P Craig, EU Administrative Law (Oxford, Oxford University Press 2006) 464. 11  Including breach of the public law general principles such as proportionality, legitimate expectation and indeed fundamental rights. 12  Parliament and Council Directive 2004/38/EC 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States (Citizenship Directive) amending Regulation (EEC) No 1612/68 and repealing Directives 64/221/EEC, 68/360/EEC, 72/194/EEC, 73/148/ EEC, 75/34/EEC, 75/35/EEC, 90/364/EEC, 90/365/EEC and 93/96/EEC [2004] OJ L158/77 and corrigenda [2004] OJ L229/35, [2005] OJ L30/27, [2005] OJ L197/34 and [2007] OJ L204/28.     7     8

324  Cathryn Costello document purports to represent a legal status that is objectively false. However, in the EU context, the legal responses and consequences of claims of fraud have been controversial. In order to facilitate mutual recognition, the Court has limited the measures permissible in order to check the genuineness of documentation.13 In addition, given the serious legal consequences that flow from fraud, the case law aims to ensure that a previous misstatement will not always undermine a subsequent EU legal status granted.14 ii.  The Purposive Enquiry Accounts of abuse usually focus on an individual’s reliance on rights for an improper purpose. In this vein, the most sophisticated articulation of an EU abuse of rights doctrine is that of Advocate-General Poiares Maduro in Halifax, concerning putative abuse of the harmonised EC VAT rules.15 In that context, building on Emsland-Stärke,16 a two-limb test emerges. The first limb concerns the ‘subjective element’ relating to the identification of the purpose of the activities in question as abusive. The second element is the ‘objective element’ or, as Advocate-General Poiares Maduro puts it, a ‘teleological element whereby the purpose and objectives of the Community rules allegedly being abused are compared with the purpose and results achieved by the activity at issue’.17 The purposive enquiry is manageable when a specific legislative purpose is discernable. For instance, in Emsland-Stärke, the legislation intended to confer a benefit on those who exported certain agricultural products outside the EU. The conduct in question involved the export and immediate reimportation of the products, meaning that the benefit was being claimed on spurious grounds. In the VAT context, abusive practices are those that lack an economic rationale, so their sole purpose is to avoid tax. In the Internal Market context, where broad economic guarantees are at stake, it is more difficult to specify well-defined permissible purposes and conversely, illicit ones. The Court has tended to emphasise that once the economic activity is genuine, the right is engaged. Building a ‘genuineness’ requirement into the economic freedom may be characterised as rooting out abuse latu sensu. However, the economic freedoms are not only individual economic rights, but also ‘essential instruments in the distribution of power within the Constitutional Order of the Union’.18 As will be explored in the next section, the Court treats these entitlements as entailing a right to circumvent national law as part of this process of allocating regulatory power across the Member States.

13  See for example, Case C-246/00 Commission v Netherlands [2003] ECR I-7485 discussed in Part IV.B.iii below. 14  See further, Case C-285/95 Suat Kol v Land Berlin [1997] ECR I-3079; Case C-36/96 Faik Günaydin v Freistaat Bayern [1997] ECR I-5143; Case C-16/05 R (on the application of Veli Tum and Mehmet Dari) v Secretary of State for the Home Department (SSHD) [2007] ECR I-7415 discussed in Part III.B below. 15   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. AG Poiares Maduro, paras [87–88] in particular. 16   Emsland-Stärke (n 3 above). 17  AG Poiares Maduro goes on to clarify that this element ‘provides a safeguard for those instances where the sole purpose of the activity might be to diminish tax liability but where that purpose is actually a result of a choice between different tax regimes that the Community legislature intended to leave open’. Emsland-Stärke (n 3 above), para [88]. 18   M Poiares Maduro, We, the Court: The European Court of Justice and the European Economic Constitution (Oxford, Hart Publishing 1998) 167.

Citizen of the Union: Above Abuse?  325 EU citizenship, as will be examined in Part IV, builds on the previous market citizenship19 embodied by the Treaty provisions on free movement. Citizenship implies, or at least promises, a transcendence of the economic purposes, as we move into a political union. Some citizenship rights seem to be more akin to fundamental (ie human) rights, rather than economic freedoms. As such, the line between use and abuse cannot depend on the results of a purposive enquiry.20 Human rights are intrinsically valuable, so do depend on instrumental benefits. Individuals may rely on them without having to demonstrate that their actions pursue any particular purpose.21 In fundamental rights treaties, most rights may be restricted in the general interest. Admittedly, in addition, we usually find an anti-abuse provision. For instance, Article 17 of the European Convention on Human Rights (ECHR), now mirrored in Article 54 of the EU Charter of Fundamental Rights, provides that: Nothing in this Convention may be interpreted as implying for any State, group or person, any right to engage in any activity or to perform any act aimed at the destruction of any of the rights and freedoms set forth herein or at their limitation to a greater extent than is provided for in the Convention.

However, this ECHR notion of abuse is qualitatively different from the purposive enquiry under EU law.22 It permits Contracting States to adopt the stance of militant democracy, denying rights-coverage to activities which would undermine the very institutional foundations of democracy and the rule of law.23 Its scope is highly contested, and for the most part, the ECHR has tended to examine States’ behaviour only under the substantive provisions rather than Article 17.24 Exceptionally and controversially, Article 17 has been relied on to deny protection to hate speech.25 Whatever account underlies Article 17, it is crucial to maintain the distinction between these two variants of ‘abuse’ lest a generalised purposive enquiry undermine human rights protection. iii. Circumvention The Internal Market case law determines the allocation of regulatory power in the Union, distributing competence across the Member States, mainly by institutionalising home State control. When EU migrants demand that host States recognise their home State nationality, educational qualifications, and so on, the invocation of EU rights is in keeping with this allocation of competence. Similarly, when migrant EU citizens seek to 19  See further, M Everson, ‘The Legacy of the Market Citizen’ in J Shaw and G More (eds), New Legal Dynamics of European Union (Oxford, Oxford University Press 1995) 73. 20  See further, A Sajo, ‘Abuse of Fundamental Rights or the Difficulties of Purposiveness’ in A Sajo (ed), Abuse: The Dark Side of Fundamental Rights (Utrecht, Eleven International Publishing 2006) 29. 21  Of course, there may well be some instrumental dimension to fundamental rights. For example, freedom of expression may foster the intrinsic value of human creativity, but may also be defended as an instrument to support democracy. 22  The Interpretative Note to the Charter confirms that Article 54 ‘corresponds to’ Article 17 ECHR, but makes no reference to any EU doctrine of abuse of rights. Analogous provisions are found in Article 30 of the Universal Declaration of Human Rights (UDHR), and Article 5 of both the International Covenant on Civil and Political Rights (ICCPR) and the International Covenant on Economic, Social and Cultural Rights (ICESCR). 23  See further, Sajo (n 20 above) 53. 24   United Communist Party of Turkey v Turkey (App no 19329/92) ECHR 30 January 1998. 25  I refer here mainly to claims deemed inadmissible on manifestly unfounded grounds, as in various cases on Holocaust denial such as: T v Belgium (App no 9777/82) (1983) DR 34, 158; H, W, P and K v Austria (App no 12774/87) (1989) 62 DR 216; Kühnen v FRG (App no 12194/86) (1988) 56 DR 205.

326  Cathryn Costello enjoy benefits in their countries of residence on a non-discriminatory basis with home nationals, there is no ‘circumvention’ to speak of. In contrast, reliance on EU law to challenge home State rules may amount to problematic circumvention. However, as Part III explores, for the most part the Court has shown itself sensitive to the comings and goings throughout the European citizen’s life-course, protecting both the right to leave and return. We must also bear in mind the distribution of regulatory power between the Member States and the Union. Focusing on nationality in Part II reveals the salience of the potential scope of EU harmonised rules to the issue of abuse: if there is no prospect of harmonised rules, abuse claims could undermine the other mode of interaction open to the EU, namely mutual recognition. In contrast, in Part III we see that in those areas where there is EU competence, permitting Member States to use the ‘abuse’ doctrine may be an interim solution, pending the adoption of EU legislative measures.26 The Internal Market permits transborder economic activity, even if, indeed particularly if, its purpose is to avoid less favourable regulatory or tax regimes. In permitting this reliance, the Court makes a choice about the scope of national regulatory competence. Circumventing national law emerges as inherent in reliance on the Internal Market freedoms, and so cannot provide a full account of which problematic behaviour ought not to benefit from EU legal protection. We must distinguish between the avoidance of national law inherent in reliance on EU rules, and ‘illicit’ circumvention. Even in relation to corporate mobility, the line between licit and illicit circumvention depends on an account of the proper role of regulatory competition in the EU.27 In attempting this line-drawing exercise, we should bear in mind the ease of mobility of capital over labour. Taken together with the slipperiness of corporate entities, this phenomenon suggests that problematic circumvention is more likely in relation to the movement of corporate entities rather than persons. The movement of persons, in particular exit to escape disfavourable domestic conditions, is much less likely to create intolerable regulatory gaps or unfair regulatory competition. Individual emigration does not loom over States exerting downward pressure on domestic regulatory standards in the way that capital flight may. iv. Artificiality In Emsland, a transaction was deemed ‘artificial’ when goods were exported and immediately reimported in order to claim a benefit under trade law, the purpose of which was to reward genuine export.28 When the Court invokes the idea of ‘wholly artificial arrangements’, as in Cadbury Schweppes,29 these are characterised as those that ‘do not reflect economic reality’. A transaction’s ‘artificiality’ in the context of freedom of establishment is revealed by the fact that it has no economic rationale, and so can only be explained in light of the aim to circumvent the law. However, chasing that ‘legal advantage’ may well be an economically viable activity with an undeniable economic  As for example in Case C-61/89 Bouchoucha [1990] ECR I-3551.  See further, G Ringe, ‘Sparking Regulatory Competition in European Company Law: The Impact of the Centros Line of Case Law and its Concept of “Abuse of Law”’, ch 8 above. 28   Emsland-Stärke (n 3 above). 29   Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 26 27

Citizen of the Union: Above Abuse?  327 purpose. Even for the economic freedoms, attempting to distinguish between reality and ‘artificiality’ is fraught. It may seem like an easier exercise when the activities of natural persons are under assessment. However, even here, the notion of ‘artificiality’ may lead to fruitless lines of inquiry. Nationality and incorporation are equally ‘artificial’ legal constructs, one no more real than the other. As is evident in the discussion in Part II, while nationality has an affective dimension, the legal notion of ‘nationality’ is one acquired through an accident of birth. The evocative physicality of the notions of blood and soil should not obscure the purely legal notion of nationality. v. Power An assessment of the relative power of the right- and duty-holder is also relevant to the issue of abuse. When corporations attempt to circumvent EU rules and undermine home State laws (in particular when there is no prospect of re-regulation at the EU level), we should be concerned. Capital is more mobile than labour, so the potential imbalance in the Internal Market should be borne in mind with the vulnerability of domestic standards to downward pressure. In addition, with the increasing penetration of the Internal Market into the private realm, a type of private law abuse of rights may helpfully limit some of the potentially detrimental effects of the invocation of EU rights, particularly when the ability of private actors to claim public interest justification in their defence is legally constrained.30 Some EU rules operate by means of a codified asymmetrical notion of abuse to temper the more powerful actors’ freedom.31 For instance, the Fixed Term Workers Directive32 prohibits employers from using fixed term employment contracts beyond a certain period, in order to protect workers’ security of employment. Absent specific legislative intervention, there are no EU rules regulating termination of employment contracts, so there would be no place for an EU principle of ‘abuse’ of contractual freedom to play a role. Moreover, no abstract principle could hope to capture the array of conflicting interests that underlie policy choices about how to reconcile job security and employer flexibility, the latter being necessary for job creation.33 When it comes to worker-protective measures, the Court has interpreted the EU rules so as to apply them in a comprehensive manner, rejecting abuse claims. For example in Busch,34 it was alleged that an employee was abusing the EC right to return to work after parental leave in an ‘artificial’ manner, in order to gain a higher maternity allowance. The Court, keen to ensure job security and mindful of the prevalence of precarious employment, did not mention abuse at all, but simply interpreted the right to return to work as applicable in the scenario. Advocate-General Colomer rejected the suggestion of abuse in light of the particular facts, noting that reliance on the right could not be understood solely in terms of the purpose of attaining the higher maternity leave payment.35 30  I am thinking in particular of scenarios such as Case C-438/05 International Transport Workers’ Federation and Finnish Seamen’s Union v Viking Line [2007] ECR I-10779; Case C-341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet, Avdelning 1 [2007] ECR I-11767. 31   ‘Abuse of dominant position’ under Article 102 TFEU [ex Article 82 EC] might be included here also. 32   Directive 99/70 on fixed-term work 1999/70/EC [1999] OJ L175/43. Case C-212/04 Konstantinos Adenelner et al v Ellinikos Organismos Galaktos [2006] ECR I-6057. 33  In EU policy jargon, the apparently oxymoronic ‘flexicurity’ seeks to do this. 34   Case C-320/01 Busch v Klinikum Neustadt GmbH & Co Betriebs-KB [2003] ECR I-2041. 35  AG Colomer in Busch (n 34 above), para [39].

328  Cathryn Costello C.  Institutional Considerations i.  The Nature of EU Rights Comparativists tend to explain approaches to abuse of rights in light of the diverse approaches to rights themselves.36 ‘Abuse of rights’ emerges as a judicial construct where codified rights are rigid and exhaustive.37 In civilian systems, the abuse doctrine emerges in the context of exhaustive codes of abstract principles. In contrast, part of the explanation of the lack of abuse on common law relates to the specificity of legislation, and the tendency to codify limits and exceptions along with specified rights. Where a code is rigid, it is important to allow judges discretion to adjudicate problematic cases going beyond the letter of the code. Does this rationale hold in the EU context? In some instances, legislative blockages are indeed prevalent at EU level. Both Treaty provisions and legislation subject to unanimity procedures are difficult to amend should problems emerge. However, the EU legislature does often respond to judicial developments; indeed, their interaction is commonplace, with legislative provisions routinely reshaped and supplemented by judicial pronouncement. In Part IV, I examine the impact of Article 24 of the Citizenship Directive, which introduces restrictions on the benefits migrant EU citizens may claim from host States. We also examine the codification of the Lair38 test for distinguishing between genuine work and students abusively trying to pass themselves off as workers. Overall, it seems difficult to rationalise the role of abuse in light of the rigidity of EU rules. Although the Internal Market rules in the Treaty have undergone constitutionalisation into broad individual economic rights, this judicial transformation has been accompanied by open-ended grounds of restriction.39 The governmental propensity to invoke ‘abuse of rights’ with such frequency, when there are already open-ended grounds of justification, seems to reflect an anxious desire to escape the rigours of proportionality analysis for assessing Internal Market restrictions, where strict scrutiny seems often to be fatal in practice. Abuse seems a tempting plea, given its comparative lack of baggage. A historical account of abuse of rights tends to emphasise its role of a corrective to the excesses of private property rights.40 The Internal Market freedoms may be in need of some correction at present, but whether a discrete doctrine of abuse is apt is doubtful. Even if a rethink is warranted, the appropriate response may simply be to redraw the limits of the Internal Market guarantee, Keck-style,41 or address specific concerns about the standard 36  See further, J Snell, ‘The Notion of and a General Test for Abuse of Rights: Some Normative Reflections’, ch 15 above. 37  See further, A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of the European Court of Justice on Circumvention, Fraud and Abuses of Community Law’ in M Andenas and W Roth (eds), Services and Free Movement in EU Law (Oxford, Oxford University Press 2003) 245; J Voyame, B Cottier and B Rocha ‘Abuse of Right in Comparative Law’ in Abuse of Rights and Equivalent Concepts: The Principle and Its Present Day Application, Proceedings of the 19th Colloquoy on European Law, Luxembourg, 6–9 November 1989 (Strasbourg, Council of Europe 1990) 23. 38   Case 39/86 Sylvie Lair v Universität Hannover [1988] ECR 3161. 39  At least for national regulation that does not discriminate against foreign workers, goods and so on. See Case 120/78 Rewe-Zentrale AG v Bundesmonopolverwaltung für Branntwein [1979] ECR 649; Case C-415/93 Union Royale Belge des Sociétés de Football Association ASBL v Jean-Marc Bosman [1995] ECR I-4921. 40  Sajo (n 20 above) 29. 41   Joined Cases C-267/91 and C-268/91 Keck & Mithourd [1993] ECR I-6097.

Citizen of the Union: Above Abuse?  329 of assessment of justifications based on provision of public goods such as health care and education by the Member States.42 A discrete doctrine of abuse, operating only on specific factual assessments in individual cases, seems unlikely to address the need for systematic correction. ii.  ‘Abuse’: Limiting the Scope of Rights, or Derogation? The case law contains contradictory dicta on the nature and status of the concept of abuse. When we are dealing with the interpretation of harmonised EU rules, it appears that the doctrine of abuse acts as an interpretative principle, an inherent part of the teleological reasoning. This approach is embodied in the Opinion of Advocate-General Poiares Maduro in Halifax, where although characterising the doctrine as a ‘general principle’43 the sense in which abuse was employed was not as an autonomous principle, but rather went to define the scope of EU entitlements.44 Drawing on Lowe’s account in public international law, we should accordingly treat abuse of rights not as a general principle stricto sensu but rather as an ‘interstitial norm’ without ‘independent normative charge of [its] own’.45 An apparently contrasting approach emerges in the same Advocate-General’s Opinion in Cartesio,46 concerning freedom of establishment. Whilst advocating a broad construction of the notion of a barrier to freedom of establishment, Advocate-General Poiares Maduro regarded abuse as a ground of derogation, mentioned as but one aspect of the general public interest derogation. Similarly, Ninni-Orasche47 confirms that abuse does not go to the issue of whether someone is a worker, but rather arises once the economic freedom has been engaged. In Centros48 the Court affirmed that inherent in freedom of establishment was the right of secondary establishment of a company incorporated in the UK, where there is no domestic legal requirement of a tangible commercial presence. It also accepted that host Member States could rely on its concerns about fraud and abuse by way of derogation, if for example, the individuals were attempting to evade creditors or the like. These apparently contradictory approaches are reconcilable if we bear in mind both the division of labour in the preliminary reference procedure and the multifarious concerns that may underlie a plea of ‘abuse of rights’. To respond to a plea of abuse by 42  See for instance, the concerns expressed by A Somek, ‘Solidarity Decomposed: Being and Time in European Citizenship’ (2007) 32 European Law Review 787. 43  AG Poiares Maduro in Halifax (n 15 above), para [64]. 44  AG Poiares Maduro in Halifax (n 15 above), para [69]: That this notion of abuse operates as a principle governing the interpretation of Community law . . . What appears to be a decisive factor in affirming the existence of an abuse is the teleological scope of the Community rules invoked, which much be defined in order to establish whether the right claims is in effect, covered by such provisions, to the extent to which it does not manifestly fall outside their scope. This explains why the Court often refers not to abuse of rights, but simply to abuse. 45   V Lowe, ‘The Politics of Lawmaking: Are the Method and Character of Norm Creation Changing?’ in M Byers (ed), The Role of Law in International Politics: Essays in International Relations and International Law (Oxford, Oxford University Press 2000) 207, 217. 46  AG Poiares Maduro in Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641. 47  Case C-413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I-13187, para [26]. For a recent application, see Joined Cases C-22/08 and C-23/08 Athanasios Vatsouras (C-22/08) and Josif Koupatantze (C-23/08) v Arbeitsgemeinschaft (ARGE) Nürnberg 900, [2009] ECR I-4585 Opinion of AG Ruiz-Colomer. 48   Case C-212/97 Centros Ltd v Erhvers-og Selskabsstyrelsen [1999] ECR I-1459. See further, Case C-208/00 Überseering BV v Nordic Construction Company Baumanagement GmbH [2002] ECR I-9919; Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155.

330  Cathryn Costello asserting that particular behaviour falls within the scope of a given EU right excludes further general inquiry into its permissibility. The Court has an interpretative hegemony over the meaning and scope of the core terms. However, permitting examination whether individual applications are exceptionally ‘abusive’ opens up a fact-specific inquiry, which under the terms of the preliminary reference procedure, leaves more to the national judge. Another reason may lead the Court to keep the scope of EU freedoms broad, leaving abuse to derogation. The existence of EU legislative competence is often dependent on deeming the diversity of national rules to constitute barriers to Internal Market freedoms.49 Article 35 of the 2004 Citizenship Directive50 seems to embody the derogation approach, permitting Member States to ‘refuse, terminate or withdraw any right conferred by this Directive in the case of abuse of rights or fraud’, subject to proportionality analysis and procedural protections. Treating ‘abuse’ as a ground for derogation is more in keeping with the well-established requirement that findings of abuse require case-by-case assessment, taking due consideration of the evidence of the particular circumstances of the individual case.51 The crucial question in practice will be whether the standards for assessing abuse derogations are the same as those for assessing other public interest derogations. If the standards of assessment are the same, then little turns on whether we regard abuse as a discrete legal doctrine or not. However, the scheme and wording of the 2004 Citizenship Directive suggest that general derogations and those on grounds of abuse are different.52 II. Nationality of a member STate and Eu citizenship

The edifice of EU citizenship stands upon the diverse nationality laws of the Member States. Article 20 TFEU [ex Article 17 EC] provides: Citizenship of the Union is hereby established. Every person holding the nationality of a Member State shall be a citizen of the Union.

A Declaration to the Maastricht Treaty affirmed that the determination of who is a national of the Member States remains for the Member States alone.53 The Treaty of Amsterdam introduced the qualification that ‘citizenship of the Union shall complement and not replace national citizenship’, the Lisbon now replacing ‘complement’ with ‘shall be additional to.’ The most obvious reading of this caveat is simply as a reassertion of national sovereignty over ‘the crucial power of definition’.54 However, it may also simultaneously 49   Case C-376/98 Germany v Parliament and Council [2000] ECR I-8419; Case C-210/03 The Queen, on the application of: Swedish Match AB and Swedish Match UK Ltd v Secretary of State for Health [2004] ECR I-11893; Joined Cases C-154/04 and C-155/04 Alliance for Natural Health [2005] ECR I-6451; and Case C-380/03 Germany v Parliament and Council [2006] ECR I-11573. 50   Citizenship Directive (n 12 above). 51  AG Jacobs in Case C-147/03 Commission v Republic of Austria [2005] ECR I-5969, para [56], citing Case C-436/00 X and Y [2002] ECR I-10829, para [42]; Centros (n 48 above), para [25]. 52  See further, the discussion below of the English case of Kenya (n 152 below). 53   Final Act to the Maastricht Treaty, Second Declaration (on nationality of a Member State): The Conference declares that, wherever in the Treaty establishing the European Community reference is made to nationals of the Member States, the question whether an individual possesses the nationality of a Member State shall be settled solely by reference to the national law of the Member State concerned. Member States may declare, for information, who are to be considered their nationals for Community purposes by way of declaration lodged with the Presidency and may amend any such declarations when necessary. 54   J Shaw, ‘Constitutional Settlements and the Citizen After Amsterdam’ Jean Monnet Working Paper 7/98 Part I, 2: www.jeanmonnetprogram.org/papers/98/98-7-.html.

Citizen of the Union: Above Abuse?  331 enhance arguments for a better fit between nationality and EU citizenship. The notion of complementarity in particular could yet come to shape nationality designation, requiring it to fit the transnational, or perhaps even postnational, context of the EU.55 For now, absent any clear competence or political prospect of formal harmonisation,56 the Court has recognised the widest discretion on the part of the Member States to confer nationality. Micheletti 57 concerned a dual Argentine/Italian national seeking recognition in Spain as an EU citizen. According to Spanish law, he would be regarded as Argentinean, as the Spanish Civil Code in cases of dual nationality treated individuals as being of the nationality of their country of habitual residence. As such, Micheletti would not be entitled to rely on the EU’s free movement guarantees. The Court rejected the Spanish argument, establishing two key propositions concerning nationality designation. The first proposition recognises Member State competence over nationality, but subject to (as yet largely unspecified) EU law constraints: Under international law, it is for each Member State, having due regard to Community law, to lay down the conditions for the acquisition and loss of nationality.

The second proposition establishes a strict mutual recognition regime, such that: [I]t is not permissible for the legislation of a Member State to restrict the effects of the grant of nationality of another Member State by imposing an additional condition for recognition of that nationality with a view to the exercise of the fundamental freedoms provided for in the Treaty.58

A.  Member State Competence over Nationality Concerning the first proposition, the Court has as yet only hinted at the possible EU constraints. In Kaur59 a British Overseas citizen (without a domestic right of residence in the UK) challenged her exclusion from EU citizenship by the UK. The Court affirmed the Member State’s prerogative to determine who its nationals were for the purposes of EU law. However, the Court hinted that had the UK wrongly sought to deprive individuals of nationality that would be potentially subject to review.60 Recall too that the Court has jurisdiction to give rulings only on questions of EU law, so in answering the question it was establishing the EU dimension to the issue.61 Garcia Avello62 confirms that dual 55   From the outset, commentators have discussed the potentially productive tension between nationality and EU citizenship. See for example, S O’Leary, ‘Nationality Law and Community Citizenship: A Case of Two Uneasy Bedfellows’ [1992] Yearbook of European Law 353; C Closa, ‘Citizenship of the Union and Nationality of the Member States’ in D O’Keeffe and PM Twomey (eds), Legal Issues of the Maastricht Treaty (London, Wiley 1994) 109. 56  EU law does impact upon nationality law, even in the absence of formal harmonisation. See further, K Rostek and G Davies, ‘The Impact of Union Citizenship on National Citizenship Policies’ (2006) 10 European Integration Online Papers: eiop.or.at/eiop/index.php/eiop/article/view/2006_005a/22. 57   Case C-369/90 Mario Vicente Micheletti et al v Delegación del Gobierno en Cantabria [1992] ECR I-4239. 58   Micheletti (n 57 above), para [10]. 59   Case C-192/99 R v Secretary of State for the Home Department ex p Kaur [2001] ECR I-1237, para [19]. 60  The UK’s actions ‘did not have the effect of depriving any person who did not satisfy the definition of a national of the United Kingdom of rights to which that person might be entitled under Community law’. Kaur (n 59 above), para [25]. 61  In contrast, Advocate-General Léger treated the situation as being wholly internal to the UK, and so did not reply to the question, although he suggested that a fundamental rights question might arise under Article 3 of the Fourth Protocol to the ECHR, which prohibits the expulsion by a state of its own nationals. 62   Case C-148/02 Garcia Avello v Belgian State [2003] ECR I-11613.

332  Cathryn Costello nationality of two EU Member States may bring individuals within the scope of the EU law. For private international lawyers, the case raises particular concerns.63 From an EU law perspective, it follows from the Micheletti propositions. However, those propositions arise only concerning the recognition of nationality of the Member States.64 The ruling in Rottmann 65 recently provided a crucial further elucidation. Rottmann was an Austrian citizen by birth, and accordingly on Austria’s EU accession, became an EU citizen. In 1998, he naturalised as a German citizen, falsely asserting in the process that he had no criminal record. By virtue of acquiring the nationality of another state, Austrian law then automatically deemed him to have lost his Austrian nationality. In the meantime, the German authorities were informed of his criminal conviction and withdrew his German nationality. The German Federal Administrative Court referred questions to the Court on whether what was then Article17(1) EC [now Article 20(1) TFEU] precluded actions of the German sort, where the effect was to render the individual stateless. AdvocateGeneral Poiares Maduro characterised nationality of a Member State and EU citizenship as ‘two autonomous, yet inextricably linked notions’.66 Not only the general principles of EC law, but potentially ‘any norm of Community law’ may apply to constrain the removal or award of nationality.67 In the instant case, he opined that Germany had clearly acted on a legitimate State interest to avoid fraud.68 The ECJ similarly found that withdrawal of naturalisation in the circumstances was subject to ‘judicial review carried out in the light of European Union law’69 emphasising in particular the EU principle of proportionality, albeit leaving it to the national court to apply it to the facts.70 The range of EU legal constraints that may be applicable goes beyond the general principles, and includes the rules on EU citizenship themselves. Space precludes an exhaustive analysis of the case,71 but for present purposes, perhaps most noteworthy is the recognition that preventing fraud is a basis for withdrawal, without any reference to the concept of ‘abuse’. As I argue that the doctrine of abuse does not have the status of a constitutional general principle, it should not be available as a matter of EU law to either require or permit withdrawal of naturalisation. An account of citizenship rights would be incomplete without reference to two recent cases concerning the right to vote in European parliamentary elections for residents of Gibraltar and Aruba in which the Court affirmed the applicability of EC general

63  P Lagarde, ‘Comments on Garcia Avello’ [2004] Revue Critique de Droit International Privé 192. G de Groot, ‘Towards European Conflict Rules in Matters of Personal Status’ (2004) 11 Maastricht Journal of European and Comparative Law 115. 64  C-179/98 Belgian State v Fatna Mesbah [1999] ECR 1-7955 concerned a dual Moroccan/Belgian national resident in Belgium, who sought to rely on the non-discrimination clause in the Association Agreement between the EC and Morocco. The Court confirmed that the situation was wholly internal to Belgium, as Belgian law alone determined that she was Belgian for EC law purposes. 65   Case C-135/08 Janko Rottman v Freistaat Bayern, Opinion of AG Poiares Maduro, 30 September 2009, CJEU Ruling, 2 March 2010 nyr. 66  AG Poiares Maduro in Rottman (n 65 above), para [23], citing C Closa, ‘Citizenship of the Union and Nationality of Member States’ (1995) 32 Common Market Law Review 487. 67  AG Poiares Maduro in Rottman (n 65 above), para [28]. 68  AG Poiares Maduro in Rottman (n 65 above), para [33]. 69   CJEU (n 65 above) para [48]. 70   CJEU (n 65 above) para [58]. 71  See further, the various contributions in J Shaw (ed) ‘Has the European Court of Justice Challenged Member State Sovereignty in Nationality Law?’ EUI Robert Schuman Centre for Advanced Studies EUDO Citizenship Observatory 2010, available at .

Citizen of the Union: Above Abuse?  333 principles.72 Spain v UK73 concerned the measures adopted by the UK to remedy the violation of the right to vote identified by the European Court of Human Rights in Matthews.74 Spain challenged the UK’s attribution of European Parliament (EP) voting rights to certain individuals resident in Gibraltar, a British Crown Colony to which EU law applies. The individuals were Qualifying Commonwealth Citizens, rather than UK citizens per se. Spain argued that voting rights should be conferred only on EU citizens, relying on the then EC Treaty statement that the EP brings together ‘representatives in the European Parliament of the peoples of the States brought together in the Community’ in Articles 189 and 190 EC [now replaced principally by Articles 14(1-3) TEU although the EU Treaty now refers to representatives of the Union’s citizens]. The Court rejected this argument, noting that ‘the term “peoples” . . . may have different meanings in the Member States and languages of the Union’.75 The ‘fundamental nature’ of EU citizenship did not ‘necessarily mean that the rights recognised by the Treaty are limited to the citizens of the Union’.76 Rather than simply stopping at the point of asserting that it was for the Member State to distribute voting rights, the Court noted that the individuals granted the right to vote had ‘close links’77 in the form of ‘a specific link with the territory’.78 The case is important for its rejection of tying the franchise exclusively to formal citizens, in contrast to some other jurisdictions.79 The Lisbon change may be a retrograde one in this respect. Spain may directly challenge the attribution of voting rights on EU law grounds, albeit not the attribution of nationality itself. In Eman & Sevinger,80 Dutch citizens in Aruba argued that the Netherlands should have accorded them voting rights.81 The Court found a violation of the general principle of equal treatment in the differential treatment of Dutch citizens resident in third countries and those resident in Aruba.82 Eman & Sevinger and Rottmann sit awkwardly with Kaur, although the earlier decision may be explained by the fact that Kaur did not hold UK nationality stricto sensu, but rather the status of British Overseas citizen. Moreover, when the free movement provisions are at stake, the Court has apparently insulated nationality attribution from scrutiny. No doubt, the Matthews background was influential in the voting rights cases, prompting the Court to review the issues in order to protect the fundamental right to vote, although that was not operative in its minimalist reasoning. For the purposes of the current discussion, the significance of Commission v Spain, Eman & Sevinger and Rottmann is the shift towards some EU law constraints on the 72  See further, L Besselink, ‘Casenotes: Case C-145/04 and C-300/04’ (2008) 45 Common Market Law Review 787; J Shaw, ‘The Political Representation of Europe’s Citizens: Developments’ (2008) 4 European Constitutional Law Review 162. 73   Case C-145/04 Spain v UK [2006] ECR I-7917. 74   Matthews v United Kingdom (App no 24833/94) (1999) 28 EHRR 361. 75   Spain v UK (n 73 above), para [71]; repeated in Eman & Sevinger (n 80 below), para [44]. 76   Spain v UK (n 73 above), para [79]. 77   Spain v UK (n 73 above), para [78]. 78   Spain v UK (n 73 above), para [79]. 79  See further, German Constitutional Court BVerfGE 83, 11, Nr 3, 37. See further, S Benhabib, The Rights of Others: Aliens, Residents and Citizens (Cambridge, Cambridge University Press 2004) 202–09. 80   Case C-300/04 MG Eman and OB Sevinger v College van burgemeester en wethouders van Den Haag (Eman & Sevinger) [2006] ECR I-8055. 81  The predicament of such nationals was highlighted by G de Groot, ‘The Relationship Between the Nationality Legislation of the Member States of the European Union and European Citizenship’ in M La Torre (ed), European Citizenship: An Institutional Challenge (Hague, Kluwer Law International 1998) 115, 130. 82   Eman & Sevinger (n 80 above), paras [58–61].

334  Cathryn Costello withdrawal of nationality and distribution of attendant EU citizenship rights. While these scope of these constraints remains vague, the absence of references to ‘abuse’ is welcome, as is any related attempt to fashion a ‘genuine’ nationality construct with nationality being distributed, say, only to those with close links. Nationality may arguably be underinclusive in many EU countries, and much could be done to include and enfranchise long-term residents.83 However, such a judicial requirement would amount to a tool to refashion entirely nationality law, a step too far under current conditions. B. Strict Mutual Recognition of Nationality in the Context of Free Movement The second Micheletti proposition creates a strict duty of mutual recognition, even in the absence of common minimum standards. Consider Collins, a jobseeker in the UK relying on his Irish nationality, although he had never lived or worked in Ireland. Dr de la Feria expresses surprise that the issue of abuse was not raised in this case.84 However, all systems of nationality by descent (jus sanguinis) may end up attributing nationality to individuals who have never set foot in their country of nationality. Many EU Member States attribute nationality even to the second generations of emigrants, reflecting a diasporic nationality. As Rainer Baübock notes: ‘The driving force for this de-territorialization of citizenship was ethnic nationalism . . . Emigration stretches the community of descent beyond the state territory.’85 Ethnic nationalism certainly has many unpalatable implications, particularly if diasporic nationality is instituted in tandem with a retraction of access to nationality for residents in the territory of ethnic minority background, as is the trend for instance in Italy, suggesting an exclusive ethnos.86 Developments in nationality law should be on the radar of EU policymakers and scholars, being crucial to any contextualised understanding of EU citizenship. However, entertaining a legal claim of abuse would involve deeming a Member State’s attribution of nationality as improper, without the possibility of a shaping a positive EU entitlement to nationality. It is difficult to see what valuable purpose would be served by handing other EU Member States a tool to undermine one another’s attributions of nationality when free movement was at stake. Nationality by birth on the territory (jus soli) also fails to map onto a genuine link with state of nationality in some instances, as birth in the territory may be sufficient to attribute nationality, without the child’s family having any established links to the country. In the EU, only Ireland maintained a system of absolute jus soli, the underlying source of controversy in Chen.87 The scenario is also peculiar in that baby Chen had acquired Irish nationality by birth in Northern Ireland, part of the United Kingdom. Baby Chen’s mother, a Chinese citizen, arranged to give birth in Northern Ireland in 83  The Tampere European Council in 1999 endorsed the objective ‘that long-term legally resident third country nationals should be offered the opportunity to obtain the nationality of the Member State in which they are resident’. Tampere Summit Conclusions, para [21]. 84  R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 416. 85  R Bauböck, ‘Citizenship and National Identities in the European Union’, Jean Monnet Working Paper 97/4, Part 2(1): www.jeanmonnetprogram.org/papers/97/97-04-.html. 86  See further, F Pastore, ‘A Community Out of Balance: Nationality Law and Migration Politics in the History of Post-Unification Italy’ (2004) 9 Journal of Modern Italian Studies 27. 87  Case C-200/02 Kunqian Catherine Zhu and Man Lavette Chen v Secretary of State for the Home Department (Chen) [2004] ECR I-9925.

Citizen of the Union: Above Abuse?  335 order to ‘create a right of residence’ for both her and her daughter in the UK. The Court reasserted the two Micheletti propositions, so the UK was obliged to recognise Ireland’s attribution on nationality, and was precluded from imposing any additional conditions for that recognition.88 In rejecting the UK’s attempt to rely on the doctrine of abuse of rights from Centros, the Court noted that: ‘None of the parties that submitted observations to the Court has questioned either the legality, or the fact, of Catherine’s acquisition of Irish nationality.’89 Would the notion of abuse have been helpful here? Dr de la Feria draws the parallel with other cases where the transaction was ‘artificial’.90 Professor Sørensen also sees the Chen case through the prism of abuse, arguing that ‘the judgment could indicate that if the Irish Government had argued that Irish legislation was abused, its claim would have been more likely to have succeeded’.91 Similarly, Advocate-General Jacobs, writing extracurially, has characterised Chen as presenting ‘circumstances which might have been regarded as an abuse of rights’.92 The notion of the ‘artificiality’ of the link with Ireland (as manifested in the decision to give birth in Northern Ireland) is particularly unhelpful. To suggest that birth in the territory, when the place of birth is chosen by the mother for legal reasons, is similarly ‘artificial’ seems to miss the crucial point that nationality by birth is automatic, consequent on either place or parentage in all cases, and does not enquire into underlying purpose in any case. As Advocate-General Tizzano opined, the UK’s argument could lead to ‘suspicions of abuse . . . in almost all cases of intentional acquisition of nationality of a Member State. And, paradoxically, that could lead to a situation in which the enjoyment of rights deriving from citizenship of the Union was subject to the condition . . . that such citizenship had to have been acquired involuntarily’93 in violation of the second Micheletti proposition. The ruling in Chen may have created a regulatory gap for the two Member States concerned. It meant that children born in Northern Ireland to parents with no link to either Member State could establish a legally protected EU right to reside in the UK, obviating home State control of immigration of such families. However, the ruling did not have any systemic impact across the EU. As Advocate-General Tizzano stated, ‘the problem, if problem there be, lies in the criterion used by the Irish legislation granting nationality [iure soli]’ and that ‘to avoid such situations, the criterion could have been moderated by the addition of a condition of settled residence of the parent’.94 In this context, we may surmise that the Court wisely refused to introduce any general exception into the system, the effects of which would be liable to go way beyond the narrow confines of the case. Instead, the Court emphasised the peculiarity of the case. Had Ireland wished 88  See further, J Carlier, ‘Annotation’ (2005) 42 Common Market Law Review 1211; B Hofstötter, ‘A Cascade of Rights, or Who Shall Care for Little Catherine? Some Reflections on the Chen Case’ (2005) 30 European Law Journal 548; A Tryfonidou, ‘C-200/02, Kunqian Catherine Zhu and Man Lavette Chen v Secretary of State for the Home Department: Further Cracks in the “Great Wall” of the European Union?’ (2005) 11 European Public Law 527; B Kunoy, ‘A Union of National Citizens: The Origins of the Court’s Lack of Avant-Gardism in the Chen Case’ (2006) 43 Common Market Law Review 179. 89   Chen (n 87 above), para [38]. 90   de la Feria (n 84 above) 429. 91   KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 448. 92   F Jacobs, ‘Citizenship of the European Union—A Legal Analysis’ (2007) 13 European Law Journal 591, 596. See further, A Arnull, The European Union and Its Court of Justice, 2nd edn (Oxford, Oxford University Press 2006) 531; Kunoy (n 88 above) 179, 185. 93  AG Tizzano in Chen (n 87 above), para [127]. 94  AG Tizzano in Chen (n 87 above), paras [124–25].

336  Cathryn Costello to contest baby Chen’s nationality, it would have needed domestic legal authority to do so. At the time of her birth, the nationality rule set out in the Irish Constitution was clear: all those born on the island of Ireland were entitled to Irish nationality. Subsequently, Ireland amended its Constitution, removing the provisions of automatic jus soli, admittedly going much further than was required to bring the Irish law in line with that of the UK.95 The Court states that Member States designate nationality ‘in accordance with international law’. However, international law does in extremis countenance legitimate refusal to recognise another State’s attribution of nationality. In Nottebohm, the International Court of Justice (ICJ) apparently required States to respect a notion of ‘genuine link’ in attributing nationality.96 Some commentators treat Nottebohm as an abuse of rights case.97 Nottebohm acquired Liechtenstein nationality in order to evade the consequences of the international law of war. The ICJ arguably saw this naturalisation as an abuse of rights, but may have hesitated to say so to avoid impugning the conduct of a sovereign state.98 The focus in public international law is on states abusing their rights to attribute nationality, thereby undermining other important rules of public international law. In contrast in the EU we have a strict duty of mutual recognition. The Micheletti propositions fit the EU context, and in effect preclude claims of abuse concerning nationality designation. We can explain the Court’s approach partly as necessary to secure free movement, and partly as reflective of higher levels of trust across the EU. In particular, any generalised requirement to demonstrate that nationality reflects a ‘genuine link’ would leave many nationals of the Member States vulnerable. The notion of abuse, if it were to embody a requirement of ‘genuine’ or ‘close links’, would allow Member States to undermine one another’s attribution of nationality, touching on politically sensitive nerves, including issues of contested territory and post-colonial ties. All designations of nationality at birth had to predict a child’s life-course and where she will develop genuine links. A ‘real link’ test would be in danger of undermining the crucial clarity and certainty of regimes of citizenship by birth. However, in reaffirming that attributing nationality is for the Member States, whilst at the same time applying the general principles and, if the Rottmann Opinion is followed, other EU constraints, to Member States’ allocation of nationality, the Court may tentatively contribute to some common precepts—a tentative process of judicial minimum harmonisation. For the moment, the determination of who is a citizen of the Union rests with the individual Member States, and the Court’s approach is fitting.

95  The Constitution of Ireland was amended so that citizenship is granted to children born on the island of Ireland if they have ‘at least one parent who is an Irish citizen or entitled to be an Irish citizen’. For a critique, see C Costello, ‘Accidents of Place and Parentage: Birthright Citizenship and Border Crossings’ in The Citizenship Referendum: Implications for the Constitution and Human Rights (Dublin, School of Law, Trinity College Dublin 2004) 5. B Ryan, ‘The Celtic Cubs: The Controversy over Birthright Citizenship in Ireland’ (2004) 6 European Journal of Migration and Law 173. 96   Nottebohm Case (Liechtenstein v Guatemala) (Second Phase), Judgment of 6 April 1955, ICJ Reports 1955, 4. 97  RD Sloane, ‘Breaking the Genuine Link: The Contemporary International Legal Regulation of Nationality’ (2009) 50 Harvard International Law Journal 1; cf Schwarzenberger (n 8 above). 98  Sloane (n 97 above) 36 citing ILC First Report on Diplomatic Protection.

Citizen of the Union: Above Abuse?  337 C.  The Analogy between Nationality and Incorporation The contrast between the citizenship cases and those concerning the attribution of nationality to legal persons is illuminating.99 Home States are entitled to impose restrictions on emigrating companies, as is evident in Daily Mail100 and most recently in Cartesio.101 In the latter case, the Court departed from the Opinion of AdvocateGeneral Poiares Maduro,102 holding that Article 48 EC [now Article 54 TFEU] did not preclude the law of the State of incorporation from preventing companies moving their commercial headquarters to another Member State. The Court explicitly drew a parallel with nationality designation by treating the rules of the State of incorporation as ‘like the question whether a natural person is a national of a Member State’ and accordingly ‘in the absence of a uniform Community law definition of the companies . . . can only be resolved by the applicable national law’.103 There seem to be two contrasting ideas here. The first concerns the prospect for EU harmonisation and the second concerns home State control and exit restrictions. On both points, the analogy may be strained. Nationality is unlike incorporation in that there is a European company, while there is no European nationality. On the second point, Cartesio seems to give too much ground to home State control, in that it precludes reliance on freedom of establishment to challenge features of home State rules that do go to prevent primary corporate emigration, in a manner that seems to suggest a sharp distinction between primary and secondary establishment.104 In contrast, nationality merely goes to status, rather than the right to emigrate. D.  Questioning Connection: A Role for Abuse? The main impact of the citizenship provisions, as is explored in the next section, is to require Member States to treat resident EU citizens like their own. The host State’s duty to afford migrant EU citizens equal treatment increases over time, with duration of residence taken as a proxy for integration. The equal treatment guarantee applies widely. For example, in Konstantinidis,105 the Court found a violation of Article 12 EC [now Article 18 TFEU] in the manner in which Germany obliged a Greek citizen to use a transliteration of his name that did not render its pronunciation accurately. The Court condemned Germany’s actions as discriminatory, given that home nationals would benefit from accurately recorded names.106 In later cases, we see that the citizenship provisions may also enable individuals to insist on recognition of names, acquired not under the law of the country of residence  See generally, Kunoy (n 88 above) 183.   Case 81/87 R v HM Treasury and Commissioners of Inland Revenue, ex p Daily Mail and General Trust plc [1988] ECR 5483. 101   Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641. 102  AG Poiares Maduro in Cartesio (n 101 above). 103   Cartesio (n 101 above), para [109]. 104  See further, WG Ringe, ‘No Freedom of Emigration for Companies?’ (2005) 16 European Business Law Review 621. 105   Case C-168/91 Christos Konstantinidis v Stadt Altensteig—Standesamt and Landratsamt Calw— 106   Konstantinidis (n 105 above), para [13]. The ECJ also condemned the measure as creating a barrier to freedom of establishment under Article 43 EC [now Article 49 TFEU], see para [15].     99 100

338  Cathryn Costello but in the country of origin. Garcia Avello107 concerned children of dual SpanishBelgian nationality who sought recognition by Belgium, their country of residence, of their Spanish surnames. The Court held that they had such a right, on the basis that it was discriminatory for Belgium to treat them as purely Belgian citizens. In effect, this created a duty to accommodate their dual nationality, reading Articles 12 and 17 EC [now Articles 18 and 20 TFEU] in tandem. In Grunkin Paul,108 the Court required the German authorities to take into account the name registered to a German child in Denmark, his country of birth. Failure to do so, although it did not amount to discrimination on ground of nationality (all Germans being subjected to the same strict rules), did violate Article 18 EC [now Article 21 TFEU], being liable to disadvantage the applicant for having exercised his rights to free movement. The Court in effect condemned the German practice of only having regard to nationality as a connecting factor, whereas the applicant, although a German national, had been born and registered in Denmark. The cases illustrate that EU citizenship takes precedence over normal national rules of private international law. As Ralf Michaels points out, in both Garcia Avello and Grunkin Paul, the children ‘successfully avoided application of the law of both their nationality and their habitual residence’, which would otherwise have been applicable.109 The Court is gradually refashioning the traditional private international law rules to fit the transnational EU context. This case law could lead to further claims for recognition of other statuses acquired under the law of other Member States, be that the country of nationality, place of birth, residence or other connection. The trend seems to require a Member State to take into account EU citizens’ transnational and multinational heritage in applying their domestic rules. Member States may want to resist such claims, but on what basis? Does the abuse doctrine have a role to play? In her Opinion in Grunkin Paul, Advocate-General Sharpston suggested that refusal to recognise a name might be justifiable where the law conferring the name was that of a State to which the connection was too tenuous. She suggested in particular that this would be so if the child was connected by birth but not nationality and ‘if the place of birth [was] shown to have been chosen simply in order to circumvent the rules of the Member State of nationality, without there being any other real connection with that place’.110 The attendant footnote continues: To allow such a justification would, admittedly, involve some tension with the Court’s judgment in [Chen] in which it rejected an argument that it was not possible to rely on nationality of a Member State acquired by virtue of a place of birth deliberately chosen for that sole purpose. However, the Court’s reasoning there was based on the right of each Member State to lay down the conditions for acquisition of nationality, and did not concern the use of nationality or any other criterion as a connecting factor for purposes of private international law.111

  Garcia Avello (n 62 above).   Case C-353/06 Stefan Grunkin and Dorothee Regina Paul (Grunkin Paul) [2008] ECR I-7639. A previous reference at an earlier stage in the same proceedings had been rejected as inadmissible by the ECJ, although AG Jacobs did examine the substance of the issue, coming to the same result as the ECJ in the later case, see AG Jacobs in Case C-96/04 Standesamt Stadt Niebull v Stefan Grunkin, Dorothee Regina Paul [2006] ECR I-3561. 109  R Michaels, ‘The New European Choice of Law Revolution’ (2008) 82 Tulane Law Review 1607, 1632–33. 110  AG Sharpston in Grunkin Paul (n 108 above), para [86]. 111  AG Sharpston in Grunkin Paul (n 108 above), fn 47, citing Case C-370/90 The Queen v Immigration Appeal Tribunal and Surinder Singh, ex p Secretary of State for Home Department (Surinder Singh) [1992] ECR I-4265, para [24], and the case law cited therein. 107 108

Citizen of the Union: Above Abuse?  339 This footnote raises an issue that will no doubt be of further concern as individuals seek to rely on statuses available under the laws of other Member States. III.  FAMILY MEMBERS OF EU CITIZENS

EU legislation has long recognised that free movement of EU workers necessitated free movement for their family members, irrespective of the latter’s nationality.112 In later cases, even in the absence of express legislative residence rights, the Court created residence rights for family members in order to avoid impediments to free movement and vindicate the fundamental right to family life, as in Chen,113 Carpenter114 and Baumbast.115 As the EU Member States tightened the conditions for family migration, this EU right became particularly valuable, encouraging attempts to invoke it in novel situations. The classic situation undoubtedly covered by EU law was an EU worker moving to another Member State along with her family members who were lawfully resident TCNs in the home Member State. Carpenter116 marked a striking extension of the scope of EU rules, with a UK citizen residing in the UK found to fall within the scope of EU law, due to the fact that he travelled to other EU Member States to provide services, so as to require a right to reside for his TCN spouse on the basis that: [T]he separation of Mr and Mrs Carpenter would be detrimental to their family life and, therefore, to the conditions under which Mr Carpenter exercises a fundamental freedom. That freedom could not be fully effective if Mr Carpenter were to be deterred from exercising it by obstacles raised in his country of origin to the entry and residence of his spouse.117

No doubt it is the broad scope of EU rules that has prompted greatest comment and critique.118 A.  Licit Circumvention of National Immigration Law Surinder Singh confirmed that the rights of the TCN family member persisted on return to the home State.119 However, in that ruling the Court also accepted en passant that Member States were permitted to prevent abuse of EC law, which it equated with evasion of national legislation.120 Clarification of the scope of abuse became a matter of practical importance as blended TCN/EU families facing immigration difficulties increasingly availed themselves of the Surinder Singh route. Various transborder flashpoints have emerged. The UK Home Office complained of the ‘Dublin hop’;121 the Dutch government  Article 10 of Regulation 1612/68 [1968] OJ SpEd L257/2.   Chen (n 87 above).   Case C-60/00 Mary Carpenter v Secretary of State for the Home Department [2002] ECR I-6279. 115   Case C-413/99 Baumbast and R v Secretary of State for the Home Department [2002] ECR I-7091. 116   Carpenter (n 114 above). 117   Carpenter (n 114 above), para [39]. 118  Editorial Comments, ‘Freedoms Unlimited? Reflections on Mary Carpenter v Secretary of State’ (2003) 40 Common Market Law Review 537; L Jadoul and F Vanneste, ‘Note on Case C-60/00 Mary Carpenter v Secretary of State for the Home Department’ (2003) 9 Columbia Journal of European Law 447, 451. 119   Surinder Singh (n 111 above), para [24]. 120   Surinder Singh (n 111 above), para [24], citing in particular the judgments in Case 115/78 Knoors v Secretary of State for Economic Affairs [1979] ECR 399, para [25], and Bouchoucha (n 26 above), para [14]. 121   G Barrett, ‘Family Matters: European Community Law and Third-Country Family Members’ (2003) 40 Common Market Law Review 369, 379–80. 112 113 114

340  Cathryn Costello disliked the ‘Belgian route’;122 and the Danish authorities introduced restrictions on movement to other EU countries by Danish/TCN blended families.123 Akrich124 limited the scope for claims of abuse. The case concerned the TCN spouse of a UK citizen, Hacene Akrich. He had repeatedly violated UK immigration law and was facing deportation. He and his British wife moved to Ireland in order to rely on Surinder Singh, she taking a job for a period of around six months. On return to the UK, the government strenuously contested the applicability of his EC right to reside, deeming the scenario to be abusive in its entirety. Advocate-General Geelhoed opined that actually installing themselves in another Member State was ‘by its nature not a misuse of law’125 and, in light of Surinder Singh, neither was the return to the home State,126 irrespective of motive. The immigration law benefit had to be regarded as ‘inherent’ in the rights to move and return.127 The Advocate-General considered whether a minimum residence period should be imposed in order to ensure that the move to the other Member State was ‘genuine’, but rejected such a requirement as being easily amenable to circumvention also.128 (Somewhat confusingly, Advocate-General Geelhoed in his subsequent Opinion in Jia characterised Akrich as a case of abuse,129 although his Opinion in Akrich had explicitly deemed the actions involved as legitimate.) The Court agreed that ‘the motives which may have prompted a worker to seek employment in another Member State are of no account as regards his right to enter and reside in a Member State, provided he pursues an effective and genuine activity’.130 In so doing, the Court agreed that motive was irrelevant, and that it was not abusive to move back relying on the EU rights of residence ‘even if the spouse did not, at the time when the couple installed itself in another Member State, have a right to remain in the Member State of which the worker is a national’.131 The Court did, however, instantiate a form of abuse, incorporating the concept of marriages of convenience from domestic

122  See J Bierbach, ‘European Citizens’ Third-Country Family Members and Community Law’ (Casenote on Case C-291/05 Eind ) (2008) 4 European Constitutional Law Review 344, in particular 358. 123  It has been reported that thousands of couples have left Denmark, choosing to settle in Sweden or in northern Germany in order to avoid the strictures of Danish family reunification law. R Jozwiak, ‘Denmark Feels Immigration Law is under Threat’ EU Observer, 1 August 2008. 124   Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607. See further, E Spaventa, ‘Casenote: Case C-109/01 Secretary of State for the Home Department v Hacene Akrich’ (2005) 42 Common Market Law Review 225. 125  AG Geelhoed in Akrich (n 124 above), para [181]. 126  AG Geelhoed in Akrich (n 124 above), para [182]. 127  AG Geelhoed in Akrich (n 124 above), paras [178–80]. 128  AG Geelhoed in Akrich (n 124 above), para [176]. 129  AG Geelhoed in Case C-1/05 Yunying Jia v Migrationsverket (Jia) [2007] ECR I-0001, para [75]: Even if there were complete harmonisation of the rules applicable to the admission of third-country nationals at the external borders of the Community, to permit certain categories of persons to gain access to the territory of a Member State and to reside there on the basis of the family relationship with a Community citizen who has happened to have exercised his right to free movement would undermine the effectiveness of these common rules and invite abuse of the type which was at the basis of the case in Akrich. To establish a right of residence merely on the chance factor of a national of a Member State having moved to another Member State is both arbitrary and unjust as it creates inequality in respect of Community citizens who have not exercised this right and in respect of third-country nationals who do not have the privilege of being the relative of a migrant Community citizen. 130   Akrich (n 124 above), para [55], citing Case 53/81 DM Levin v Staatssecretaris van Justitie [1982] ECR 1035, para [23]. 131   Akrich (n 124 above), para [56].

Citizen of the Union: Above Abuse?  341 legal systems.132 It also stated, conversely, that where the marriage was genuine, Member States were to have regard to the fundamental right to family life.133 The Court’s key holding in Akrich was that the EU legislation covered only freedom of movement within the Community and was silent on the rights of a TCN spouse to enter the territory of the EC for the first time. Accordingly, the EU rights only applied to spouses ‘lawfully resident in a Member State when he moves to another Member State to which the citizen of the Union is migrating or has migrated’.134 In Metock, the Court explicitly abandoned the prior lawful residence requirement,135 affirming that TCN family members had an EC right of entry and residence. Accordingly, Member States may not impose their normal immigration laws on such family members. The Court precluded claims of abuse arising out of past infractions of domestic immigration law, singling out for assessment under Article 27 of the 2004 Citizenship Directive (rather than Article 35 thereof ) instances ‘where the Member State wishes to penalise the national of a non-member country for entering into and/or residing in its territory in breach of the national rules on immigration before becoming a family member of a Union citizen’.136 Accordingly, the notion of abuse appears to have been whittled down mainly to abuse of family law, in particular, via sham marriage. B.  Contrast with Cases Where National Authorisation is Decisive A useful contrast may be drawn with the role the doctrine of abuse plays in cases where TCNs derive residence rights from other EU rules, but where national competence over the authorisation of entry remains, such as under the Europe Agreements137 and the Association Agreement with Turkey.138 We will focus on the latter for reasons of space. 132   Akrich (n 124 above), para [57]: [T]here would be an abuse if the facilities afforded by Community law in favour of migrant workers and their spouses were invoked in the context of marriages of convenience entered into in order to circumvent the provisions relating to entry and residence of nationals of non-Member States. 133   Akrich (n 124 above), para [58]. 134   Akrich (n 124 above), para [50]: In order to benefit in a situation such as that at issue in the main proceedings from the rights provided for in Article 10 of Regulation No 1612/68, the national of a non-Member State, who is the spouse of a citizen of the Union, must be lawfully resident in a Member State when he moves to another Member State to which the citizen of the Union is migrating or has migrated. 135   Metock (n 1 above), para [58], citing Case C-459/99 Mouvement contre le racisme, l’antisémitisme et la xénophobie ASBL (MRAX) v Belgian State [2002] ECR I-6591, para [59], and Case C-157/03 Commission v Spain [2005] ECR I-2911, para [28]. 136   Metock (n 1 above), para [96]. 137  Case C-63/99 The Queen v Secretary of State for the Home Department, ex p W Gloszczuk and E Gloszczuk [2001] ECR I-6369. The ECJ confirmed that Member States were permitted to decline applications of individuals who sought to rely on the Europe Agreement in an abusive manner (citing Centros). ‘Abuse’ would arise if ‘when that application was made, the applicant was residing illegally within its territory by reason of false representations made to those authorities for the purpose of obtaining initial leave to enter that Member State on a separate basis or of the failure to comply with an express condition attached to that entry and relating to the authorised duration of his stay in that Member State’. (Para [77].) 138  The Agreement does not confer a right to freedom of establishment, but rather in Article 41(1) of the Additional Protocol of 1970 contains a standstill clause, precluding any additional restrictions on freedom of establishment, which does apply to conditions of entry and residence. Case C-16/05 R (on the application of Veli Tum and Mehmet Dari) v Secretary of State for the Home Department [2007] ECR I-7415.

342  Cathryn Costello Suat Kol139 concerned an attempt to rely on residence rights under the Association Agreement with Turkey. Mr Kol’s initial residence permit was acquired via false statements concerning his marriage to a German woman. Advocate-General Elmer and subsequently the Court stated that the periods of residence after an initial permit had been secured by fraudulent means could not be regarded as lawful. In contrast, Günaydin140 concerned a Turkish worker whose initial residence in Germany was under a temporary permit, and under which he had undertaken to leave the country on expiry. The German government sought to preclude his reliance on that initial period of residence as the basis for a secure right of residence under the Association Agreement, arguing that otherwise it would amount to an abuse. Advocate-General Elmer sought to distinguish between the fraudulent case of Suat Kol, and the case of Günaydin where an individual gave a bona fide undertaking at the time of initial residence, but his position subsequently changed. The Court explicitly ruled out deeming situations where the initial right of residence was temporary as inherently abusive. Only when the worker made false statements with the ‘sole purpose’ of inducing the authorities to issue a residence permit that the subsequent periods of residence could be disregarded under the Association Agreement. A claim of abuse also arose in Tum and Dari.141 The UK sought to argue that the right of residence under the Association Agreement should not be available to failed asylum seekers. Although Advocate-General Geelhoed seemed receptive to this argument,142 the Court ruled out the possibility of deeming reliance by failed asylum seekers on the rights under the Association Agreement as ipso facto ‘abusive’. ‘Abuse’ would require an individual assessment143 to establish that the individual has the ‘sole aim of wrongfully benefiting from advantages provided under Community law’.144 The remaining scope for ‘abuse’ claims has been fleshed out by national courts, reacting to Tum and Dari, for instance in Aldogan.145 Collins LJ accepted that although asylum seekers were not all engaged in abuse when they sought to rely on the Association Agreement, he stated that: [I]f there [was] fraud, if there [was] dishonesty in the application, and furthermore if the business in question had been built up in breach of a condition of entry or, if not of entry, of temporary admission, then again that is an abuse of the immigration laws of this country and, as I have said, on the basis of consistent Court jurisprudence, should not be allowed to benefit a particular claimant.146

The delicate line set out in Günaydin,147 which allows room for change in the individuals’ circumstances to explain breaches in immigration conditions, seems to have been obscured.

  Suat Kol (n 14 above). See further, Kjellgren (n 37 above) 245, 250–51.   Günaydin (n 14 above). See further, Case C-294/06 The Queen on the application of Payir, Akyuz, Ozturk v Secretary of State for the Home Department [2008] ECR I-203. 141   Tum and Dari (n 138 above). 142  AG Geelhoed in Tum and Dari (n 138 above), paras [70–72]. 143   Tum and Dari (n 138 above), para [64]. 144   Tum and Dari (n 138 above), para [66]. 145   R (Aldogan) v Secretary of State for the Home Department [2007] EWHC 2586 (Admin). 146   Aldogan (n 145 above), para [12]. 147   Günaydin (n 14 above). See further, Payir (n 140 above). 139 140

Citizen of the Union: Above Abuse?  343 C.  Marriages of Convenience The concept of marriage of convenience looms as an instantiation of abuse in both EU case law and legislation. Conceptually, it differs from the notion of abuse in the Internal Market case law, as it has emerged in the purely domestic context as a reliance on domestic marriage law to avoid immigration restrictions. In effect, the notion permits one domestic regime, namely immigration law, to ignore a status granted by another, namely family law. The legal status of a ‘marriage of convenience’ in family law is unaffected by its being viewed as a sham by the immigration authorities. EU law is inevitably imbricated in these issues as it too confers rights to reside directly on the basis of family connection, in particular, for TCN spouses of EU citizens.148 Long before the issue was litigated before the Court, the Council resolved on the matter,149 not only permitting Member States to take action on marriages, but requiring them to do so in light of the need for ‘equivalent’ approaches to the problem. The definition refers to: [A] marriage concluded between a national of a Member State or a third-country national legally resident in a Member State and a third-country national with the sole aim of circumventing the rules on entry and residence of third-country nationals and obtaining for the third-country national a residence permit or authority to reside in a Member State. (Emphasis added.)

The Resolution then lists various factors that may provide grounds for suspecting that a marriage is one of convenience. If there are such suspicions, national authorities are to issue residence permits only once they have taken steps to ensure that the marriage is genuine. The Resolution also stipulates that its objective ‘is not to introduce systematic checks on all marriages with third-country nationals. Checks will only be carried out where there are well-founded suspicions’.150 Advocate-General Geelhoed cited the Resolution in Akrich. The reference to ‘sole purpose’ is a crucial constraint on the concept of marriage of convenience. Earlier non-binding measures, in particular the 1997 Resolution on Family Reunification referred to the ‘sole or principal purpose’. However, the current binding measures have taken on the ‘sole aim’ formulation. Recital 28 of the 2004 Citizenship Directive recalls the text of the Resolution when it refers to marriages of convenience entered into ‘for the sole purpose’ of enjoying a right to free movement or residence.151 Article 35 mentions marriage of convenience as an example of ‘abuse or fraud’ without specifying which. It is unlikely to be regarded as fraudulent per se, as although a couple entering into a marriage of convenience is dishonest in that their purpose is not to share married life together, the legal institution of marriage is often availed of in order to accrue legal advantages—for example, in tax law. Marriage law does not generally inquire into the underlying purpose of the marriage. The concept of ‘marriage of convenience’ singles out a category of legal advantage and deems it to be legally problematic to marry with the sole purpose of securing that category of legal advantage. While the marriage may not be fraudulent, the couple is likely to make fraudulent representations in their dealings with 148   B de Hart, ‘The Marriage of Convenience in European Immigration Law’ (2006) 8 European Journal of Migration and Law 251, 260. 149   Council Resolution 97/C382/01 of 4 December 1997 on measures to be adopted on the combating of marriages of convenience. 150   Council Resolution 97/C382/01 (n 149 above), para [6]. 151  See also Article 16(2)(b) of the Family Reunification Directive, employing a similar formulation.

344  Cathryn Costello immigration authorities. Again though, any such finding must be fact-specific, rather than inferred from the nature of the marriage. The UK Court of Appeal recently interpreted Article 35 in a case where the applicants disputed the findings of a marriage of convenience.152 The applicants argued that for Article 35 to apply, the conditions of Articles 27 and 28 of the Directive would also have to be complied with, being the conditions that normally attach to the restrictions on residence rights in public policy cases. The Court rejected this interpretation, treating the abuse exception as being of a different order. The express wording of Article 35 refers to two procedural safeguards: Article 30 on notification of decisions and Article 31 on judicial review. Lady Justice Arden explained the difference in light of the nature of abuse of rights. While a citizen whose right to reside was withdrawn on the basis of public security has strong substantive and procedural protections, by contrast, a person who has been a party to a marriage of convenience has had his rights terminated after the marriage of convenience has been proven.153 Many excessive national restrictions will require revision to comply with the EU approach. Indeed, current national practices are also in tension with fundamental rights protection. For example, in Baiai,154 the House of Lords recently condemned the UK system for preventing sham marriages155 as a violation of Article 12 ECHR. The UK system required all couples planning to marry in non-Anglican ceremonies,156 where one or both parties were subject to immigration control, to obtain specific permission, on payment of a substantial application fee. In effect, marriages would not be permitted if the TCN’s migration status in the UK was precarious, including, for example, if the TCN was an asylum seeker.157 Ultimately, the House of Lords held that none of the conditions ‘has any relevance to the genuineness of the proposed marriage’158 and so found the system to amount to a disproportionate interference with the right to marry. IV. EXPLORING THE CASE LAW ON FREE MOVEMENT OF CITIZENS

In its Internal Market case law, the Court has used the doctrine of abuse of rights, admittedly somewhat inconsistently, to determine the scope of the economic freedoms themselves. The relationship between citizenship of the Union and the pre-existing economic bases for movement and residence rights in the EU is dynamic and evolving. Undoubtedly, citizenship has permeated the case law on free movement of persons thereby securing residence rights159 and expanding the range of benefits that migrants may claim on a non-discriminatory basis with nationals of the host State.160 The Court has made   TC (Kenya) v Secretary of State for the Home Department [2008] EWCA Civ 543, CA.  Citing Akrich (n 124 above), para [61]. 154   R (Baiai) v Secretary of State for the Home Department [2009] 1 AC 287 (HL). 155   Under section 19 of the Asylum and Immigration (Treatment of Claimants etc) Act 2004. 156   Baroness Hale in Baiai (n 154 above), para [37] on the discriminatory nature of the scheme. 157   Baroness Hale in Baiai (n 154 above), para [42]. 158  Lord Bingham in Baiai (n 154 above), para [31]. 159   Case C-85/96 María Martínez Sala v Freistaat Bayern [1998] ECR I-2691; Baumbast (n 115 above). 160  S Besson and A Utzinger, ‘Future Challenges of European Citizenship: Facing a Wide-Open Pandora’s Box’ (2007) 13 European Law Journal 573; A Epiney, ‘The Scope of Article 12 EC: Some Remarks on the Influence of European Citizenship’ (2007) 13 European Law Journal 611; D Kostakopoulou, ‘European Union Citizenship: Writing the Future’ (2007) 13 European Law Journal 623; F Jacobs, ‘Citizenship of the European Union—A Legal Analysis’ (2007) 13 European Law Journal 591. 152

153

Citizen of the Union: Above Abuse?  345 the tantalising statement that ‘Union citizenship is destined to be a fundamental status of nationals of the Member States, enabling those who find themselves in the same situation to enjoy the same treatment in law irrespective of their nationality, subject to such exceptions as are expressly provided for.’161 However, the EC legislature in the 2004 Directive has undermined any notion that citizenship has transcended its market origins.162 The Directive replaces the previous patchwork of legislation dealing separately with different modes of transnational economic activity,163 and aims to ‘strengthen the right of free movement and residence of all Union citizens’.164 However, it is conservative, perhaps even regressive, in three respects. Concerning the material scope of rights, the Directive confirms a transnational conception, Article 3(1), stating that it applies to ‘all Union citizens who move to or reside in a Member State other than that of which they are a national, and to their family members . . . who accompany or join them’.165 Secondly, it reasserts the resource and insurance conditions on the right to reside of the non-economically active.166 Thirdly, Article 24 permits Member States to differentiate amongst EU citizens according to their economic status and nature of the benefit claimed, reasserting the relevance of economic activity as a determinant of the non-discrimination guarantee.167 However, this economic differentiation is time-bound, and duration of residence in the host State brings with it a secure residence status irrespective of economic activity and a greater range of benefits available on a non-discriminatory basis. A.  Claims against the Host State: Limited Solidarity with Students and Jobseekers The Court commonly stipulates that economic activity must be ‘genuine’ in order to benefit from the Internal Market provisions. The pre-citizenship cases of Levin168 and Lair169 set the tone for later developments. In Levin, the Court stated that motive was irrelevant to the issue of whether individuals could rely on the provisions on free movement of workers, provided the work undertaken was ‘effective and genuine’ rather than ‘marginal and ancillary’. The application of this test remains particularly important when certain benefits are confined to workers rather than other migrants, a feature that has survived the introduction of citizenship of the Union. 161   Case C-184/99 Rudy Grzelczyk v Centre public d’aide sociale d’Ottignies-Louvain-la-Neuve [2001] ECR I-6193, para [31]. 162   Citizenship Directive (n 12 above). 163  Regulation 1612/68 and Directives 64/221/EEC, 68/360/EEC, 72/194/EEC, 73/148/EEC, 75/34/EEC, 75/35/ EEC, 90/364/EEC, 90/365/EEC and 93/96/EEC. 164  Recital 3, Citizenship Directive (n 12 above). 165  Article 3(1), Citizenship Directive (n 12 above). 166  Article 7, Citizenship Directive (n 12 above) on the right of residence for more than three months. 167  Article 24(2) states: By way of derogation from paragraph 1, the host Member State shall not be obliged to confer entitlement to social assistance during the first three months of residence or, where appropriate, the longer period provided for in Article 14(4)(b), nor shall it be obliged, prior to acquisition of the right of permanent residence, to grant maintenance aid for studies, including vocational training, consisting in student grants or student loans to persons other than workers, self-employed persons, persons who retain such status and members of their families. 168   Levin (n 130 above). 169   Lair (n 38 above).

346  Cathryn Costello Lair concerned a benefit (a student maintenance grant) which migrant workers, but not those who were students simpliciter, could claim on a non-discriminatory basis as a ‘social advantage’. The Court accepted that individuals may continue to rely on their status of worker to claim student grants even if they are no longer in an employment relationship, provided there is ‘some continuity between the previous occupational activity and the course of study’.170 In addition, if the worker became involuntarily unemployed, no such link would be required. However, in order to ensure that individuals did not claim to be workers in an abusive manner, it accepted that if a worker had ‘entered a Member State for the sole purpose of enjoying, after a very short period of occupational activity, the benefit of the student assistance system in that State’ then such abusive conduct was ‘not covered by the Community provisions in question’.171 Thus, the notion of abuse allowed national authorities to deem the situation ‘not covered’ by the free movement of workers provisions. However, the Court rejected the imposition of any fixed period of prior occupational activity to identify abuses. The Court effectively legislated a tailored context-sensitive abuse rule, taking into account various policy considerations, including the precariousness in the labour market, and the desirability of workers’ enhancing or indeed replacing skills. That rule has now been codified in Article 7(3)(d) of the 2004 Directive.172 The distinction between workers and students continues to be of legal significance. Ninni-Orasche,173 decided in 1993, concerned an Italian citizen who had worked part-time in Austria for a time, and then sought to rely on her status as a worker when she commenced studies in Austria. Although leaving the final factual assessment for the national court, the Court suggested that the Lair abuse caveat was not applicable on the facts, emphasising that the applicant had entered the host country with her husband and worked for a time. Following Advocate-General Geelhoed, the Court insisted the abuse issue only arose once it was established that the applicant was indeed a ‘worker’.174 The later citizenship cases have displayed a different approach to claims brought by migrants against their host States. The Court took a deductive leap from the legislative statement that migrants should not become an ‘unreasonable burden’ on the host State, which led to the conclusion that they had a license to become a ‘reasonable burden’, and which in turn was seen as underpinning the notion of transnational solidarity.175 Accordingly, cases decided from the early 2000s onwards suggested that the line between an economically active worker and a student was less important than the degree of

170   Lair (n 38 above), para [37]. The ECJ also admitted that ‘[s]uch continuity may not, however, be required where a migrant has involuntarily become unemployed and is obliged by conditions on the job market to undertake occupational retraining in another field of activity’. Lair (n 38 above), para [37]. 171   Lair (n 38 above), para [43]. 172  It states that ‘the status of worker or self-employed person is maintained if the worker embarks on vocational training. Unless he/she is involuntarily unemployed, the retention of the status of worker shall require the training to be related to the previous employment.’ 173   Ninni-Orasche (n 47 above), para [26]. For a recent application, see Vatsouras (n 47 above). 174   Ninni-Orasche (n 47 above), para [31]. 175   For example, Grzelczyk (n 161 above), para [44] states that the applicable ‘Directive’s preamble envisages that beneficiaries of the right of residence must not become an ‘unreasonable’ burden on the public finances of the host Member State. [The Directive] thus accepts a certain degree of financial solidarity between nationals of a host Member State and nationals of other Member States, particularly if the difficulties which a beneficiary of the right of residence encounters are temporary.’

Citizen of the Union: Above Abuse?  347 integration in the host State.176 In Grzelczyk,177 the Court used this approach to bring the payment of the Belgian minimum subsistence allowance to a French student within the scope of EC law, overruling Brown.178 The decisive factor leading to the conclusion that Rudy Grzelczyk was not an ‘unreasonable burden’ was that he had undertaken threeyears of self-financed study before seeking to avail himself of the minimum subsistence allowance. D’Hoop179 concerned a Belgian student who sought a Belgian ‘tideover allowance’. She was ineligible as the grant was only available to those who had undertaken their secondary education in Belgium, whereas she had done her schooling in France. The Court held that the grant fell within the scope of the Treaty, and that D’Hoop was being disadvantaged for having exercised her right to free movement, relying on the deterrence rationale from Surinder Singh. While it was permissible for States to ensure that there was ‘a real link between the applicant for that allowance and the geographic employment market concerned’180 the condition imposed was ‘too general and exclusive in nature’.181 Bidar 182 developed this approach further. A migrant French student was entitled to certain UK financial support on a non-discriminatory basis, as he had only been resident in the host country for three years and undertaken his secondary schooling there. Nonetheless, the Court emphasised that he was integrated into that host country and so entitled to equal treatment with home nationals. i.  Förster: Residue or Resurgence of Market Citizenship? In contrast, the Citizenship Directive permits Member States to impose a five-year waiting period before migrant students may access student maintenance grants.183 Recently in Förster,184 the Court seems to have remoulded its case law to fit the Directive. The contrast between the Opinion and Judgment is striking, and suggests that the doctrine of abuse has been relegated as the notion of ‘integration’ is retooled to permit the waiting period. Advocate-General Mazak found the five-year requirement to be too blunt in the circumstances, in particular as there was ‘no evidence of abuse in the present case’ as the applicant had been in ‘substantial paid employment relationships for more than three years before ceasing to work’.185 Accordingly, Förster’s sole purpose in working was not to avail of student assistance, so the Lair instantiation of abuse was inapplicable.186

176  See further, D Kostakopolou, ‘Ideas, Norms and European Citizenship: Explaining Institutional Change’ (2005) 68 Modern Law Review 233. 177   Grzelczyk (n 161 above). 178   Case C-197/86 Brown v Secretary of State for Scotland [1988] ECR 3205. 179   Case C-224/98 Marie-Nathalie D’Hoop v Office national de l’emploi [2002] ECR I-6191. 180   D’Hoop (n 179 above), para [38]. 181   D’Hoop (n 179 above), para [39]. 182   Case C-209/03 R (on the application of Dany Bidar) v London Borough of Ealing, Secretary of State for Education and Skills [2005] ECR I-2119. See further, C Barnard, ‘Note on Bidar’ (2005) 42 Common Market Law Review 1465. 183  Article 24(2), Citizenship Directive (n 12 above). 184   Case C-158/07 Jacqueline Förster v Hoofddirectie van de Informatie Beheer Groep [2008] ECR I-8507. 185  AG Mazak in Förster (n 184 above), para [86]. 186  AG Mazak in Förster (n 184 above), para [86], citing Case 197/86 Steven Malcolm Brown v The Secretary of State for Scotland [1988] ECR 3205, paras [27–28].

348  Cathryn Costello While the Opinion is informed by a reading of the previous case law as undergoing a ‘process of emancipation of Community rights from their economic paradigm’187 the Court seems to take a different view. It did not mention abuse at all, confirming both the applicability of the non-discrimination guarantee under Article 12 EC [now Article 18 TFEU] and the permissibility of the five-year residence period, as set out in Article 24(2) of the Citizenship Directive, being apt to ensure that students were ‘integrated into the society of the host Member State’.188 Whether Förster can truly be reconciled with Bidar is doubtful. Despite the Citizenship Directive’s avowed aim of ‘remedying this sector-bysector, piecemeal approach to the right of free movement and residence’189 the Court has permitted the categorical differentiation in Article 24(2). Prior to the introduction of citizenship of the Union, although the Court had extended the right to reside to jobseekers,190 it treated unemployment benefits as falling outside the scope of the non-discrimination guarantee.191 Post-citizenship, Collins192 brought the jobseekers’ allowance within the scope of EC law, but its availability could be limited to those with a ‘genuine link’ with the employment market in the host State, and tested by the proportionate application of a residence requirement.193 The passing references to ‘abuse’ in both the Judgment and Opinion in Collins assume that abusive conduct is that which the genuine link criterion roots out.194 However, if the Förster approach is applied to jobseekers, and the text of Article 24(2) of the Citizenship Directive given full sway, a different result would emerge with the host Member State permitted to deny all entitlement to social assistance for the first three months of residence. The case law predating the entry into force of the 2004 Directive (Grzelczyk, D’Hoop, Bidar and Collins in particular) evidences a waning of the notion of abuse as the lines between jobseekers, students and workers faded, to be replaced with the context-sensitive assessment of the migrant’s degree of integration. At first glance, the integration approach has much to recommend it over the notion of ‘abuse’, which offers little guidance in the case of benefits claims by migrant EU citizens in their host Member States. By linking integration and solidarity, the Court captures the nature of the relationships underlying the welfare state, without an appeal to nationalism. It prompts an examination of whether the migrant EU citizen has developed a real link with the host Member State, such as to warrant inclusion within the scope of the solidaristic community. Instrumentally, the solidarity enquiry is apt to ensure equal access for migrant EU citizens who have an established link with the host country, whilst preventing ‘welfare tourism’. Institutionally, its benefits include that it requires case-by-case individual assessments and as a distinctively communautaire principle, national administrations and courts must seek guidance on its 187  AG Mazak in Förster (n 184 above), para [54], citing AG Cosmas in Case C-378/97 Wijsenbeek [1999] ECR I-6207, paras [84–86], and AG Poiares Maduro in Case C-72/03 Carbonati Apuani [2004] ECR I-8027, paras [68–69]. 188  AG Mazak in Förster (n 184 above), para [52]. 189  Recital 4, Citizenship Directive (n 12 above). 190   Case C-292/89 R v Immigration Appeal Tribunal ex p Antonissen [1991] ECR I-745. 191   Case 316/85 Centre public d’aide sociale de Courcelles v Marie-Christine Lebon [1987] ECR 2811, paras [25–26]. Cf Case C-90/97 Swaddling v Adjudication Officer [1999] ECR I-1075. 192   Case C-138/02 Brian Francis Collins v Secretary of State for Work and Pensions [2004] ECR I-2703. See further, G Davies, ‘“Any Place I Hang My Hat?” or: Residence is the New Nationality’ (2005) 11 European Law Journal 43, in particular 54–55. 193   Collins (n 192 above), paras [69–73]. 194  AG Ruiz-Jarabo Colomer in Collins (n 192 above), para [75]; Collins (n 192 above), para [50] referring to the Commission’s argument regarding ‘abuse by work-seekers who are not genuine’.

Citizen of the Union: Above Abuse?  349 meaning and scope. Some commentators have raised concerns that this case law does not allow Member States to maintain the integrity of their welfare systems.195 Nevertheless, the integration approach has much to recommend it over the crude notion of abuse of rights. However, the 2004 Directive, and its application in Förster, seems to cut across this debate and suggest a new phase of judicial deference to Member State restrictions legitimated by the wording of the Directive. The Directive underscores the lines between student, jobseeker and worker. The Förster ruling reasserts the privileged status of workers, such that the relevance of Lair remains, to prevent instrumental working in order to claim benefits not available to students per se. B.  Abusive U-Turns? i.  The Right to Return The above cases, with the exception of D’Hoop,196 concern the typical scenario of an EU migrant claiming rights in the host State on a non-discriminatory basis with citizens. The attempt to claim benefits available in the host State does not disturb the allocation of regulatory competence in the Internal Market. Accordingly, Member State governments’ claim that migrants’ ‘benefit tourism’ should be characterised as ‘abusive’ tend to be rejected.197 As Advocate-General Jacobs and later the Court emphasised in Commission v Austria,198 foreign students going to study in Austria were acting precisely in accordance with the aim of the Treaty provisions, so their actions ‘cannot therefore [of themselves] constitute an abuse of that right’.199 However, the EU free movement rules also apply in other contexts, notably when individuals return to their host States. The invocation of EU rights against the home State may, in contrast, unsettle the allocation of regulatory competence underpinning the Internal Market. Recall that in the absence of harmonised EU rules, Member States are required to recognise product standards, educational qualifications and so on that comply with the home State’s laws, unless the host State demonstrates an overriding objective that requires the application of its domestic standards.200 A consequence of the preservation of national regulatory capacity of home States is that they are permitted to 195  Somek (n 42 above). Cf C O’Brien, ‘Real Links, Abstract Rights and False Alarms: The Relationship between the ECJ’s “Real Link” Case Law and National Solidarity’ (2008) 33 European Law Review 643. She argues that the entitlements based on demonstration of a genuine link are flimsy, and that the case law permits diverse national restrictions, acting as a mere rationality filter. 196   D’Hoop was a Belgian claiming rights in Belgium, but the ECJ emphasised that she was being disadvantaged for having exercised her free movement rights by undertaking her schooling in France. 197  AG Geelhoed in Bidar (n 182 above), para [66]: Finally, it was submitted by various intervening Governments that Member States have a legitimate interest in preventing abuse of their student support schemes and in preventing ‘benefit tourism’. I do consider that this is indeed a legitimate concern of the Member States, but the manner in which this should be ensured should not be such as to undermine the fundamental rights of EU citizens residing lawfully within their territory. 198   Case C-147/03 Commission v Republic of Austria [2005] ECR I-5969. C Rieder, ‘Casenote: Case C-147/03 Commission v Austria’ (2006) 43 Common Market Law Review 1711; Somek (n 42 above). 199   Commission v Austria (n 198 above), para [70]. 200   Mutual recognition is sometimes treated as synonymous with home State control (eg, in the Commission’s post-Cassis communication). Cf K Armstrong, ‘Mutual Recognition’ in C Barnard and J Scott (eds), The Law of the Single European Market: Unpacking the Premises (Oxford, Hart Publishing 2002) 225.

350  Cathryn Costello regulate domestic production or the activities of their own citizens more harshly than imports or migrant EU citizens in some instances. As Damian Chalmers et al point out: The presence of a dual regime, with one law for imports and another for domestic products, provides incentives for operators to cheat by passing off domestic products as imports. To counter this, an ‘abuse of rights’ doctrine exists, preventing Article 28 EC [now Article 34 TFEU] from being applied to goods produced in a Member State and then re-imported into it purely for the purposes of setting up a legal challenge to a national law [as is evident in Leclerc v Au blé vert].201

However, when it comes to the movement of persons, the Court has taken a more modulated approach. As previously mentioned, Surinder Singh202 established that EU rights persist on return to the home Member State. In contrast to the Leclerc abuse scenario concerning exporting goods and immediately reimporting them so as to avoid domestic regulation, individuals may leave their home State, acquire EU rights, and return home without being disadvantaged. Any potential analogy between transport of widgets and human life-courses fails here, and the Court regards the continuation of rights enjoyed on the basis of EU law as necessary, both to encourage free movement203 and protect the migrant’s right to return to her home State.204 However, in Surinder Singh too the Court permitted Member States to prevent abuse. In response to the UK government’s concerns about ‘fraud’, the Court stated, citing Knoors205 and Bouchoucha,206 that ‘the facilities created by the Treaty cannot have the effect of allowing the persons who benefit from them to evade the application of national legislation and of prohibiting Member States from taking the measures necessary to prevent such abuse’. Akrich, as previously discussed,207 demonstrates that individuals may deliberately rely on the free movement provisions to benefit from EU rules, a benefit which will persist on return to the home State. The next two sections illustrate that proposition and its limits. ii.  Mutual Recognition of Educational Qualifications In Knoors,208 the Court clarified that it was permissible for an individual to rely on EC law to demand recognition by his Member State of origin of qualifications obtained in another Member State. However, it also acknowledged that Member States had a legitimate interest in its nationals, ‘by means of facilities created under the Treaty, from attempting wrongly to evade the application of their national legislation as regards training for a trade’.209 Nonetheless in the particular case, the conditions set out in the applicable Directive210 regarding the duration of the vocational activity excluded the risk 201   D Chalmers et al, European Union Law: Text and Materials (Cambridge, Cambridge University Press 2006) 674, citing Case 229/83 Leclerc et al v SARL ‘Au Blé Vert’ et al [1985] ECR 1, para [27]. 202   Surinder Singh (n 111 above). 203  The original rationale emphasised in Surinder Singh. 204  This was emphasised in Case C-291/05 Minister voor Vreemdelingenzaken en Integratie v Eind [2007] ECR I-10719. 205   Knoors (n 120 above), para [25]. 206   Bouchoucha (n 26 above), para [14]. 207   Akrich (n 124 above). 208   Knoors (n 120 above). 209   Knoors (n 120 above), para [25]. 210  Article 3 of Council Directive of 7 July 1964, laying down detailed provisions concerning transitional measures in respect of activities of self-employed persons in manufacturing and processing industries falling within ISIC Major Groups 23–40 (Industry and Small Craft Industries) [1964] OJ 1863/64.

Citizen of the Union: Above Abuse?  351 of abuse.211 Once the general legislative framework for mutual recognition was in place, the practice of home citizens undertaking courses of study abroad (or even at home), thereby gaining diplomas awarded in other Member States, was protected. For instance, in various cases concerning Spanish212 and Greek recognition of foreign diplomas,213 the Court confirmed that the effect of the Directive on Mutual Recognition of Diplomas214 was precisely to permit individuals to acquire professional qualifications wherever they wished in the EU (which included the home State), in studies leading to the award of a foreign diploma. In contrast, the Court identified an ‘abusive’ scenario in Bouchoucha,215 where the activity in question (osteopathy) was not subject to the EC mutual recognition legislation. In light of the absence of harmonisation, the Court stated that it was permissible for the home Member State to restrict the activity exclusively to persons holding the qualification of doctor of medicine, and accordingly refuse to permit a French national who had acquired an osteopathy qualification in the UK to practice his trade. iii.  Mutual Recognition of Driving Licences In several cases on the mutual recognition of drivers’ licences under Directive 91/439/ EEC,216 Member States have attempted to argue that their duties to recognise foreign licences were tempered by the requirement to combat fraud or abuse. For the most part, such arguments fail and the duty of mutual recognition is strict. Once a licence is issued by the country of residence, it should be recognised by others.217 For example, in Commission v Netherlands218 the Court rejected the Netherlands’ argument that requiring migrant EU drivers to register after a period of residence was necessary in order to prevent fraud, holding that the measure was disproportionate to the aim of enhancing road safety and liable to discourage free movement. Two recent cases show that the mutual recognition system does have limits, and that the Directive cannot be relied on for fraudulent purposes. Both cases concern claims brought by German drivers against their home State. The German drivers, when facing German restrictions for driving offences, obtained Czech licences and then attempted to compel Germany to recognise those licences under the Directive. The circumvention of home State rules could not be more vivid. In Wiedemann, the Court confirmed that Germany was not obliged to recognise the Czech licence if it was issued during the period for   Knoors (n 120 above), para [26].   Case C-286/06 Commission v Spain [2008] ECR I-8025. 213   Case C-274/05 Commission v Greece [2008] ECR I-7969; Case C-151/07 Theologos-Grigoris Khatzithanasis v Ipourgos Igeias kai Kinonikis Allilengiis [2008] ECR I-9013; Case C-84/07 Commission v Greece [2008] ECR I-171. 214   Council Directive 89/48/EEC of 21 December 1988 on a general system for the recognition of higher education diplomas awarded on completion of professional education and training of at least three years’ duration [1989] OJ L19/16, as amended by Directive 2001/19/EC of the European Parliament and of the Council of 14 May 2001 [2001] OJ L206/1. 215   Bouchoucha (n 26 above). 216   Directive 91/439 on driving licences [1991] OJ L237/1, as amended by Regulation 1882/2003, [2003] *EU Focus 68 OJ L284/1. 217  Case C-227/05 Halbritter v Freistaat Bayern [2006] ECR I-49; C-340/05 Kremer [2006] ECR I-98; Commission v Netherlands (n 13 above); Case C-408/02 José Antonio da Silva Carvalho [2004] OJ C94/15–16; Case C-476/01 Kapper [2004] ECR I-5205. 218   Commission v Netherlands (n 13 above). 211 212

352  Cathryn Costello which Germany had withdrawn the German licence.219 Similarly, in Weber,220 the Court accepted that there was no duty to recognise a licence obtained during the period in which a licence issued in the first Member State was suspended, where both the suspension and the withdrawal were based on grounds existing at the date of issue of the second driving licence. The Court did not mention abuse or fraud in either case, basing its rulings on a purposive interpretation of the Directive. In contrast, in his Opinions in both cases, Advocate-General Bot explicitly categorised the behaviour as fraudulent. In Wiedemann he stated that insisting on recognition of the Czech licence would ‘have the effect of facilitating or upholding fraud’.221 Accordingly: Not only the law but also plain common sense requires, in such a situation, that fraud be treated as destroying mutual trust and that Member States be afforded the right to investigate the circumstances in which the issue of a driving licence has been obtained.222

In Weber, Advocate-General Bot stated that Mr Weber ‘although having committed a road traffic offence, tried to avoid the resulting sanctions by going to another Member State to obtain a new driving licence there’.223 As a result, Germany should not be obliged to recognise the Czech licence, as it would ‘have the consequences of authorising a person in a potentially dangerous state to drive, of allowing that person to avoid the criminal sanction which they incurred, and of encouraging fraud’.224 The Court’s approach is preferable, in that there is no need to rely on the concept of abuse in these cases. V. CONCLUSIONS

Citizenship of the Union is conferred on the nationals of the Member States. At present, the Member States are individually competent to define who their own nationals are. Although EU law may gradually come to play a role over the attribution of nationality, or at least in guiding the allocation of the rights attaching to EU citizenship, in principle, nationality law remains national. Accordingly, Member States are under a strict duty to recognise each other’s designations of nationality, precluding claims that an EU citizen acquired nationality in an abusive manner. The Court’s stance is appropriate, as requiring nationality to track a ‘genuine link’ would allow widespread questioning of nationality attribution across the EU. Rottmann opens up a new vista, but thankfully, omits reference to abuse, confirming only the licence to remove nationality in cases of deception or fraud. TCN family members who derive residence rights from their family relationships with EU citizens are in a privileged category, and the exclusively EU nature of their rights to reside appears to insulate them from objections that their reliance on EU law is abusive. They may ‘licitly’ circumvent restrictions under domestic immigration law, in contrast to TCNs who derive rights from other EU guarantees where a limited doctrine of abuse prevails, thus permitting Member States to have regard to their conditions of entry in determining whether they should be able to rely on the economic rights under the various 219   Joined Cases C-329/06 and C-343/06 Arthur Wiedemann v Land Baden-Württemberg (C-329/06), and Peter Funk v Stadt Chemnitz (C-343/06) [2008] ECR I-4635, para [65]. 220   Case C-1/07 Frank Weber [2008] ECR I-8571. 221  AG Bot in Wiedemann (n 219 above), para [81]. 222  AG Bot in Wiedemann (n 219 above), para [82]. 223  AG Bot in Weber (n 220 above), para [48]. 224  AG Bot in Weber (n 220 above), para [49].

Citizen of the Union: Above Abuse?  353 association agreements considered. The concept of ‘marriage of convenience’, defined in EU law according to its sole illicit purpose of circumventing immigration law, remains as a different form of abuse, which in effect nullifies the immigration benefit if—and only if—acquiring it was the sole purpose of the marriage. Once individuals bring themselves within the personal scope of the provisions on Citizenship of the Union, we see that their rights to movement and residence attract particularly strong judicial protection, with the Court attuned to the need positively to encourage free movement of persons. Accordingly, as one would expect, the Citizenship case law evidences a more limited role for abuse than the other free movement provisions. The migration of persons, in particular, exit to escape disfavourable domestic conditions, is less likely to create intolerable regulatory gaps or unfair regulatory competition than corporate flight. When individuals seek to enjoy benefits in their countries of residence on a non-discriminatory basis with home nationals, claims of abuse are inapt. ‘Abuse’ plays a limited role as the law continues to differentiate between migrant EU citizens according to economic status (for example, by preventing students from passing themselves off as genuine workers). When EU citizens rely on the free movement rules against their home States, abuse claims may arise to prevent the circumvention of the home State rules. However, these claims will not succeed if there is a harmonised framework for mutual recognition, in which case only fraud will undermine the duty to recognise a status acquired elsewhere.

24 Some Comments on the Idea of a General Principle of Union Law Prohibiting Abuses of Law in the Field of Free Movement for Union Citizens Michael Dougan*

I

have two sets of reservations about the proposition that there is a general principle of Union law prohibiting abuses of law in the context of the free movement of Union citizens: the first concerns the very idea of a general principle of Union law prohibiting abuses of law; the second concerns the potential application of any such principle within the specific context of free movement for Union citizens. I. A general principle of UNION law prohibiting abuses of law?

The term ‘general principle of Union law’ has a specific technical meaning within the Union legal order. It does not refer simply to some general juridicial concept which can help fill certain ‘gaps’ in EU law arising from the supposedly ‘incomplete’ character of the legal system created pursuant to the Treaties. It refers instead to the system of unwritten rules, developed in its case law by the Court of Justice and of (at least) equal status to the Treaties themselves, which furnish the Union legal order with its basic system of administrative law: for example, as regards the principles of proportionality,1 legal certainty2 and legitimate expectations.3 In that specific technical sense, the case law has established that a general principle of Union law is legally binding in two categories of situation: first, where the Union institutions exercise the (legislative, executive or judicial) powers conferred upon them through the Treaties;4 secondly, where a Member State acts   * This short chapter is based on the author’s written notes as discussant for Cathryn Costello’s paper on ‘Abuse of Law and Citizenship of the Union’ as presented at the Oxford symposium on Prohibition of Abuse of Law: A New General Principle of EC Law? on 3 October 2008. See the contribution by C Costello, ‘Citizenship of the Union: Above Abuse?’, ch 23 above. 1  For example, Case 240/78 Atalanta Amsterdam BV v Produktschap Voor Vee en Vlees [1979] ECR 2137. 2  For example, Case C-143/93 Gebroeders van Es Douane Agenten BV v Inspecteur der Invoerrechten en Accijnzen [1996] ECR I-431. 3  For example, Case 120/86 Mulder v Minister van Laudbouw en Visserij [1988] ECR 2321. 4  For example, Case C-453/03 ABNA Ltd et al v Secretary of State for Health and Food Standards Agency [2005] ECR I-10423; Case 181/84 R v Intervention Board ex p Man (Sugar) Ltd [1985] ECR 2889; Case C-185/95P Baustahlgewebe v Commission [1998] ECR I-8417.

356  Michael Dougan ‘within the scope of the Treaties’, either when implementing EU law through the exercise of its own (legislative, executive or judicial) competences,5 or when derogating from the Treaties by seeking to restrict the exercise of rights created under EU law.6 In this second category of situation, the general principles of Union law serve essentially as grounds of judicial review against the Member States and, in particular, as a source of public law obligations for the relevant national authorities. Against that background, it is difficult to see where a ‘general principle of Union law prohibiting abuses of law’ might be located within the basic conceptual structures of EU law in general, or of free movement law in particular. After all, the concept of abuse of law is primarily concerned with and directed against the conduct of private individuals, ie where the latter seek to rely on particular provisions of EU law for reasons other than for their intended purpose.7 That is clearly not a matter of administrative law and therefore seems an inappropriate candidate to qualify as a general principle of Union law stricto sensu. True, the interaction between the general principles of Union law, the exercise of national competences within the scope of the Treaties and the legal response to purely private conduct can sometimes be difficult to disentangle. Consider, by way of illustration, the dispute in Preston, which concerned the calculation of when a national limitation period should begin to run for the purposes of bringing an equal pay claim based on ex Article 141 EC [now Article 157 TFEU].8 The principle that national law cannot render the exercise of Treaty rights virtually impossible or excessively difficult constitutes a general principle of Union law, and is binding upon the Member State when it acts within the scope of the Treaties, including when it furnishes the procedural framework for exercising the right to equal pay under Article 157 TFEU. Disapplication of the disputed limitation period in Preston was a necessary response, pursuant to the principle of effectiveness, to the conduct of a private employer relying (legitimately, if cynically) upon purely national law. That may look like a manifestation of a concept of ‘abuse of law’—albeit in a situation where the individual is seeking not to rely on Union law other than for its intended purpose, but rather to hide behind certain provisions of domestic law so as to evade his/her EU obligations—but it is clearly not a general principle of Union law in its proper sense. The duties flowing from the principle of effectiveness in Preston were clearly public law obligations concerning the nature of judicial protection available for the exercise of Union rights, and they were firmly directed against the Member State when it was acting within the scope of the Treaties. There was no ‘abuse of law’ by the private employer—only a deficient procedural system as furnished by the Member State itself, of which the private employer had then taken advantage for its own gain. It seems fair to admit that the scope of application of the general principles of Union law has been thrown into a certain degree of confusion following the notorious ruling in 5  For example, Case 222/84 Johnston v Chief Constable of the RUC [1986] ECR 1651; Case 5/88 Hubert Wachauf v Bundesamt für Ernährung und Forstwirtschaft [1989] ECR 2609; Joined Cases C-31/91 to C-44/91 SpA Alois Lageder et al v Amministrazione delle Finanze dello Stato [1993] ECR I-1761. 6  For example, Case 261/81 Walter Rau Lebensmittelwerke v De Smedt PVBA [1982] ECR 3961; Case C-254/98 Schutzverband gegen unlauteren Wettbewerb v TK-Heimdienst Sass GmbH [2000] ECR I-151; Joined Cases C-482 and 493/01 Georgios Orfanopoulos et al (C-482/01) and Raffaele Oliveri (C-493/01) v Land BadenWürttemberg [2004] ECR I-5257. 7   Cf Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, paras [52–53]. 8  Case C-78/98 Preston et al v Wolverhampton Healthcare NHS Trust et al [2000] ECR I-3201.

Comments on Abuses of Law and the Free Movement of Citizens  357 Mangold.9 There, it will be recalled, the Court held that the general principle of Union law guaranteeing equal treatment on grounds of age could be relied upon by an employee against his private employer, so as to challenge certain terms and conditions of his fixedterm contract of employment which discriminated against older workers. Mangold has been heavily criticised on various grounds: first, because the evidence to support the existence of a general principle of Union law prohibiting age discrimination was thin or even non-existent; secondly, because the Court seemed to ignore the limits placed on the potential direct effect of ex Article 13 EC [now Article 19 TFEU], the legal basis for enacting general measures guaranteeing equal treatment on grounds including age, as introduced by the Treaty of Amsterdam; thirdly, because the Court’s invocation of the general principles of Union law also permitted it to overcome various restrictions on the potential direct effect of Directive 2000/7810 (itself enacted under ex-Article 13 EC), given that the dispute in Mangold had arisen before expiry of the deadline for transposition of the Directive into national law and concerned relations between two private parties where one would have expected ‘horizontal direct effect’ to be entirely ruled out. For present purposes, Mangold might seem even more controversial, insofar as it suggests that the general principles of Union law are indeed capable of going further than controlling the (public law) actions of the Union institutions and Member States within the scope of the Treaties, so as also to serve as a direct source of legally binding (private law) obligations for individuals. If correct, that interpretation of Mangold could indeed provide the legal basis for elaborating a general principle of Union law prohibiting abuses of law by private parties, in the specific technical sense of the term. Upon reflection, however, that interpretation of Mangold does not seem at all persuasive. Two features of the dispute in Mangold explain why the general principle of Union law prohibiting age discrimination, qua administrative law obligation, was nevertheless capable of having legal effects in relations between two private parties. First, the disputed terms and conditions of the claimant’s fixed-term contract of employment were directly based upon legislation adopted by the relevant Member State pursuant to its own regulatory powers. The unlawful discrimination was therefore not derived from the exercise by two private parties of their contractual autonomy, but had a clear public law basis, the latter providing the necessary administrative wrongdoing against which the general principles of Union law were ultimately directed. Secondly, the discriminatory legislation adopted by the relevant Member State was intended to transpose into its national legal system another Union measure, ie Directive 1999/70 giving effect to the Framework Agreement on Fixed-Term Work.11 The fact that the Member State was implementing Directive 1999/70 meant that the situation in Mangold actually fell within the scope of the Treaties—thus providing the necessary legal context within which the general principles of Union law could be considered binding. By contrast, in the more recent Barstch dispute, the Court refused to enforce the general principle of Union law guaranteeing equal treatment on grounds of age, because the relevant dispute had arisen in a field of employment law (concerning qualification for an occupational survivor’s

 Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9981.  Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation [2000] OJ L303/16. 11  Council Directive 1999/70/EC of 28 June 1999 concerning the framework agreement on fixed-term work concluded by ETUC, UNICE and CEEP [1999] OJ L175/43.     9 10

358  Michael Dougan pension) not already governed by other measures of Union secondary legislation.12 In the absence of a trigger such as Directive 1999/70, the Member State was not at all bound to respect the general principle of Union law guaranteeing equal treatment on grounds of age. By implication, the relevant discrimination could only be considered unlawful after the deadline for transposition of Directive 2000/78 had fully expired—though it remains unclear whether, in such cases, the prohibition on age discrimination will find its legal basis only in Directive 2000/78 itself (and thus subject to the usual conditions for the latter’s direct effect within the national legal system) or also in the general principles of Union law (because the post-deadline Directive 2000/78 somehow provides the necessary trigger to bring the dispute within the scope of application of the Treaties). In other words, Mangold illustrates no more than the simple proposition that, where the general principles of Union law are binding upon a Member State acting within the scope of the Treaties, public law acts which are tainted by illegality in accordance with the standards of administrative conduct laid down by the Court of Justice may be challenged, not only directly in an action for judicial review against the delinquent public authorities themselves, but also indirectly where such public law acts are challenged collaterally in a dispute between two private parties. That proposition is already well established in case law concerning Treaty provisions such as ex Article 28 EC [now Article 34 TFEU], which impose obligations only upon Member States and other public bodies, but can be invoked also in horizontal disputes between individuals.13 Nor would it have made much sense for the Court in Mangold to have attempted to extend the general principles of Union law beyond administrative conduct by the Member State, so as also to cover the purely private actions of individuals: after all, why should private parties be held to the same standards of behaviour, under principles such as legal certainty and legitimate expectations, as those imposed upon public entities entrusted with the exercise of power in the general good? Those considerations suggest that any proposal for a general principle of Union law prohibiting abuse of law seems inherently difficult to square with the nature and purpose of general principles of Union law within the scheme of the Treaties. But even accepting for the sake of argument the possibility that a general principle of Union law prohibiting abuse of law could arise in other contexts, there are good reasons to doubt whether such a notion could comfortably find a conceptual nest within the specific intellectual structure of disputes concerning the free movement of persons. As we shall see in the next section, allegations of abusive behaviour by a migrant Union national can of course arise in certain free movement cases: it is perfectly possible that a Member State might call upon certain public policy considerations related to improper conduct by a private individual, either when the national authorities seek to explain why they are refusing to implement certain provisions of EU law, or when those authorities wish to justify why they are actively derogating from the claimant’s rights under the Treaties. But in the former situation, ‘abuse of law’ must be acting as an interpretative tool, defining the scope and content of, and conditions for exercising, a particular EU right derived from secondary legislation; whereas in the latter situation, ‘abuse of law’ surely performs the role of 12  Case C-427/06 Birgit Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH [2008] ECR I-7245. 13  For example, Case 74/76 Iannelli and Volpi SpA v Ditta Paolo Meroni [1977] ECR 557; Case C-315/92 Verband Sozialer Wettbewerb eV v Clinique Laboratories SNC et Estée Lauder Cosmetics GmbH [1994] ECR I-317; Case C-33/97 Colim NV v Bigg’s Continent Noord NV [1999] ECR I-3175.

Comments on Abuses of Law and the Free Movement of Citizens  359 an express derogation/imperative requirement in the process of objectively justifying the Member State’s infringement of the individual’s primary Treaty rights. Clearly, in neither situation could the idea of an ‘abuse of law’ be described as the manifestation of a general principle of Union law in the specific technical sense of the term. II. Abuse of Law in the Context of Union Citizenship?

It has been suggested that several recent free movement disputes illustrate the potential relevance of a general concept of ‘abuse of law’ within the specific context of Union citizenship. However, further reflection suggests that such cases have in fact little to do with individuals somehow making improper use of otherwise good law, and rather more to do with some of the difficult regulatory problems and policy controversies created by EU law for itself when it comes to the free movement of Union citizens. For example, the ruling in Collins has been cited as an example of individual conduct amounting to some sort of ‘abuse’ of the right to free movement under ex-Article 18 EC [now Article 21 TFEU], on the basis that an economically inactive Union citizen made ‘unwarranted’ demands for financial support from the welfare resources of the host society.14 Yet Collins is merely one judgment in a burgeoning line of case law— including rulings such as Grzelczyk, Trojani, Bidar and Förster (on relations within the migrant’s host State)15 and D’Hoop, Tas-Hagen, Hendrix and Morgan and Bucher (on relations with the migrant’s country of origin)16—which addresses the complex problems generated by the interaction between the Union’s free movement and equal treatment rights as interpreted by the Court of Justice, and each Member State’s restrictions on access to welfare benefits by economically inactive migrants. The claimant in Collins did not create those problems. He merely took advantage of the possibilities created for him by Union law itself, thanks to the ECJ’s decision to invent a limited duty of financial solidarity towards migrant Union citizens, which positively invited individuals to test the boundaries of the Member State’s new-found welfare generosity. In fact, the claimant in Collins should be thanked for doing EU law a service in clarifying the impact of the expanded equal treatment rights associated with Union citizenship upon the legal status of migrant work-seekers as traditionally defined under what is now Article 45 TFEU [ex Article 39 EC]. Similarly, the ruling in Chen has also been cited as an example of individual conduct amounting to a form of ‘abuse’ under what is now Article 21 TFEU [ex Article 18 EC], on the basis that the third country national claimants deliberately gave birth to their child in part of the territory of one Member State (the UK) which, for particular historical and constitutional reasons, automatically qualified the baby for nationality of another  Case C-138/02 Brian Francis Collins v Secretary of State for Work and Pensions [2004] ECR I-2703.  Case C-184/99 Rudy Grzelczyk v Centre public d’aide sociale d’Ottignies-Louvain-la-Neuve [2001] ECR I-6193; Case C-456/02 Michel Trojani v Centre public d’aide sociale de Bruxelles (CPAS) [2004] ECR I-7573; Case C-209/03 R (on the application of Dany Bidar) v London Borough of Ealing and Secretary of State for Education and Skills [2005] ECR I-2119; Case C-158/07 Jacqueline Förster v Hoofddirectie van de Informatie Beheer Groep [2008] ECR I-8507. 16  Case C-224/98 Marie-Nathalie D’Hoop v Office national de l’emploi [2002] ECR I-6191; Case C-192/05 Tas-Hagen and Tas v Raadskamer WUBO van de Pensioen- en Uitkeringsraad [2006] ECR I-10451; Case C-287/05 D P W Hendrix v Raad van Bestuur van het Uitvoeringsinstituut Werknemersverzekeringen [2007] ECR I-6909; Joined Cases C-11/06 and C-12/06 Rhiannon Morgan v Bezirksregierung Köln (C-11/06) and Iris Bucher v Landrat des Kreises Düren (C-12/06) [2007] ECR I-9161. 14 15

360  Michael Dougan Member State (Ireland), while simultaneously satisfying the need for a cross-border element to trigger the Treaty, all with a view to generating a secure right of residence for the parents themselves under EU law.17 Yet again, the alleged ‘abuse’ in Chen amounted to no more than one family taking advantage of the opportunities afforded to them under Union law, given the dependent nature of Union citizenship upon the idiosyncrasies of certain Member State nationality rules, and the almost boundless willingness of the Court of Justice to recognise a cross-border element capable of triggering the application of the Treaty. Moreover, Chen can also be linked up with a whole series of disputes about the residency rights of third country national family members, such as Surinder Singh, Akrich and Metock,18 where the relevant claimants sought to rely on free movement rights under the Treaty as a means of avoiding the application of national immigration restrictions. It seems a little harsh to describe the conduct of the claimants in such cases as an ‘abuse’ of EU law. They were merely arguing for an interpretation of the scope of application of their Treaty rights, which the Court of Justice was then free to ratify or reject as representing an accurate understanding of the current state of Union law. Fair enough, one might consider that, in cases like Akrich or Chen, the ‘subjective element’ generally involved in ‘abuse of law’ cases was indeed present, insofar as the claimants intended to obtain some advantage from Union law by ‘artificially’ creating the conditions necessary to qualify for that benefit.19 However, the ‘objective element’ required for a true ‘abuse of law’ was entirely lacking.20 In fact, the real cause for complaint in cases like Akrich or Chen—or indeed in rulings like Collins—lies in our uneasiness about the law itself, ie in terms of how the Union’s institutions, and especially the Court, have reconciled certain competing values within the operation of the Union legal system, and as regards the latter’s interaction with the domestic legal orders. One might well disapprove of individuals who appear to engage in welfare tourism, or seek to bypass national immigration controls, but their behaviour is merely a consequence of the policy decisions which flow from granting universal rights to free movement across the Union’s highly diverse regulatory environment, and the legal response to such behaviour lies entirely in the hands of the Union institutions when implementing and interpreting the relevant provisions of primary and secondary EU law. Having said all that, there is nevertheless some scope for employing a general concept of ‘abuse of law’ within the context of free movement for Union citizens. However, such a concept would manifest itself not in the guise of a general principle of Union law in any specific technical sense, merely as an interpretative tool for the purposes of defining and enforcing certain autonomous concepts of Union law. Three examples should illustrate that proposition. First, consider the fact that there are significant differences under EU law as regards the legal rights relating to residency and equal treatment afforded (on the one hand) to economically active migrants and (on the other hand) to economically inactive Union citizens. Those differences clearly create an incentive for certain individuals to attempt to 17  Case C-200/02 Kunqian Catherine Zhu and Man Lavette Chen v Secretary of State for the Home Department [2004] ECR I-9925. 18  Case C-370/90 R v Immigration Appeal Tribunal and Surinder Singh ex p Secretary of State for the Home Department [1992] ECR I-4265; Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607; Case C-127/08 Blaise Baheten Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I-6241. 19   Cf Emsland-Stärke (n 7 above), para [53]. 20   Cf Emsland-Stärke (n 7 above), para [52].

Comments on Abuses of Law and the Free Movement of Citizens  361 pass themselves off as workers or self-employed persons, particularly so as to gain access to a wider range of social and welfare benefits within the host society.21 Such behaviour— whether innocently opportunistic or downright fraudulent—obviously requires EU law to establish, and the national authorities to enforce, a clear and coherent borderline between the various categories of migrant. Moreover, it will be recalled that similar problems have long arisen when it comes to defining the borderline between established persons and service providers for the purposes of determining the legal nature of economic relations with a host State (since it is more likely that established persons will have to submit to professional supervision by the host State, than service providers who already satisfy the regulatory standards of their home State).22 But in both situations, any idea of ‘abuse of law’ arises and is resolved merely as a matter of correctly classifying an individual migrant and thus identifying the particular body of legal rules to which he/she should be subject under EU law. Secondly, the Council seems set to express formally its concerns about the prospect of individuals attempting to evade the national immigration restrictions ordinarily applicable to third country nationals, not through the ‘artificial’ exercise of free movement rights by a Union citizen so as to trigger protection under the Treaty for their family members (as in cases like Akrich or Chen), but rather through the simple forging of relevant documents or the contracting of sham marriages.23 Of course, the importance of acting against individuals engaged in such fraudulent attempts to invoke the benefits of EU law so as to bypass ordinary national immigration restrictions has been recognised both by the Court of Justice in rulings such as Metock,24 and by the Union legislature itself when enacting Directive 2004/38.25 However, it seems clear that the concept of ‘abuse of law’ in this context should again be understood as an interpretative tool, not this time for the purposes of slotting the claimant into the correct legal sub-category within the system of variegated free movement rights for Union citizens and their family members, but rather so as to remove altogether those individuals engaged in the use of fraudulent documentation or the contracting of sham marriages from the proper scope of application of the Treaty and/or the protection of the relevant Union secondary legislation. Thirdly, the Council also seems set to make clear its disdain for Union citizens who ‘abuse’ or ‘misuse’ their free movement rights by breaking the law ‘in a sufficiently serious manner by committing serious and repeated offences’.26 In the Council’s view, the right to free movement may be a fundamental freedom for Union citizens, but it also entails certain responsibilities, including respect for the rule of law within the host State. Once more, the rhetoric surrounding ‘abuse of law’ as a political and moral concept in such situations must be properly translated into the legal and conceptual framework of free movement 21  Consider for example, Case C-413/01 Franca Ninni-Orsache v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I-13187; Trojani (n 15 above). 22  For example, Case 33/74 Johannes Henricus Maria van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299; Case C-55/94 Reinhard Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano [1995] ECR I-4165. 23  See draft Council conclusions on abuses and misuses of the right to free movement of persons 16151/1/08 REV 1 (26 November 2008). 24   Metock (n 18 above). 25  See Art 35, Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States amending Regulation (EEC) No 1612/68 and repealing Directives 64/221/EEC, 68/360/EEC, 72/194/EEC, 73/148/EEC, 75/34/EEC, 75/35/EEC, 90/364/EEC, 90/365/EEC and 93/96/EEC [2004] OJ L158/77. 26  See draft Council conclusions (n 23 above).

362  Michael Dougan under EU law. The Council is clearly referring to the Member State’s competence, under the primary Treaty provisions,27 to terminate an individual’s exercise of their free movement rights on grounds of public policy or public security, in compliance with the substantive and procedural safeguards now enshrined in Directive 2004/38.28 Here, in other words, the concept of ‘abuse of law’ serves merely as a layperson’s shorthand for the Member State invoking a legitimate ground of derogation so as to restrict or revoke an individual’s rights under the Treaties. Apart from such situations, however, it is difficult to see how the idea of ‘abuse of law’—surely when conceived as a general principle of Union law, but even when deployed merely as an interpretative tool to guide the construction of EU legislation, or as an express derogation/imperative requirement for Member States attempting to restrict the exercise of Treaty rights—can apply in the specific context of free movement for Union citizens. After all, one of the core elements underpinning ‘abuse of law’ is that the individual is using EU law for reasons other than its intended purpose.29 The problem is that, since the introduction of Union citizenship, rights to free movement for Union nationals have been divorced from their purely economic rationale, decoupled from their largely instrumental nature in the construction of the Single Market, and transformed instead into autonomous rights for the benefit of all Union citizens. It is true that not all Union citizens will necessarily benefit from full free movement rights in practice, in particular, since rights to residency for more than a relatively short period of time remain subject to qualifying conditions based on the claimant’s active economic contribution or sufficient financial resources. Nevertheless, the fact remains that free movement now pertains to Union citizenship per se: beneficiaries may exercise their rights under Article 21 TFEU [ex Article 18 EC], regardless of their economic or social motivation, in pursuit simply of whatever opportunities for personal betterment/fulfilment that each citizen defines for him or herself. Indeed, it seems constitutionally difficult to argue that any right pertaining to the status of citizenship should only serve a given ‘purpose’, or else its exercise will be deemed to constitute an ‘abuse’. Surely none of the rights specifically associated under the Treaties with the status of Union citizenship—from the right to vote and stand in local and European elections, to the right to seek diplomatic protection in third countries, and also the right to free movement across the territory of the Member States—can be described as instrumental to any given end. They now constitute inherent elements of the citizens’ legal patrimony, as determined by the Member States when they agreed and/or amended the Treaties. It is one thing to determine each claimant’s proper place within the complex nomenclature of free movement law, or to exclude from the protective scope of the Treaty those third country nationals whose (non-existent) status was based on fraudulent conduct, or to rescind the residency rights of certain individuals who have contravened fundamental public policy interests. But beyond that, the question arises: how can a Union citizen possibly ‘abuse’ a right which serves no legally defined (or definable) purpose?

 Especially ex Arts 18(1), 39(3), 46(1) and 55 EC [now Arts 21(1), 45(3), 52(1) and 62 TFEU].  See, in particular, Ch VI, Directive 2004/38 (n 25 above). 29   Cf Emsland-Stärke (n 7 above), para [52]. 27 28

25 The Anatomy of Tax Avoidance Counteraction: Abuse of Law in a Tax Context at Member State and European Union Level judith FREEDman* I. Introduction

C

oncepts of abuse of rights, abuse of law and anti-avoidance in the field of taxation differ between jurisdictions and even within jurisdictions, depending upon the legal culture and the particular context. The concepts may vary even if they are superficially described in the same language; sometimes the language is different yet the concepts are similar. In the UK the phrase used in this area of tax law is generally ‘avoidance’—perhaps ‘aggressive’ avoidance or ‘unacceptable’ avoidance—rather than ‘abuse’. Elsewhere, we find a range of phrases which sound similar but are subtly different, such as: abus de droit par fraude á la loi, abus de droit par simulation, acte anormal de gestion, fraus legis, sham, substance over form and their equivalents in other languages.1 Quite apart from the linguistic differences, there are important variations in meaning so that, for example, simulation in France is wider than the UK notion of a sham, and the US notion of sham is also more extensive than that in the UK, although the word used is exactly the same.2  * The author would like to thank a number of colleagues for discussions on this topic and in particular Dr Rita de la Feria for her very helpful comments. Responsibility for the chapter is that of the author alone. 1   It is not the purpose of this chapter to provide a detailed description of the origins and nature of these concepts referred to in each Member State. There are several general discussions of this topic available: see D Ward et al, ‘The Business Purpose Test and Abuse of Rights’ [1985] British Tax Review 68 and especially the literature cited at fn 1 thereof; F Vanistendael, ‘Judicial Interpretation and the Role of Anti-Abuse Provisions in Tax Law’ in G Cooper (ed), Tax Avoidance and the Rule of Law (Amsterdam, IBFD Publications 1997); International Fiscal Association, Form and Substance in Tax Law: Cahiers de droit fiscal international, vol 87a (The Hague, Kluwer Law International 2002 )(hereafter IFA 2002); M Greggi, ‘Avoidance and abus de droit: The European Approach in Tax Law’ (2008) 6 eJournal of Tax Research 23. On the UK terminology see J Freedman, ‘Interpreting Tax Statutes: Tax Avoidance and the Intention of Parliament’ (2007) 123 Law Quarterly Review 53. For the distinction in France between abus de droit par simulation and abus de droit par fraude á la loi see L Leclercq, ‘Interacting Principles: The French Abuse of Law Concept and the EU Notion of Abusive Practices’ (2007) 61 Bulletin for International Taxation 235. 2   In the UK a sham is an act done or document executed by the parties which are intended by them to give to third parties or to the court ‘the appearance of creating between the parties legal rights and obligations different from the actual legal rights and obligations (if any) which the parties intend to create’. Snook v London & West Riding Inv [1967] 2 QB 786, per Diplock LJ. See R Ballard and P Davison in IFA 2002 (n 1 above) 571. For the

366  Judith Freedman It has been argued by Rita de la Feria3 and others that current developments in the case law of the Court of Justice seem to be creating a new European Union (EU) concept or, possibly, a set of concepts: a notion of abuse of EU law4 and perhaps also a related, but more general, principle of abuse. Thus there is being added a further layer of meaning or meanings. Inevitably, the ideas emerging in the EU have been derived in part from national statute and case law and coloured by the way in which the questions are posed to the Court of Justice by national courts. Although what emerges are EU concepts, these concepts not only govern the answers to questions relating to EU law with some resulting level of standardisation across the Member States,5 but also, as discussed below, developments are beginning to suggest that these concepts may be influencing the evolution of abuse concepts in national tax systems more generally, even, perhaps, in some areas where EU law is not directly relevant. In this way there might emerge, very slowly and over many years, a more widespread harmonisation of ideas about the tests for abuse or avoidance, although this might more properly be called ‘seepage’ of legal ideas. As Malcolm Gammie has commented, in comparing the UK judicial anti-avoidance principle with the EU abuse of law principle, although the starting points might differ, the problems being dealt with are similar, so it is not surprising if the results are similar too.6 While cross-fertilisation of legal ideas across jurisdictions is nothing new, it is speeded up by the interaction between the law of the EU and that of Member States and the traffic is not one-way but multidirectional.7 Whether there is one coherent EU abuse principle, or rather a related but varied set of concepts which are applied to meet a range of similar, but not identical, difficulties is, of course, a central issue for this book.8 Frans Vanistendael, for example, has remarked that: [T]he ECJ may finally come around to a coherent general theory of abuse of law that extends in a consistent way the notion of abuse of Community law to all tax matters. At the same time these theories may provide an instrument to protect the integrity of national tax systems.9

Part II of this chapter examines briefly the notion of abuse at EU level in a tax context and from the point of view of interaction with national laws. Other chapters in this book wider US meaning, see W Streng and L Yoder in IFA 2002 (n 1 above) 596. There is another range of terminology for illegal acts such as evasion, fraude fiscale and Steuerhinterziehung. This type of illegal transaction is not considered in this chapter. 3   R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. 4  Note that the term ‘abuse of law’ is used in preference to the term ‘abuse of rights’ in the current context. See AG Poiares Maduro in Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, para [70] and fn 66. 5  See F Vanistendael, ‘Halifax and Cadbury Schweppes: One Single European Theory of Abuse in Tax Law?’ (2006) 15 EC Tax Review 192. 6  M Gammie, ‘The Judicial Approach to Avoidance: Some Reflections on BMBF and SPI’ in J Avery Jones, P Harris and D Oliver (eds), Comparative Perspectives on Revenue Law: Essays in Honour of John Tiley (Cambridge, Cambridge University Press 2008) 25. 7  On the concept of the ‘judicial transplant’ in a tax avoidance connection, see C Garbarino, ‘The Development of a Judicial Anti-Abuse Principle in Italy’ [2009] British Tax Review 186. On the more general point see J Usher, ‘The Reception of General Principles of Community Law in the United Kingdom’ (2005) 16 European Business Law Review 489. 8  De la Feria (n 3 above). The issue is discussed by many contributors to this book and reflected in its title. 9  Vanistendael (n 5 above). In his contribution to this book, however, the same author recognises that more than one concept of abuse may be needed and may be being developed by the Court of Justice (see F Vanistendael, ‘Cadbury Schweppes and Abuse from an EU Tax Law Perspective’, ch 28 below).

The Anatomy of Tax Avoidance Counteraction  367 deal with this in more detail10 but the discussion here highlights themes of importance in relation to the national law of Member States. Part III considers the special difficulties of abuse and avoidance encountered in the area of taxation. It does not seek to examine the details of national tax systems but considers in general terms the common problems that have been encountered by the Court of Justice and national systems seeking to devise mechanisms to deal with these issues. Part IV considers how the EU concepts are likely to impact on the development of national concepts which seek to control tax avoidance. In conclusion, in Part V, it is argued that not only is there a formal influence on national tax systems from the Court of Justice’s decisions, but also that the discussion of various concepts at the level of the Court of Justice has already stimulated general national discussion and lawmaking around the concept of abuse in the area of taxation, and will continue to do so. The essence of a principle of abuse of law is, however, flexibility. Whilst guidelines and a reasonably certain framework are needed, attempts to turn the principle into a set of detailed rules through over-reliance by national courts on the specific wording of Court of Justice decisions could prove to be very unhelpful in controlling abuse.11 II. Concepts of abuse and artificiality developed by the Court OF JUSTICE in a tax context

It seems that the argument for a coherent theory in the area of tax law requires us to equate at least two rather different conceptual situations. These differences have been widely recognised by the commentators and variously described.12 There are two paradigm cases for the purposes of this discussion. First, there are cases arising in harmonised areas of law, such as the value-added tax (VAT) case of Halifax,13 where the relevant questions are whether a tax advantage is being obtained by the taxpayer contrary to the purpose of the specific EU legislation and whether it is apparent from objective factors that the essential aim of the transactions concerned is to obtain a tax advantage, (the two part test). In such circumstances, the Halifax decision states that an abusive practice exists and the transactions involved must be redefined so as to re-establish the situation that would have prevailed in the absence of the transactions constituting that abusive practice. This is a very familiar situation for tax lawyers used to dealing with what is argued to be abusive use of national legislation and raises a similar set of questions. The only difference here is that what is in question is the purpose of EU legislation or of national legislation derived from EU legislation, and which is to be interpreted in accordance with an EU directive or regulation. The issues at both EU and national level are very similar. Key problems are the extent to which the test is objective; whether the accrual of a tax advantage has to be a principal, essential or a sole aim of the transaction; and whether there is an overriding legal principle in operation or merely a form of statutory interpretation. These are difficult but familiar questions   For this reason this chapter will not analyse the facts of the Court of Justice’s cases in detail.  This preference for a principle over rules is discussed further in J Freedman, ‘Defining Taxpayer Responsibility: In Support of a General Anti-Avoidance Principle’ [2004] British Tax Review 332. 12  De la Feria (n 3 above) distinguishes an abuse of law from an abuse of EU law; Wolfgang Schön talks of the different levels of legislation, see W Schön, ‘Abuse of Rights and European Tax Law’ in Avery Jones et al (n 6 above) 75. See also B Kiekebeld, ‘Anti-Abuse in the Field of Taxation: Is There One Overall Concept?’ (2009) 18 EC Tax Review 144, as well as the distinctions drawn in a number of the other tax chapters in this book. 13   Halifax (n 4 above). 10 11

368  Judith Freedman at both national and EU levels and are discussed further in Part III. One thing clearly stated in relation to this type of case by the Court of Justice, however, is that in applying the relevant legislation, the courts may take into account whether the transactions in question have a purely artificial nature.14 The second type of case presents greater difficulties and appears, at least initially, to be of a completely different type from the first. In this model, as illustrated by the Cadbury Schweppes case,15 there is a piece of national anti-avoidance legislation, but the issue is not whether that national legislation is abused by the taxpayer. The question is, rather, whether that legislation may properly apply to the taxpayer: that is, whether and to what extent that legislation may legitimately prevent the taxpayer from exercising his right to freedom of establishment under EU law. In other words, in the Cadbury Schweppes context, is the attempt to establish a subsidiary in another Member State for tax purposes abusive of the freedom of establishment, so that the national legislation may properly inhibit it, or does that national legislation go too far? It is firmly established in the case law that setting up in a Member State for the purpose of enjoying a favourable tax regime in that State is not of itself an abuse of the freedom of establishment.16 Thus the question is whether there are any circumstances in which national anti-avoidance legislation can legitimately apply to prevent the exercise of this freedom or whether the national legislation is itself in breach of the freedom. On one view, no question arises on breach of the freedom by the national legislation where the establishment in question is not in genuine and actual pursuit of an economic activity in the host Member State. The freedom simply does not apply in such circumstances.17 On another analysis, as explained by the Advocate-General in Cadbury Schweppes, there is a restriction imposed on the freedom, and the question is whether the restriction is justifiable or proportional because the taxpayer has relied on the freedom abusively by setting up an artificial arrangement in order to avoid tax.18 The issue in this second type of case is linked to that of abuse of law because the test used to establish the circumstances in which there is no breach, or alternatively where the restriction is justified, is that of whether there are ‘wholly artificial arrangements’. This is said to be the same test as that developed in abuse cases such as Halifax, derived from earlier cases in other areas of law (the Halifax model).19 This seems to be a linguistic link rather than an assertion that the cases are conceptually identical. Advocate-General Léger in his Opinion in Cadbury Schweppes referred to a ‘formula, the language of which reproduces that of the doctrine of abuse of rights’ (emphasis added).20 The Advocate-General pointed out that it had been established by the jurisprudence of the ECJ21 that ‘hindrance to a freedom guaranteed by the Treaty can only be justified on   Halifax (n 4 above), para [81].  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 16  AG Léger in Cadbury Schweppes (n 15 above), para [39]. 17  AG Léger in Cadbury Schweppes (n 15 above), para [54], citing Case C-221/89 R v Secretary of State for Transport, ex p Factortame Ltd et al [1991] ECR I-3905. See also the chapters by Julian Ghosh and Richard Lyal in this book: J Ghosh, ‘Cadbury Schweppes: Breach, Abuse Justification and Why They Are Different’, ch 31 below; R Lyal, ‘Cadbury Schweppes and Abuse: Comments’, ch 29 below. 18  AG Léger in Cadbury Schweppes (n 15 above), paras [86–88]. 19  Originating with Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569 as discussed by de la Feria (n 3 above). 20  AG Léger in Cadbury Schweppes (n 15 above), para [88], referring to Emsland-Stärke (n 19 above). 21  Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector 14 15

The Anatomy of Tax Avoidance Counteraction  369 the ground of counteraction of tax avoidance if the legislation in question is specifically designed to exclude from a tax advantage wholly artificial arrangements aimed at circumventing national law’.22 He then went on to state that ‘application of Community law may be refused only when the company in question relies on it abusively because it has set up an artificial arrangement in order to avoid tax’.23 Thus the language used in the two types of cases is similar. At the core is artificiality and how to identify that artificiality. Nevertheless the cases have different bases and potentially very different impacts. In the first type of case, the Halifax model, it can be seen that it is easy to argue that the concept of abuse of EU law is simply an extended rule of statutory interpretation24 (albeit a very special one that can run counter to the literal meaning of the legislation), since the question is how to apply a specific piece of legislation, and whether it applies where there is an artificial transaction. In the second type of case, the Cadbury Schweppes model, there is a different exercise undertaken. First of all, it is the national law rather than the Union law (the freedom of establishment) that has to be interpreted to see if it satisfies the test of restricting only abusive transactions. In one sense this involves interpretation of EU law—that is, of the freedom itself—since the Court of Justice must decide what nature of establishment the freedom attempts to protect. But when it comes to deciding the case at a national level, the detailed discussion is of the meaning of the national law. Statutory interpretation may be involved insofar as the national courts attempt to cut down the national anti-avoidance legislation in question by reference to the freedom in order to make it consistent with the Treaty, under the Marleasing principle (or principle of ‘indirect effect’).25 The general process here, however, looks more like one of an overriding principle than a rule of construction. The freedom of establishment is an overriding principle and if the national legislation cannot be interpreted so as to comply with it, the national legislation must be changed. A third type of situation is illustrated by the Kofoed case,26 which dealt with the Merger Directive27 in Denmark. Under the Directive, which had been implemented in national legislation, an exchange of shares of the kind undertaken in the case was not prima facie taxable, but the national court found that the exchange of shares in issue ‘was not carried out for any commercial reason whatsoever but solely for the purpose of achieving tax savings’.28 Denmark had not chosen to enact a specific anti-avoidance clause as was provided for by that Directive. This would have applied where tax evasion or tax avoidance was a ‘principal objective or one of the principal objectives’ of the transaction. Nevertheless the Danish revenue authorities sought to apply general Danish anti-avoidance law to of Taxes) [1998] ECR I-4695; Case C-436/00 X & Y v Riksskatteverket [2002] ECR I-10829; Case C-324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I-1779; Case C-9/02 Hughes de Lasteyrie du Saillant v Ministère de l’Economie, des Finances et de l’Industrie [2004] ECR I-2409. 22  AG Léger in Cadbury Schweppes (n 15 above), para [87]. 23   Cadbury Schweppes (n 15 above), para [88]. 24  See AG Poiares Maduro in Halifax (n 4 above), para [71], where he refers to ‘this Community law principle of interpretation’. 25  Case C-106/89 Marleasing SA v La Comercial Internacional de Alimentacion SA [1990] ECR I-4135 and see, in the UK, Vodafone 2 v The Commissioners for her Majesty’s Revenue & Customs (HMRC) (No 2) [2009] STC 1480 (CA), [2009] EWCA Civ 446, discussed further below. 26  Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795. 27  Council Directive 90/434/EEC of 23 July 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States (Merger Directive) [1990] OJ L225/1–5, as amended. 28   Kofoed (n 26 above), para [36].

370  Judith Freedman counteract the transaction.29 The ECJ held that the national rules on ‘abuse of rights, tax evasion or tax avoidance’ could be applied if they could be interpreted in accordance with the abuse article in the Directive (despite the fact that the national legislature had not included this when implementing the Directive). Thus, in the national court, the revenue authorities can argue that the national anti-abuse law should be interpreted so as to accord with the unimplemented Article in the Directive.30 Although this is a different route of travel, it shows the way in which the Court of Justice will be able to influence the interpretation of national abuse legislation and judicial law in the future. These concepts of abuse of law have fundamentally different starting points. The first allows a strongly purposive and contextual interpretation of specific Union-based legislation and of the surrounding facts. It enables Member States to apply their national implementing legislation with no further legislative amendment being necessary to prevent abuse. The second and third models deal with the interpretation and development of national laws in such a way as to conform to the Court of Justice’s interpretation of the Treaty and the directives. If the national legislation can be interpreted so as to conform with the Court of Justice’s view of the EU law—be it the directive or the freedom—this may be done by the national court, but this type of case is more likely than the first to lead to redrafting and new legislation from national legislatures in terms which satisfy the Court of Justice’s restrictions. The Court of Justice’s decisions in the first type of case seem likely to have an influence on the development of judicial decisions in Member States. Those in the second type of case will have an impact on the shaping of anti-avoidance provisions introduced by the legislatures in the Member States. The third type of case underlines the interaction between legislative and judicial law and may influence both. What binds the decisions in these cases is language, and the attempt to draw the line between genuine economic activity and the artificial; between activity intended to be treated one way by the legislation and activity which, in the opinion of the court, is intended to be treated in some other way. These are distinctions with which all tax systems struggle and therefore we turn now to a discussion of this problem. III. Common problems encountered in dealing with tax avoidance

As pointed out above, since the underlying issues involved with tax avoidance are common to all systems, it is not surprising to see shared approaches evolving.31 Concepts developed  As to which see J Pedersen in IFA 2002 (n 1 above) 233.   Kofoed (n 26 above), para [46]. In the case of Modehuis A Zwijnenburg BV (Case C-352/08 20 May 2010 nyr) the Court of Justice went one step further and considered tax avoidance in the context of the Merger Directive and the domestic law of the Netherlands as it applied to two Netherlands companies. The Merger Directive did not apply to this purely domestic merger, but the domestic law that did apply was based on the Merger Directive in order to apply equal treatment to domestic and cross-border mergers. The national court asked whether the anti-abuse provision taken from the Directive could be applied to an attempt to avoid a form of tax not covered by the Directive. The Court of Justice ruled that the favourable treatment introduced under the Directive could not be withheld from a taxpayer who has sought to avoid a tax not covered by the Directive. In the context of this chapter the most important point about the decision was the willingness of the Court of Justice to consider the interpretation of the anti-avoidance provision despite the fact that only domestic companies were involved in the particular case. The Court of Justice commented at para [33] ‘it is clearly in the European Union’s interest that, in order to forestall future differences of interpretation, provisions or concepts taken from European Union law should be interpreted uniformly, irrespective of the circumstances in which they are to apply’. 31   Gammie (n 6 above). See also the various discussions of national systems in IFA 2002 (n 1 above) and in 29 30

The Anatomy of Tax Avoidance Counteraction  371 at national level can be seen to be both feeding into the Court of Justice’s decisions and being influenced by them, not least because questions for the Court of Justice are framed by national courts. There are several key issues: all involve different methods of reaching the same question. First, there is the problem of attempting to assess ‘economic reality’ within what is often a complex and artificial tax system based on legal principles rather than economics. Second, there is the issue of the relevance of the taxpayer’s purpose. Should this be tested subjectively or by using objective criteria, and should the test be that tax avoidance is the sole purpose, the essential aim or the principal purpose?32 The third major issue arising is whether the anti-abuse principle is an overriding principle or merely a guide to interpretation. This relates to ascertainment of the intention of the legislature, and the extent to which the courts can go beyond clear words of the legislation to some supposed purpose of the legislative body. Without ascertaining the purpose of the legislator, it is not possible to assess whether there is a tax advantage, the grant of which would be contrary to that purpose (as required by the ECJ in Halifax33). It is important to stress that the factors listed here are interlinked and, as we shall see, it may be misleading to see them as separate tests. One of the difficulties with some of the discussions of the Court of Justice’s jurisprudence is the tendency of commentators, practitioners and national courts to try to apply the tests as if they were completely disconnected, which can lead to attempts at manipulation and excessive reliance on the wording of the tests as detailed rules rather than as an overriding principle.34 Some would argue that the first two issues, referred to above, fold into the third, in that they are merely ways of approaching the issue of interpretation, whilst others would say that the first two approaches can result in a test which creates an overriding principle which is stronger than the approach that can be achieved by interpretation of the legislation alone. This would make the abuse of law approach a substantive one of form over substance, which the European Commission suggests is the case in its Communication on anti-abuse measures.35 This latter approach makes the finding of ‘economic reality’ or substance over form the overriding aim of the exercise, but even this takes us back to J Freedman (ed), Beyond Boundaries: Developing Approaches to Tax Avoidance and Tax Risk Management (Oxford, Oxford University Centre for Business Taxation 2008). 32  These different tests are all referred to in ECJ decisions. AG Poiares Maduro in Halifax (n 4 above) refers to a purpose to be ‘objectively determined on the basis of the absence of any other economic justification for the activity other than that of creating a tax advantage’ (para [87]) but this becomes ‘essential’ in the ECJ decision (para [75]) although the ECJ seemed to have thought it was following the AG. The Kofoed case refers to ‘principal objective or one of the principal objectives’ but only because it is concerned with the wording of Directive 90/434. (Kofoed (n 26 above), para [3].) In Part Service (Case C-425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I-897), however, it was established that the test in VAT cases, at least, is whether the principal aim is the accrual of a tax advantage. 33   Halifax (n 4 above), para [86]. 34   For an example of what is arguably an over-prescriptive approach, see the questions posed to ECJ by the UK Court of Appeal in Case C-103/09 HMRC v Weald Leasing Limited, ECJ Reference for a preliminary ruling made on 13 March 2009 [2009] OJ C129/13, as to the meaning of ‘normal commercial operations’, where the question is whether these words from the Halifax case add a further hurdle to be satisfied before the abuse of law principle can be applied or are merely part of the other two tests. The Advocate-General in the Weald Leasing case in an Opinion delivered on 26 October 2010 (nyr) has stated the view [para 33] that ‘normal commercial operations’ refers to the second part of the two part test in Halifax and is not a further requirement. Hopefully this common sense view will be followed by the Court of Justice. 35  See Commission (EC), ‘The Application of Anti-Abuse Measures in the Area of Direct Taxation’ (Communication) COM (2007) 785 final 4, 10 December 2007, which states: ‘Artificial arrangements must be detected via a comparative analysis of form and substance.’

372  Judith Freedman the question of statutory interpretation, since the economic reality has to be seen in the context of the objectives of the legislation. Thus it is a mistake to try to create a series of discrete tests. Nevertheless, for ease of discussion, these issues will now be discussed under broad headings. A.  Economic Reality Tax law tries to deal with real economic outcomes but it does so through the medium of private law concepts that do not always lend themselves to judgment by economic effect.36 The resulting conflict makes it difficult to legislate in a way that takes economic reality into account and judges need to be given a basis on which they can override the legislation to assess economic substance, if this is what they are required to do. At what point does one switch from the legal code to economic outcomes? ‘Reality’ and economic substance can be very difficult to discern if one is not dealing with legal transactions which have legal reality. Discussion of ‘form and substance’ needs to be seen in this light. Legal form is not always devoid of substance. Legal form may impart significant substance, though this will be legal substance and not economic substance. This means we may need a device set out in legislation or case law to help us draw a line between the legal (sometimes apparently ‘artificial’) realities we use as the foundation of our legal and therefore our tax system, and the legal realities we are to ignore to get the right economic result. For example, generally the existence of a subsidiary company cannot be ignored in a legal system which endows companies with separate legal personality, even though the corporation is a legal construct and economists would normally ignore the difference between, say, a branch and a subsidiary.37 The debt/equity distinction is in many ways entirely artificial, and economists would counsel that we should ignore it in designing a tax system, but if the system has not in fact ignored it and so treats debt and equity differently for tax purposes, we cannot decide to recharacterise debt to equity without disturbing that system. It follows that distinguishing ‘wholly artificial’ activities from ‘genuine’ ones is going to be problematic for any legal system if the ‘artificiality’ of a transaction is to be measured in terms of economic results which are not the initial basis for the legal classifications governing the tax system. Specific anti-abuse legislation can define elements of economic reality, but where general case law is required to determine this it will need to do so in terms of the surrounding circumstances, the purpose of the legislation, and the purpose of the taxpayer, judged by objective factors, as discussed below.38 Thus all the tests, formulated in the Court of Justice and in national contexts, aim at defining what is ‘genuine’ and simply use different routes to get there. The UK has traditionally rejected the concept of abuse of rights.39 It has no statutory general anti-avoidance provision and no judicial substance-over-form doctrine in a tax context.  A point made by R Ijzerman in IFA 2002 (n 1 above) 455.  Commission Communication (n 35 above) 3.   Halifax (n 4 above), para [75]; Cadbury Schweppes (n 15 above), para [65]. See also Barclays Mercantile (n 41 below), discussion below. 39   In the tax context, see Baylis v Gregory [1986] STC 22 (Ch) 43, per Vinelott J: ‘The doctrine of abuse of right by a taxpayer which obtains in some continental countries has no place in our jurisprudence.’ 36 37 38

The Anatomy of Tax Avoidance Counteraction  373 The case law has produced, however, the so-called Ramsay principle, which can be applied in tax avoidance cases.40 This has had many manifestations but is now firmly stated by the courts to be nothing more than a principle of statutory construction.41 Even so, despite express denials of this by the courts and commentators, what remains of the Ramsey principle seems to take account of what some might consider to be economic substance. Lord Hoffmann, writing extra-judicially, has explained: We said that the formula [in the Ramsay line of cases] was not a freestanding principle but rather the effect of construing a taxing provision in a particular way. If the statute required a transaction which had a business purpose . . . why then, steps which had no business purpose would not satisfy the statute. If the statute required something which had real commercial existence, like a profit or loss, then a series of preordained transactions which taken together produced no profit or loss would not satisfy the statute. On the other hand, if all that the statute required was something which had a particular legal effect, like discharging a debt or passing title to property, then a transaction which had that effect satisfied the statute even if it had no business purpose.42

Another formulation of this position, put forward by Ribeiro PJ, in the Hong Kong case of Arrowtown,43 was cited by the House of Lords with approval in the leading UK case of Barclays Mercantile:44 [T]he driving principle in the Ramsay line of cases continues to involve a general rule of statutory construction and an unblinkered approach to the analysis of facts. The ultimate question is whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.45

The reference to viewing the transaction realistically may be seen as another way of referring to economic substance, albeit that this is described as a rule of statutory construction, and not substance over form. It is not every statute that can be applied in this way; it depends on its wording and context. Nevertheless this principle looks somewhat different from a normal principle of statutory construction because of the way it is applied to the facts. Wolfgang Schön has described a similar debate around the German statutory general anti-avoidance tax provision, even though the starting point of this legislative provision looks very different from that of the UK case law.46 The German provision states: The tax law cannot be circumvented by abusive constructions. In case of an abuse the tax claim arises in the same way as it would have if a reasonable/adequate legal construction reflecting the true economic nature of the event would have been executed.47

The meaning of abuse in this German provision has recently been clarified by further legislation, which imports tests related to taxpayer purposes and the intention of the legislation. The provision cited above expressly links statutory construction and the concept of economic reality. Professor Schön comments that amongst German experts, 40   WT Ramsay Ltd v IRC [1982] AC 300. There is a vast literature on this principle but for some recent examples, see Lord Hoffmann, ‘Tax Avoidance’ [2005] British Tax Review 197; Gammie (n 6 above); Freedman (n 1 above). 41   Barclays Mercantile Business Finance Ltd v Mawson [2005] 1 AC 684, [2004] UKHL 51. 42  Hoffmann (n 40 above) 203. 43   Collector of Stamp Revenue v Arrowtown Assets Ltd [2003] HKCFA 46, para [35]. 44   Barclays Mercantile (n 41 above). 45   Arrowtown (n 43 above), para [36]. 46   W Schön, ‘Statutory Avoidance and Disclosure Rules in Germany’ in Freedman (ed), (n 31 above) 47. 47   § 42, para [1] Abgabenordnung cited in and translated by W Schön (n 46 above) 48.

374  Judith Freedman the majority opinion is that this provision goes beyond pure interpretation, but there is a strong minority view that it does not. Part of the exercise the courts undertake, however, is to review whether ‘reasonable parties’, having the same economic goal in mind, would have chosen this structure. Whilst this appears to be an objective test, subjectivity is introduced at a subsequent stage, since non-tax motives of the individual taxpayer are also considered. Just like the case law approach in the UK, this seems to result in a special sort of principle of statutory construction which does permit the court to look at the surrounding economic and commercial circumstances, ‘viewed realistically’. In the Halifax case, the ECJ quite properly left the responsibility for determining the ‘real substance and significance of the transactions concerned’ with the national court, but laid down that ‘the purely artificial nature of those transactions and the links of a legal, economic and/or personal nature between the operators involved in the scheme’ may be taken into account in this process.48 Thus the Court of Justice has given a steer as to how this is to be done and, in doing so, links the question of substance and economic reality clearly with taxpayer purpose. The level of artificiality is simply one way of judging taxpayer purpose in relation to the objectives of the legislation.49 B.  The Taxpayer’s Purpose It is established across the globe that tax minimisation per se is a permissible purpose, and this is clear from the Court of Justice’s cases also.50 In the UK this is expressed as the principle in the Duke of Westminster case51; in the USA it is found in Gregory v Helvering.52 Elsewhere it is enshrined in the constitution.53 It is deep in the psyche of the courts in most jurisdictions that the taxpayer is entitled to some protection when acting to minimise tax within the law. This may be seen as a principle of certainty: taxpayers are entitled to rely on the intention of the legislature as expounded in the clear wording of legislation. At the same time, it is clear that the development of sophisticated tax avoidance industries in most jurisdictions has resulted in curbs being developed on this principle where the taxpayer attempts to abuse it. There is some sub-category of tax minimisation which is to be restricted and the taxpayer’s objectives may be relevant in defining this sub-category. Yet if tax minimisation is generally acceptable, the relevance of these objectives or purpose is hard to discern, since it can be argued that it is the transaction that should be taxed, and the way in which similar transactions are taxed should not vary with the purpose of the taxpayer. In fact it appears that the purpose test is a proxy for economic reality. If the essential reason for the transaction is not commercial then it can be said to be ‘artificial’ or ‘not genuine’.   Halifax (n 4 above), para [81].   If this is correct, it follows that different levels of artificiality may be needed depending on the objective of the legislation, as pointed out by Kiekebeld (n 12 above). 50   Generally see Vanistendael (n 1 above). On the Court of Justice’s jurisprudence, see AG Poiares Maduro in Halifax (n 4 above), para [84], where he states that: ‘[t]he basic principle is that of the freedom to opt for the least taxed route to conduct business in order to minimise costs’, citing Joined Cases C-487/01 and C-7/02 Gemeente Leusden v Staatssecretaris van Financiën and Holin Groep BV cs v Staatssecretaris van Financiën [2005] ECR I-5337, para [79]. See also Cadbury Schweppes (n 15 above), paras [36–37]. 51   Commissioner of Inland Revenue v Duke of Westminster (1936) 1 AC 19. 52   Gregory v Helvering (1934) 69 F 2d 809. 53   For example, Art 42 of the Italian Constitution protects the freedom to conduct private economic activities. See Garbarino (n 7 above). 48 49

The Anatomy of Tax Avoidance Counteraction  375 One problem with this test is that even highly artificial arrangements can often be presented by well-advised taxpayers in such a way as to show a ‘commercial’ non-tax purpose. Requiring tax to be the sole purpose is, therefore, a test of limited value to the tax authorities, which is why a main purpose test is often preferred in national systems.54 Even the Halifax test of essential aim is a relatively narrow one and permits considerable scope for argument around the weighting of aims, which can be very difficult. It is not surprising, therefore, that we now see the Court of Justice interpreting the Halifax test to mean that there can be a finding of an abusive practice where the accrual of a tax advantage constitutes the principal aim of the transaction or transactions at issue, in the subsequent case of Part Service.55 This is a somewhat more stringent test for the taxpayer and we can expect to see further pressure and development around this test in future litigation. The other main difficulty is whether the test is objective or subjective. It is clear from Halifax that the Advocate-General struggled with the subjective element of the test derived from Emsland.56 The objective/subjective issue is one which always arises where there is an issue about the intention of the parties.57 The Advocate-General’s solution was to inject objectivity into the test by assessing the purpose of the activities in question objectively on the basis of the absence of any other economic justification for the activity other than that of creating a tax advantage.58 The practical answer is that a purely subjective test is fairly clearly unworkable as a method of controlling tax avoidance. Thus inferences must be drawn from objective circumstances to reach a conclusion about the aim of the transactions.59 This takes us back to the test expounded in Arrowtown, of whether the relevant statutory provisions, construed purposively, were intended to apply to the transaction, viewed realistically.60 It also takes us on to a discussion of the test as one of statutory construction. C.  Principle or Statutory Interpretation? As has been noted, the Court of Justice’s abuse of law principle can be understood as one of interpretation. This is an unusual form of interpretation, however, that requires 54   In the UK, for example, the normal formulation in anti-avoidance legislation is ‘main purpose or one of the main purposes’. In the view of the UK revenue authorities, even very artificial tax-driven transactions can involve a minor or incidental non-tax purpose which would be enough to prevent a sole purpose test from applying and therefore policymakers should not use a ‘sole’ purpose test if at all possible. Even a dominant purpose test involves the problem that the weight of purposes can be manipulated. See HMRC, ‘Simplifying Unallowable Purpose Tests’ (HMRC Discussion Document) (31 July 2009) 18. 55   Part Service (n 32 above). 56   Emsland-Stärke (n 19 above). The subjective test had already been criticised by other AGs in non-tax cases, in particular AG Lenz in Case C-23/93 TV10 SA v Commissariaat voor de Media [1994] ECR I-4795, para [61]. See de la Feria (n 3 above) 423. 57   In the UK, the test of main object has been held to be subjective and one of fact, despite attempts by HMRC to argue to the contrary: IRC v Brebner [1967] 2 AC 18. 58  AG Poiares Maduro in Halifax (n 4 above), paras [70] and [86]. 59  As commented in the UK in 1955 by the Royal Commission on Taxation when discussing whether trading was taking place for tax purposes: There are cases in which the purpose of the transaction of purchase and sale is clearly discernible. Motive is never irrelevant in any of these cases. What is desirable is that it should be realised clearly that it can be inferred from surrounding circumstances in the absence of direct evidence of the seller’s intentions and even, if necessary, in the face of his own evidence . . . There are many fields in which the law has to concern itself with ascertaining motive but a tax appeal is not well suited to this kind of enquiry. 60   Arrowtown (n 43 above), para [36].

376  Judith Freedman the legal provision at issue to be interpreted ‘contrary to its literal meaning, as actually not conferring the right [that runs counter to its purposes]’.61 This goes beyond the understanding many lawyers have of pure statutory interpretation. The real difficulty lies with ascertaining the purposes of the statute, although this may be easier with EU legislation than with UK provisions; in the case of the latter the very detailed drafting style adopted can sometimes seem devoid of any principle whatsoever.62 But even EU legislation may not always be founded on a coherent and obvious economic principle and thus it can be difficult for the courts to decide these issues, as Dr de la Feria has argued, making the point that, for example, there can be a clash between VAT rules and VAT principles, both of which have the EU legislature as their source.63 Whether an anti-abuse rule is legislative or judicial, questions arise about its relationship with the specific legislation under consideration. Where a general anti-avoidance provision is legislative it may more easily be seen as operating as an overriding provision, but even in jurisdictions with such provisions the question of statutory construction versus overriding rule is problematic, as can be seen in Wolfgang Schön’s discussion of the German situation referred to above.64 In the Netherlands, the fraus legis doctrine, which is a doctrine of general application that is used in tax cases, can only be considered after the normal interpretation methods have been fully utilised but have not led to an outcome that can be regarded as consistent for the purpose and intent of the law.65 This clearly goes beyond normal interpretation rules and involves some level of override. If it is a rule of interpretation, then it is a very special one. The process of a general principle of interpretation gathering an accretion of substantive rules can be seen very clearly in the UK in the Ramsay line of cases. This new approach to the interpretation of tax legislation, intended to give the courts a measure of flexibility, began to be encrusted with rules. The House of Lords has attempted to remove these in recent cases66 and has declared that the Ramsay principle is nothing more than an application of modern, purposive statutory interpretation. The application of the case law seems to go beyond this, however, by virtue of the strong focus on the facts of the transaction and the way in which they, and not just the statute, should be viewed.67 What is more, there seems to be a tendency to build back the detail of the rules, especially in the lower courts. In a similar vein, the Court of Justice’s concept of abuse of law requires a provision to be interpreted contrary to its formal application if certain circumstances exist and the transactions can be analysed in a certain way.68 As with the UK case law, so with the Court of Justice: the guidance set out by the Court of Justice (in a tax context, as  AG Poiares Maduro in Halifax (n 4 above), para [70].  As pointed out by Lord Hoffmann (see n 40 above).   R de la Feria, ‘The European Court of Justice’s Solution to Aggressive VAT Planning—Further Towards Legal Uncertainty?’ (2006) 1 EC Tax Review 27. 64  Schön (text to n 46 above). 65   Ijzerman (n 36 above) 455. 66   Macniven (Her Majesty’s Inspector of Taxes) v Westmoreland Investments Ltd [2003] 1 AC 311, [2001] UKHL 6; Barclays Mercantile (n 41 above). 67  The author has argued elsewhere that there remains a judicial principle which goes beyond purposive statutory interpretation. See Freedman (n 1 above) 69. Gammie has also argued that this combination of a purposive construction and the factual reality of a composite transaction is something more than statutory construction. See M Gammie, ‘Sham and Reality: The Taxation of Composite Transactions’ [2006] British Tax Review 294, 310. 68  AG Poiares Maduro in Halifax (n 4 above), para [70]. 61 62 63

The Anatomy of Tax Avoidance Counteraction  377 to how to ascertain whether the aim of a transaction is to obtain a tax advantage or how to decide whether an entity is wholly artificial) becomes the basis for interpretation and for requests for clarification from the courts of Member States. This could result in greater rigidity and too many signposts being made available to those wishing to work around the principle and certainly starts to move away from a pure principle of statutory interpretation. For those who argue that the Court of Justice’s abuse of law concept is a new general principle of EU law, once again it would seem to be significantly more than an ordinary principle of interpretation. Dr de la Feria suggests that this new principle can act as an overriding rule of law to strike down a domestic provision in a form of judicial review, as in Cadbury Schweppes.69 There are limits on the extent to which EU law can be applied to modify the domestic law of Member States, however, and it would seem that a general principle of EU law, like a rule imposed by a directive, is subject to those limitations.70 In the Vodafone 2 case in the UK, the domestic courts accepted that the duty imposed on them to interpret domestic legislation on Controlled Foreign Companies (CFCs) to comply with EC law under the Cadbury Schweppes decision is ‘both broad and farreaching’ and ‘not constrained by conventional rules of construction’.71 Application of Community law ‘permits departure from the strict and literal application of the words which the legislature has elected to use’ and ‘it permits the implication of words necessary to comply with Community law obligations’.72 So there is flexibility beyond normal rules of interpretation, but not to interpret domestic legislation in a way that is incompatible with its underlying thrust or contra legem.73 Despite these limitations, the Court of Appeal in Vodafone 2 felt able to read into the UK CFC legislation concepts taken from the Court of Justice’s case law, as discussed in Part IV below. Whether this is a general principle that can override rules of law subject to principles or a very special rule of interpretation, it can result in modification of the normal approach to statutory interpretation adopted by national courts. IV.  Impact of the Court OF JUSTICE’s concepts on national systems

As has been noted above, there are both varied concepts and different uses of language across the Member States of the European Union in relation to tax abuse and avoidance and we have seen that the thread which links the different tax cases discussed here appears to be a linguistic one.74 In an interesting opinion statement on the emerging case law of the Court of Justice in the field of tax law, the Confédération Fiscale Européenne (CFE) has taken up this theme, commenting that linguistic and conceptual discrepancies arising from the Court of Justice’s judgments has made the concept of abuse far from clear. They are  See de la Feria (n 3 above) 438.   For a discussion of this and other matters pertaining to general principles, see A Arnull, ‘What is a General Principle of EU Law?’, ch 2 above. 71   Vodafone 2 (n 25 above), para [37] 72   Pickstone v Freemans plc [1989] AC 66 (HL); Litster v Forth Dry Dock & Engineering Co Ltd (in receivership) [1990] 1 AC 546 (HL); Ghaidan v Godin-Mendoza [2004] AC 557, [2004] UKHL 30; HMRC v IDT Card Services Ireland Ltd [2006] STC 1252, [2006] EWCA Civ 29. 73   Marleasing (n 25 above); Case C-334/92 Teodoro Wagner Miret v Fondo de Garantía Salarial [1993] ECR I-6911. 74  See text and footnote (n 20 above). 69 70

378  Judith Freedman particularly concerned that a clear distinction should be maintained between tax evasion and tax fraud on the one hand, and abuse on the other. They also comment ‘that the European institutions should overcome the linguistic and conceptual lack of clarity that currently surrounds the scope of abuse [in the field of tax law]’.75 The CFE comment that the current position is that Member States are allowed to apply their domestic and treaty anti-abuse measures in the field of tax law that have not been harmonised or coordinated, subject to the condition that such measure should not give rise to unjustified restriction on the exercise of the fundamental freedoms, but they question whether leaving Member States free to decide how to counter abuse is in line with the principles and goals of the Internal Market.76 As we have seen, the condition that the freedoms must not be restricted is itself a force that will result in modification of domestic anti-abuse laws, and thus the capacity of Member States to enact domestic anti-abuse law has been modified already. Not only will this affect cross-border anti-abuse laws, but it is predictable that general concepts and language used in tax avoidance situations will also develop in a way consistent with the Court of Justice’s thinking. Judges, and eventually legislatures, will be influenced by the reasoning and the approach they see in the Court of Justice’s cases and this may carry across to purely domestic cases. Examples are already beginning to emerge that bear out this prediction. In Italy, a series of decisions in the Corte di Cassazione has introduced a new judicial anti-abuse principle into the Italian tax system, side by side with the Italian legislative general anti-avoidance rule, which, according to Italian commentators, is explicitly based, in part, on the Court of Justice’s decisions.77 In addition, the Italian Court has relied upon the law of many other jurisdictions to find that a concept of abuse of law in the tax field forms a ‘common core’ of tax systems. This has been used to create a principle at the domestic level. The Italian court has shown itself willing to apply the Court of Justice’s case law across the board to harmonised and non-harmonised areas alike, and commentators suggest that the influence of the Court of Justice’s decisions has been very great, even though there has, in addition, been direct influence from other individual jurisdictions.78 In France too, the impact of the Court of Justice’s decisions on the domestic approach to abuse of law is very evident. The wording used by the ECJ in Halifax was reflected in the decision of the Conseil d’État in Janfin, a purely domestic case.79 In this case the French Court held that even if the general anti-abuse provision contained in Article L64 of the Tax Procedure Code did not apply, a general doctrine of abuse of rights could do so. The court’s definition of such abuse of rights was ‘greatly inspired by the definition adopted by the ECJ in the Halifax case’.80 The wording of the test refers to ‘instruments 75  Confédération Fiscale Européenne (CFE), ‘Opinion Statement of the CFE ECJ Task Force on the Concept of Abuse in European Law, Based on the Judgments of the European Court of Justice Delivered in the Field of Tax Law—November 2007’ (2008) 1 European Taxation 33. (See also n 1 above.) 76  CFE (n 75 above), para [44]. 77   Garbarino (n 7 above); M Greggi of the University of Ferrara has written (in letters to the author dated 28 September 2008 and 21 December 2008) that the Italian courts have accepted the Court of Justice’s notion of abuse and extended its application to cases and taxes not covered or discussed by the Court of Justice. The abuse concept has become a ‘catch-all’ in the tax context in Italy, according to Greggi. The author is grateful to Professors Greggi and Garbarino for discussing this issue with her. 78   Garbarino (n 7 above). 79   Conseil d’État, 27 September 2006, No 260050, Société Janfin.. 80   Leclercq (n 1 above) 235.

The Anatomy of Tax Avoidance Counteraction  379 which, seeking the benefit of a literal application of the laws to the detriment of the objectives sought by their makers, could not have been motivated by any purpose other than that of escaping or reducing the tax liability which, but for these instruments, the interested party would normally have borne in light of its actual situation and activities’.81 This extended case law definition has now been enacted into an amended form of Article L64 by the Finance Law 2008. It applies to all taxes and this seems to be widely accepted to have been the result of the influence of the Court of Justice’s cases.82 The novelty is that the new provision looks beyond fictitious acts (as covered under the previous provision) to those that ‘seek to benefit from a literal application of the laws against their stated aims with the exclusive intention of avoiding or reducing the tax charge’.83 Lefèvre Pelletier point out that the problem is for the taxpayer to know the intentions of the originator of the tax law and take comfort in the fact that the test requires the tax intention to be ‘exclusive’.84 Given developments in the Court of Justice as discussed above, it can be imagined that the definition of an exclusive intention may be a contentious one and that the French courts and legislators could even turn once again to the case law of the Court of Justice for inspiration and modification, given the continuing discussions in the Court of Justice.85 If, on the other hand, this does not happen, the national and the Court of Justice tests might start to diverge once again. This could cause a problem in a cross-border setting and possibly even give rise to cases of discrimination if the domestic rule was less stringent than the Court of Justice test or one derived from an EU Directive.86 As discussed above, the Zwijnenburg case is also an example of the influence of the Court of Justice’s jurisprudence on the development of domestic law, in this case as the result of a direct decision by the legislature in the Netherlands to apply the same rules to domestic as to cross-border mergers within Europe. Moreover, the Court of Justice expressly stated that the ECJ could answer a request for a preliminary ruling on a domestic case because it was in the interest of the Community legal order that the Community provision should be given uniform interpretation irrespective of the circumstances in which it was to be applied.87 These are but three examples. It is not possible within the scope of this chapter to survey every member of the EU, but it seems likely that a similar effect can be seen, more or less pronounced, elsewhere. There will be a clear pressure throughout the Member States to conform their general abuse rules in the area of tax abuse to avoid further problems. In addition, the formulations and language used by the Court is having a discernible impact on the thinking of judiciaries in Member States. The UK, which currently has no general anti-avoidance provision and whose case law at the highest level continues to assert that there is only an issue of statutory interpretation, rather than a judicial anti-abuse rule in domestic cases, seems likely to be rather more resistant to the Court of Justice jurisprudence than are other jurisdictions. The Halifax decision will of course be followed in relation to VAT, but probably the courts will feel able to keep that quite separate from cases of domestic direct tax avoidance.   English wording taken from Leclercq (n 1 above).  See Taxand, Taxand Quarterly ( January 2009) 7, www.taxand.com; Lefèvre Pelletier & associés, Panorama Fiscal 2008 (January 2009) 34, www.lpalaw.com. 83  Translation taken from Panorama Fiscal (n 82 above). 84   Panorama Fiscal (n 82 above). 85  See n 55 above and text thereto. 86   For a full discussion of this suggestion see E Picq, ‘“Abuse” of EU Holding Companies: Fundamental Freedoms, EC Parent-Subsidiary Directive and the French Constitution–Part 1’ (2009) 10 European Taxation 471. 87   Zwijnenburg (n 30 above). 81 82

380  Judith Freedman The Cadbury Schweppes decision has had a more dramatic impact, however, and brings in the cross-border issues referred to above, which cannot be ignored by the courts or the legislature. That this is so can be seen in the UK in the Vodafone 2 decision,88 where the Court of Appeal89 has shown itself prepared to apply the EC jurisprudence to the UK’s CFC legislation in a detailed way in order to conform it to Community law.90 The Court of Appeal held that it was possible to read the UK CFC legislation as though it had an exception inserted in it, based on the wording in Cadbury Schweppes, and therefore conformed with Community law.91 In this way, the wording taken from the case law of the Court of Justice was read directly into the UK legislation. It was thought that the case might have to return to the First Tier Tribunal for it to establish whether on its facts there were ‘wholly artificial arrangements intended to escape the national tax normally payable’ but it has now been settled.92 V. Conclusion

It has been shown in this chapter that the development of various concepts of abuse of law in a tax context at the level of the Court of Justice has already stimulated a considerable amount of general discussion, judicial activity and legislative activity in the area of taxation at a national level. This looks set to continue. We are some way away from having Frans Vanistendael’s ‘coherent general theory of abuse of law that extends the notion of abuse of Community law to all tax matters’.93 The conceptual differences between the different types of case discussed in Part II above make it unlikely that there will ever be one simple test dealing with all such situations. Nevertheless, there are similarities in the issues arising in these different situations and in different jurisdictions. The language of artificiality in particular is common to all these types of case and we can expect to see further developments around this at national and EU levels. Further jurisprudence is needed to provide guidance on as yet inchoate concepts and in the interests of the Internal Market and equality of treatment of domestic and crossborder transactions, it is desirable that the influence across jurisdictions continues to operate. In this process the Court of Justice has a positive role to play. Flexibility remains essential to the essence of a principle of abuse of law, however. Attempts to press the Court of Justice into providing ever more detailed guidance on the meaning of the general principle or principles of abuse of law in a tax context in the name of harmonisation might simply provide ammunition to those who wish to restrict the principle. Finding the difficult balance between certainty for taxpayers and the control of abuse of tax law will remain a juggling act at a European Union and Member State level.   Vodafone 2 (n 25 above).  The Supreme Court has refused leave to appeal in this case. 90  See n 72 above and n 73 above and text thereto. This is despite the fact that there was no directive for guidance, only case law. The judge in the High Court, who was overturned by the Court of Appeal, had not been prepared to hold that the legislation could be conformed in this way without a directive: Vodafone 2 v HMRC (No 2) [2008] STC 2391, [2008] EWHC 1569 (Ch). 91   Vodafone 2 (n 25 above), para [39]. 92  Dorsey & Whitney, ‘CFCs—Vodafone 2—Supreme Court Refuses Leave to Appeal’ Lexology, 4 February 2010. On the settlement, which has resulted in protests in the UK, see A Goodall ‘Vodafone: HMRC statement fails to halt protests’ Tax Journal 1 November 2010. 93   Vanistendael (n 5 above); see also n 9 above and text thereto. 88 89

26 Abuse of Law in the Context of Indirect Taxation: From (Before) Emsland-Stärke 1 to Halifax (and Beyond) Pasquale Pistone I. Introduction

V

arious decades of case law by the Court of Justice have turned abuse of law into a true principle of European law which applies in all EU Member States and throughout the different branches of law. Derived from the French principle of abus de droit and the German principle of Rechtsmissbrauch, abuse of law is not a synonym for abuse of rights, though English-speaking tax literature has often unduly put them together in the past. The main difference lies in their respective objects. There is an abuse of law when persons seek the improper application of a statute to atypical situations, ie situations other than those which the lawmaker intended to regulate, and which are in conflict with the rationale of such measure. By contrast, an abuse of right generally indicates the use of such right for an improper purpose, such as harming someone else’s right. The object of abuse in taxation is more frequently law than actual rights, possibly because legal relations in tax matters neither involve the attribution of rights with erga omnes effects, nor are interpersonal, but normally involve the taxpayer, on the one hand, and the Revenue, on the other hand. In this context, even when taxpayers abuse their rights for tax-saving purposes, the effects arising therefrom would not harm a specific person, but rather the State, by depriving the Revenue of such tax. The fact that abuse of law is a principle of European law does, however, not imply that its dimension should be limited to cases in which the object of abuse is European law itself, but rather include all situations in which EU Member States apply measures to counter the circumvention of their tax rules. Such situations actually indicate what practices are regarded as abusive from the perspective of national law, thus making it possible to single out cases in which no actual right can be claimed on the basis of European law. This chapter will focus on whether this common framework for abuse in European tax law may lead to a single concept of abuse, applicable to all types of taxes, thus including cases in which the object of abuse is a provision of European law, as well as those that implement it into national law or are just provisions of national law itself.

382  Pasquale Pistone II. Three issues of paramount importance

In particular, three particular issues seem of paramount importance when carrying out such analysis. First, may, or should the concept of abuse in tax law differ from the general pattern provided for by European law? In other words, is it conceivable or desirable that general principles—such as abuse of law—should take on a new shape or dimension when imported into tax law because of the peculiarity of this legal branch? Second, may there be a different legal dimension of general principles within different areas of tax law—such as for direct and indirect tax purposes—or according to whether the principle applies to primary or secondary law? Third, may there be a different dimension for abuse according to whether a specific domain has been the object of common policy, as in the case of value-added tax (VAT), or only exceptionally includes common provisions, as for secondary law in some specific areas of direct taxation?

III. The basic framework for the analysis

An answer to such questions is only possible to the extent that one overcomes the obstacles that have so far hindered a proper analysis of such issues. Such obstacles are mainly due to prejudices, a lack of dialogue and a mutual problem of understanding among the different branches of law. On the one hand, experts of general European law are often not inclined to deal with technicalities required for the proper understanding and application of tax law; on the other hand, in most countries—and especially within the community of those international tax experts, which mostly focus on direct tax issues falling within the scope of double tax conventions—indirect taxation has turned into a world apart, ie separate from the other tax domains. Furthermore, international tax experts have traditionally shown a limited understanding of general principles of European law, believing that national tax sovereignty under the traditional rules of international tax law would have not been significantly affected by the development of case law by the Court of Justice. In fact, they underestimated such phenomenon for many decades without really contributing to shaping up its principles. This trend has not too long ago changed, especially taking into account that as many as twelve and a half per cent of all completed cases by the European Court of Justice in 2006 dealt with tax problems. Such difficulties may be overcome on the basis of the following elements. Although a clear understanding of the mechanism of taxation is an important tool to gather the structural issues raised by the facts of a case and the consequences of a decision, or of the application of a principle, this does not imply that tax law should remain a reserved domain for tax experts only. If a principle of law has a general nature, which is normally the case, then it should apply to all legal domains, none being excluded. Of course, principles may suffer limitations, just as rules have exceptions. However, whether and to what extent there should be such limitations and specifications of general principles, it is a matter to handle with care, without reaching superficial conclusions that may be the mere outcome of misunderstanding. Even in the early days of its history, international

Abuse of Law in the Context of Indirect Taxation  383 taxation was perceived as a highly technical and intricate domain. This reason led the League of Nations to reject in 1927 the Four Tax Treaty Models drafted by a committee of the four tax experts (Professors Bruins, Einaudi, Seligman and Stamp) and set up a new committee to prepare new documents that could be best reconciled with public international law, perhaps marking a first sign in the separation as such of tax law from public international law. Such separation then gradually consolidated within the OECD history of tax treaties and the clauses that allowed Contracting States to interpret terms not defined by the treaty according to their domestic law (unless the context otherwise requires). Such context is, however, significantly different from the one object of this chapter, mainly because the supremacy of European law is structurally linked to the need to apply its principles consistently in all branches of law, well beyond the differences that each of them has. Furthermore, unlike international taxation, there is one single body competent to interpret and apply European law, ie the Court of Justice. An additional issue specifically arises in the field of indirect taxation and, in particular, value-added tax, whereas national rules are merely allowed to implement the provisions of the common system contained in the VAT Directives, subject to the limits of ensuring a neutral functioning of such regime throughout the European Union. Consequently, tax technicalities should not be an obstacle to the full application of the principles of European law, but they require a more thorough consideration of the factual pattern. Nor can the supremacy of European law—including the principle of prohibition of abuse of law—be made subject to the direct effect of a provision contained in its secondary law, which requires implementation by the Member States.1 This chapter will analyse how abuse of law arose and developed in the field of valueadded tax, taking into account the parallel developments in other fields of tax law and relating both such developments to the general dimension of abuse of law within European law. IV. The starting point of abuse in European law: the Emsland-Stärke 1 decision

Many regard the starting point of the development of abuse of law as the EmslandStärke 1 decision—a well-known case on export restitutions due in the framework of the European common agricultural policy—and, as far as value-added tax is concerned, the point of arrival in the Halifax decision. This chapter will challenge this view by critically analysing the development also in the light of the Court’s cases on value-added tax decided before Emsland-Stärke 1 and after Halifax. Nevertheless, this chapter acknowledges that both domains are the object of a common policy and that the Emsland-Stärke 1 decision represents a milestone in the construction of a European concept of abuse of law. The Emsland-Stärke 1 decision was based on the need to prevent European law from being relied upon in respect of transactions (or a series of transactions) that were not 1  This was instead accepted by the European Court of Justice in the Kofoed decision (Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795), where the ECJ stated that the absence of Danish national measures implementing Art 11(1)(a) of the EU Merger Tax Directive kept such provision without a direct effect, thus excluding that abusive schemes falling under the scope of this Directive could be countered by a Member State.

384  Pasquale Pistone genuine and in substance countered the European legal order.2 This goal indeed puts together all decisions on abuse, including those that apply this principle within taxation and make it flank the traditional tax categories of tax avoidance and evasion. Such structural similarity has a particular importance for tax purposes and, especially, in the context of value-added tax. Before exploring it in detail, a similarly important point in the Emsland-Stärke 1 decision deserves attention: namely, the different way in which the European Commission and the Court of Justice have framed abuse. The European Commission invoked the existence of an unwritten principle of abuse of law in European law from its existence in the legal order of most EU Member States,3 whereas the Court of Justice, since that very decision, has focused on abusive practices, though in fact, has followed the three elements indicated by the European Commission.4 These elements are: (i) the creation of artificial conditions for the transaction (the so-called ‘objective element’); (ii) the possibility of justifying such a transaction on the basis of another advantage (the so-called ‘subjective element’), this time in conflict with the goals of European law; and (iii) the evidence of such a situation (the so-called ‘procedural element’) by the administration of the Member State, to be reversed only in exceptional cases. The reason the Court of Justice has focused on abusive practices rather than on abuse of law is probably due to the need to avoid entering into the sphere of competence of national Courts. In other words, leaving it up to the national Court to ascertain the actual facts, the Court of Justice nevertheless stated abuse as a principle of European law and defined its contour. Being the focus of this chapter on value-added taxation, this principle, as defined in the Emsland-Stärke 1 decision, should now be compared with the position of the Court of Justice in cases that specifically involve such tax. V. The true starting point of the evolution in the field of valueadded taxation: the Direct Cosmetics decision

Our analysis departs from the Direct Cosmetics case,5 decided by the ECJ some twelve years before Emsland-Stärke 1 and so far unduly neglected by commentators. This decision does not actually use the term ‘abuse’, but refers to tax avoidance and evasion. However, being the tax dimension of abuse essentially related to a saving of taxes unduly obtained by the taxpayer,6 it is often related to tax avoidance and/or evasion both in literature and official texts. For such reason, a comparison will be drawn between the two decisions, taking into account the evolution of all such concepts in further tax decisions of the Court of Justice on abuse. Such reconstruction aims at understanding the precise relation between abuse and the two traditional tax categories of tax avoidance and evasion. Like all VAT decisions on this specific issue, the interpretation of the European Court of Justice in Direct Cosmetics also focused on Article 27 of the Sixth VAT Directive, which allows Member States to derogate from the ordinary rules on the determination of 2  Accordingly, this decision reflects a position held by the Court of Justice since the decision in Case 33/74 Van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299, para [13]. 3  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, para [38]. 4  See Emsland-Stärke (n 3 above), paras [52–53]. 5   Joined Cases 138/86 and 139/86 Direct Cosmetics Ltd and Laughtons Photographs Ltd v Commissioners of Customs & Excise [1988] ECR 3937. 6   See further on this, P Pistone, Abuso del diritto ed elusione fiscale (Padua, Cedam 1995) 5.

Abuse of Law in the Context of Indirect Taxation  385 the taxable base (set by Article 11A(1)(a)) for the purpose of countering tax evasion and avoidance. Being a single European concept of tax avoidance needed to keep a uniform tax base for VAT purposes and exclude tax distortions within the Common Market, the ECJ concluded that Member States could apply their national measures to derogate from Article 11A(1)(a) also in the presence of business exercised not with the intention of obtaining a tax advantage, but for commercial reasons.7 The Court added three important points in this decision that are well worth a mention. First, the inherently objective nature of tax avoidance does not require the intention as a condition for its existence, as instead in the case of tax evasion.8 Second, the ECJ here first addressed the discrepancy arising in the text of the Sixth VAT Directive across several official languages and concerning the improper reference to tax fraud and evasion to mean, in fact, what in the technical terminology is commonly referred to as tax avoidance and evasion. This decision successfully overcame the problem—possibly caused by a ‘false friend’ translation with the French text: the apparent equivalents of évasion et fraude fiscales when drafting the original text of the Directive—by looking at the technical features of tax avoidance and using the tools of multilingual interpretation of the Directive. Nevertheless, the limited use of this decision in the following two decades turned this linguistic discrepancy into an evergreen of the entire European tax law, giving rise to possible conceptual uncertainties on tax avoidance, evasion and their relations to abuse that well exceed the boundaries of value-added tax. Third, this decision was among the first to state the need to measure the reaction of the Member States to tax avoidance using the principle of proportionality as the cornerstone of the analysis, thus anticipating a judicial trend that would gradually become a milestone in the Court’s tax case law on abuse. The question now turns into whether and to what extent a structural similarity may be deemed to exist between the Direct Cosmetics and the Emsland-Stärke 1 decisions. An exhaustive answer to such a question is unfortunately not possible in the light of the wording of both decisions. However, both deal with phenomena that are not tolerated by European law, insofar as abusive practices in the field of taxation generally give rise to situations related to tax avoidance or evasion schemes. The Emsland-Stärke 1 decision defines abuse, whereas the Direct Cosmetics decision does not give a clear and precise definition of tax avoidance, despite requiring its uniform concept. Nevertheless, two elements contained in the latter decision give some useful hints to understanding what avoidance should be. First, by making reference to the avoidance of tax on the value added by the dealer to the final consumer, the Court in fact implies a circumvention of the taxable rules that would otherwise apply in respect of a specific transaction. Furthermore, the Court explicitly draws a distinction from tax evasion when it excludes the relevance of the subjective element in respect of an inherently objective phenomenon such as tax avoidance, thus giving a second useful element to indirectly reconstruct the latter concept and coming close to the elements that the Court of Justice would clearly define decades later in the Halifax decision. In other words, after Direct Cosmetics, it was clear that tax avoidance and evasion—at least in the field of value-added taxation—were two different phenomena.   Direct Cosmetics (n 5 above), para [24].   Direct Cosmetics (n 5 above), paras [21–22].

7 8

386  Pasquale Pistone VI. The path towards the Halifax decision

The path of tax avoidance and evasion in the case law of the European Court of Justice between the Direct Cosmetics and the Halifax decisions is nevertheless a long and intricate one. In 1997 the decision on the Garage Molenheide cases gave the ECJ one more occasion to draw a dividing line between tax avoidance and four different situations giving rise to tax evasion,9 though an unfortunate linguistic mismatch does not allow us to duly appreciate such nuance in the English wording of the decision.10 Besides the lack of clarity, this decision marked in my opinion an important step forward in the evolution of the case law of the Court of Justice on value-added tax, according to which the principle of proportionality applies to both tax avoidance and evasion despite their different structural features,11 whereas the latter phenomenon contains elements that can, in fact, turn into forms of fraud. Roughly during the same period, the European Court of Justice analysed problems of tax avoidance and evasion in the field of direct taxation (ICI) and company law (Centros). Despite not being VAT cases, such decisions are relevant for the purposes of this chapter, considering their importance in defining abuse. In particular, the ICI decision is well known for having introduced into European tax law a notion of tax avoidance based on wholly artificial transactions set up to circumvent the liability to tax.12 This definition in fact reflects the position taken by the ECJ in the Centros13 decision on company law, where it followed a similar reasoning with a different wording, which explicitly used the terms ‘abusive’ and ‘fraudulent’ sometimes together and other times separated by an ‘or’. Despite being clear that European law may not be invoked for ‘abusive’ or ‘fraudulent’ purposes, it is not likewise clear whether or not the two expressions may be regarded as synonyms. This chapter suggests that the two expressions should in principle indicate two close, but different, phenomena in the field of taxation, whereas abuse should involve cases of tax avoidance and fraud may occur in some cases of tax evasion. However, the arguments and wording used by the Court of Justice in its tax case law make this conclusion far from being undisputable.

    9  Joined Cases C-286/94, C-340/95, C-401/95 and C-47/96 Garage Molenheide BVBA (C-286/94), Peter Schepens (C-340/95), Bureau Rik Decan-Business Research & Development NV (BRD) (C-401/95) and Sanders BVBA (C-47/96) v Belgische Staat [1997] ECR I-7281. In particular, such situations involved a failure to comply with reporting obligations (Garage Molenheide), a carousel fraud (Schepens), a litigation on the entitlement to refund of a VAT credit (Decan) and a purchase without invoice (Sanders). 10  See Garage Molenheide et al (n 9 above), para [22], which recorded ‘serious grounds for presumption related in particular to a type of fraud known as circular sales, not involving evasion of VAT but creating fictitious VAT excesses, in particular on intra-Community transactions’. The French version instead reported the three terms underlined above respectively as fraude, vente de carrousel and éluder, whereas the term fraude is the French equivalent of the English ‘evasion’ and the term éluder clearly indicates ‘avoidance’ and not ‘evasion’. 11  Accordingly, this decision extended to VAT the position reached by the ECJ a few months earlier on direct taxes in the decision in Case C-28/95 A Leur-Bloem v Inspecteur der Belastingdienst/Ondernemingen Amsterdam 2 [1997] ECR I-4161, paras [44–45]. 12  Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I-4695, para [26]. 13  Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459.

Abuse of Law in the Context of Indirect Taxation  387 VII. The milestone of abuse in value-added taxation: the Halifax decision and the definition of abuse for value-added tax purposes

The decisions of the ECJ on abuse in the field of value-added and direct taxation kept mixing up tax avoidance, evasion, abuse and fraud with recurrent linguistic inconsistencies for almost a decade until a major breakthrough occurred in the Halifax decision. Most experts regard it as the landmark decision on abuse for value-added tax purposes and do so for at least three good reasons. First, in this decision the European Court of Justice gave a definition of abuse in the field of value-added taxation, which characterises abuse as a circumvention of taxing rules through transactions essentially driven by tax reasons. Second, it confirmed the position already taken by the Court in Direct Cosmetics on tax avoidance almost two decades before, by saying that the existence of objective factors was sufficient for detecting the existence of abusive practices. Third, based on a technically correct application of the principle of proportionality, the Court excluded that a reaction to such phenomenon could legitimately exceed the reestablishment of the situation that would have otherwise arisen in the absence of the abusive practices, thus once more confirming the less harmful nature of abusive practices than that of schemes of tax evasion and fraud. These three elements are, in fact, the milestones of the definition of abuse in the sphere of value-added taxation and lead us to raise three questions, which partly guide us in answering some of the basic issues of this chapter. First, does such definition of abuse reflect the structural elements of the general prohibition of abuse within European law? Second, may there be different definitions of abuse within taxation? Third, does this definition of abuse in fact confine abuse within the traditional tax category of tax avoidance? The answer to the latter question seems simpler than the answers to the remaining two, possibly because the definition of abuse pivots on the same elements that normally characterise tax avoidance. However, the certainty of an affirmative answer to such a question could fade away because of the linguistic discrepancies and the wording of other tax decisions of the Court of Justice, such as the Cadbury Schweppes case. Although this decision always refers to abusive practices without actually using the terms ‘tax avoidance’ or ‘tax evasion’, the examples brought for such practices within the text of the decision may give the impression that the Court considers both phenomena indistinctively. Such is the case, for instance, in the reference to ‘letter-box companies’,14 which are mere appearances and as such indicate—at least for most legal systems—the existence of tax evasion rather than avoidance. Furthermore, Halifax considers as abusive transactions those essentially aimed at the circumvention of taxing rules, whereas the Cadbury Schweppes decision considers abusive practices as ‘wholly artificial arrangements’, thus following instead the traditional formulation used in direct taxation since the ICI decision. Are such elements sufficient to justify a different notion of abuse in two different domains of taxation, whereas in VAT such phenomenon would be equated with tax avoidance and elsewhere remain at a more 14  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995, para [68].

388  Pasquale Pistone general level together with tax evasion and fraud? The problem no longer concerns the use of the different expressions ‘essentially’ and ‘wholly’, since the Court reconciled this point in the Part Service decision, by stating that the national court may take into account the wholly artificial nature of the transactions to reach the conclusion that their essential aim is to obtain a tax advantage.15 By contrast, a more important element should be taken into account. Certainly in VAT, the dividing line between abuse and tax avoidance, on the one hand, and fraud and tax evasion, on the other hand, was further strengthened in the case law of the Court of Justice, as various decisions on carousel fraud may confirm.16 The vulnerability of the VAT intra-European transitional regime has shown the importance of differentiating between such highly harmful techniques and the ones which are also not tolerated, for but merely circumventing the taxable event. By contrast, a corresponding flaw may not be recorded in the field of direct taxation. For such reason, other decisions on abuse in the field of direct taxation—as in, for instance, Kofoed—simply put together the words ‘improperly’ (which in fact is a synonym of abusive) and ‘fraudulently’, and refer to both as abusive practices without further differentiating between them. Surprisingly though, both this and the Cadbury Schweppes decision include an explicit reference to abuse as defined in the Halifax decision. Such reference should not tempt us to invoke Halifax as a general precedent for two direct tax decisions on abuse. Making reference to one or more specific paragraphs of a decision, especially when involving a different context, may imply the possibility to invoke the principles expressed in it, though not without adapting it to the context in which it is applied. For our purposes this could mean that abuse, as defined in value-added tax, could influence a broader area of direct taxation, which includes tax avoidance and evasion, but does not mean that both latter phenomena could be expressions of abuse for value-added tax purposes. Furthermore, another important point arising from the tax case law of the Court of Justice should be taken into account for our purposes. The Court of Justice frequently makes reference to paragraphs and decisions that in fact present some discrepancies from the one that contains such reference. Although one may in principle support the view that this judicial practice aims at streamlining possible conflicts among the various decisions, it in fact increases the difficulties of national courts to reconstruct the exact boundaries of an interpretation given by the Court of Justice and often forces them to refer cases that in a clearly-defined context could otherwise be decided by the same national court. A third element could then be represented by the fact that the measures at stake in the Cadbury Schweppes case do, in fact, have cases of tax avoidance as their object. All such elements support the conclusion that abuse has a common core applicable throughout different areas of taxation, which nevertheless allows for some differentiation in function of the different factual context and needs arising respectively in direct and indirect taxation. In particular, being the core of abuse related to the application of a statute in a situation not explicitly provided by it and irreconcilable with its rationale  Case C-425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I-897, para [62].  See among others, Joined Cases C-354/03, C-355/03 and C-484/03 Optigen Ltd (C-354/03), Fulcrum Electronics Ltd (C-355/03) and Bond House Systems Ltd (C-484/03) v Commissioners of Customs & Excise [2006] ECR I-483; Joined Cases C-439/04 and C-440/04 Axel Kittel v Belgian State (C-439/04) and Belgian State v Recolta Recycling SPRL (C-440/04) [2006] ECR I-6161. 15 16

Abuse of Law in the Context of Indirect Taxation  389 and very spirit, abusive practices should match in all cases the ones that give rise to tax avoidance. Besides this common core, the reconstruction of abuse in the field of taxation based on the case law of the Court of Justice should currently exclude it as an equivalent to evasion and fraud in the field of value-added tax, whereas the same conclusion could perhaps be uncertain from the perspective of direct taxation. However, from a truly dogmatic perspective I wonder whether such a conclusion should not change in the future. Abuse in the field of taxation arises when the taxpayer seeks to apply statutes in atypical situations giving rise to tax saving, whereas fraud is intrinsically linked to evidence of the taxpayer’s behaviour to create an artificial situation and evasion likewise does, though with a minor intensity in the subjective element. Accordingly—and hereby the conclusion on the third question previously raised—I believe that abusive practices should be fully confined within the traditional tax category of tax avoidance: not only for value-added tax purposes (as it happens today), but also in the field of direct taxation, there not being any reasonable ground for the differentiation arising in the current scenario. Should any difference arise in this context, then it may be due to the application of the same concept of abuse in a different factual context, such as, from the perspective of European law, that of different tax rules, or of different taxes. However, a complete answer to the second question should take into account additional factors that could hinder the idea that a common concept of abuse within European law achieves a uniform dimension of abuse within Europe, at least in the field of taxation. Previously in this chapter I have mentioned that the concept of abuse may be influenced by whether or not a common policy exists in respect of a specific field of law. Accordingly, abuse could be confined within the very same dimension in Emsland-Stärke 1 and Halifax despite the diversity of their object (agricultural export restitutions and value-added tax, respectively) as both cases deal with domains that are objects of a common policy.

VIII. Abuse: deviation upon a theme? Comparing the Halifax and Kofoed decisions

If we compare instead the Halifax and Kofoed decisions, we find that discrepancies arise not only as a consequence of the reference to fraudulent situations by the Kofoed decision, but also due to some additional elements which are incompatible with the reasoning of the ECJ in the Halifax decision. In particular, after making reference to abuse as in Halifax, the Kofoed decision accepts that it is for the Member States to decide whether or not to include anti-abuse provisions that may limit the application of the Merger Directive. This automatically shows that, in the absence of such provision, the principle of abuse in European law could not be invoked for the purpose of countering practices that in fact abuse provisions set by secondary law. Accordingly, there seems to be a common core concept of abuse developed in respect of tax avoidance schemes within value-added taxation, whose scope in direct taxation is on the one hand broader, because it has so far usually included tax evasion and fraud arising in cases of national anti-abuse measures that affect European fundamental freedoms, and on the other hand narrower, because it keeps abusive practices out when affecting secondary law in domains that are not objects of a common policy and the Member State has not made use of the right to introduce anti-abuse measures, as was the case for Denmark in the Kofoed case on the Merger Directive.

390  Pasquale Pistone In other words, a common core concept of abuse in European law exists even if European tax rules (namely those contained in the EU Merger Tax Directive on direct taxation) keep the right of Member States to introduce anti-abuse measures in respect of cases falling within their scope. Nor would problems arise in reconciling such a conclusion with the need for the unitary application of European tax rules in the field of valueadded tax, which was supported by the European Court of Justice when interpreting abuse in the Halifax decision. The difference between these two situations, both of which involve secondary law, lies in the difference between its function in direct taxation and value-added tax. Common rules on direct taxation are still merely sporadic and aimed at removing specific obstacles to the functioning of the Common Market, whereas a systematic harmonisation of value-added tax is the outcome of a common policy, thus requiring its uniform application. However, a question arises from a general policy perspective: does it make sense that, in the actual pattern, we counter abuse on the basis of an almost common concept at the European level, almost homogeneously applied to secondary law in the field of VAT and primary law in respect of direct taxes, while carving out secondary law in the latter field? Is this really what Article 11(1)(a) of the Merger Directive17 and Article 1.2 of the ParentSubsidiary Directive18 really meant to achieve? Although an answer to this question only incidentally falls within the scope of this chapter, I submit that when the two Directives on direct taxation (ie 90/434/EEC, also known as the Merger Directive,19 and 90/435/ EEC, also known as the Parent-Subsidiary Directive20) were conceived, back in the late 1960s and early 1970s there was no possible idea of developing a common principle of abuse in European law, and especially no idea that, even if such a concept was starting to develop, that it could have affected a domain like taxation—especially direct taxation.21 I suspect that instead the only function of such carve-out provisions was to avoid that possible issues of legal interpretation based on the predominance of European law on the basis of the lex superior argument could deprive Member States of their right to counter situations in which European law itself was an instrument to carry out abusive practices. After all, one may wonder whether it really makes sense to allow Member States in general terms to counter abusive practices on the basis of a general principle of  The English version of Art 11(1)(a) of the Merger Directive (n 19 below) reads: A Member State may refuse to apply or withdraw the benefit of all or any part of the provisions of Titles II, III, IV and IVb, where it appears that the merger: (a) Has as its principal objective or as one of its principal objectives tax evasion or tax avoidance; the fact that one of the operations referred to in Article 1 is not carried out for valid commercial reasons such as the restructuring or rationalization of the activities of the companies participating in the operation may constitute a presumption that the operation has tax evasion or tax avoidance as its principal objective or as one of its principal objectives. 18  The English version of Art 1(2) of the Parent-Subsidiary Directive (n 20 below) reads: ‘This Directive shall not preclude the application of domestic or agreement-based provisions required for the prevention of fraud or abuse.’ 19  Council Directive 90/434/EEC of 23 July 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States (Merger Directive) [1990] OJ L225/1–5. 20  Council Directive 90/435/EEC of 23 July 1990 on the common system of taxation applicable in the case of parent companies and subsidiaries of different Member States (Parent-Subsidiary Directive) [1990] OJ L225/6–9. 21  Both Directives were amended (for the Parent-Subsidiary Directive see Directive 2003/123/EC of 22 December 2003 and for the Merger Directive see Directive 2005/19/EC of 17 February 2005), though without affecting the powers of Member States in respect of abusive practices. 17

Abuse of Law in the Context of Indirect Taxation  391 European law, while subordinating the application of such principle to the existence of a specific national measure in the case of directives on direct taxation. An alternative position could be that the European principle of abuse of law may not automatically apply out of domains that are objects of a common European policy. However, should this view be correct, we would end up limiting the application of this principle in the field of direct taxation. Accordingly, we would in fact introduce an element of differentiation between direct taxation and value-added taxation, where no such limit would be allowed and where it is appropriate to conclude that all domesticbased anti-abuse principles and rules, especially after the Halifax decision, could soon become redundant. However, neither of these alternatives is fully acceptable in my view. European law may not be invoked for abusive purposes and—to answer the first question previously raised— therefore, the principle of abuse of law should apply regardless of further circumstances: one single concept of abuse in European law (based on the elements indicated in the Halifax decision and previously described in this chapter) with one single framework of application, regardless of what tax is involved and with the safeguard of possible different factual elements. Nevertheless, until the Court of Justice clarifies its position on the actual dimension of abuse in European tax law, such conclusions may not be drawn by national courts22 because this would imply an actual interpretation of European law, which is by no means admitted under the rules of European procedural law and the strict separation of competences between the Court of Justice and national courts. Especially taking into account the specific technical elements arising within tax law, abuse and fraud are two different phenomena. Although European law may not be relied upon in respect of either of them, fraud—which includes tax evasion—harms the Internal Market in a significantly more severe way (the cases of carousel fraud clearly prove this) and should thus be ascertained with a higher degree of precision, thus excluding any involvement of taxpayers acting in good faith or not intentionally being related to it.

22  This happened instead in several recent decisions of the Tax Chamber of the Italian Corte di Cassazione (Cass, sez trib, 4 April 2008, n 8772; 21 April 2008, n 10257; 17 October 2008, n 25374; and 21 January 2009, n 1465). For some criticism on this see P Pistone, ‘Il divieto di abuso come principio del diritto tributario comunitario e la sua influenza sulla giurisprudenza tributaria nazionale’ in M Guglielmo (ed), Quaderni della Rivista di diritto tributario (Milan, Giuffré 2009) 311. The Grand Chamber of the Corte di Cassazione acknowledged the improper nature of this interpretation by the Tax Chamber and suggested instead (Cass, SSUU civili, 23 December 2008, nn 30055 and 30057) that a general prohibition of abuse could be derived at the interpretative level (following a similar approach to that used by Austrian scholars—see for all W Gassner, Interpretation und Anwendung der Steuergesetze (Vienna, Wirtschaftsverl Orac 1972) 89—and shared by the Austrian Verfassungsgerichtshof, unlike the Austrian Verwaltungsgerichtshof ) by a proper interpretation of the circumvented taxing rule in the light of the ability-to-pay principle, stated by Art 53 of the Italian Constitution. This conclusion could also, however, be challenged insofar as the current evolution of European tax law (see in particular Case C-527/06 RHH Renneberg v Staatssecretaris van Financiën [2008] ECR I-7735, para [64]) acknowledges ability to pay as a principle of European law. Considering that this principle is not regarded by Italian scholars and courts as a fundamental principle of the Constitution (see F Gallo, ‘Ordinamento comunitario e principi costituzionali tributari’ (2006) 2 Rassegna Tributaria 415), its national dimension could not undermine the supremacy of the European law on the basis of the doctrine of controlimiti (F Gallo, this note above). I believe that for such reason the ability-to-pay principle should be interpreted in a way that is consistent with its scope under European law. See on this P Pistone, this note above.

392  Pasquale Pistone IX. The judicial path of abuse in value-added tax after Halifax

Shortly before concluding the previous section of this chapter I submitted that in some cases, national anti-abuse provisions, or equivalent judicial techniques, could soon become redundant, especially in the field of value-added tax. This conclusion represents in my opinion the ideal point of arrival of the evolution started some time ago by the Court of Justice, but is still far from being reached, as some recent decisions may prove. For such purpose I will make reference to the decisions on the Part Service,23 Ampliscientifica-Amplifin24 and Sosnowska25 cases, each of which bring a relevant element to the conclusion and all of which were rendered in 2008. Although the Sosnowska decision is the more recent one among the three cases, I would like to briefly touch upon it now, considering that its contribution to the evolution of abuse has been mainly in providing for an additional element to refine the application of the principle of proportionality, by measuring it not just in terms of the damage caused to a taxpayer, but also in reasonableness, thus excluding that the extension of the statutesof-limitation rule for VAT auditing purposes was legitimate and could be avoided through the payment of a guarantee.26 The Part Service decision, already mentioned in Part VII of this chapter, shows one of the problems faced by civil law courts of last instance in the aftermath of the Halifax decision. If we compare the substance of questions referred by the UK and Italian courts, the core element is clearly similar: considering that VAT is an object of common policy and therefore interpreted in a way that does not necessarily match all other fields of taxation, may abuse be countered in the absence of rules (or equivalent judicial techniques) applicable within this domain? Both national courts showed a remarkably correct attitude: the UK court for not applying to VAT the judicial anti-avoidance techniques developed since the 1980s in the field of direct taxation, and the Italian Supreme Court for referring the case, six months after the Halifax decision, to ascertain whether it would still be allowed to apply characterisation based on civil law within the sphere of VAT. The answer given by the Court of Justice somehow understates the dilemma of the national court, by treating it as one of the problems left to the exclusive competence of the national court. In other words, being that the Court is competent to interpret European law only, it should not interfere with any other factual element, be it actual facts or national law. However, the statement of the European Court of Justice27 in Part Service essentially gives the green light to characterisation based on civil law in countries whose legal system follows such tradition, thus opening up a potential differentiation between two categories of countries—namely, civil and common law countries—that nevertheless have the same common system of value-added taxation and that should therefore be bound by similar rules of interpretation and characterisation, based on the autonomy of European law. Accordingly, a clearly designed concept in the Halifax decision takes on a   Part Service (n 15 above).  Case C-162/07 Ampliscientifica Srl and Amplifin SpA v Ministero dell’Economia e delle Finanze and Agenzia delle Entrate [2008] ECR I-4019. 25  Case C-25/07 Alicja Sosnowska v Dyrektor Izby Skarbowej we Wrocławiu Ośrodek Zamiejscowy w Wałbrzychu [2008] ECR I-5129. 26  Accordingly, the Court of Justice has followed a reasoning already applied respectively in Case C-55/98 Skatteministeriet v Bent Vestergaard [1999] ECR I-7641, and (for guarantee) Case C-9/02 Hughes de Lasteyrie du Saillant v Ministère de l’Économie, des Finances et de l’Industrie [2004] ECR I-2409. 27   Part Service (n 15 above), in particular, see para [62]. 23 24

Abuse of Law in the Context of Indirect Taxation  393 different application in the light of the Part Service decision, despite the absolutely correct position taken by the Italian national court of last instance. Although I fully agree that interpretation of European law should not interfere with how Member States implement the common rules into their respective national legal orders, I nevertheless submit that possible distortions caused by national systems should be kept under the control of the single body competent to interpret European law. The legacy of the Part Service decision proves that, even after Halifax and even in the field of value-added taxation, there are still significant differences among anti-abuse provisions and principles within the European Union. I submit and hope that the future evolution in the case law of the Court of Justice may soon fill these gaps, or at least provide some more appropriate criteria to prevent the further differentiation of—or separate common and civil law approaches to—valueadded taxation systems. The Ampliscientifica-Amplifin decision adds an interesting point to our analysis and goes beyond the object of the case, which was the sphere of payments of VAT within a group of companies. In particular, when we draw a comparison with how abuse was defined in the Halifax decision, we find an interesting reference to ‘wholly artificial arrangements not reflecting the economic reality’ which characterised the wording of abuse in the Cadbury Schweppes decision. Considering that the latter decision in turn contained a reference to abuse as defined in the Halifax decision, one may reasonably come to the conclusion that the Court of Justice is slowly, but effectively, moving in the direction of a common concept of abuse that links up different branches of tax law despite possible differences due to factual elements or context. The more references we see to tax cases in non-tax decisions, the more difficult it will be in future to argue that tax law and taxes in general, thus including value-added tax, are a world apart from other areas of law. There is not and there should not be any such distinction within European law if we really want to remove all barriers—including those caused by taxation and still subject to the unanimity requirement for the deliberations of the European Council— within the Internal Market. Predictably Ampliscientifica-Amplifin won’t be the last word for long. Already pending at the Court of Justice there are two news cases concerning the application of the principle of prohibition of abuse of law within the field of VAT. Interestingly both cases originate in the UK courts. The first is Weald Leasing, a complex case involving a leasing arrangement by a partially exempt body designed with the sole purpose of deferring the payment of irrecoverable VAT.28 The questions referred by the UK High Court focussed upon three aspects: interpretation of the first element of the abuse of law test; meaning and significance of the expression ‘normal commercial operations’ in the context of the test; and re-definition of abusive transactions. The second is RBS Deutschland Holdings, also concerning a leasing arrangement by a financial institution.29 The questions referred in this case concentrate on the application of the two-part abuse test to the said arrangement.

28  Case C-103/09 Reference for a preliminary ruling from Court of Appeal (United Kingdom) made on 13 March 2009—The Comissioners for Her Majesty’s Revenue & Customs v Weald Leasing Limited [2009] OJ C129/8. 29  Case C-277/09 Reference for a preliminary ruling from Court of Session (Scotland), Edinburgh (United Kingdom) made on 21 July 2009—The Comissioners for Her Majesty’s Revenue & Customs v RBS Deutschland Holdings GmbH [2009] OJ C267/30.

27 Abuse of Law in the Context of Indirect Taxation: Why We Need the Subjective Intention Test, When is Combating Abuse an Obligation and Other Comments Dennis Weber I. introduction

A.  Points of Assumption

A

ccording to the case law of the Court of Justice, Member States have the possibility under EU law to combat abuse. In addition to the specific conditions in various areas of law with which the combating of abuse will have to comply, it should be recalled that there are a number of general preconditions to the combating of abuse. First, I should like to refer to Kefalas1 in which the Court stated that: [T]he application of such a national rule must not prejudice the full effect and uniform application of Community law in the Member States . . . In particular, it is not open to national courts, when assessing the exercise of a right arising from a provision of Community law, to alter the scope of that provision or to compromise the objectives pursued by it.

In addition, it is important to recognise that the principle of legal certainty is the starting point in the application of the rule of law. This principle entails that rules of law must be made known and that situations of law governed by EU law must be foreseeable. One consequence of this is that it must be possible for a taxpayer to choose the fiscal means most favourable to him on the basis of the EU law known to him. In the value-added tax (VAT) case law, the Court has acknowledged the acceptability of choosing the most favourable fiscal means a number of times, as in Halifax:2 Moreover, it is clear from the case-law that a trader’s choice between exempt transactions and taxable transactions may be based on a range of factors, including tax considerations relating to the VAT system (see, in particular, BLP Group, paragraph 26, and Case C-108/99 Cantor Fitzgerald International [2001] ECR I-7257, paragraph 33). Where the taxable person chooses one of two transactions, the Sixth Directive does not require him to choose the one which involves 1  Case C-367/96 Alexandros Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I-2843, para [22]. 2  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609.

396  Dennis Weber paying the highest amount of VAT. On the contrary, as the Advocate General observed in point 85 of his Opinion, taxpayers may choose to structure their business so as to limit their tax liability.3

The principle of legal certainty entails that if a taxpayer can rely on (ie can foresee) the fact that if he is eligible for a certain tax advantage, he can make use of it. The principle of legal certainty can only be set aside if there is a matter of abuse. In this circumstance, the abuser, namely, can no longer rely on the foreseeability of the law.

II. Objective and subjective tests

A. General It follows from Emsland-Stärke4 that abuse of EU law must be established on the basis of: i) the objective circumstances from which it appears that the envisioned objective of EU law cannot be attained (objective test); and ii) the subjective abuse intention (subjective test). i.  Objective Test In the Emsland-Stärke judgment, with respect to the objective test for abuse, the Court required ‘a combination of objective circumstances in which, despite formal observance of the conditions laid down by the Community rules, the purpose of those rules has not been achieved’.5 Under the objective test, there should be awareness that in principle— on the basis of the letter of the law—a taxpayer has a right to favourable treatment. If, however, the objective circumstances are assessed in the light of the objective of the applicable EU law, then the conclusion can be reached that the envisioned objective has not been attained on account of the rule. In such a situation, this could constitute abuse, and this is often described by the Court such that the objective circumstances do not mean ‘bona fide commercial transactions’, or ‘create artificial conditions’ or ‘wholly artificial arrangements’. Where there is, in fact, a matter of abuse, it must also be established that there has been a subjective abuse intention (see hereafter). The objective test, whereby the objective circumstances must be qualified in the light of the objective of EU law, are in fact basically an interpretation of EU law, for which in the first instance, the Court has jurisdiction. In this framework, the national court must put preliminary questions to the Court about the objective of the relevant EU law.6

  Halifax (n 2 above), para [73].  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 5   Emsland-Stärke (n 4 above), para [52]. 6  See also A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of the European Court of Justice on Circumvention, Fraud and Other Misuses of Community Law’ (2000) 11 European Business Law Review 179, 190; and DM Weber, Tax Avoidance and the EC Treaty Freedoms: A Study on the Limitations under European Law to the Prevention of Tax Avoidance (Eucotax Series on European Taxation), Vol 11 (The Hague, Kluwer Law International 2005) 186. 3 4

Abuse of Law in the Context of Indirect Taxation: Comments  397 ii.  Subjective Test The Court describes the subjective test as ‘the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it’.7 In direct taxation cases, the Court mentions ‘wholly artificial arrangements’.8 If an interested party does not openly disclose that he is trying to avoid the national legislation via invoking EU law, his actual intention will have to be derived from the objective circumstances (the objectified intention).9 The artificiality of a transaction would thus have to be able to be redirected to a subjective abuse intention of an interested party. Accordingly, it is often the case that when applying both the objective test and the subjective test, the objective circumstances will have to be examined. The result is that these two tests can become somewhat intertwined and difficult to separate from each other.10 The Court has applied the Emsland-Stärke test in VAT cases,11 and under the Capital Tax Directive,12 and has recalled this case law in Commission v Greece.13 In his chapter, Professor Pistone14 wonders whether a difference exists (or should exist) between the combating of abuse within tax law and outside tax law. Should abuse within tax law be combated on the basis of the general pattern (the Emsland-Stärke test) or must tax law deviate from this? I am of the opinion that the Court must follow in general the EmslandStärke test and that in light of the specific objective and the specific scope of European law in the various areas of law, the Court must define in further detail the general criteria from Emsland-Stärke on this. In tax cases in the area of direct tax law—consider here ICI, Hoechst, LankhorstHohorst, X & Y, De Lasteyrie, Cadbury Schweppes and Test Claimants in the Thin Cap Group Litigation—the Court permitted in abstracto that ‘wholly artificial arrangements’ were to be combated.15 In Cadbury Schweppes a link was made in this framework to Emsland-Stärke and Halifax. It must be remarked, however, that the subjective test for the VAT in cases such as Halifax and more clearly in Part Service16 is more broadly interpreted by the Court whereby not only is there a matter of abuse in wholly artificial arrangements, but also when ‘it is apparent from a number of objective factors that the essential aim   Emsland-Stärke (n 4 above), para [53].   Cf AG Tesauro in Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I-4695.     9   Cf AG Poiares Maduro in Cases C-255/02, C-419/02 and C-223/03 Halifax plc et al, BUPA Hospitals Ltd et al, and University of Huddersfield v Commissioners of Customs & Excise [2006] ECR I-1609, para [70]. 10  Thus under both tests, the Court often mentions the ‘artificiality’ of a construction. 11   Halifax (n 2 above). 12   Directive 69/335/EEC of the Council of 17 July 1969 concerning the indirect taxes on the raising of capital [1969] OJ L249/25. 13  Case C-178/05 Commission v Greece [2007] ECR I-4185. 14   P Pistone, ‘Abuse of Law in the Context of Indirect Taxation: From (Before) Emsland-Stärke 1 to Halifax (and Beyond)’, ch 26 above. See Part II of that chapter. 15   ICI (n 8 above); Joined Cases C-397/98 and C-410/98 Metallgesellschaft Ltd et al (C-397/98), Hoechst AG and Hoechst (UK) Ltd (C-410/98) v Commissioners of Inland Revenue and Her Majesty’s Attorney General [2001] ECR I-1727; Case C-324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I-11779; Case C-436/00 X & Y v Riksskatteverket [2002] ECR I-10829; Case C-9/02 Hughes de Lasteyrie du Saillant v Ministère de l’Économie, des Finances et de l’Industrie [2004] ECR I-2409; Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995; Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107. 16  Case C-425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I-897.     7     8

398  Dennis Weber of the transactions concerned is to obtain a tax advantage’ (emphasis added). It can be derived from Part Service that if a tax benefit was the essential aim for a transaction, this subjective criterion has also been satisfied if in addition, non-fiscal economic objects have played a role (eg in Part Service, the ECJ mentioned marketing, organisation and guarantee). It thus does not necessarily have to be so that there is only a matter of abuse if exclusively the obtaining of a tax benefit is envisioned with a certain transaction. In VAT, it must concern the ‘essential aim’, which strongly resembles the ‘principal objective’. By whatever means, this description of abuse gives the Member States more scope in the combating of abuse, particularly in situations in which the ultimate objective of a certain transaction in itself is non-fiscal, but the route thereto was inspired particularly by fiscal purposes. Although both in VAT case law and in direct taxation case law17 the Court explicitly demands the subjective test,18 doubts have been raised in the literature19 and by the Advocates-General20 about the usefulness of the subjective test. Professor Lang21 also resists the subjective test. Professor Lang is of the opinion that an interpretation of objective and scope of a provision is sufficient in order to combat abuse.22 I hold another view: it must be considered that in abuse situations, the point of assumption is that an interested party according to the letter of the law, thus according to the ‘formal conditions’, is entitled to a certain favourable tax treatment. If the wording of the law is clear, we do not need a teleological interpretation. When, however, there is abuse of law, it is permitted that a certain case be decided contrary to the wording of the law.23 In light of the fact that normally, an interested party must be able to rely on the principle of legal certainty (and thus on the letter of the law), such an opportunity must be handled with extreme caution. In other words, that it can be established that a certain treatment is in conflict with the objective and scope of a provision is not yet sufficient to come to the conclusion that abuse is present. A law can be badly formulated and allow certain (tax) advantages which, in light of the objective of the legislation, were not the intention. The principle of legal certainty and the principle of legality oppose the setting aside of legislation which is textually clear merely by invoking the fact that the advantage is in conflict with the objective of the legislation. A person who has no subjective intention to abuse can simply make use of such an advantage. This is otherwise, however, when the  See for example, Cadbury Schweppes (n 15 above), paras [63–64].  In the literature, doubts are sometimes raised—incorrectly in my view—as to the existence of the subjective test. See for example, KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 451. 19  See Sørensen (n 18 above) 457; M Lang and S Heidenbauer, ‘Wholly Artificial Arrangements’ in L Hinnekens and P Hinnekens (eds), A Vision of Taxes Within and Outside European Borders: Festschrift in Honour of Prof Dr Frans Vanistendael (The Hague, Kluwer Law International 2008) 507, 608–09. 20  In Case C-23/93 TV 10 v Commissariaat voor de Media [1994] ECR I-4795, AG Lenz defends not applying the subjective test to legal entities given that a legal person cannot develop a subjective attitude. AG Geelhoed remarks in Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607, para [173] that ‘subjective criteria are not useful’. See as a critique on this, D Weber, ‘Abuse of Law—European Court of Justice, 14 December 2000, Case C-110/99, Emsland-Stärke’ (2004) 31 Legal Issues of Economic Integration 51, 52. AG Poiares Maduro endeavours in Halifax (n 2 above), paras [70-71] to explain the subjective test such that it has to concern the artificiality of certain acts or transactions on the basis of objective circumstances and not the subjective intentions of persons. 21  See M Lang, ‘Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective’, ch 30 below, Part II C. 22  See Lang and Heidenbauer (n 19 above) 608–09. 23  See also J Snell, ‘The Notion of and a General Test for Abuse of Rights: Some Normative Reflections’, ch 15 above. 17 18

Abuse of Law in the Context of Indirect Taxation: Comments  399 advantage is in conflict with the objective and scope of the law and the interested party also has a subjective intention to abuse. In such a case, by invoking abuse of law, the principle of legal certainty can be set aside given that a person who has the intention to abuse cannot invoke legal certainty. Only in such a situation can clear law be interpreted with the aid of the teleological method of interpretation. In addition, it must be realised that in practice, the objective and the subjective tests in fact imply a division of the burden of proof. It will have to be in the hands of the Member State (in this case, the tax authorities) to demonstrate that the objective test has been satisfied; if it has been satisfied, then it would appear that there is a suspicion of abuse which can be refuted by the interested party by demonstrating that he had a non-fiscal reason for his transaction.24 III. General principle of EU law

For a long time, it was unclear whether the combating of abuse implied a general principle of EU law.25 For the first time, in Kofoed the Court was clear and mentioned ‘the general Community law principle that abuse of rights is prohibited’.26 The Court referred to various cases in various areas of law in which abuse had been under discussion, such as Centros (freedom of establishment/incorporation conditions for a company), Halifax (VAT), Agip Petroli (carriage by sea within the Member States) and Cadbury Schweppes (right of establishment/direct taxes).27 Accordingly, after this judgment, it was clear that the prohibition on abuse of law was a general principle of Community law in the view of the Court. One important consequence of this would seem to be that if the Member State itself does not provide for a general concept for abuse (such as in the United Kingdom and Denmark), this Member State can still combat abuse on the basis of the prohibition on abuse of law as a general principle of EU law.28 29 This occurred, for example, in Halifax in which the Court gave the English Inland Revenue the possibility to maintain the general Community law principle that abuse of rights is prohibited, whereas a similar national principle is not provided for in the United Kingdom. A second consequence, which has not yet been fully clarified, is the question whether abuse of law as a general principle also implies that Member States have an obligation to combat abuse.30 In my view, there can only be an obligation to combat abuse in cases in which in a certain area, the Member States have assigned their jurisdiction on combating abuse to the Union. In situations 24   Cf PJ Wattel, ‘Nothing Either Good or Bad, But Thinking Makes It So’ in RPC Cornelisse and PJ Wattel (eds), Dat is verder geen probleem: Vriendenbundel Jaap Zwemmer (Amersfoort, SDU Uitgevers 2006) 209, 215–16. 25  See extensively on this R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. 26  Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795, para [38]. 27  Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459; Halifax (n 2 above); Case C-456/04 Agip Petroli SpA v Capitaneria di porto di Siracusa et al [2006] ECR I-3395; Cadbury Schweppes (n 15 above). 28  See also Sørensen (n 18 above) 440. 29   An exception to this is where a directive contains a provision which permits the Member States to combat abuse. On the basis of the fact that directives must be transposed into national law and on the basis of the principle of legal certainty, it is not possible to merely fall back on the Union principle in the case of abuse of law, but the anti-abuse provision from the directive must be transposed into national law, see Kofoed (n 26 above), paras [40–46]. 30  See also Sørensen (n 18 above) 440; Wattel (n 24 above) 219.

400  Dennis Weber in which there is unification or harmonisation, the Member States will often (but not always) assign their rights and obligations (including the right on combating abuse) to the Union, and in doing so, restrict their sovereignty.31 Particularly where in a certain area, a Union policy is in place and the non-combating of abuse should interfere with this policy, an obligation to combat abuse will exist. This obligation to combat abuse will more readily be at issue in the event the proceeds of a certain levy belong to the resources of the EU. The principle of sincere cooperation in the Union would appear to entail this. Such a situation would appear to have arisen in Emsland-Stärke. With respect to the levying of VAT, it is also defendable that there is an obligation to apply the general principle of EU law with respect to abuse of law, given that to a far-reaching degree, VAT has been harmonised and forms part of the general resources of the Union.32 In my view, this is otherwise a case in which EU law is applicable but the adoption of national law still belongs to the sovereignty of the Member States. This occurs, for example, in direct taxation. In such a situation, the Member States have not assigned their sovereignty with respect to the combating of abuse to the Union. There can be no obligation to combat abuse.33 If, however, a Member State does combat abuse (voluntarily) and in doing so, free movement is hindered, the combating of abuse will have to comply with the requirements laid down by the Court.34 IV.  Different levels of abuse

Together with Professor Wattel,35 I am of the opinion that abuse takes place in general at two levels: i) at EU level; and ii) at national level. In the case of abuse at EU level, endeavours are made to make direct use of the EU rules (by, for example, invoking a certain exemption in a directive) which in fact are not intended for that person. Such a form of abuse often occurs in situations in which the law has been made uniform or has been harmonised (consider, for example, VAT). By abuse at national level, endeavours are made to avoid certain national legislation (for example, the non-deductibility of interest) by invoking EU law (for example, the right of establishment). Such a form of abuse occurs, for example, in non-harmonised areas such as direct taxation, where taxpayers devise all kinds of structures in order to avoid the non-deductibility of interest. It is striking that in the course of time, the Court has developed two different formula­ tions in the case law on the basis of which abuse can be combated. The first formulation can be found, amongst others, in Emsland-Stärke. In this, the Court considers that for the question whether there is abuse, an objective test and a subjective test must be satisfied. The Court maintains this formulation each time there is an issue of abuse at EU level.36 In the second formulation, the Court does not refer directly to the objective and the subjective tests, but considers more generally that the Member States may impede abuse. The Court maintains this more general formulation of abuse where there is abuse at   Cf Weber (n 6 above), para [4.3.1].  See also Pistone (n 14 above), Part VIII and Lang (n 21 above), Part III B.  See also Lang (n 21 above), Part III C. 34  See also Weber (n 6 above), para [4.3.1]. 35   Wattel (n 24 above) 212. 36  See for example, for capital tax Commission v Greece (n 13 above). 31 32 33

Abuse of Law in the Context of Indirect Taxation: Comments  401 national level, as for example in Centros.37 In cases concerning direct taxation in which the Member States invoke the combating of tax avoidance as justification for a restriction of the free movement, the Court refers, not directly to the objective and subjective tests, but mentions more generally the combating of ‘wholly artificial arrangements’. With this difference in formulation, the question arises whether the Court draws a distinction between combating abuse at EU level and at national level. In my view, this should make no difference because the core question persists of whether EU law allows the presumed combating of abuse, independent of the question of whether the abuse takes place at EU level or at national level.38 For that matter, we see that the Court refers more and more to cases from various areas of law39 and accordingly, it is already active in bringing the case law more on one line. It is recommended, however, that the Court be more consistent in the formulation of what can be considered abuse.40 An alignment as close as possible with the objective and subjective tests from Emsland-Stärke has my preference, and we see from the doctrine and in the views of the Advocates-General that this description is considered a general guideline.41 V.  Further definition: how specific should an anti-abuse provision be?

Restrictive national rules which are applicable ‘to every situation’ (in other words, measures which do not take into account specific circumstances that indicate abuse) ‘for whatever reason’ (thus without taking account of the subjective intention to abuse) are, according to the Court, too general in order to serve to combat abuse. Their restrictive effect can then also not be justified with the argument that tax avoidance must be combated. Emigration levies at the emigration of a natural person (the Lasteyrie case), levies of capital tax when relocating the corporate seat (an issue in Commission v Greece) and the refusal of an interest deduction because the shareholder is established abroad (the Lankhorst-Hohorst case, which concerned the interest deduction restriction under German thin capitalisation rules), are examples of measures which are too general in order to justify this restrictive effect with the argument that tax avoidance is being combated. It is clear from the case law that anti-abuse measures which are only applicable in the case of tax avoidance (such as the doctrine of fraus legis) are, in principle, permitted under EU law. In recent judgments, the Court has also given the Member States more scope to maintain more general anti-abuse provisions. In Cadbury Schweppes the British Controlled Foreign Companies (CFC) legislation was at issue. In short, this legislation made it possible to   Centros (n 27 above). The ECJ considered in para [24]: It is true that according to the case-law of the Court a Member State is entitled to take measures designed to prevent certain of its nationals from attempting, under cover of the rights created by the Treaty, improperly to circumvent their national legislation or to prevent individuals from improperly or fraudulently taking advantage of provisions of Community law. 38  See also Wattel (n 24 above) 212. 39  In a VAT case to a direct tax case see for example, Case C-162/07 Ampliscientifica Srl and Amplifin SpA v Ministero dell’Economia e delle Finanze and Agenzia delle Entrate [2008] ECR I-4019, para [28] or vice versa Kofoed (n 26 above). 40  See also de la Feria (n 25 above) 439 and Wattel (n 24 above) 216. 41  See AG Stix-Hackl in Case C-452/04 Fidium Finanz AG v Bundesanstalt für Finanzdienstleistungsaufsicht [2006] ECR I-9521, para [95]; AG Poiares Maduro in Case C-311/06 Consiglio Nazionale degli Ingegneri v Ministero della Giustizia, Marco Cavallera (CNDI) [2009] ECR I-415, para [38]; and AG Geelhoed in Akrich (n 20 above), para [96]. 37

402  Dennis Weber levy corporation tax in the United Kingdom on profits that were attained by subsidiaries established outside the United Kingdom, the shares of which were held by British holding companies. The profits of such foreign subsidiaries were attributed to the parent company that was established in the United Kingdom and taxed with corporation tax (whereby the foreign corporation tax was set-off). The CFC legislation is applicable in the event the foreign subsidiary is subject to a ‘lower level of taxation’ elsewhere as compared to the British level of levy. There is a ‘lower level’ if the foreign tax paid amounts to less than seventy-five per cent of the British tax which would have been due on the profit of the subsidiary. This rule is of particular interest for companies which are developing financial activities. There are exceptions to this CFC legislation. One of these exceptions is the ‘motive test’: the taxpayer must demonstrate that a lowering of the British tax was not (one of) the principle reason(s) to create the foreign subsidiary and furthermore, that the effected lowering of British tax was not the consequence of a ‘re-routing’ of profits. In Cadbury Schweppes, the profit of two subsidiaries established in Ireland was attributed to the parent company established in the United Kingdom, Cadbury Schweppes plc. The two Irish subsidiaries performed group financing activities and in Ireland, made use of what is called the International Financial Services Centre (IFSC) regime whereby only ten per cent Irish corporation tax was paid (the rate in the United Kingdom was thirtythree per cent). The Court ruled that the CFC legislation restricted the free movement of establishment, because the CFC legislation was never applicable to domestic companies, nor was it applicable if the company was established in a Member State where the profit was not taxed at a rate that was lower than seventy-five per cent of the British rate. The question subsequently was if the restriction was justified because this was specifically targeted at the combating of wholly artificial constructions. The Court then examined what the objective of the right of establishment was42 and recalled that an establishment implied ‘the actual pursuit of an economic activity through a fixed establishment in that State for an indefinite period’. Subsequently, the Court considered: It follows that, in order for a restriction on the freedom of establishment to be justified on the ground of prevention of abusive practices, the specific objective of such a restriction must be to prevent conduct involving the creation of wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory.43

The Court then went on to examine whether the British CFC legislation could be justified by the combating of wholly artificial arrangements and if this restriction was proportional. In this framework, the Court considered first that the CFC legislation was appropriate to attain this objective. With respect to the necessity of the CFC legislation, the Court found first that there were a number of exceptions to the CFC legislation, but considered that the CFC legislation could only be justified if it were only applicable in the case of an ‘artificial arrangement’ and the taxpayer was given the opportunity to provide   Cadbury Schweppes (n 15 above), para [53]: That objective is to allow a national of a Member State to set up a secondary establishment in another Member State to carry on his activities there and thus assist economic and social interpenetration within the Community in the sphere of activities as self-employed persons (see Case 2/74 Reyners [1974] ECR 631, para [21]). To that end, freedom of establishment is intended to allow a Community national to participate, on a stable and continuing basis, in the economic life of a Member State other than his State of origin and to profit therefrom. 43   Cadbury Schweppes (n 15 above), para [55]. 42

Abuse of Law in the Context of Indirect Taxation: Comments  403 evidence to the argument that the subjective and objective conditions from EmslandStärke and Halifax had not been satisfied on the basis of ‘objective factors which are ascertainable by third parties with regard, in particular, to the extent to which the CFC physically exists in terms of premises, staff and equipment’. It appears from Cadbury Schweppes that the Court is prepared to allow in principle more general measures such as the CFC legislation (such legislation prompts the suspicion that in these situations, there is a matter of abuse), under the condition that the taxpayer is given the opportunity to prove that in his situation, there had been no question of abuse. We see the same in Test Claimants in the Thin Cap Group Litigation (Thin Cap).44 In this case, British thin capitalisation rules were up for discussion whereby interest paid within a group under certain conditions could not be deducted (the deductible interest was then re-qualified to a non-deductible profit distribution). According to the Court, the rules constituted a restriction on the right of establishment because they were only applicable if the lender was established abroad. The Court subsequently recalled that such a restriction could be justified: [T]he specific objective of such a restriction must be to prevent conduct involving the creation of wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory.45

The Court first established that the British thin capitalisation rules were appropriate to attain this objective. The Court then examined whether these were also necessary and considered that the statutory rule with regard to thin capitalisation could be justified: where it provides that interest paid by a resident subsidiary to a non-resident parent company is to be treated as a distribution only if, and in so far as, it exceeds what those companies would have agreed upon on an arm’s-length basis, that is to say, the commercial terms which those parties would have accepted if they had not formed part of the same group of companies.46

The Court went on to consider: The fact that a resident company has been granted a loan by a non‑resident company on terms which do not correspond to those which would have been agreed upon at arm’s length constitutes, for the Member State in which the borrowing company is resident, an objective element which can be independently verified in order to determine whether the transaction in question represents, in whole or in part, a purely artificial arrangement, the essential purpose of which is to circumvent the tax legislation of that Member State. In that regard, the question is whether, had there been an arm’s-length relationship between the companies concerned, the loan would not have been granted or would have been granted for a different amount or at a different rate of interest.47

On the basis of the principle of proportionality, the Court then made two conditions to the thin capitalisation rules: 1. ‘[O]n each occasion on which the existence of such an arrangement cannot be ruled out, the taxpayer is given an opportunity, without being subject to undue administrative constraints, to provide evidence of any commercial justification that there may have been for that arrangement.’   Thin Cap (n 15 above).   Thin Cap (n 15 above), para [74]. 46   Thin Cap (n 15 above), para [80]. 47   Thin Cap (n 15 above), para [81]. 44 45

404  Dennis Weber 2. ‘[T]hat, where the consideration of those elements leads to the conclusion that the transaction in question represents a purely artificial arrangement without any underlying commercial justification, the re-characterisation of interest paid as a distribution is limited to the proportion of that interest which exceeds what would have been agreed had the relationship between the parties or between those parties and a third party been one at arm’s length.’48 It also thus appears from Thin Cap that in itself, the Court is allowing more general anti-abuse measures, such as an at arm’s-length test as an anti-abuse measure, provided the taxpayer always has the opportunity to demonstrate that the transaction had taken place for commercial reasons. Also here, the ‘arm’s-length’ test prompts the suspicion of abuse, unless the taxpayer can demonstrate that he had no subjective intention of abuse. Although it would appear that the Court always requires that an anti-abuse measure provides that a taxpayer can prove that in his situation there was no question of abuse,49 the Court is not always consistent in this. A recent example is Ampliscientifica. At issue in this case was whether, prior to entering into a ‘fiscal unity’ in VAT, the requirement could be made that a holding period of one to two years must be satisfied. After having referred to the case law concerning the prohibition on abuse of law (Halifax, Cadbury Schweppes, etc), the Court considered that: [A rule] which requires economic operators to demonstrate, by means of a certain continuity in their activities and operations, that recourse to a mechanism to simplify VAT declarations and payments is not simply motivated by the intention to obtain a tax advantage, inter alia by entering a VAT debit or credit in the accounts of the parent company or body, which would have the effect, in the first case, of reducing its taxable income and, in the second, of securing it an immediate tax credit, but is the result of a more long-term economic decision, is not contrary to the principle prohibiting the abuse of rights.50

The Court also considered that the holding period respected the principle of proportionality. In this framework, the Court noted: Conversely, if no time‑limit had been imposed, the effect might have been to permit individual operations to be carried out justifying the ad hoc formation of legal structures. Such national legislation is likely to encourage abuse and evasion, the prevention of which is precisely one of the objectives pursued by Community law.51

In this case, we see that the Court permits a holding period prior to entering into a fiscally favourable arrangement (fiscal unity), given that in doing so, it is ascertained that the arrangement is not abused exclusively for a tax advantage and legal ad hoc structures are avoided. It appears from the judgment that, according to the Court, the ad hoc setting up of legal structures can encourage abuse. In itself, that is not surprising, and all the more so given the fact that in Leur-Bloem52 the Court had already indicated that a merger for a structure for a limited period of time and not on a permanent basis may constitute evidence of tax avoidance (see paragraph [42]). It is however striking that in Ampliscientifica, the Court did not require proof to the contrary for situations in good faith. In Leur-Bloem,   Thin Cap (n 15 above), paras [82–83].  See also Lang (n 21 above), Part II C.   Ampliscientifica (n 39 above), para [30]. 51   Ampliscientifica (n 39 above), para [31]. 52  Case C-28/95 A Leur-Bloem v Inspecteur der Belastingdienst/Ondernemingen Amsterdam 2 [1997] ECR 1-4190. 48 49 50

Abuse of Law in the Context of Indirect Taxation: Comments  405 the Court considered, with respect to a (similar) holding period for a shares merger:53 ‘[I]t is nevertheless possible that a merger by exchange of shares with the aim of creating a specific structure for a limited period of time and not on a permanent basis may have valid commercial reasons.’ The Court subsequently denied, with reference to the principle of proportionality, a general holding period. The Court is not consistent here: on the basis of the prohibition on abuse of law/the principle of proportionality, a general holding period for a shares merger is denied, but a more rigid one to two-year holding period for the fiscal unity in VAT is in fact permitted. VI. Conclusion

The principle of legal certainty is the underlying assumption in the combating of abuse. This principle entails that a taxpayer can rely on (can foresee) the fact that he is eligible for a certain tax advantage he can also make use of. The principle of legal certainty may only be set aside if there is a matter of abuse (see Part I). The abuser, namely, can no longer rely on the foreseeability of the law. In this framework, for the question of whether there is abuse, it is important to institute not only an objective test (establishing conflict with the objective and scope of EU law), but also a subjective test. Under abuse of law, it is permitted that a certain case is decided contrary to the wording of the law. In light of the fact that an interested party should normally be able to rely on the principle of legal certainty (and thus on the letter of the law), such an opportunity must be handled with extreme caution. Only interested parties who have a subjective abuse intention may be refused a certain favourable treatment upon invoking abuse of law (see Part II). The prohibition on abuse of law is a general principle of EU law, and this means that a Member State which does not provide for such a general principle in national law, can invoke, for the combating of abuse, the Union principle. In my view, the assumption is that the Member States, in principle, do not have the obligation to combat abuse, unless the Member States have assigned their jurisdiction in a certain area on combating abuse to the Union. This would particularly be the case where there is EU policy in a certain area and the non-combating of abuse could interfere with the objectives of this policy. The VAT can serve as an example here (see Part III). Abuse occurs at two levels: i) at EU level; and ii) at national level. In the case of abuse at EU level, endeavour is made to make direct use of EU rules (by, for example, invoking a certain exemption from a directive) which in fact, is not designated for that person. This form of abuse often occurs in situations in which the law has been made uniform or has been harmonised (consider, for example, the VAT). In the case of abuse at national level, endeavour is made to avoid certain national legislation (for example, the nondeductibility of interest), by invoking EU law (for example, the right of establishment). This form of abuse occurs, for example, in non-harmonised areas such as direct taxation, where taxpayers devise all kinds of structures in order to avoid (for instance) the nondeductibility of interest. The Court uses different formulations for abuse at EU level and abuse at national level. This, therefore, gives rise to the question whether the Court draws a distinction between the combating of abuse at EU level and the combating of abuse 53   According to the then-prevailing Netherlands legislation, an exchange of shares was only effected when undertakings were amalgamated ‘in the financial and economic sense for the long term in one entity’.

406  Dennis Weber at national level. In my view, there should be no difference, because the core question remains whether EU law permits the assumed combating of abuse, independent of the question of whether the abuse has taken place at EU level or at national level (see Part IV). It is, however, to be recommended that the Court be more consistent in the formulation of what can be considered abuse. Hereby, an alignment as close as possible to the objective and subjective tests from Emsland-Stärke has my preference. Finally, I have also found that the prohibition of abuse of law and the applicability of the principle of proportionality means a refusal of a holding period for a shares merger in direct taxation, whereas a more rigid one to two-year holding period for the fiscal unity in VAT is in fact permitted. In my view, the Court is inconsistent in this and again here, better harmonisation of the case law is desirable (see Part V).

28 Cadbury Schweppes and Abuse from an EU Tax Law Perspective Frans Vanistendael I. INTRODUCTION

T

his paper deals with the concept of abuse in tax law as it has been developed recently in the case law of the Court of Justice. The discussion centres on a few leading tax cases: Halifax1 and Part Service 2 (PS) in the area of value-added tax (VAT); Cadbury Schweppes 3 (CS) and Columbus Container 4 (CC) in the area of fundamental freedoms; and Kofoed 5 in the area of specific anti-abuse provisions in Directives. All these cases were decided within a two-year period between 21 February 2006 and 21 February 2008. The discussion will focus on three questions: (1) is the abuse concept developed in the VAT cases and in Kofoed the same as the abuse concept in CS; (2) is the decision in CC relevant for the abuse concept in CS; and (3) if the answer to both questions is negative, is it conceivable to apply two different concepts of abuse in EU tax law? II.  CADBURY SCHWEPPES

A.  Facts, Questions and Answers in Cadbury Schweppes In a nutshell, the facts in Cadbury Schweppes (CS) were as follows: CS was a UK resident company, parent of the CS group, which consisted of companies in the UK, other Member States, and countries outside the EU. The group included two subsidiaries, Cadbury Schweppes Treasury Services (CSTS) and Cadbury Schweppes Treasury International (CSTI) which were established in the International Financial Services Centre (IFSC) in Dublin, where they were subject to a special tax rate of ten per cent. Both subsidiaries were used to raise finance for the companies of the CS group. 1   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 2   Case C-425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I-897. 3   Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 4   Case C-298/05 Columbus Container Services BVBA & Co v Finanzamt Bielefeld-Innenstadt [2007] ECR I-10451. 5   Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795.

408  Frans Vanistendael Three questions were raised in this case. The first was a very simple one: whether the establishment by a parent company of a subsidiary in another Member State for the avowed purpose of enjoying a more favourable tax regime than the one in effect in the state of its establishment constituted, in itself, an abuse of the freedom of establishment. The second question was whether the UK Controlled Foreign Companies (CFC) legislation hindered the exercise of that freedom. The third was whether the UK legislation was justified as a measure against such an abuse, and in particular, under what specific criteria such a measure would be justified and proportional. The answer of the ECJ to the first question was ‘no’, to the second it was ‘yes’ and to the last question it was ‘no’, unless the measure ‘relates only to wholly artificial arrangements intended to escape the national tax normally payable’.6 This was a decision of major importance in which both the reasoning of the Advocate-General and of the Court itself were clear and fully in line with the concept of the Common Market as defined in Articles 2, 3(1)(c) and 14(1) and (2) of the EC Treaty.7 B.  Question One: Is Moving for the Sole Purpose of Benefiting from a More Favourable Tax Regime an Abuse? In dealing with that question the ECJ reiterated the doctrine of abuse of Community law: ‘nationals of a Member State cannot attempt, under cover of the rights created by the Treaty, improperly to circumvent their national legislation. They must not improperly or fraudulently take advantage of provisions of Community law’.8 This also applies to tax law as indicated in the general European Association of Tax Law Professors (EATLP) report of the 2004 EATLP annual congress in Paris: Yet there is one anti-abuse concept that could be developed to shore up the defences of the Member States and that is the concept of abuse of Community law. This concept has been established in several non-tax cases, but has rather sparingly been mentioned in taxation . . . It is intriguing that the Member States have not tried more often this anti-abuse doctrine, which the ECJ itself has developed, rather than to press the case of their national anti-abuse views.9

The holding of the Court in CS confirms the fundamental doctrine of abuse of EU law. It provides reassurance to national tax administrations that abuse of EU law will not be tolerated as a way to escape the consequences of national tax law. The doctrine of abuse of EU law is not peculiar to tax law, however. It has been developed in earlier non-tax cases, as is amply illustrated by the other contributions in this volume, but apparently the doctrine has found new life because of the problems of cross-border taxation.10 However, after acknowledging the right of the Member States to protect the effectiveness of their national tax laws, the Court in CS immediately stepped in to protect the exercise of the fundamental freedom of establishment: ‘[a]s to the freedom of establishment, the   Cadbury Schweppes (n 3 above), para [75].  This concept is now to be found—in essence—in Articles 3(3) TEU and Articles 3(1)b, 4(2)a, 26(1) and (2) TFEU, which refer to the term ‘Internal Market’ instead.     8   Cadbury Schweppes (n 3 above), para [35].     9   F Vanistendael (ed and general reporter), EU Freedoms and Taxation: General Report on the Fundamental Freedoms and National Sovereignty in the European Union (Amsterdam, IBFD 2006) 212, 213. 10   For the origins and developments of the abuse of EU Law doctrine, see the article by R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395.     6     7

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  409 Court has already held that the fact that a company was established in a Member State for the purpose of benefiting from more favourable legislation does not in itself suffice to constitute an abuse of that freedom’.11 The Court cited the decisions in Centros and Inspire Art.12 In Centros the founders of the company wanted to circumvent restrictions of Danish company law in favour of the more liberal regime of English company law through incorporation in the UK. For tax purposes, however, the Centros case would have had no consequences, because the business of the Danish branch office would have been fully taxable in Denmark, in the same way as if the company had been established in Denmark. In Inspire Art, the Amsterdam Chamber of Commerce attempted, among other things, to impose minimal capital requirements of Dutch company law on a company validly established under English law and active in the Netherlands. In both cases, the ECJ decided that the incorporation under English law was valid in spite of the minimal legal requirements and that the host State was not entitled to add other conditions for the recognition of the English companies. For tax purposes in CS, the ECJ went in the same direction stating that: [I]t is settled case law that any advantage resulting from low taxation to which a subsidiary established in a Member State other than the one in which the parent company was incorporated is subject cannot by itself authorise a Member State to offset that advantage by less favourable treatment of the parent company.13

Hence the Court decided that, faced with regimes of substantially lower taxation in other Member States, an EU Member State cannot take the law into its own hands and compensate through heavier taxation on outbound investments in the home State, tax advantages granted by the host State. Such course of action is prohibited, unless there is abuse in the sense of EU law. The Court confirms this settled case law once more by stating that ‘the mere fact that a resident company establishes . . . a subsidiary in another Member State cannot set up a general presumption of tax evasion and justify a measure which compromises the exercise of a fundamental freedom guaranteed by the Treaty’.14 C.  Question Two: Is the UK CFC Regime Hindering the Exercise of the Right of Establishment? The next question was whether the CFC rules as applied in the UK contained a tax disadvantage so as to constitute an effective barrier to the right of outbound establishment. Prima facie there does not seem to be a problem. The profits of the foreign subsidiary that are taken into consideration for taxation in the UK benefit from a tax credit that is identical to the tax credit available on profits distributed by a domestic subsidiary. Her Majesty’s Commissioner and six other Member States pointed out that the total tax burden on foreign and domestic profits was identical and therefore there was no tax disadvantage and consequently no barrier on the exercise of the outbound right of establishment.15 The responses by the Advocate-General and the Court to this argument   Cadbury Schweppes (n 3 above), para [37].   Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459; Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155. 13   Cadbury Schweppes (n 3 above), para [49]. 14   Cadbury Schweppes (n 3 above), para [50]. 15   Cadbury Schweppes (n 3 above), para [45]. 11 12

410  Frans Vanistendael are most instructive on the opinion of the Court about what Member States can and cannot do in the Internal Market. The Advocate-General makes an in-depth analysis of the impact of the CFC rules on all decisions to exercise the right to outbound establishment and comes to the conclusion that: Even if the legislation at issue were tax neutral compared to a purely domestic situation . . . that would not call into question the existence of unequal treatment and the disadvantage to Cadbury in comparison with the position of a resident company which has established a subsidiary in another Member State which has a less favourable tax regime than that in effect in the International Financial Services Centre. Unlike the UK, I do not see why Cadbury’s situation should not be compared to that of such company. I take the view that the assessment of the compatibility with Community law of the legislation in question must examine all ramifications of that legislation. As we know ‘discrimination’ is defined as the application of different rules to comparable situations or the application of the same rule to different situations. The only question to be asked in order to determine whether different treatment of two situations is discriminatory is therefore whether those two situations are comparable. I take the view that that is the case of Cadbury’s position and that of a resident company which has established a subsidiary in another Member State having a less favourable tax regime than that in effect in the International Financial Services Centre because, in either case, a UK resident company has established a subsidiary in another Member State.16 (Emphasis added.)

In other words, in the opinion of the Advocate-General, the problem is not that the CFC legislation makes it impossible to benefit from the lower tax rate of the Dublin Docks by the combined application of the CFC rules and the UK credit system. The problem is that the CFC legislation dictates in which Member States a subsidiary of a UK parent company is entitled, as long as the profits are not distributed, to benefit from a tax rate lower than the one applicable in the UK, and in which Member States the subsidiary is not entitled to benefit from such lower tax rate. The Court holds that this distinction in tax treatment constitutes a discrimination, because it boils down to a different treatment in a similar situation. The argument that the difference in tax rates between Member States constitutes an objective difference allowing for a treatment differentiated on the basis of the tax rates of the Member States is set aside by the Advocate-General in the following terms: If that argument were to be followed through, that would be tantamount to conceding that a Member State is entitled, without infringing the rules of the Treaty, to choose the other Member States in which its domestic companies may establish subsidiaries with the benefit of the tax regime applicable in the host State . . . such a situation would manifestly lead to a result contrary to the very notion of ‘single market’.17

In its decision the Court does not take on board the elaborate reasoning of the AdvocateGeneral, but it consistently refers to the situation in other Member States, however, in doing so but it does not concentrate on the rate differential between the Dublin Docks and the UK. The basis of its decision is the timing disadvantage of immediate taxation of undistributed profits in Ireland compared to the deferred taxation in the UK, but also in the Member States which do maintain a lower level of taxation in the UK, but not a lower level of taxation than the one in the Dublin Docks:  AG Léger in Cadbury Schweppes (n 3 above), paras [77, 78].  AG Léger in Cadbury Schweppes (n 3 above), para [80].

16 17

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  411 That [timing] difference18 in treatment creates a tax disadvantage for the resident company to which the legislation on CFCs is applicable. Even taking into account . . . the fact referred to by the national court that such resident company does not pay, on the profits of the CFC within the scope of application of that legislation, more tax than that which would have been payable on those profits if they had been made by a subsidiary established in the UK, the fact remains that under such legislation the resident company is taxed on profits of another legal person. That is not the case for a resident company with a subsidiary taxed in the UK, or a subsidiary established outside that Member State which is not subject to a lower level of taxation.19 (Emphasis added.)

The conclusion of the Court is clear: [T]he separate tax treatment under the legislation on CFCs and the resulting disadvantage for resident companies which have a subsidiary subject, in another Member State, to a lower level of taxation are such as to hinder the exercise of the freedom of establishment by such companies, dissuading them from establishing . . . a subsidiary in a Member State in which the latter is subject to such [lower] level of taxation. They therefore constitute a restriction on the freedom of establishment within the meaning of Articles 43 EC and 48 EC [now Articles 49 and 54 TFEU].20 (Emphasis added.)

Although the conclusions of the Advocate-General and the outcome of the decision are exactly the same, there is a difference in the emphasis of their arguments. The AdvocateGeneral does not dwell very long on the vertical comparison between the tax regime in the home State and the host State, because he seems to accept the argument made by Her Majesty’s Government that in that comparison, the ultimate tax burden between the tax regime of the home State and the CFC regime applied to the host State is basically not different and therefore not discriminatory. He rather concentrates on the horizontal comparison between low tax Member States, which fall within the ambit of the CFC regime, and other Member States with higher tax rates, but one that does not exceed the tax rate of the UK, which are not within the scope of the CFC rules. He considers that this fragmentation of the Internal Market between ‘good’ Member States and ‘bad’ Member States is a prohibited discrimination. The Court also refers to the horizontal difference in treatment between Member States with low and high tax rates but attaches more importance on the dissuading effect of the CFC regime as a purely non-discriminatory restriction. D.  Question Three: Is the Hindrance Justified? The Court accepts that CFC regimes may be a suitable instrument to counter abusive tax practices making use of low tax jurisdictions: By providing for the inclusion of the profits of a CFC subject to a very favourable tax regime in the tax base of the resident company, the legislation on CFCs makes it possible to thwart practices which have no purpose other than to escape the tax normally due on the profits generated by activities carried on in the national territory.21

 The timing difference described in the prior para [44].   Cadbury Schweppes (n 3 above), para [45]. 20   Cadbury Schweppes (n 3 above), para [46]. 21   Cadbury Schweppes (n 3 above), para [59]. 18 19

412  Frans Vanistendael The question is then one of proportionality: ‘whether that legislation goes beyond what is necessary to achieve that purpose’.22 Before going into the detailed analysis of the CFC rules in the UK, the Court had already set out the conditions that must be fulfilled by anti-abuse provisions. These conditions are very strict when it comes to evaluating measures that restrict the exercise of the fundamental freedoms: ‘a national measure restricting the freedom of establishment may be justified where it specifically relates to wholly artificial arrangements aimed at circumventing the application of the legislation of the Member State concerned’23 (emphasis added). And further: It is necessary, in assessing the conduct of the taxable person, to take particular account of the objective pursued by the freedom of establishment . . . the concept of establishment within the meaning of the Treaty provisions . . . involves the actual pursuit of an economic activity through a fixed establishment . . . for an indefinite period.24 (Emphasis added.)

The conclusion of the Court does not leave any doubt: It follows that, in order for a restriction on the freedom of establishment to be justified on the ground of prevention of abusive practices, the specific objective of such a restriction must be to prevent conduct involving the creation of wholly artificial arrangements which do not reflect economic reality.25 (Emphasis added.)

In answering the question of the wholly artificial arrangement the Court enters into a detailed analysis of the CFC regime applicable in the UK. After finding that the safe harbour rules do not apply in this case, the Court concentrates on the motive test, which is the means of last resort to save the taxpayer. Under the motive test the taxpayer must submit evidence that: (1) the reduction in tax was not one of the main purposes of the transaction, and that (2) the achievement of the tax reduction by a diversion of profits was not the main reason for incorporating the CFC. Since the intention to obtain tax relief is sufficient to prompt the application of the CFC regime under the motive test, it is clear that prima facie, this test is not based on wholly artificial arrangements. The Court held: In order to find that there is such an arrangement there must be, in addition to a subjective element consisting in the intention to obtain a tax advantage, objective circumstances showing that, despite the formal observance of the conditions laid down by Community law, the objective pursued by the freedom of establishment . . . has not been achieved.26 (Emphasis added.)

The Court refers to a case on the regulation for export refunds for agricultural products27 and a VAT case28 and takes the road of these two cases in order to assess whether the objective circumstances are fulfilled: ‘[t]hat incorporation must correspond with an actual establishment intended to carry on genuine economic activities in the host Member State’ and ‘that finding must be based on objective factors which are ascertainable by third parties with regard . . . to the extent to which the CFC physically exists in terms of premises, staff and equipment’.29   Cadbury Schweppes (n 3 above), para [60].   Cadbury Schweppes (n 3 above), para [51]. 24   Cadbury Schweppes (n 3 above), paras [52, 54]. 25   Cadbury Schweppes (n 3 above), para [55]. 26   Cadbury Schweppes (n 3 above), para [64]. 27   Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 28   Halifax (n 1 above). 29   Cadbury Schweppes (n 3 above), paras [66, 67]. 22 23

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  413 The Court concludes that it is for the national judge to evaluate the motive test and to determine whether: the motive test, as defined by the legislation on CFCs, lends itself to an interpretation which enables the taxation provided for by that legislation to be restricted to wholly artificial arrangements or whether, on the contrary, the criteria on which that test is based mean that, where none of the exceptions . . . applies and the intention to obtain a reduction in UK tax is central to the reasons for incorporating the CFC, the resident parent comes within the scope of application of that legislation, despite the absence of objective evidence such as to indicate the existence of an arrangement of that nature.30

In the first case, the motive test as restricted to wholly artificial arrangements is compatible with the fundamental freedoms and in the second case, it is not. The conclusion is clear when there is no objective evidence of a wholly artificial arrangement: the conclusion should be that there is no abuse even when tax is the main reason for the construction. The way the Court has formulated its holding also clearly indicates that the burden of proof is on the tax administration and not on the taxpayer, who, in the absence of objective evidence of a wholly artificial arrangement, goes scot free. E.  Conclusions About Cadbury Schweppes The conclusions of the Opinion of the Advocate-General and of the decision in CS can be summarised as follows: a. The exercise of the right of secondary establishment to take advantage of a substantially more beneficial tax regime in another Member State does not constitute in itself an abuse. b. The EC Treaty permits Member States to take action preventing abuse of Community law by circumventing national tax legislation. c. Within the Treaty framework, national anti-abuse legislation must specifically target wholly artificial arrangements. d. An establishment in another Member State can never be treated as an abuse if it consists of an effective establishment in order to conduct a genuine economic activity. e. It is for the national court to determine whether the motive test in the CFC legislation is formulated in such a way that its application is limited to wholly artificial arrangements. III.  COLUMBUS CONTAINER

A.  Facts and Questions in Columbus Container Columbus Container (CC) was a limited partnership established under Belgian law and owned by eight individuals resident in Germany, each owning a ten per cent share, and a German partnership owning the remaining twenty per cent. CC fulfils the conditions   Cadbury Schweppes (n 3 above), para [72].

30

414  Frans Vanistendael of a Belgian coordination centre, providing financial services for a multinational group associated with the coordination centre and subject to corporate income tax in Belgium at a substantially preferential level of taxation. Partnerships (foreign and domestic) are treated in Germany as transparent entities and profits are taxed directly to the individual partners. Under the Belgian-German tax treaty, profits from permanent establishments are exempt from income tax in the country of residence of the taxpayer. Hence the profits of the Belgian coordination centre, flowing immediately to the individual partners under the German transparency rule, were not subject to tax in Germany. On 21 December 1993 the Missbrauchsbekämpfungs- und Steuerbereinigungsgesetz (law combating abuse and harmonising taxation) introduced a switch-over provision in the Aussensteuergesetz (foreign tax law), according to which, double taxation on passive income from foreign establishments of resident taxpayers would be eliminated by a foreign tax credit rather than by way of exemption as provided under the tax treaty (where, if that establishment were a separate foreign corporation, the CFC rules would apply). The application of the switchover clause increased considerably the tax burden of the German individual partners. In this case there was one essential question: whether the switch-over from exemption to credit as an anti-abuse measure for passive income was permitted under the freedom of establishment or free movement of capital provision: whether freedom of establishment and the free movement of capital preclude a Member State . . . for the purpose of avoiding double taxation of the income and capital . . . derived from particular investments in another Member State, from unilaterally replacing the exemption method by the set-off method.31

Basically it was the same question as in CS—eg whether an anti-abuse measure consisting in switching from exemption to credit, when the permanent establishment would have been in the same conditions as a subsidiary to which CFC would apply, constitutes an infringement of the right of establishment, except that in CC there was no discussion of the motive test. B.  Opinion of the Advocate-General: Is the Switch-over from Exemption to Credit Allowed? While in the CS case there was a close parallel between the Opinion of the AdvocateGeneral and the decision of the Court, the Opinion and the decision in CC were miles apart. In his analysis, Advocate-General Mengozzi followed closely the reasoning of Advocate-General Léger in CS. In doing so he reiterated a few truths about EU tax law, clearly indicating that the Court of Justice is not always on the side of the taxpayer. He first confirmed that the different tax treatment of partnerships under Belgian and German tax law was not contrary to the fundamental freedoms. Taxation of a partnership as an entity in one Member State and transparent tax treatment of the same partnership in another Member State can perfectly coexist: ‘At the current state of development of Community law, it does not require Member States to recognize . . . the legal and tax status afforded by the domestic law of the other Member States to entities which carry out their economic activities there’.32  AG Mengozzi in Columbus Container (n 4 above), para [4.]  AG Mengozzi in Columbus Container (n 4 above), para [41].

31 32

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  415 Second, he indicated in paragraph [76] of his Opinion that the increase in tax burden on an outbound moving taxpayer is in itself permitted under EC law: [A] difference in treatment under Community law is not measured in the light of a factual or legal change in the situation of one and the same person. It requires a comparison between the situation of persons who have exercised one of the freedoms . . . and that of persons who have not done so.

In making that comparison between domestic investors and outbound investors he stated that there was no disadvantage and therefore no discrimination as ‘neither the referring court . . . nor Columbus have identified any difference in treatment between the situation of the latter’s partners and a domestic situation’.33 Finally, he confirmed that the use of the credit method as an instrument to eliminate double taxation remained compatible with the EC Treaty: ‘The use of methods to avoid double taxation according to the nature of income . . . cannot be criticized per se.’34 From then on, however, the Advocate-General followed closely the reasoning of CS with emphasis on the argument that Member States were not allowed under the Treaty to design their tax rules in such a way that they would result in the fragmentation of the Internal Market. This is backed up by two crucial arguments: The unfavourable tax treatment in the present case does not result purely from the application of the different tax legislations of the Member States, but from the choice made in the German tax legislation to set the mechanism for off-setting tax levied abroad on the income in question . . . where that tax is below . . . 30%.35

And: [A] Member State of residence cannot restrict the freedom of establishment of its nationals to part of the common market . . . Thus is the obligation on the . . . State of residence . . . to ensure in addition to respect for equal treatment among its residents as regards whether they have or have not exercised their freedom of movement, that they are not deterred from establishing themselves in the Member State of their choice, inter alia by means of tax measures.36 (Emphasis added.)

The conclusion of Advocate-General Mengozzi is, not surprisingly, that there is a violation with the freedom of establishment, because the anti-abuse provision is not limited to wholly artificial arrangements. In this case, the economic effectiveness of the establishment was indeed never in doubt. C.  Decision of the Court: Is the Switch-over Allowed? In reaching its decision, the Court follows a path that is completely different from that taken by the Advocate-General. The word ‘abuse’ is not used on one single occasion. The Court did not see the switch-over provision as an anti-abuse measure, but merely as a general instrument to avoid double economic taxation. This results from the rephrasing of the question referred by the Finanzgericht of Münster for a preliminary ruling. While  AG Mengozzi in Columbus Container (n 4 above), para [76].  AG Mengozzi in Columbus Container (n 4 above), para [96] 35  AG Mengozzi in Columbus Container (n 4 above), para [128]. 36  AG Mengozzi in Columbus Container (n 4 above), para [133]. 33 34

416  Frans Vanistendael the Münster tax tribunal still referred to ‘passive income’ and ‘CFC income’37 which have a clear connotation of specific anti-abuse rules that are not part of the general rules of income tax, the Court rephrased the issue as one of purely elimination of double taxation without any indication of abuse: [U]nder which the income of a resident national derived from capital invested in an establishment having its registered office in another Member State is, notwithstanding the existence of a double taxation convention concluded with the Member State in which that establishment has its registered office, not exempted from national income tax, but is subject to national taxation against which the tax paid in the other Member State is set off.38

The holding of the Court at the end of the decision literally copies the same words without any reference to CFC or abuse. The fact that the ECJ did not perceive as an antiabuse measure, the selective switch-over on passive income that would be subject to CFC rules if earned by a separate subsidiary, is the only explanation for the wide divergence between the conclusion of the Advocate-General and the decision of the ECJ. The remainder of the discussion of the CC decision can be very short. Citing KerckhaertMorres,39 the Court did not see any disadvantage for the outbound investors because they were treated exactly in the same way as domestic investors: ‘Since partnerships such as Columbus do not suffer any tax disadvantage in comparison with partnerships established in Germany there is no discrimination.’40 The Court concluded that the coexistence of the two different national tax systems were the root of all evil and that this evil should be taken care of in double tax conventions: ‘[T]he adverse tax consequences which might arise from the application of the system of taxation of profits such as that put in place by the AstG result from the exercise in parallel by two Member States of their fiscal sovereignty’—and further: ‘In this respect double taxation conventions such as those envisaged by Article 293 EC [now repealed after the Treaty of Lisbon] are designed to eliminate or mitigate the negative effects on the functioning of the internal market resulting from the coexistence of national tax systems.’41 (Emphasis added.) That is clear and plain language with respect to the elimination of double taxation, but it does not help us any further with respect to the definition of abuse for the application of a national tax rule that was clearly intended as an anti-abuse measure. D. Does Columbus Container Revoke or Nullify Cadbury Schweppes? The answer to this question can be very short. Apart from the similarity of the facts of both cases, there is no logical connection whatsoever on the legal grounds on which they have been decided. In that sense, CS cannot be invoked as a precedent against CC, and CC cannot be invoked as saying anything about the doctrine of abuse of European law and the concept of abuse in EU tax law. The only conclusion can be that on similar facts we now not only have the parallel exercise of fiscal sovereignty in the legislation of the Member States, but also parallel case law by the Court of Justice, in which sometimes anti-abuse legislation is recognised and dealt with as such, and sometimes it is not.  AG Mengozzi in Columbus Container (n 4 above), para [25].   Columbus Container (n 4 above), para [26]. 39   Case C-513/04 Mark Kerckhaert and Bernadette Morres v Belgian State [2006] ECR 1-10967. 40   Columbus Container (n 4 above), para [40]. 41   Columbus Container (n 4 above), paras [43, 44]. 37 38

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  417 IV. The VAT cases: Halifax and Part Service

A.  Why Discuss VAT Cases for the Application of the Treaty Freedoms? One may wonder: why we should discuss the concept of abuse in VAT cases in order to determine the concept of abuse in indirect tax law under EU law? There are several reasons for this. The doctrine of ‘abuse of EU law’ is part of a wider non-tax doctrine developed by the Court in many areas of European law outside the area of taxation.42 In Halifax the existence of a theory of abuse of rights was discussed in principle, because the taxpayer denied the existence of such principle in the VAT system.43 The Court held that: The application of Community legislation cannot be extended to cover abusive practices by economic operators, that is to say transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law. That principle of prohibiting abusive practices also applies to the sphere of VAT.44

Finally, in CS the Court referred specifically to Emsland-Stärke and Halifax and used wording found in these two cases to define the concept of abuse: In order to find that there is such an arrangement there must be, in addition to a subjective element consisting of the intention to obtain a tax advantage, objective circumstances showing, that despite formal observance of the conditions laid down by Community law, the objective pursued by the freedom of establishment . . . has not been achieved.45

This led me to formulate the question: ‘Halifax and Cadbury Schweppes: one single European theory of abuse in tax law?’46 B.  The Concept of Abuse in Halifax i.  Opinion of the Advocate-General In his very elaborated Opinion, Advocate-General Poiares Maduro confirmed the existence of what he calls the ‘abuse’ doctrine: [T]his notion of abuse operates as a principle governing the interpretation of Community law . . . What appears to be a decisive factor in affirming the existence of an abuse is the teleological scope of the Community rules invoked, which must be defined in order to establish whether the right claimed is, in effect, conferred by such provisions, to the extent to which it does not manifestly fall outside their scope. This explains why the Court often refers not to abuse of rights, but simply to abuse.47 (Emphasis added.)

  See de la Feria (n 10 above).   Halifax (n 1 above), para [62]. 44   Halifax (n 1 above), paras [69, 70]. 45   Cadbury Schweppes (n 3 above), para [64]. 46   F Vanistendael, ‘Halifax and Cadbury Schweppes: One Single European Theory of Abuse in Tax Law?’ (2006) 4 EC Tax Review 192. 47  AG Poiares Maduro in Halifax (n 1 above), para [69]. 42 43

418  Frans Vanistendael He then continues on the question of the applicability of the concept of abuse of Community law in VAT stating: [T]he prohibition of abuse of Community law, seen as a principle of interpretation, does not give rise to derogations from the provisions of the Sixth Directive. The result of its application is that the legal provision interpreted cannot be regarded as conferring the right at issue because the right claimed is manifestly beyond the aims and objectives pursued by the provision abusively relied upon.48 (Emphasis added.)

He also brushes aside the argument that the Community concept of abuse cannot be applied, because VAT is national law: ‘The right to deduct (input tax) is conferred by the Sixth Directive is a Community right whose legal basis is Article 17 of the Sixth Directive and whose content, does not . . . leave the Member States any discretion as regards its implementation.’49 He then goes on to define the Community law concept of abuse: [T]he Community law notion of abuse, applicable to the VAT system, operates on the basis of a test comprising two elements. Both elements must be present in order to establish the existence of an abuse of Community law in this area. The first corresponds to the subjective element . . . but it is subjective only in so far as it aims at ascertaining the purpose of the activities in question. That purpose—which must not be confused with the subjective intention of the participants in those activities—is to be objectively determined on the basis of the absence of any other economic justification for the activity than that of creating the tax advantage . . . when applying it, the national authorities must determine whether the activity at issue has some autonomous basis which, if tax considerations are left aside, is capable of endowing it with some economic justification in the circumstances of the case.50 (Emphasis added.) The second element . . . corresponds to the so-called objective element . . . whereby the purpose and objectives of the Community rules allegedly abused are compared with the purpose and results achieved by the activity at issue. This second element . . . provides the standard upon which the purpose and results of the activity in question are to be assessed. It also provides a safeguard for those instances where the sole purpose of the activity might be to diminish tax liability but where that purpose is actually a result of a choice between different tax regimes that the Community legislature intended to leave open. Therefore, where there is no contradiction between recognition of the claim made by the taxable person and the aims and results pursued by the legal provision invoked, no abuse can be asserted.51

The defining moment in the Advocate-General’s concept of abuse of Community law is in paragraph [89] of his Opinion: The prohibition of abuse, as a principle of interpretation, is no longer relevant where the economic activity carried out may have some explanation other than the mere attainment of tax advantages against the tax authorities. In such circumstances, to interpret a legal provision as not conferring such an advantage on the basis of an unwritten general principle would grant an excessively broad discretion to the tax authorities in deciding which of the purposes of a given transaction ought to be considered predominant.52 (Emphasis added.)

The Advocate-General allows for statutory national anti-abuse provisions which would give a wider discretion to the tax authorities to operate within the national context, but  AG Poiares Maduro in Halifax (n 1 above), para [79].  AG Poiares Maduro in Halifax (n 1 above), para [81]. 50  AG Poiares Maduro in Halifax (n 1 above), para [87]. 51  AG Poiares Maduro in Halifax (n 1 above), para [88]. 52  AG Poiares Maduro in Halifax (n 1 above), para [89]. 48 49

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  419 such discretion cannot be tolerated as an unwritten principle of interpretation in the European context: There can be little doubt that the possibility must be recognized that also in such cases, where activities are accounted for by a mixture of tax and non-tax considerations, further restrictions could be introduced for claims arising from activities, which, to varying extents, predominantly seek to achieve tax advantages. This however, will require the adoption of appropriate national legislative measures. Mere interpretation will not suffice.53 (Emphasis added.)

ii.  Decision by the Court As usual, the decision of the Court is much shorter than the lengthy and learned Opinion of the Advocate-General, but by and large, the Court followed the Advocate-General, although the formulation of its holding is slightly different. The Court confirms the right of the taxpayers to base their decisions on ‘a range of factors, including tax considerations’ and that they ‘may chose to structure their business so as to limit their tax liability’.54 It then goes on to define the concept of abuse on the basis of two elements: [I]n the sphere of VAT, an abusive practice can be found to exist only if first, the transactions concerned, notwithstanding formal application of the conditions laid down by the relevant provisions of the Sixth Directive and the national legislation transposing it, result in the accrual of a tax advantage the grant of which would be contrary to the purpose of those provisions. Second it must also be apparent from a number of objective factors that the essential aim of the transaction concerned is to obtain a tax advantage. As the Advocate-General observed in point 89 of his Opinion, the prohibition of abuse is not relevant where the economic activity carried out may have some explanation other than the attainment of tax advantages.55 (Emphasis added.)

Prima facie this seems to consecrate the ‘defining moment’ of the Opinion of AdvocateGeneral Poiares Maduro. Yet one word makes the difference. The Advocate-General is clearly indicating that there can only be abuse when the tax consideration is the sole consideration for the transaction. Abuse only exists in the absence of any economic consideration or in the absence of any other explanation. The wording used by the Court is that there is abuse when the tax advantage is the essential aim of the transaction. Essential, according to the Oxford English Dictionary, is ‘extremely important’ or ‘fundamental’. Sole, on the other hand, means the ‘one and only’. This seems to indicate a difference in nuance, which in the area of tax abuse may be highly significant. However, one and only does not mean that the taxpayer is entitled to allege any non-tax motive to escape the application of the notion of abuse. There must be objective factors to buttress the non-tax motives. Sole means that the non-tax consideration, although being ascertainable on the basis of objective factors, may not be the predominant reason for the transaction in comparison to the importance of the tax considerations. This clearly follows from the statement by the Advocate-General that in cases where there is a mix of motives, of which tax may be one of the more important factors, the concept of abuse of EU law does not apply. Essential, to the contrary, means that when there are different motives, the tax motive may not be preponderant. That is a position that is irreconcilable with the position of the Advocate-General. However, because the Court refers specifically  AG Poiares Maduro in Halifax (n 1 above), para [90].   Halifax (n 1 above), para [73]. 55   Halifax (n 1 above), paras [74, 75]. 53 54

420  Frans Vanistendael to paragraph [89] of the Advocate-General’s Opinion and it also literally refers to ‘some explanation other than the mere attainment of tax advantages’56 one still could argue that what the Court meant by ‘essential’ was in reality ‘sole or one and only’. C.  The Concept of Abuse in Part Service The concept of abuse in relation to VAT was once more discussed in the case of Part Service. The reasoning of the Court in this decision was rather succinct. The referring court requested guidance on the question whether a finding could be made of an abusive practice ‘when the accrual of a tax advantage is the principal aim of the transaction . . . or if such finding can only be made if the accrual of that tax advantage constitutes the sole aim pursued, to the exclusion of other economic objectives’57 (emphasis added). In other words, the national court was asking whether the essential aim mentioned in the Halifax decision meant the most important aim or the one and only aim of the transaction. The Court referred to paragraphs [74] and [75] of the Halifax decision including the essential aim and the reference to the abuse concept as explained by Advocate-General Poiares Maduro, including the statement that if there was some explanation other than the mere attainment of tax advantages there was no abuse. It even pointed out that the reference in Halifax to the sole purpose of the transaction as obtaining a tax advantage was only intended to describe the factual situation of the case and to guide the national court in reaching its decision. All this reasoning seemed to confirm the narrow interpretation of abuse of Community law as defended by Advocate-General Poiares Maduro. But then suddenly, and without further ado, the Court peremptorily stated that: ‘[T]he Sixth Directive must be interpreted as meaning that there can be a finding of an abusive practice when the accrual of a tax advantage constitutes the principal aim of the transaction or transactions at issue.’58 In answering the second question whether the splitting of a lease contract in a rental contract and a financial contract could be considered as an abusive practice under the Sixth Directive, the ECJ maintained its position. Referring to Halifax (paragraph [81]) it stated: [T]he national court, in the assessment which it must carry out, may take account of the purely artificial nature of the transactions and the links of a legal, economic and/or personal nature between the operators involved . . . those aspects being such as to demonstrate that the accrual of a tax advantage constitutes the principal aim pursued, notwithstanding the possible existence, in addition, of economic objectives arising from, for example, marketing, organization or guarantee considerations.59 (Emphasis added.)

This means that from now on, at least for VAT purposes, the Court considers that there is an abusive practice when the tax advantage in the opinion of the judge is the principal reason for engaging in the transaction. It is not clear on which criteria a judge should evaluate the importance of the tax advantage to the non-tax advantages. It is clear, however, that this concept is not very much different from the abuse concepts to be found   Halifax (n 1 above), para [75].   Part Service (n 2 above), para [40]. 58   Part Service (n 2 above), para [45]. 59   Part Service (n 2 above), para [62]. 56 57

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  421 in many of the national tax legislations and tax doctrines of the Member States. The question is whether this still can be considered a European concept of abuse. V. The Kofoed case

In the discussion about the abuse of EU law, the Kofoed 60 decision takes a special position, because it is neither based on the VAT Directives, nor is it directly based on the Treaty freedoms. It is rather an interpretation of the Merger Directive, and in particular, the anti-abuse provision of the Merger Directive contained in Article 11. The case turned on the question whether a complicated transaction of an exchange of shares followed by a dividend distribution complied with the conditions set in the Tax Merger Directive for such exchange. In particular, the question was whether a dividend distribution to the shareholders should be considered as a payment of cash disqualifying the transaction for a tax exemption under the provisions of the Tax Merger Directive. However, that part of the decision is not relevant for the anti-abuse discussion. It is the second part of the decision which dealt with the doctrine of abuse of Community law. The Court starts its analysis from the text of Article 11 of the Tax Merger Directive which states that: ‘A Member State may refuse to apply or withdraw the benefit of all or any part of the provisions . . . [w]here it appears that the merger . . . a) has as its principal objective or one of its principal objectives tax evasion or tax avoidance’61 (emphasis added). The Court goes then on to say: Thus Article 11(1)a of Directive 90/434 reflects the general Community principle that abuse of rights is prohibited, individuals must not improperly or fraudulently take advantage of provisions of Community law. The application of Community legislation cannot be extended to cover abusive practices, that is to say, transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law.62 (Emphasis added.)

This type of reasoning is illustrative for the rather loose and very flexible use of concepts which the Court has been following in its anti-abuse case law. There should be a clear distinction between: (1) an anti-abuse doctrine prohibiting tax advantages when tax considerations were clearly the only way to explain the taxpayer’s legal behaviour and there was no other credible explanation based on objectively ascertainable facts; and (2) an anti-abuse doctrine in which tax considerations may be the principal or one of the principal reasons for explaining the taxpayer’s behaviour. In Kofoed the Court does not seem to make this distinction, because it considers the rule of ‘principal’ or ‘one of the principal’ tax reasons as formulated in Article 11 of the Tax Merger Directive as the expression of the ‘one and only’ doctrine of Cadbury Schweppes. For the implementation of the anti-abuse provision under Article 11 of the Directive, the Court then gives a very flexible interpretation of how this article could be applied under Danish national law, even when the national legislation did not specifically incorporate this rule in a formal national statute:   Kofoed (n 5 above).  Art 11, Council Directive 90/434 EEC of 23 July 1990 on the common system of taxation applicable to mergers . . . as amended by Council Directive 2005/19/EC of 17 February 2005 [2005] OJ L58/19. 62   Kofoed (n 5 above), para [38], followed by a specific reference to Cadbury Schweppes (n 3 above), para [35]. 60 61

422  Frans Vanistendael Although it is true that the requirement of a directive-compliant interpretation cannot reach the point where a directive, by itself and without national implementing legislation, may create obligations for individuals or determine or aggravate the liability in criminal law of persons who act in contravention of its provisions, a Member State may nevertheless . . . impose a directivecompliant interpretation of national law on individuals . . . It is therefore for the national court to ascertain whether there is, in Danish law, a provision or general principle prohibiting abuse of rights or other provisions on tax evasion or tax avoidance, which might be interpreted in accordance with Article 11(1)a of Directive 90/434 and thereby justify taxation of the exchange of shares in question . . . if so, it will be for the national court to determine whether the conditions for the application of those provisions are satisfied in the main proceedings.63 (Emphasis added.)

This is ‘le monde à l’envers’. Instead of controlling whether the way anti-abuse doctrine applied in matters of EU law conforms with the standards set by the Court in its own milestone decisions like Cadbury Schweppes, there is a delegation of decision-making power to the national judge who, if he finds a doctrine (even in non-statutory form) of abuse applied in national case law, is invited to apply that doctrine in accordance with the loose interpretation which the Court has been giving to that doctrine. There is not even the proviso in the Kofoed decision, that that national doctrine should be in conformity with the Court’s case law. VI. Is there more than one EU concept of abuse?

By now it should be clear to anyone with enough courage to continue reading this chapter that the concept of abuse as it has evolved in Kofoed and Part Service is miles apart from the same concept as developed in Cadbury Schweppes. Yet both concepts claim to be the EU concept of abuse. Since Kofoed and Part Service, the small crack that was already visible in Halifax has widened into a chasm. The difference is that under the CS ruling, any other reason than a tax advantage is sufficient to keep at bay the abuse concept, while under the PS holding there may be a mix of tax and non-tax reasons, but if the tax reasons are more important or even one of the important reasons, abuse is established. In dealing with this problem we have to keep in mind that the EU concept of abuse also applies outside the area of taxation. The non-tax example to keep in mind is the Centros decision64 in which the ECJ decided that the mere registration of a private limited company under English law was sufficient to legitimise its legal existence and capacity to do business in the country where it maintained a branch office: [T]he fact that a national of a Member State who wishes to set up a company chooses to form it in the Member State whose rules of company law seem to him the least restrictive and to set up branches in other Member States cannot, in itself, constitute an abuse of the right of establishment.65

That holding is compatible with the decision in CS, but is incompatible with the one in PS. The question then is: is there room for more than one notion of abuse under ‘European law’? I would dare to argue, yes, maybe there is room for more than one concept of   Kofoed (n 5 above), paras [45–47].   Centros (n 12 above). 65   Centros (n 12 above), para [27]. 63 64

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  423 abuse. Immediately after the CS decision I signalled that Halifax was dealing with an interpretation problem that was very much akin to the interpretation of a national tax rule, while CS was dealing with a cross-border problem involving two national tax systems under the fundamental freedoms.66 This difference may justify a different approach in interpreting the abuse concept. Under national law there are many abuse concepts. One that is frequently used in case law or under statutory general anti-avoidance rules is one stating that when there are different motives for a particular transaction and tax minimisation is one of them, tax authorities are allowed to challenge such a transaction when obtaining the tax advantage was the principal, or one of the principal reasons for engaging in such a transaction. The wider the provision is, the more discretionary power is given to the tax authorities. This is something for the national legislator to decide. Technically, VAT is secondary EU law, but it functions very much like a national tax law. Halifax and Part Service did not deal with cross-border issues that are central to the fundamental freedoms. The issues are revenue issues and comparisons of the relative tax burdens of different categories of transactions just as in domestic income tax. The major part of the transactions are domestic transactions. There are some cross-border issues of discrimination and protection of domestic markets under Article 110 TFEU [ex Article 90 EC], but they hardly involve abuse of those provisions. The cross-border abuses have more to do with VAT carousels, which could be easily classified as an abuse, because there is no redeeming Union virtue in these carousels, except for the naked desire for a tax advantage. Therefore an argument could be made to apply a wider concept of abuse for that part of VAT which has to do with purely domestic transactions such as in Halifax and PS. Cross-border issues under the fundamental freedoms are another matter. The fundamental freedoms are primary EU law, which takes precedence over national tax law also in matters of abuse and the fight against tax avoidance. The list of cases in which national tax authorities have been arguing that the exercise of the fundamental freedoms amounted to tax avoidance or tax evasion and therefore one form or another of abuse, and in which the ECJ decided otherwise, is almost endless. In setting a special standard for abuse under the EC Treaty (ie the famous ‘wholly artificial arrangement’) the ECJ was entirely justified. In modern economic life, and in particular in modern international activities, tax is always at least one of the considerations. As all other barriers to trade are abolished or are receding it is the tax barriers that become more important, and therefore it is of paramount importance to protect the fundamental freedoms. A classification as a ‘wholly artificial arrangement’ would very often be considered as an inadequate antiavoidance formula for domestic taxation purposes. In the domestic context, the balance is between the legal ground for the assessment and the effectiveness of the tax rule. In the European context, there is not only the balance between legality and effectiveness, as there is no general statutory anti-avoidance provision in EU tax law, but there is also the balance between the fundamental freedoms and the effectiveness of the national tax system. If the Court would have to balance the importance of various reasons for exercising the fundamental freedoms and have to decide whether obtaining the tax advantage was the principal objective or one of the principal objectives of a transaction, it is clear that the   Vanistendael (n 46 above) 195.

66

424  Frans Vanistendael Court would necessarily exercise wide discretionary powers and interfere directly in the tax systems of the Member States. If, however, evaluating the balance of the various motives were left to national courts, the Court could lose complete control, thus making it difficult to maintain the supremacy of the Treaties over national tax law. Therefore, in interpreting the EU concept of abuse in questions involving the fundamental freedoms and other Treaty provisions with direct effect, the Cadbury Schweppes doctrine should be followed. There is also the argument of authority. In his Opinion in Halifax, the AdvocateGeneral points out that ‘the absence of a unitary test for the operation in every field of Community law of the principle of the prohibition of abuse must be regarded as perfectly natural in Community law, as it is in any national legal system’67 (emphasis added). There are two final remarks. In CS the ECJ held that ‘such a tax measure [ie CFC rules] must not be applied where it is proven, on the basis of objective factors which are ascertainable by third parties, that despite the existence of tax motives that CFC is actually established in the host Member State and carries on a genuine economic activity there’.68 (emphasis added). The national courts should take care not to turn this condition into an instrument whereby national tax authorities indirectly would be able to evaluate the importance of the tax motive in comparison with other motives. In the case of CS, there is some evidence that Inland Revenue would like to go down that road, by making an analysis of economic activities carried out in the IFSC and calculating the amount of profit that should be attached to the functions performed in the Dublin Docks. In that respect, the Court has clarified its position by stating: [T]he fact that the activities which correspond to the profits of the CFC could just as well have been carried out by a company established in the territory of the Member State in which the resident company is established does not warrant the conclusion that there is a wholly artificial arrangement.69

The only thing that is required is evidence of a genuine establishment, keeping in mind that in the Centros case, the mere incorporation of a company in one Member State in order to circumvent company law was enough to establish a genuine presence for company law purposes. Finally, there is the question whether the abuse concept as it has now been elaborated by the Court for VAT purposes should be characterised as an abuse concept of EU law, or rather as a uniform national concept borrowed from EU law. I agree that we are very close to discussing the gender of the angels. In his Opinion in Halifax, the AdvocateGeneral pointed out, however, that the ‘prohibition of abuse has to be balanced against the principles of legal certainty and the protection of legitimate expectations that also form part of the Community legal system’.70 This leads to the position that without an EU statutory base the EU concept of abuse must be applied with restraint. In particular, when the abuse concept would involve a balancing concept of various tax and non-tax motives, the Advocate-General would require ‘the adoption of appropriate national legislative measures’71 (ie that concept of abuse is, in his opinion, too wide to be classified  AG Poiares Maduro in Halifax (n 1 above), para [83].   Cadbury Schweppes (n 3 above), para [75]. 69   Cadbury Schweppes (n 3 above), para [69]. 70  AG Poiares Maduro in Halifax (n 1 above), para [84]. 71  AG Poiares Maduro in Halifax (n 1 above), para [90]. 67 68

Cadbury Schweppes and Abuse from an EU Tax Law Perspective  425 as an EU concept of abuse). It is clear that this position has already been made obsolete by the PS decision. The question is why the Court should not be in a position to develop different concepts of abuse and have one concept that is specific to domestic situations applied in accordance with secondary EU tax legislation, giving more leeway to national tax authorities. The only objections against following this route would be the principle of legal certainty and legitimate expectation under EU law, as well as the national constitutional provisions on the legal basis of taxation. However, this simply takes us back to the balancing act that courts attempt between different legal principles.

29 Cadbury Schweppes and Abuse: Comments Richard Lyal*

T

he following remarks seek to respond to the views expressed by Professor Frans Vanistendael in his chapter of this volume.1 Professor Vanistendael examines the concept of abuse developed in the value-added tax (VAT) case law of the Court of Justice (in particular Halifax2 and Part Service3) and in its case law on direct tax (in particular Cadbury Schweppes4). He concludes that there are two different concepts, corresponding to two distinct needs. I.  Halifax and Part Service—abuse of law and VAT

To an English audience there may, on first impressions, seem something foreign about the discussion of the concept of abuse of law in Halifax and subsequent cases in the tax field (a discussion which has been going on for some time in other areas of EU law). After all, it is usually said that such a concept does not exist in the common law. Nevertheless, in the field of tax, the general notion of substance over form is not very far removed from that concept, at least in practical terms. A particularly close analogy may be drawn with the concept of fiscal nullity, a well-established doctrine in relation to direct tax.5 That doctrine is considered to be a rule of statutory construction (purposive interpretation). There does not seem to be unanimity over the nature of the concept of abuse in EU law (canon of construction or rule of law), but in their content it is hard to see any difference between the two concepts. In Burmah Oil,6 Lord Diplock identified the core of the Ramsay principle, pointing out that Ramsay concerned: a pre-ordained series of transactions (whether or not they include the achievement of a legitimate commercial end) into which there are inserted steps that have no commercial purpose apart from the avoidance of a liability to tax which in the absence of those particular steps would have been payable.7

 *  The opinions expressed are those of the author and should not be attributed to the European Commission.

 F Vanistendael, ‘Cadbury Schweppes and Abuse from an EU Tax Law Perspective’, ch 28 above.   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 3   Case C-425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I-897. 4   Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 5   Ramsay v Inland Revenue Commissioners [1982] AC 300; MacNiven v Westmoreland Investments [2003] 1 AC 311; Barclays v Mawson [2005] 1 AC 684. 6   Inland Revenue Commissioners v Burmah Oil Co Ltd [1982] STC 30. 7   Burmah Oil (n 6 above) 32. 1 2

428 Richard Lyal What the courts were thus able to do was ‘to ignore the intermediate circular book entries and to look at the end result’.8 For a reader of the Halifax judgment, that formulation has a familiar echo. One of the ironies of Halifax is that the courts in the United Kingdom refrained from taking the same approach to VAT on the ground that the latter was a Community system, to which purely national tax law doctrines should not be applied. Accordingly, in advocating the application of the concept of abuse of law in the field of VAT in the Halifax case and its successors, the Commission was able to derive some comfort from the consideration that its argument mirrored an established doctrine in the United Kingdom. That line of argument was not free from controversy. First, there was some concern that while the concept of abuse of law had been used in order to prevent persons from obtaining benefits from Community funds in a manner not intended by the applicable legislation,9 its extension into VAT would affect financial relations between taxpayers and Member States, and not merely, as in previous cases, between economic operators and the EU institutions. The important factor, however, was not the destination of the tax revenue but the fact that VAT is governed by a uniform system: it is a code established by EU legislation and applied, at least in theory, in a uniform manner throughout the Union. Just as economic operators should not be allowed to benefit in an unintended and abusive manner from Union funds, so taxpayers should not be allowed to manipulate EU VAT law in an unintended and abusive manner. There was also some concern that the introduction of an open-ended rule of this kind could have harmful consequences for legal certainty. That worry merited close attention, though problems of legal certainty can be minimised by formulating a narrow rule aimed at clear cases of manipulation. Perhaps more importantly in practical terms, the other options available to the Court—if it is agreed that there should be a way of preventing manipulation of the VAT system of the kind seen in Halifax—were not attractive. The UK tax authorities argued before the national courts that a series of transactions of the kind entered into by the various members of the Halifax group of companies should be treated as falling outside the scope of VAT altogether, on the ground that they lacked economic substance and thus did not constitute supplies. That would have been an even worse result from the perspective of legal certainty, since it could have consequences not only for the targeted taxpayers but also for other actors in the chain of supply (that was an issue that arose also in the ‘carousel fraud’ cases10). Moreover, the Court had to take into account the fact that some Member States did apply a theory of abuse of law or an equivalent general anti-abuse measure in VAT (in Ireland, for example, see section 811 of the Taxes Consolidation Act 1997). In order to protect legal certainty—and for that matter, the uniform application of the common system of VAT—it was necessary, or at least desirable, to have a single solution. Either the Member States which already applied an anti-abuse rule in VAT had to stop doing so, or a general Community rule had to be introduced. In the event, the Court of Justice concluded that the right to deduct input tax could   Burmah Oil (n 6 above) 33.  For example: Case 125/76 Enterprise Peter Cremer v Bundesanstalt für landwirtschaftliche Markordnung (‘BALM’) [1977] ECR 1593; Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 10   Joined Cases C-354/03, C-355/03 and C-484/03 Optigen Ltd, Fulcrum Electronics Ltd and Bond House Systems Ltd v Commissioners of Customs & Excise [2006] ECR I-483 and subsequent cases.     8     9

Cadbury Schweppes and Abuse: Comments  429 be denied to taxpayers who engaged in transactions intended to procure an undue tax advantage. In paragraphs [74–75] of the judgment it defined a two-fold test, holding that an abusive practice could be found only where: 1.  The transactions concerned, while meeting the formal requirements of the applicable provisions, lead to a result contrary to their purpose. 2. The essential aim of those transactions is to obtain a tax advantage. In analysing the existence of abuse, and in order to determine the real substance and significance of the transactions concerned, the courts may have regard to the purely artificial nature of those transactions and the links of a legal, economic or personal nature between the operators involved in the scheme for reduction of the tax burden (see paragraph [81] of the judgment). That formulation is aimed at circumscribing the scope of the rule in order to preclude excessively broad application of the concept of abuse. Professor Vanistendael is concerned that the rule laid down in Halifax may not be expressed narrowly enough. As he points out, Advocate-General Poiares Maduro argued in his Opinion that recourse may be had to the prohibition of abuse only in relation to transactions carried out solely in order to procure a tax advantage; it has no reason to operate where a transaction may have some other explanation (see paragraphs [86– 90]). To give broader scope to the prohibition, he said, would introduce too much legal uncertainty for economic operators. The Court, however, used language which, at least in appearance, gives more flexibility to tax administrations. Where Advocate-General Poiares Maduro argues that abuse may arise where the sole purpose of a transaction (or series of transactions) is tax avoidance, the Court speaks of the essential aim of the activity. Professor Vanistendael maintains, moreover, that the Court went on, in the Part Service case, to lay down a different, even broader rule from that laid down in Halifax; that view is based on the use of the expression ‘principal purpose’ in Part Service. To start with the last point, the answer is simple, though regrettable. In French, the internal working language of the Court, the same expression is used in both Halifax and Part Service: but essentiel. And the Italian version (Italian was the language of the case in Part Service) uses scopo essenziale. All the Court set out to do in Part Service, therefore, was to reiterate the point made in Halifax that avoidance or minimisation of tax liability need not be the sole purpose of the transactions in question—it suffices that they be their essential purpose. In so doing it was replying to what it saw as the underlying concern of the national court, namely the perceived contradiction in Halifax between passages such as paragraph [60] which referred to the ‘sole aim’ of the transactions and others which referred to the ‘essential aim’ (see paragraphs [29–31] of the Part Service judgment). The apparent divergence between the two judgments is the result of a difference in the English translation for which there is no obvious reason. More generally, this focus on the precise words used by the Advocate-General and by the Court may be an instance of over-interpretation. It is worth bearing in mind that a judgment of the Court is not a piece of legislation, in which each word must be given its proper meaning. The explanation for the Court’s use of the expression ‘essential aim’ rather than ‘sole purpose’ in Halifax may simply be that it did not want to make it too easy for taxpayers to argue that even though the real purpose of their transaction was tax avoidance, there was also some more or less plausible business motive. It should also be noted that in the very paragraph where it established the test of ‘essential aim’ the Court

430 Richard Lyal expressly approved the reasoning of Advocate-General Poiares Maduro, stating that ‘the prohibition of abuse is not relevant where the economic activity carried out may have some explanation other than the mere attainment of tax advantages’ (paragraph [75] of the judgment). That suggests that its intention was to confine very narrowly the possible application of the abuse doctrine. Professor Vanistendael is certainly right, nevertheless, in drawing attention to the need to avoid allowing national administrations to give too broad an interpretation to the expression ‘essential’. If taxpayers cannot be allowed to adduce largely fictitious business motives, it is equally necessary to prevent the administration from ignoring real business motives on the pretext that tax minimisation was an important or even the main factor in the taxpayer’s decision. The key to this debate perhaps lies in the importance to be given to subjective intention. Just to what extent and in what circumstances is it legitimate to deny a taxpayer the favourable tax consequences of a particular series of business transactions on the ground that he carried out those transactions in order to minimise his tax burden? In analysing the reasons for a business decision there will very frequently be a question to which the answer will be: ‘I did it for tax reasons’. To take the circumstances of another very familiar case, why would one establish a company to make and sell confectionery and soft drinks? Presumably to make profits. Why would a business group with a range of operations establish a holding and financial subsidiary? To make the business of the group more efficient—that is to say, to optimise profits. Why would it establish that holding or financial subsidiary in one country rather than another? One of the considerations taken into account must inevitably be the tax environment. As the Court pointed out in Halifax, it is entirely legitimate for a taxpayer to take tax considerations into account in structuring his business. Where a taxpayer has the option of two different types of transaction in order to arrive at a desired result, there is no obligation on him to choose the one which entails the greater tax liability (see paragraph [73] of the judgment and paragraph [85] of the Opinion of Advocate-General Poiares Maduro). A striking example of this proposition, also in the field of VAT, may be seen in Heerma.11 The tax motive may be more or less prominent among the various reasons for structuring a business transaction in a certain way, and it is only in extreme cases that there is any justification for applying tax treatment which differs from that normally governing the transaction which has been carried out. In this context, the subjective intention of the taxpayer in itself is an unreliable criterion, for it tells us little or nothing about the legitimacy of his choice of transaction or, more importantly, the reality of that transaction. It may well be that, from a range of options, a taxpayer chooses the option that entails the least tax, and that he does so solely for tax reasons. There is no reason to deny him that choice simply because it is motivated by a tax saving. The fact remains that he has chosen that option, he has carried out that transaction, and that transaction normally attracts a certain, favourable tax treatment. Instead of subjective intention, or in addition to it, there is a need for a clearly ascertainable objective criterion, and just such a criterion is set out by the Court at paragraph [81] of the judgment: the ‘purely artificial nature’ of the transactions concerned (coupled, in that passage of the Court’s reasoning, with the existence of personal or organic links between the operators involved).   Case C-23/98 Staatssecretaris van Financiën v J Heerma [2000] ECR I-419.

11

Cadbury Schweppes and Abuse: Comments  431 It is only by focusing on the artificial character of the arrangements made by taxpayers that a clearly defendable basis for unfavourable tax treatment can be identified. Insofar as the subjective intention of the taxpayer is considered relevant on dogmatic grounds, it may easily be inferred from that artificiality. Such an approach is consistent with what may be thought to be the correct role to be attributed to subjective intention. It is not a criterion in its own right so much as one of the reasons underlying a workable, fair and solidly justifiable rule. Reference to the artificial character of the taxpayer’s transactions also reflects an essential aspect of abuse of law: the failure of the transaction to fulfil the purpose of the rule in question. Abuse of law lies in a contrived course of conduct which sets out to comply with the formal requirements of a rule while procuring a result which is quite different from that intended by it. It is this aspect, the emphasis on the presence of an artificial construction, that most strongly links the VAT case law with the case law on direct tax. II.  Cadbury Schweppes—abuse and the Treaty freedoms

While Cadbury seems to be regarded as the new locus classicus of the concept of abuse in direct tax, it does not in fact represent an innovation. It was in ICI12 that the Court enunciated the artificiality test, holding that national measures aimed at preventing tax evasion and avoidance were compatible with the Treaty freedoms if they had ‘the specific purpose of preventing wholly artificial arrangements, set up to circumvent [national] tax legislation, from attracting tax benefits’.13 In his Opinion in Test Claimants ACT14 (at paragraph [73]), Advocate-General Geelhoed attached the term ‘abuse of law’ to the approach adopted by the Court both in ICI and in Lankhorst-Hohorst.15 The Court had already examined the concept of ‘abuse’ of the Treaty freedoms in X and Y16 without, however, laying down a clear set of criteria. Curiously, the expression ‘abuse’ is used in the Cadbury judgment only in relation to the first issue raised by the UK Special Commissioners: whether the establishment of a subsidiary in another Member State solely because of a more favourable tax regime available there amounts to the abuse of the Treaty freedoms. That is, with all respect, an extraordinary question. It would not occur to anyone to raise the same question in respect of the establishment of a subsidiary in a Member State in order to take advantage of low wages, or for that matter, to benefit from some of the things that paying taxes buys, such as good infrastructure and an educated workforce. Nor, it may be hoped, would anyone have asked that question if Cadbury had established a chocolate factory in Ireland. It should be clear that none of these things can be regarded as being in some way a circumvention of the normal rules which would be applicable in the home State. Accordingly, in paragraphs [36–38] of the judgment, the Court essentially repeats the idea set out in paragraph [73] of the Halifax judgment: the mere fact of choosing a business 12   Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I-4695. 13   ICI (n 12 above), para [26]. 14   Case C-374/04 Test Claimants in Class IV of the ACT Group Litigation [2006] ECR I-11673. 15   Case C-324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I-11779. 16   Case C-436/00 X and Y v Riksskatteverket [2002] ECR I-10829.

432 Richard Lyal configuration which attracts favourable tax treatment does not amount to abuse of law (or abuse of the Treaty freedoms). Yet the incongruity of this debate is helpful, for it highlights the correct distinction between abuse and the legitimate exercise of rights under EU law. Moving to another country or setting up ancillary operations there in order to take advantage of the tax climate is a legitimate business choice in the same way as moving or establishing business operations in order to exploit any other local advantage. It is in relation to the second issue raised by the Special Commissioners, on the application of the Treaty freedoms to national rules regarding controlled foreign companies (CFC), that the Court examines the criterion of artificiality. Here the Court notes that EU law does not allow Member States to seek to neutralise the favourable tax consequences that undertakings may derive from cross-border activity. The mere fact that an undertaking has established itself in another Member State does not create a presumption of tax evasion, even where it has obtained tax advantages in doing so. Member States may, however, take measures to counter clear instances of tax evasion in the form of ‘wholly artificial arrangements aimed at circumventing the application of [their] legislation’.17 When the Court sets out to explain what a wholly artificial arrangement might be, it begins by describing the opposite: the objective pursued by the freedom of establishment is to allow a business to be set up and carried on in another Member State, as part of economic life in that second Member State. In essence, that means the actual pursuit of an economic activity there. An artificial construction, by contrast, is one which creates the form of a business establishment without its substance, a simulacrum which does not represent the economic reality. This is nothing more than an interpretation of the expression ‘establishment’. One might therefore ask what added value is to be found in the use of the expression ‘abuse of law’ here. From one perspective it is legitimate to say that there is none. If there is a wholly artificial construction, then there is no establishment and no exercise of the freedom. The concept of abuse of law is not itself of any utility in defining or circumscribing the manner in which Union citizens may benefit from the freedoms. Yet this case is a telling example of the distinction between the exercise and the abuse of a freedom. In that sense, it is correct to regard Cadbury as a case on abuse of law, even though the Court does not use that expression in the relevant section of its judgment: the essence of abuse is creating the appearance or form of a situation which attracts a certain legal treatment without truly creating its substance. The expression ‘abuse of law’ is a convenient label to attach to a type of conduct which does not merit the protection of the Treaty freedoms. Where is the importance of motive in this context? Subjective intention is a dangerous criterion. The natural tendency to disapprove of tax avoidance may tempt tax authorities or for that matter legislators to attribute a key role to motive. The United Kingdom’s CFC rules at stake in Cadbury are an example. However, as already noted, there is always likely to be a tax motive when an undertaking establishes itself (or wishes to give the appearance of having established itself) in a low-tax jurisdiction. Tax is one of the factors to be taken into account, along with many other things. Thus the motive of avoiding or minimising tax is not in itself a practical or reliable criterion. It proves too much. That is not to say that subjective intention is devoid of importance: it is surely one of the features 17   Cadbury Schweppes (n 4 above), para [51], citing ICI (n 12 above) and subsequent case law. Cf AG Geelhoed in Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107, paras [62–67].

Cadbury Schweppes and Abuse: Comments  433 that justifies departure from the normal tax treatment. The great virtue of the ‘wholly artificial’ test is that it provides clear and unequivocal evidence of intention at the same time as demonstrating that the purpose of the tax rule in issue is not being fulfilled. In such circumstances, the justification for refusing favourable tax treatment is equally unequivocal. At the same time, the test to be applied must be a rigorous one, since what is at stake is conduct by a State which may conflict with a Treaty freedom. As Professor Vanistendael notes, that is a consideration which justifies a conclusion different from what it might be in a strictly internal context. Artificial constructions and genuine establishments are conceptual opposites, but it may not always be easy, in the particular circumstances of a given case, to determine what corresponds to economic reality. Abuse of establishment is one thing, adequacy of the establishment quite another. Just what sort of operation can be said to amount to a genuine establishment? That is likely to be the next focus of debate in this area, perhaps beginning with the detailed analysis of the Cadbury case itself in the English courts. It may be supposed that a single person with a computer transferring large sums of money around the globe will not be regarded as sufficient to constitute an establishment, but it may also be that group service companies can be run with relatively small staff. One underlying concern of the UK authorities appears to be the desire to arrive at some degree of correspondence between the type and size of the operation carried on in a place and the level of profits ostensibly generated there. While that is a legitimate preoccupation, which reflects the concept of balanced allocation of taxing rights, the question is how to arrive at a satisfactory reconciliation of the competing interests of revenue gathering and free movement. It is likely to be difficult to find a set of criteria that are not arbitrary. There is no obvious way of determining the level or type of establishment that should give rise to any given amount of profit. It is submitted that here too a test based on the presence of a ‘wholly artificial construction’ is the appropriate means of dealing with the problem. Such a test should be quite adequate once it is borne in mind that the notion of ‘wholly artificial’ can extend not simply to the establishment itself but also to each of the transactions that the establishment ostensibly carries out. In this regard we may apply the logic deployed by the Court in, for example, Test Claimants Thin Cap.18 There the Court held, citing Cadbury, that thin capitalisation rules applied under national legislation were compatible with the Treaty freedoms insofar as their ‘specific objective’ was to ‘prevent conduct involving the creation of wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory’ (paragraph [74]). It is worth noting, incidentally, that the Court speaks of conduct having the ‘sole purpose’ (paragraph [77]) and the ‘essential purpose’ (paragraph [81]) of tax avoidance, surely a deliberate echo of the Halifax formulae. In the particular case of thin capitalisation, the Court approves the use of an arm’slength criterion, subject to the possibility for the taxpayer of demonstrating that a transaction carried out on terms other than those which would have been applied in an arm’s-length relationship nevertheless has a commercial justification. In his Opinion, drawing on experience in the field of indirect tax and in non-tax spheres, AdvocateGeneral Geelhoed points out (at paragraph [66]) that ‘the arm’s length test represents in this context an objective factor by which it can be assessed whether the essential aim of   Test Claimants Thin Cap (n 17 above), paras [74–82].

18

434 Richard Lyal the transaction concerned is to obtain a tax advantage’. Other objective criteria may need to be developed in relation to different methods used by taxpayers to shift profits from one jurisdiction to another in a manner which does not correspond to economic reality. The underlying theme in this debate is a simple one: Member States must not discriminate against cross-border transactions, but their right to tax profits generated in their jurisdiction should be respected. Ultimately, the aim must be to strike an appropriate balance between ensuring that people can do business across frontiers with a minimum of interference and allowing Member States to protect themselves against leakage of their tax base. III.  Separate tests for VAT and for the Treaty freedoms?

It will be apparent from the foregoing discussion that in the present writer’s view, there is a single test for abuse of law, whether it be in the field of direct or of indirect tax: a wholly artificial construction aimed at circumventing the applicable tax legislation. There are no doubt valid reasons for which one might wish to advocate a more expansive concept of abuse in relation to VAT than to the Treaty freedoms; it is true that the common system of VAT may be considered to be analogous to an internal tax system, so that the concern for freedom of cross-border trade is of less immediate importance. However, the Court has—it is submitted rightly—applied the same test in both fields (as it has done, moreover, in other areas of EU law). Insofar as the opposite impression may be created by close reading of certain judgments of the Court, it is because the debate is polluted by the language of subjective intention. The subjective intention of the taxpayer, however, is an imperfect guide. Only if that intention is made manifest in the creation of a wholly artificial situation is there a justification for interfering with the freedom of economic operators to do business in the manner they please.

30 Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective Michael Lang* I. Introduction

O

ne of the most difficult and controversial legal questions is how to deal with the attempt to circumvent the law. Within Europe, completely different legal traditions and concepts exist in this respect. Not surprisingly, these issues play an important role in European law as well. The Court of Justice has had to deal with attempts to circumvent European law on several occasions. In recent years, these issues have become increasingly relevant in tax cases. One important judgment in this respect is Cadbury Schweppes,1 to which a lot of attention has been paid by academics. Several authors consider the reasoning the ECJ developed in Cadbury Schweppes as a separate line of case law. It is my task to analyse whether this line of case law has impact on anti-abuse doctrines of the Member States. I will deal with this question in Part III of this chapter. Before dealing with that question, one has to clarify what the approach of the Court concerning abuse of law is and whether there is a specific line of Cadbury Schweppes case law at all. These issues will be dealt with in Part II of this chapter. II.  Cadbury Schweppes Case Law

A.  ‘Wholly Artificial Arrangements’ and the Court’s Case Law In Cadbury Schweppes the ECJ referred to ‘wholly artificial arrangements’ to describe situations in which a less favourable treatment for intra-Community cross-border situations could be justified in order to reduce the risk of tax avoidance.2 It is not the first time the  * The author finalised this chapter in October 2008. It was brought in line with the Treaty on the Functioning of the European Union, but subsequent literature and case-law of the Court of Justice were not taken into consideration. The author would like to thank his research assistant Elisabeth Titz for her support. 1   Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 2  See for example, Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I-4695, para [26]; Case C-324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I-11770, para [37]; Case C-446/03 Marks & Spencer plc v David Halsey (Her Majesty’s Inspector of Taxes) [2005] ECR I-10837, para [57]; Cadbury Schweppes (n 1 above), para [51]; Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107, paras [71, 72]. See also the Opinion of AG Mengozzi in Case C-298/05 Columbus Container Services BVBA & Co v Finanzamt Bielefeld-Innenstadt [2007] ECR I-10451, paras [169–84].

436  Michael Lang Court used this phrase.3 But Cadbury Schweppes is special, as for the first time, the ECJ defined what it understood by this phrase. In its previous case law, the ECJ saw no need for that because the domestic measures which had been examined by the Court had such a broad scope that it was clear that they would not merely cover ‘wholly artificial arrangements’ and were therefore not proportionate. However, in order to understand the Cadbury Schweppes judgment, one has to include the predecessors of this judgment in one’s analysis. ICI4 was the first judgment where the phrase ‘wholly artificial arrangement’ was used. ICI was about a UK tax provision denying tax relief with respect to losses incurred by a resident subsidiary. The majority of the subsidiaries controlled by the holding company had their seat outside the United Kingdom, while tax relief was granted where the majority had their subsidiaries within the UK. The ECJ found it necessary to determine whether there was a justification for such an inequality of treatment and concluded that there was none:5 As regards the justification based on the risk of tax avoidance, suffice it to note that the legislation at issue in the main proceedings does not have the specific purpose of preventing wholly artificial arrangements, set up to circumvent United Kingdom tax legislation, from attracting tax benefits, but applies generally to all situations in which the majority of a group’s subsidiaries are established, for whatever reason, outside the United Kingdom. However, the establishment of a company outside the United Kingdom does not, of itself, necessarily entail tax avoidance, since that company will in any event be subject to the tax legislation of the State of establishment.

Lankhorst-Hohorst was about German thin capitalisation rules. Under German law, interest paid by a resident subsidiary on loan capital provided by a non-resident parent company was taxed as a hidden dividend at a rate of thirty per cent, whereas, in the case of a resident subsidiary whose parent company is also resident and receives a tax credit, interest paid was treated as expenditure and not as a hidden dividend. The ECJ held that this unequal treatment infringed the freedom of establishment:6 As regards more specifically the justification based on the risk of tax evasion, it is important to note that the legislation at issue here does not have the specific purpose of preventing wholly artificial arrangements, designed to circumvent German tax legislation, from attracting a tax benefit, but applies generally to any situation in which the parent company has its seat, for whatever reason, outside the Federal Republic of Germany. Such a situation does not, of itself, entail a risk of tax evasion, since such a company will in any event be subject to the tax legislation of the State in which it is established.

X and Y was about Swedish rules that allowed for a transfer of shares from one legal entity to another legal entity at less than market value if the transferor had a holding in the other legal entity, except when the other legal entity was either a foreign legal entity or a Swedish legal entity under foreign control. The ECJ held this distinction to be in conflict with the freedom of establishment and the free movement of capital:7     3  For a description of the Court’s case law on wholly artificial arrangements and for the following deliberations see already, M Lang and S Heidenbauer, ‘Wholly Artificial Arrangements’ in L Hinnekens and P Hinnekens (eds), A Vision of Taxes Within and Outside European Borders: Festschrift in honor of Frans Vanistendael (The Hague, Kluwer Law International 2008) 597.     4   ICI (n 2 above).     5   ICI (n 2 above), para [26]. See also M Lang, ‘CFC Legislation and Community Law’ (2002) 42 European Taxation 374, 375.     6   Lankhorst-Hohorst (n 2 above), para [37].     7   Case C-436/00 X and Y v Riksskatteverket [2002] ECR I-10829, paras [61, 62].

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  437 The provision at issue here is not specifically designed to exclude from a tax advantage purely artificial schemes designed to circumvent Swedish tax law, but concerns, generally, any situation in which, for whatever reason, the transfer at undervalue is to a company established under the legislation of another Member State or a branch set up in the Kingdom of Sweden by such a company. [ . . . ] However, tax evasion or tax fraud cannot be inferred generally from the fact that the transferee company or its parent company is established in another Member State and cannot justify a fiscal measure which compromises the exercise of a fundamental freedom guaranteed by the Treaty.

In Lasteyrie du Saillant, the Court dealt with a French exit tax provision:8 As regards justification based on the aim of preventing tax avoidance, referred to by the national court in its question, it should be noted that Article 167a of the CGI is not specifically designed to exclude from a tax advantage purely artificial arrangements aimed at circumventing French tax law, but is aimed generally at any situation in which a taxpayer with substantial holdings in a company subject to corporation tax transfers his tax residence outside France for any reason whatever.

Marks & Spencer was about cross-border group relief. The ECJ accepted that the UK may distinguish between cross-border situations and domestic situations. The Court referred to three different grounds of justifications that—‘taken together’—allowed for different treatment.9 One of them was the ‘risk of tax avoidance’.10 However, the ECJ mentioned this justification again, this time without having been asked:11 It is also important, in that context, to make clear that Member States are free to adopt or to maintain in force rules having the specific purpose of precluding from a tax benefit wholly artificial arrangements whose purpose is to circumvent or escape national tax law.

In Cadbury Schweppes, which was about the UK Controlled Foreign Companies (CFC) tax rules, the Court explained its concept of ‘wholly artificial arrangements’ in more detail; we will take a closer look at this concept below. However, it is worth mentioning at this point that the Court considered itself to be in line with previous case law:12 It is also apparent from case-law that the mere fact that a resident company establishes a secondary establishment, such as a subsidiary, in another Member State cannot set up a general presumption of tax evasion and justify a measure which compromises the exercise of a fundamental freedom guaranteed by the Treaty. [ . . . ] On the other hand, a national measure restricting freedom of establishment may be justified where it specifically relates to wholly artificial arrangements aimed at circumventing the application of the legislation of the Member State concerned.     8   Case C-9/02 Hughes de Lasteyrie du Saillant v Ministère de l’Économie, des Finances et de l’Industrie [2004] ECR I-2409, para [50].     9   See also F Vanistendael, ‘The ECJ at the Crossroads: Balancing Tax Sovereignty against the Imperatives of the Single Market’ (2006) 46 European Taxation 413, 415. 10   Marks & Spencer (n 2 above), paras [49, 50]. 11   Marks & Spencer (n 2 above), para [57]; critical M Lang, ‘The Marks & Spencer Case—The Open Issues Following the ECJ’s Final Word’ (2006) 46 European Taxation 54, 58; M Lang, ‘Direct Taxation: Is the ECJ Heading in a New Direction?’ (2006) 46 European Taxation 421, 427. 12   Cadbury Schweppes (n 1 above), paras [50, 51]. See also F Vanistendael, ‘Halifax and Cadbury Schweppes: One Single European Theory of Abuse in Tax Law?’ (2006) 15 EC Tax Review 192, 193; and J Kokott and T Henze, ‘Ist der EuGH—noch—ein Motor für die Konvergenz der Steuersysteme?’ (2007) 62 Betriebs-Berater 913, 916. For the difficulty in drawing the borderline between the use and abuse of a fundamental freedom, also see N Vinther and E Werlauff, ‘Tax Motives Are Legal Motives—The Borderline between the Use and Abuse of the Freedom of Establishment with Reference to the Cadbury Schweppes Case’ (2006) 46 European Taxation 383, 384.

438  Michael Lang In Test Claimants in the Thin Cap Group Litigation the Court confirmed what it had previously held and further developed its ideas. As a starting point, the ECJ referred to its previous judgments:13 It must be pointed out that, according to established case-law, a national measure restricting freedom of establishment may be justified where it specifically targets wholly artificial arrangements designed to circumvent the legislation of the Member State concerned [ . . . ] . The mere fact that a resident company is granted a loan by a related company which is established in another Member State cannot be the basis of a general presumption of abusive practices and justify a measure which compromises the exercise of a fundamental freedom guaranteed by the Treaty.

This case law has been confirmed in NV Lammers & Van Cleeff:14 In this respect, it must be pointed out that, according to established case-law, a national measure restricting freedom of establishment may be justified where it specifically targets wholly artificial arrangements designed to circumvent the legislation of the Member State concerned (Test Claimants in the Thin Cap Group Litigation, paragraph 72 and the case-law cited). [ . . . ] The mere fact that a resident company is granted a loan by a related company which is established in another Member State cannot be the basis of a general presumption of abusive practices and justify a measure which compromises the exercise of a fundamental freedom guaranteed by the Treaty (Test Claimants in the Thin Cap Group Litigation, paragraph 73 and the case-law cited). [ . . . ] In order for a restriction on the freedom of establishment to be justified on the ground of prevention of abusive practices, the specific objective of such a restriction must be to prevent conduct involving the creation of wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory (Test Claimants in the Thin Cap Group Litigation, paragraph 74 and the case-law cited). [ . . . ] At paragraph 80 of its judgment in Test Claimants in the Thin Cap Group Litigation, the Court held that legislation of a Member State may be justified by the need to combat abusive practices where it provides that interest paid by a resident subsidiary to a non-resident parent company is to be treated as a distribution only if, and in so far as, it exceeds what those companies would have agreed upon on an arm’s-length basis, that is to say, the commercial terms which those parties would have accepted if they had not formed part of the same group of companies. [ . . . ] The fact that a resident company has been granted a loan by a non‑resident company on terms which do not correspond to those which would have been agreed upon at arm’s length constitutes, for the Member State in which the borrowing company is resident, an objective element which can be independently verified in order to determine whether the transaction in question represents, in whole or in part, a purely artificial arrangement, the essential purpose of which is to circumvent the tax legislation of that Member State. In that regard, the question is whether, had there been an arm’s-length relationship between the companies concerned, the loan would not have been granted or would have been granted for a different amount or at a different rate of interest (Test Claimants in the Thin Cap Group Litigation, paragraph 81). [ . . . ] In the present case, it is apparent from the order for reference that the interest payments made by the Belgian subsidiary on a loan granted by a non‑resident company which is a director were reclassified as dividends because the limit laid down in the second indent of Article 18 (1), point 3, of the ITC 1992 had been exceeded, that is to say, at the beginning of the taxable period the total of the interest-bearing loans was higher than the paid‑up capital plus taxed reserves. [ . . . ]   Thin Cap (n 2 above), paras [72, 73].   Case C-105/07 NV Lammers & Van Cleeff v Belgische Staat [2008] ECR I-0173, paras [26–34].

13 14

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  439 It is clear that, even if the application of such a limit seeks to combat abusive practices, it goes in any event beyond what is necessary to attain that objective. [ . . . ] As the Commission of the European Communities stated in its submissions, the limit laid down in the second indent of Article 18 (1), point 3, of the ITC 1992 also affects situations in which the transaction concerned cannot be regarded as a purely artificial arrangement. If interest payments made to non‑resident companies are reclassified as dividends as soon as they exceed such a limit, it cannot be ruled out that that reclassification will also apply to interest paid on loans granted on an arm’s length basis. [ . . . ] Consequently, the answer to the question submitted must be that Articles 43 EC and 48 EC [now Articles 49 and 54 TFEU] must be interpreted as precluding national legislation, such as that at issue in the main proceedings, under which interest payments made by a company resident in a Member State to a director which is a company established in another Member State are reclassified as dividends and are, on that basis, taxable, where, at the beginning of the taxable period, the total of the interest‑bearing loans is higher than the paid-up capital plus taxed reserves, whereas, in the same circumstances, where those interest payments are made to a director which is a company established in the same Member State, those payments are not reclassified as dividends and are, on that basis, not taxable.

In its Order in The Test Claimants in the CFC and Dividend Group Litigation the ECJ summarised its case law:15 [A] national measure restricting freedom of establishment may be justified where it specifically targets wholly artificial arrangements designed to circumvent the legislation of the Member State concerned (Cadbury Schweppes and Cadbury Schweppes Overseas, paragraph 51 and the caselaw cited). [ . . . ] It follows that, for a restriction on the freedom of establishment to be justified on the ground of prevention of abusive practices, the specific objective of such a restriction must be to prevent conduct involving the creation of wholly artificial arrangements, which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory (Cadbury Schweppes and Cadbury Schweppes Overseas, paragraph 55). [ . . . ] In order to find that there is such an arrangement there must be, in addition to a subjective element consisting in the intention to obtain a tax advantage, objective circumstances showing that, despite formal observance of the conditions laid down by Community law, the objective pursued by freedom of establishment has not been achieved (Cadbury Schweppes and Cadbury Schweppes Overseas, paragraph 64 and the case-law cited). [ . . . ] In those circumstances, in order for the legislation on CFCs to comply with Community law, the taxation provided for by that legislation must be excluded where, despite the existence of tax motives, the incorporation of a CFC reflects economic reality. That finding must be based on objective factors which are ascertainable by third parties with regard, in particular, to the extent to which the CFC physically exists in terms of premises, staff and equipment (Cadbury Schweppes and Cadbury Schweppes Overseas, paragraphs 65 and 67). [ . . . ] In this case, it is for the national court to determine whether, as maintained by the United Kingdom Government, the motive test, as defined by the legislation on CFCs, lends itself to an interpretation which enables the taxation provided for by that legislation to be restricted to wholly artificial arrangements or whether, on the contrary, the criteria on which that test is based mean that, where none of the exceptions laid down by that legislation applies and the intention to obtain a reduction in United Kingdom tax is central to the reasons for incorporating the CFC, the resident parent company comes within the scope of application of that legislation, 15   Case C-201/05 The Test Claimants in the CFC and Dividend Group Litigation v Commissioners of Inland Revenue [2008] ECR I-2875, paras [76–82], [84].

440  Michael Lang despite the absence of objective evidence which could indicate the existence of an arrangement of that nature. In the first case, the legislation on CFCs should be regarded as being compatible with Articles 43 EC and 48 EC. In the second case, on the other hand, the view should be taken that that legislation is contrary to Articles 43 EC and 48 EC (Cadbury Schweppes and Cadbury Schweppes Overseas, paragraphs 72 to 74). [ . . . ] In the light of the preceding considerations, Articles 43 EC and 48 EC must be interpreted as precluding the inclusion in the tax base of a resident company established in a Member State of profits made by a CFC in another Member State, where those profits are subject in that State to a lower level of taxation than that applicable in the first State, unless such inclusion relates only to wholly artificial arrangements intended to escape the national tax normally payable. Accordingly, such a tax measure must not be applied where it is proven, on the basis of objective factors which are ascertainable by third parties, that despite the existence of tax motives, that CFC is actually established in the host Member State and carries on genuine economic activities there (Cadbury Schweppes and Cadbury Schweppes Overseas, paragraph 75). [ . . . ] As regards, secondly, compliance requirements to which the exemption for a CFC’s profits in the hands of a resident company is subject, it is appropriate to point out, first, that in Cadbury Schweppes and Cadbury Schweppes Overseas, the Court held that the resident company is best placed to establish that it has not entered into wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory and that it must be given an opportunity to produce evidence that the CFC is actually established and that its activities are genuine (Cadbury Schweppes and Cadbury Schweppes Overseas, paragraph 70). [ . . . ] Secondly, in Test Claimants in the Thin Cap Group Litigation, the Court held that national legislation which provides for a consideration of objective and verifiable elements in order to determine whether a transaction represents a purely artificial arrangement, entered into for tax reasons alone, is to be considered as not going beyond what is necessary to prevent abusive practices where, on each occasion on which the existence of such an arrangement cannot be ruled out, the taxpayer is given an opportunity, without being subject to undue administrative constraints, to provide evidence of any commercial justification that there may have been for that arrangement (Test Claimants in the Thin Cap Group Litigation, paragraph 82).

In Jobra Vermögensverwaltungs-Gesellschaft mbH the ECJ referred to its case law on wholly artificial arrangements as well.16 Under Austrian law, a special tax incentive (‘investment growth premium’) was only granted for goods used in Austria. The Austrian Government also invoked the need to prevent abuse: The legislation at issue in the main proceedings aims to prevent wholly artificial arrangements involving transfers for remuneration. If it were not for that provision, an asset allocated to a lessor would be eligible for an investment premium irrespective of where the lessee took that asset. A concern would be that the lessor could hand over all or part of that premium to the lessee which, for its part, could use that asset to generate profits in other Member States. Thus, it would be possible to circumvent the fact that that advantage is limited to Austria.

The ECJ did not accept that justification:17 As regards the justification alleging that there is a need to prevent abuse, it must be held that a national measure restricting the freedom to provide services can be justified where it specifically targets wholly artificial arrangements which do not reflect economic reality and whose only 16   Case C-330/07 Jobra Vermögensverwaltungs-Gesellschaft mbH v Finanzamt Amstetten Melk Scheibbs [2008] ECR I-9099, paras [29], [35–39]. 17   Jobra (n 16 above).

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  441 purpose is to obtain a tax advantage (see, to that effect, Case C-196/04 Cadbury Schweppes and Cadbury Schweppes Overseas [2006] ECR I-7995, paragraphs 51 and 55, and Case C-524/04 Test Claimants in the Thin Cap Group Litigation [2007] ECR I-2107, paragraph 74). [ . . . ] In the present case, it cannot be claimed that it actually constitutes abuse for an undertaking that can claim the investment premium to hire out assets for remuneration to another undertaking which uses them primarily in other Member States. [ . . . ] Such hiring out cannot be the basis of a general presumption of abusive practice and justify a measure which compromises the exercise of a fundamental freedom guaranteed by the Treaty (see, to that effect, Test Claimants in the Thin Cap Group Litigation, paragraph 73, and Case C-105/07 Lammers & Van Cleef [2008] ECR I-0000, paragraph 27). [ . . . ] In that context, it must be noted that the legislation at issue in the main proceedings affects every lessor eligible for the investment premium which hires out assets for remuneration to undertakings carrying out cross-border activities, and does so even where nothing points towards the existence of such an artificial arrangement. Furthermore, the legislation does not allow lessors to adduce evidence that no abuse is taking place. [ . . . ] Therefore, it must be held that the legislation at issue in the main proceedings does not make it possible to limit the refusal to grant the investment premium to cases involving wholly artificial arrangements. Moreover, it has not been claimed before the Court of Justice that such an arrangement exists in the case in the main proceedings.

B.  Other Case Law on Abuse In order to develop the criteria under which it considers an arrangement to be wholly artificial, the Court referred to its judgments in Emsland-Stärke and Halifax.18 Neither judgment concerned direct taxation or the application of the fundamental freedoms. However, in Emsland-Stärke and Halifax, the Court carefully summarised its thenexisting case law and developed it further. By referring to these judgments in Cadbury Schweppes, the Court gives the impression that it is willing to apply a uniform abuse concept in different areas of law. Emsland-Stärke GmbH exported several consignments of a product based on potato starch to Switzerland. The recipients of the goods were established in Switzerland. On an application by Emsland-Stärke, and in the light, inter alia, of Swiss customs clearance certificates and freight papers, the German customs authorities granted the company an export refund. Subsequent inquiries conducted by the German customs investigation service revealed that, immediately after their release for home use in Switzerland, the exported consignments were transported back to Germany unaltered and, by the same means of transport under an external Community transit procedure, were released for home use in that Member State on payment of the relevant import duties. In respect of those consignments, the German customs authorities revoked the decisions granting an export refund and demanded repayment.19 The ECJ held that: [I]t is clear from the case-law of the Court that the scope of Community regulations must in no case be extended to cover abuses on the part of a trader. [ . . . ] The Court has also held that the fact that importation and re-exportation operations were not realised as bona fide commercial   Cadbury Schweppes (n 1 above), para [64]; see already, Lang and Heidenbauer (n 3 above) 597.   See Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569, paras [7–12]. 18 19

442  Michael Lang transactions but only in order wrongfully to benefit from the grant of monetary compensatory amounts, may preclude the application of positive monetary compensatory amounts. [ . . . ] A finding of an abuse requires, first, a combination of objective circumstances in which, despite formal observance of the conditions laid down by the Community rules, the purpose of those rules has not been achieved. [ . . . ] It requires, second, a subjective element consisting in the intention to obtain an advantage from the Community rules by creating artificially the conditions laid down for obtaining it. The existence of that subjective element can be established, inter alia, by evidence of collusion between the Community exporter receiving the refunds and the importer of the goods in the non-member country.20

Halifax is a more recent decision than Emsland-Stärke. The judgment concerned the application of the Sixth VAT Directive.21 The ECJ held that: [I]t must be borne in mind that, according to settled case-law, Community law cannot be relied on for abusive or fraudulent ends. [ . . . ] The application of Community legislation cannot be extended to cover abusive practices by economic operators, that is to say transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law.22

The Court continued by acknowledging that the: [P]rinciple of prohibiting abusive practices also applies to the sphere of VAT. [ . . . ] Preventing possible tax evasion, avoidance and abuse is an objective recognised and encouraged by the Sixth Directive. [ . . . ] However, as the Court has held on numerous occasions, Community legislation must be certain and its application foreseeable by those subject to it. [ . . . ] That requirement of legal certainty must be observed all the more strictly in the case of rules liable to entail financial consequences, in order that those concerned may know precisely the extent of the obligations which they impose on them. [ . . . ] Moreover, it is clear from the case-law that a trader’s choice between exempt transactions and taxable transactions may be based on a range of factors, including tax considerations relating to the VAT system. [ . . . ] Where the taxable person chooses one of two transactions, the Sixth Directive does not require him to choose the one which involves paying the highest amount of VAT. On the contrary, as the Advocate General observed in point 85 of his Opinion, taxpayers may choose to structure their business so as to limit their tax liability.23

The Court concluded by holding that: In view of the foregoing considerations, it would appear that, in the sphere of VAT, an abusive practice can be found to exist only if, first, the transactions concerned, notwithstanding formal application of the conditions laid down by the relevant provisions of the Sixth Directive and the 20   Emsland-Stärke (n 19 above), paras [51–53]. See also C Böing, Steuerlicher Gestaltungsmissbrauch in Europa—Eine rechtsvergleichende und gemeinschaftsrechtliche Untersuchung von Konzeptionen zur Bekämpfung des Gestaltungsmissbrauchs (Hamburg, Verlag Dr Kovaˇc 2006) 299. 21   Sixth Council Directive 77/388/EEC of 17 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes—Common system of value-added tax: uniform basis of assessment (Sixth VAT Directive) [1977] OJ L145/1, as amended. 22   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, paras [68, 69]. See also Vanistendael (n 12 above) 192; and O Rousselle and H Liebman, ‘The Doctrine of the Abuse of Community Law: The Sword of Damocles Hanging over the Head of EC Corporate Tax Law?’ (2006) 12 European Taxation 559, 561. 23   Halifax (n 22 above), paras [70–73].

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  443 national legislation transposing it, result in the accrual of a tax advantage the grant of which would be contrary to the purpose of those provisions. [ . . . ] Second, it must also be apparent from a number of objective factors that the essential aim of the transactions concerned is to obtain a tax advantage. As the Advocate General observed in point 89 of his Opinion, the prohibition of abuse is not relevant where the economic activity carried out may have some explanation other than the mere attainment of tax advantages.

The Court added that: [I]t must be borne in mind that it is the responsibility of the national court to determine the real substance and significance of the transactions concerned. In so doing, it may take account of the purely artificial nature of those transactions and the links of a legal, economic and/or personal nature between the operators involved in the scheme for reduction of the tax burden.24

In Part Service the ECJ was forced to explain its Halifax judgment in more detail. The Italian court referred the following question to the ECJ:25 ‘Does the concept of abuse of rights defined in the judgment of the Court of Justice in [Halifax and Others] as transactions, the essential aim of which is to obtain a tax advantage, correspond to the definition transactions carried out for no commercial reasons other than a tax advantage, or is it broader or more restrictive than that definition?’ The ECJ provided the following explanation:26 In paragraphs 74 and 75 of Halifax and Others, the Court first held that, in the context of interpreting the Sixth Directive, an abusive practice can be held to exist where:

·

the transactions concerned, notwithstanding formal application of the conditions laid down by the relevant provisions of the Sixth Directive and the national legislation transposing it, result in the accrual of a tax advantage the grant of which would be contrary to the purpose of those provisions; it is apparent from a number of objective factors that the essential aim of the transactions concerned is to obtain a tax advantage.

·

When, subsequently, it provided the referring court with details for guidance in interpreting the transactions in the case in the main proceedings, the Court once again referred, at paragraph 81, to transactions essentially seeking to obtain a tax advantage. [ . . . ] Therefore, when it stated, in paragraph 82 of that judgment, that in any event, the transactions at issue had the sole purpose of obtaining a tax advantage, it was not establishing that circumstance as a condition for the existence of an abusive practice, but simply pointing out that, in the matter before the referring court in that case, the minimum threshold for classifying a practice as abusive had been passed. [ . . . ] The reply to the first question therefore is that the Sixth Directive must be interpreted as meaning that there can be a finding of an abusive practice when the accrual of a tax advantage constitutes the principal aim of the transaction or transactions at issue.

C.  Objective and Subjective Criteria In Cadbury Schweppes, the ECJ referred to Emsland-Stärke and Halifax and held that—in order to describe the notion ‘wholly artificial arrangement’—‘there must be,   Halifax (n 22 above), paras [74, 75, 81].   Case C-425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I-897, para [32]. 26   Part Service (n 25 above), paras [42–45]. 24 25

444  Michael Lang in addition to a subjective element consisting in the intention to obtain a tax advantage, objective circumstances showing that, despite formal observance of the conditions laid down by Community law, the objective pursued by freedom of establishment, as set out in paragraphs 54 and 55 of this judgment, has not been achieved’.27 The Court went on to explain in more detail:28 In those circumstances, in order for the legislation on CFCs to comply with Community law, the taxation provided for by that legislation must be excluded where, despite the existence of tax motives, the incorporation of a CFC reflects economic reality. [ . . . ] That incorporation must correspond with an actual establishment intended to carry on genuine economic activities in the host Member State. [ . . . ] [T]hat finding must be based on objective factors which are ascertainable by third parties with regard, in particular, to the extent to which the CFC physically exists in terms of premises, staff and equipment.

The key words seem to be ‘economic reality’, ‘actual establishment’, ‘genuine economic activities’, and ‘the extent to which the CFC physically exists’. The Court seems to require that the complete facts have to be established. If an arrangement turns out not to reflect economic reality, not to correspond to an actual establishment, or not to exist physically, it may be ignored. This does not come as a surprise: tax authorities are well advised to establish the actual fact pattern—not only when applying the fundamental freedoms. In most Member States, the principle of legality requires the tax authorities to put much effort into carefully establishing the facts. It is rather obvious that they are not prevented from fulfilling their obligations when operating within the scope of EU law. It is remarkable that in Cadbury Schweppes the Court refers to the term ‘letterbox’.29 If checking the objective factors ‘leads to the finding that the CFC is a fictitious establishment not carrying out any genuine economic activity in the territory of the host Member State, the creation of that CFC must be regarded as having the characteristics of a wholly artificial arrangement. That could be so in particular in the case of a “letter-box” or “front” subsidiary (see Case C-341/04 Eurofood IFSC [2006] ECR I-0000, paragraphs 34 and 35).’ On the one hand, the use of the term ‘letter-box’ can be misleading.30 One might get the impression that a legal entity that has no premises, staff, or equipment may be described as a letter-box company and thus may, in a cross-border situation, be treated less favourably.31 However, the requirement imposed by the Court that the legal entity ‘physically exists in terms of premises, staff and equipment’ does not necessarily mean that legal entities without premises, staff or equipment can be completely ignored. The size of premises and the amount of staff and equipment that is required cannot be determined without taking into account the type of business a corporation operates. There does not seem to be a need for an intermediate holding company or a group financing company to have a large office space or lots of full-time employees. In a mere domestic   Cadbury Schweppes (n 1 above), para [64]; see Lang and Heidenbauer (n 3 above).   Cadbury Schweppes (n 1 above), paras [65–67]. See also Vanistendael (n 12 above) 194; and F Wassermeyer and J Schönfeld, ‘Die EuGH-Entscheidung in der Rechtssache Cadbury Schweppes und deren Auswirkungen auf die deutsche Hinzurechnungsbesteuerung’ [2006] GmbHRundschau 1065, 1066. 29   Cadbury Schweppes (n 1 above), para [68]. 30   See Lang and Heidenbauer (n 3 above) 597. 31   See H Loukota, ‘Das erste Treaty-Shopping-Urteil des VwGH’ [1998] Steuer und Wirtschaft International 105; H Loukota, ‘Einschaltung ausländischer Basisgesellschaften’ [2005] Steuer und Wirtschaft International 205, 209. 27 28

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  445 context there is no doubt that such legal entities without office space or staff are not ignored for tax purposes. On the contrary, in many tax jurisdictions, such companies are liable to minimum corporate income tax, even if they only receive tax-exempt dividends. Therefore, one cannot assume that the creation of such legal entities should be considered as wholly artificial arrangements in general.32 A closer look at Eurofood IFSC,33 which the ECJ referred to when mentioning the example of letter-box companies,34 confirms this result. In Eurofood IFSC, the ECJ had to interpret Article 3(1) of Regulation 1346/2000.35 This provision is relevant to determine the Member State where insolvency proceedings may be opened (international jurisdiction). According to this provision, in ‘the case of a company or legal person, the place of the registered office shall be presumed to be the centre of its main interests in the absence of proof to the contrary’. The Court arrived at the following conclusions:36 It follows that, in determining the centre of the main interests of a debtor company, the simple presumption laid down by the Community legislature in favour of the registered office of that company can be rebutted only if factors which are both objective and ascertainable by third parties enable it to be established that an actual situation exists which is different from that which locating it at that registered office is deemed to reflect. [ . . . ]

That could be so in particular in the case of a ‘letterbox’ company not carrying out any business in the territory of the Member State in which its registered office is situated. [...] By contrast, where a company carries on its business in the territory of the Member State where its registered office is situated, the mere fact that its economic choices are or can be controlled by a parent company in another Member State is not enough to rebut the presumption laid down by the Regulation.

From this reasoning, one could derive that a letter-box company is not a company lacking office space or staff but a company ‘not carrying out any business in the territory of the Member State in which its registered office is situated’. A company that fulfils holding or finance functions, even exclusively within a group of companies, cannot be considered a letter-box company. ‘The mere fact that its economic choices are or can be controlled by a parent company in another Member State’ is not enough to ignore the legal existence of this company as a resident of the other Member State. Thus, whenever the facts establish that the legal entity concerned carries out whatever type of business in another Member State, its activities cannot be considered a ‘wholly artificial arrangement’. A careful finding of the facts quite often proves that the conditions for applying a specific beneficial rule are not met. The Court has already dealt with cases of fraud, of the manipulation of a product to obtain export restitutions, of false declarations, or sham transactions.37 However, there are cases where the Court applies a substance-over-form approach and where it is not completely clear whether the real facts are established or   See the Opinion of AG Mengozzi in Columbus Container (n 2 above), paras [181–184].   Case C-341/04 Eurofood IFSC Ltd [2006] ECR I-3813. 34   Cadbury Schweppes (n 1 above), para [68]. 35   Council Regulation (EC) No 1346/2000 of 29 May 2000 on insolvency proceedings [2000] OJ L160/1, as amended. 36   Eurofood (n 33 above), paras [34–36]. 37   See L De Broe, ‘International Tax Planning and Prevention of Abuse under Domestic Law—Tax Treaties and EC Law: A Study of the Use of Conduit and Base Companies’ (Doctoral Thesis, Katholieke Universiteit Leuven 2007) 567. 32 33

446  Michael Lang whether the facts are recharacterised in order not to be covered by the rule which would be applied otherwise. Test Claimants in the Thin Cap Group Litigation seems to be such a borderline case. For the Court, the arm’s-length character of the transaction seems to be crucial and it remains unclear whether the Court permits the Member State to recharacterise the facts under its domestic law:38 The fact that a resident company has been granted a loan by a non-resident company on terms which do not correspond to those which would have been agreed upon at arm’s length constitutes, for the Member State in which the borrowing company is resident, an objective element which can be independently verified in order to determine whether the transaction in question represents, in whole or in part, a purely artificial arrangement, the essential purpose of which is to circumvent the tax legislation of that Member State. In that regard, the question is whether, had there been an arm’s-length relationship between the companies concerned, the loan would not have been granted or would have been granted for a different amount or at a different rate of interest. [ . . . ] As the Advocate General stated at point 67 of his Opinion, national legislation which provides for a consideration of objective and verifiable elements in order to determine whether a transaction represents a purely artificial arrangement, entered into for tax reasons alone, is to be considered as not going beyond what is necessary to prevent abusive practices where, in the first place, on each occasion on which the existence of such an arrangement cannot be ruled out, the taxpayer is given an opportunity, without being subject to undue administrative constraints, to provide evidence of any commercial justification that there may have been for that arrangement. [ . . . ] In order for such legislation to remain compatible with the principle of proportionality, it is necessary, in the second place, that, where the consideration of those elements leads to the conclusion that the transaction in question represents a purely artificial arrangement without any underlying commercial justification, the re-characterisation of interest paid as a distribution is limited to the proportion of that interest which exceeds what would have been agreed had the relationship between the parties or between those parties and a third party been one at arm’s length.

Other parts of the Court’s reasoning in Cadbury Schweppes put emphasis on the object and purpose of the provisions at stake:39 It is necessary, in assessing the conduct of the taxable person, to take particular account of the objective pursued by the freedom of establishment. [ . . . ] That objective is to allow a national of a Member State to set up a secondary establishment in another Member State to carry on his activities there and thus assist economic and social interpenetration within the Community in the sphere of activities as self-employed persons. [ . . . ] To that end, freedom of establishment is intended to allow a Community national to participate, on a stable and continuing basis, in the economic life of a Member State other than his State of origin and to profit therefrom. [ . . . ] Having regard to that objective of integration in the host Member State, the concept of establishment within the meaning of the Treaty provisions on freedom of establishment involves the actual pursuit of an economic activity through a fixed establishment in that State for an indefinite period.

Since the ECJ had to apply the freedom of establishment, it examined the object and purpose of this freedom. One might wonder whether the Court would have arrived at different results had it had to apply other fundamental freedoms, such as the free movement of capital. It would probably refer to the object of assisting ‘economic and social   Thin Cap (n 2 above), paras [81–83].   Cadbury Schweppes (n 1 above), paras [52–54].

38 39

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  447 interpenetration within the Community’ as well. However, the free movement of capital is not necessarily intended to allow someone to participate ‘on a stable and continuing basis’ in the economic life of another Member State. Furthermore, this does not necessarily involve the actual pursuit of an economic activity ‘through a fixed establishment in that State for an indefinite period’. One can therefore speculate whether the Court would define the phrase ‘wholly artificial arrangement’ in an even more restrictive way if it ever had to apply such deliberations under the free movement of capital and payments. In intragroup situations, the freedom of establishment prevails.40 However, one cannot exclude that situations occur in which the free movement of capital is still applicable. Overall, the approach followed by the ECJ is convincing. The Court identified the applicable provision—not only did it look at the letter of Article 43 EC [now Article 49 TFEU] but also at its object and purpose. In order to comprehensively determine the scope of the freedom of establishment, the Court actually took into account considerations inferred from the object and purpose of the provision. This corresponds to traditional legal methodology and is in line with general means of interpretation. The wording of a provision is only the starting point of its interpretation. One has to look at other aspects—like its object and purpose, its history, or its context—as well.41 However, in the light of these deliberations, it is remarkable that the Court has dealt with the artificiality of the arrangement at the level of justifications and proportionality and not when applying the freedom as such.42 On the contrary, the Court even found it necessary to emphasise that: [T]he fact that a Community national, whether a natural or a legal person, sought to profit from tax advantages in force in a Member State other than his State of residence cannot in itself deprive him of the right to rely on the provisions of the Treaty. [ . . . ] As to freedom of establishment, the Court has already held that the fact that the company was established in a Member State for the purpose of benefiting from more favourable legislation does not in itself suffice to constitute abuse of that freedom (see, to that effect, Centros, paragraph 27, and Case C-167/01 Inspire Art [2003] ECR I-10155, paragraph 96). [ . . . ] As noted by the applicants in the main proceedings and the Belgian Government, and by the Cypriot Government at the hearing, it follows that the fact that in this case CS [= Cadbury Schweppes] decided to establish CSTS and CSTI in the IFSC for the avowed purpose of benefiting from the favourable tax regime which that establishment enjoys does not in itself constitute abuse. That fact does not therefore preclude reliance by CS on Articles 43 EC and 48 EC (see, to that effect, Centros, paragraph 18, and Inspire Art, paragraph 98).43

It is not entirely clear why the ECJ did not consider the setting up of a wholly artificial arrangement an abuse of the fundamental freedoms, although in its own view, the object and 40   See for example, Cadbury Schweppes (n 1 above), para [33], where the ECJ refers to Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-9609, para [27] (freedom to provide services vs free movement of goods); Thin Cap (n 2 above), para [34]. See also Case C-452/04 Fidium Finanz AG v Bundesanstalt für Finanzdienstleistungsaufsicht [2006] ECR I-9521, para [48] (freedom to provide services vs free movement of capital); Case C-492/04 Lasertec Gesellschaft für Stanzformen mbH v Finanzamt Emmendingen [2007] ECR I-3775, paras [18–26] (freedom of establishment vs free movement of capital). 41   See for example, W Gassner, Interpretation und Anwendung der Steuergesetze (Vienna, Wirtschaftsverlag Orac 1972) 13; M Lang, ‘Der Normgehalt des § 22 BAO’ (2001) 4 Österreichische Steuer-Zeitung 65, 68. 42  In favour of distinguishing between the level of the application of the freedoms and the level of justification and proportionality, D Weber, Tax Avoidance and the EC Treaty Freedoms: A Study of the Limitations under European Law to the Prevention of Tax Avoidance (The Hague, Kluwer Law International 2005) 76. 43   Cadbury Schweppes (n 1 above), paras [36–38].

448  Michael Lang purpose of the freedom were relevant in this respect. In any case, because of this reasoning, the Court saw no need to distinguish its Cadbury Schweppes case law from judgments like Centros and Inspire Art. Although the level of analysis should not play a crucial role— the issues regarding the scope of the freedom, the search for a comparable situation and for an appropriate justification, and the proportionality test are closely interrelated44— examination of this issue at the justification level45 seems to have made it easier for the Court to actually look at the least restrictive measure. Had the ECJ prevented taxpayers having set up ‘wholly artificial arrangements’ from benefiting from Community law at all, it might have had difficulties including the proportionality test in its deliberations. However, there is a big difference between Centros and Inspire Art, on the one hand, and Cadbury Schweppes and Test Claimants in the Thin Cap Group Litigation, on the other hand. In Centros, ‘the provisions of national law, application of which the parties concerned have sought to avoid, [were] rules governing the formation of companies and not rules concerning the carrying on of certain trades, professions or businesses’.46 Denying a corporation access to the Internal Market by already considering the formation of a corporation a wholly artificial arrangement would have been more severe than examining the specific economic performance of the corporation. There is not necessarily a contradiction between these judgments. Accepting the formation of a legal entity under EU law does not prevent the Court from considering it a wholly artificial arrangement if specific activities are conducted through this entity. On the contrary, if the Court did not accept the validity of the formation of a legal entity that merely has its registered office in a Member State and that pursues its activities in another Member State, this would have prevented the Court from applying a differentiating approach when looking at these activities. Therefore, the assessment of the same arrangement could differ under tax law and under other areas of law. Considering an arrangement wholly artificial for tax purposes, and thus allowing CFC or thin capitalisation rules to apply, does not necessarily mean that the arrangement may as well be ignored for labour law or consumer protection law purposes.47 While the ECJ’s reasoning in Cadbury Schweppes with respect to the objective circumstances is convincing, it is less convincing that the Court asks for a subjective element as well.48 Although the Court in Cadbury Schweppes does not spend much effort in explaining what is meant by the subjective criterion, it nevertheless mentions this additional requirement for considering an arrangement ‘wholly artificial’. Accounting for intentions is in line with Emsland-Stärke and Halifax as well.49 44   See also M Lang, ‘Eine Wende in der Rechtsprechung des EuGH zu den direkten Steuern?’ in M Hebig, K Kaiser, K-D Koschmieder and M Oblau (eds), Aktuelle Entwicklungsaspekte der Unternehmensbesteuerung: Festschrift für Wilhelm H Wacker (Berlin, Erich Schmidt Verlag 2006) 365; M Lang, Die Rechtsprechung des EuGH zu den direkten Steuern (Frankfurt, Verlag Peter Lang GmbH 2007) 19; M Lang, ‘Die Rechtsprechung des EuGH zu den direkten Steuern—Planungssicherheit für den nationalen Gesetzgeber?’ in A Wagner and V Wedl (eds) Bilanz und Perspektiven zum europäischen Recht—Eine Nachdenkschrift anlässlich 50 Jahre Römische Verträge (Vienna, OGB Verlag 2007) 113, 123. 45  For a discussion of the level at which the Court scrutinises the issue of tax avoidance, see also C Böing, ‚Der Begriff des steuerlichen Gestaltungsmissbrauchs im Gemeinschaftsrecht’ (2007) 18 Europäisches Wirtschaftsund Steuerrecht, 55, 58. 46   Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459, para [26]. 47  In Cartesio, AG Poiares Maduro had difficulty bringing the ECJ’s reasoning in Cadbury Schweppes, on the one hand, and Centros and Inspire Art, on the other hand, in line with each other. See the Opinion of AG Poiares Maduro in Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641, para [26]. 48   See M Lang, ‘Rechtsmissbrauch und Gemeinschaftsrecht im Lichte von Halifax und Cadbury Schweppes’ [2006] Steuer und Wirtschaft International 273, 283. 49  For a careful analysis of the Court’s case law on the subjective criterion, see Weber (n 42 above) 190.

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  449 In academic writing it had been observed that there has been a shift in emphasis in the Court’s case law from a subjective application of the doctrine of abuse towards a more objective approach.50 In Halifax, both the Advocate-General and the ECJ restricted the subjective test in Emsland-Stärke by elaborating the objective nature of the second limb: the state of mind of the taxpayer was no longer considered relevant. Instead, the ‘essential aim’ of the transactions must be apparent from a number of objective factors. I have difficulty seeing a shift in case law. It is never possible to find out what the real intention of a human being is. One always has to look at the factual situation and try to discern why a human being acted in the way he or she did, but in the end, there is always a certain amount of guesswork involved. The well-advised taxpayer is always in a better situation, since he is able to structure his affairs in a way that establishes additional objective factors from which tax authorities or courts may draw different explanations. However, it is doubtful whether intentions should indeed play a role as an additional requirement in order to assume that an arrangement is ‘wholly artificial’. Taking into consideration the object and purpose of a law should be a regular part of any interpretation process. The relevance of this teleological factor should not depend on an additional subjective requirement. Irrespective of possible intentions, attention should be paid to the object and purpose of the law under consideration. The underlying assumption of the Court’s judgment seems to be that the letter of the law should prevail over its object and purpose. From Halifax, one gets the impression that—in the Court’s view—teleological interpretation and legal certainty contradict each other. The Court emphasised that the ‘requirement of legal certainty must be observed [ . . . ] in order that those concerned may know precisely the extent of the obligations which they impose on them’51 and thus defined abusive situations rather narrowly. However, focusing on the mere letter of the law does not provide ultimate legal certainty. On the contrary, the wording of a provision, if analysed carefully enough, usually leaves much room for heterogeneous results of interpretation. Taking into account the object and purpose of a provision, together with other means of interpretation, leads to a limitation of the number of possible different meanings. Therefore, the object and purpose of a provision should not only be taken into account if the subjective criterion is met but also in all other cases. Consequently, intention should not play a role as a separate criterion when determining the artificiality of an arrangement. It is not at all convincing to rely on the subjective element. If an arrangement is considered to be artificial, either because its real substance is different from what the taxpayer pretends it to be, or because—according to the object and purpose of the fundamental freedom—the situation should not be covered, it is difficult to understand why the legal consequence should depend on the taxpayer’s intention. Again, this would mean that in other situations that would not be covered by the fundamental freedom concerned—according to the object and purpose of the EU law provision—that the benefit had to be granted, despite its artificiality, simply because no intention of the taxpayer to escape from the tax that is normally due was on hand or could be proven. However, it should be taken into account that one does not necessarily have to conclude from cases like Emsland-Stärke that the intention of the economic operator is an independent criterion in order to determine an abusive situation. The Court expressed 50  HL McCarthy, ‘Abuse of Rights: The Effect of the Doctrine on VAT Planning’ (2007) 2 British Tax Review 160, 163. 51   Halifax (n 22 above), para [72].

450  Michael Lang the view that ‘the existence of that subjective element can be established, inter alia, by evidence of collusion between the Community exporter receiving the refunds and the importer of the goods in the non-member country’.52 In the context of this judgment, determining the intention of the taxpayer was part of the necessary exercise of properly establishing the facts. In the case of evidence of collusion between the exporter and the importer, the whole transaction can be considered a sham transaction. Thus, one could conclude from carefully analysing the facts that there was neither an export nor an import of goods. The conditions necessary to be granted an export refund had not been met at all. Since the relevance of the subjective criterion is controversial and, in my view, one cannot infer more from the remaining ‘objective’ criterion than that the fact pattern of a case has to be determined carefully and that the relevant provisions have to be interpreted in line with their object and purpose, it seems to be doubtful whether there is a specific Cadbury Schweppes line of case law. It is more convincing to take the position that ‘wholly artificial arrangements’ have to be ignored since they are not covered under the relevant provisions of EU law, either because a thoroughful characterisation of the facts shows that the factual situation is different than originally assumed, or the provisions, interpreted in the light of their object and purpose, are not applicable. This leads to the result that there is no specific abuse concept in European law. Attempts to circumvent the law have to be caught in exactly the same way as has been done for centuries in other areas of law: by means of proper interpretation. However, if one, contrary to the position taken here, assumes that there is something like a specific abuse concept in EU law, then one has to deal with the question of its impact on the domestic law of the Member States. This will be done, as a mere thought experiment, in Part III below. III. Impact of Cadbury Schweppes Case Law on the Member States

A. EU Law and Domestic Law The answer to both the question whether the Cadbury Schweppes line of case law has impact on the domestic law of the Member States and the question whether EU law has impact on the domestic law of the Member States is identical: ‘that depends’. It is clear that domestic law has to be in accordance with EU law. However, not all domestic rules are predetermined by European law. As far as EU law precisely describes what the context of domestic law has to be, there is no leeway for the domestic legislator to make its own policy decisions. In some areas, directives are so precise that they do not leave any room for domestic policy decisions. As far as EU law does not contain precise rules which have to be implemented by the domestic legislator, the domestic legislator is free to make its policy decision, as long as it pays attention to EU law requirements. These EU law requirements might stem from directives as well as from primary European law, eg the freedoms or the state aid provisions. Where the Court’s case law on ‘wholly artificial arrangements’ is relevant to the application of a precise EU law rule which has to be implemented by the domestic 52   Emsland-Stärke (n 19 above), para [53]. See also Case C-279/05 Vonk Dairy Products BV v Productschap Zuivel [2007] ECR I-0239, para [33], a case that also concerns the abuse of rights under secondary Community legislation in the field of agriculture.

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  451 legislator and does not leave room for the domestic legislator to make a policy decision of its own, this case law—whatever its content really is—has to be implemented into domestic law as well. On the other hand, if the case law on ‘wholly artificial arrangements’ has the effect of an exception to the principal EU law requirements that have to be met by domestic law, it is up to the domestic legislator to decide what to do. If the domestic legislator refuses to grant a certain benefit to a foreign taxpayer, or grant a benefit for foreign income which is usually granted to domestic taxpayers or domestic income, such a treatment does not constitute any infringement of the freedoms in the case of a ‘wholly artificial arrangement’. However, if the domestic legislator is generous enough to grant these benefits even in cases of ‘wholly artificial arrangements’, this does not constitute any infringement of EU law either. Therefore the domestic legislator is free to exclude ‘wholly artificial arrangements’ when granting these benefits but is not forced to do so. The same is true when courts, in finally deciding the case they had referred to the Court, are forced to implement a judgment of the Court on the basis of its preliminary ruling.53 Whenever a judgment of the Court leaves room for more than one solution, it is up to the domestic court to decide on one of the solutions to create domestic law that is in line with the Union requirement. Of course, the domestic legislator may decide for another solution, equally in line with EU law, for the future. However, for the case which was the reason for the referral, and for all other cases which are not yet covered by an explicit rule of the legislator, the domestic court has room to decide on the solution that is in line with the constitutional and other requirements of the domestic system of the Member State and which fits best with other rules of that State. In Gebrüder Lück the ECJ repeated the request of the domestic court and provided immediately the answer:54 The third question seeks a clarification of the consequences of the precedence of Community law, that is to say, in the present case, Article 95 of the [EEC] Treaty [later Article 90 EC and now Article 110TFEU], with regard to the provisions of national law incompatible with it. The point of the question is in particular whether the national court must hold such provisions inapplicable to the extent to which they are incompatible with Community law or whether it must declare them void as from the expiry of the period prescribed by the third paragraph of Article 95. [. . .] Although Article 95 of the Treaty has the effect of excluding the application of any national measure incompatible with it, the article does not restrict the powers of the competent national courts to apply, from among the various procedures available under national law, those which are appropriate for the purpose of protecting the individual rights conferred by Community law. Particularly when an internal tax is incompatible with the first paragraph of Article 95 only beyond a certain amount, it is for the national court to decide, according to the rules of its national law, whether the illegality affects the whole tax or only so much of it as exceeds that amount. Accordingly, it is for the court making the reference to choose a solution from among those suggested in the question and, indeed, any others.

This had been confirmed in Iannelli & Volpi:55 It is nevertheless for the national court within the framework of its own legal system to decide whether the whole of any internal taxation which is discriminatory within the meaning of Article 95 or only that part of it which exceeds the tax assessed on the domestic product is to be regarded as not payable. 53   See also S Douma, ‘Doorwerking van rechtspraak van het HvJ EG in de nationale rechtsorde’ (2008) 137 Weekblad Fiscaal Recht 1178 and oral presentation at PwC-WU Seminar on 4 December 2008 in Vienna. 54   Case 34/67 Firma Gebrüder Lück v Hauptzollamt Köln-Rheinau [1968] ECR 245, 251. 55   Case 74/76 Iannelli & Volpi SpA v Ditta Paolo Meroni [1977] ECR 557.

452  Michael Lang In IN.CO.GE the ECJ developed this line of reasoning further:56 It cannot therefore, contrary to the Commission’s contention, be inferred from the judgment in Simmenthal that the incompatibility with Community law of a subsequently adopted rule of national law has the effect of rendering that rule of national law non-existent. Faced with such a situation, the national court is, however, obliged to disapply that rule, provided always that this obligation does not restrict the power of the competent national courts to apply, from among the various procedures available under national law, those which are appropriate for protecting the individual rights conferred by Community law (see Case 34/67 Lück v Hauptzollamt KölnRheinau [1968] ECR 245).57

B.  Obligation to Implement the Court’s Abuse Concept Although the Court in Halifax mentions the phrase ‘abusive practice’ and not ‘wholly artificial arrangements’, this judgment may serve as an example for a situation where Member States are obliged to implement an abuse concept the Court developed. In Halifax, the referring court wanted to ascertain under what conditions VAT may be recovered where an abusive practice has been found to exist. The ECJ noted at the outset that no provision of the Sixth Directive deals with the recovery of VAT: That directive merely defines, in Article 20, the conditions which must be complied with in order that deduction of input taxes may be adjusted at the level of the person to whom goods or services have been provided (see the order of 3 March 2004 in Case C-395/02 TransportService [2004] ECR I-1991, paragraph 27). [ . . . ] It is therefore, as a rule, for the Member States to lay down the conditions under which the tax authorities may recover VAT after the event, while remaining within the limits imposed by Community law (TransportService, paragraph 28). [ . . . ] It is important, however, to note in that respect that the measures which the Member States may adopt under Article 22(8) of the Sixth Directive in order to ensure the correct levying and collection of the tax and for the prevention of fraud must not go further than is necessary to attain such objectives (see Gabalfrisa and Others, paragraph 52, and the order in Transport Service, paragraph 29). They may not therefore be used in such a way that they would have the effect of undermining the neutrality of VAT, which is a fundamental principle of the common system of VAT established by the relevant Community legislation (see Case C-454/98 Schmeink & Cofreth and Strobel [2000] ECR I-6973, paragraph 59). [ . . . ] It must also be borne in mind that a finding of abusive practice must not lead to a penalty, for which a clear and unambiguous legal basis would be necessary, but rather to an obligation to repay, simply as a consequence of that finding, which rendered undue all or part of the deductions of input VAT (see, to that effect, Emsland-Stärke, paragraph 56). [ . . . ] It follows that transactions involved in an abusive practice must be redefined so as to reestablish the situation that would have prevailed in the absence of the transactions constituting that abusive practice. [ . . . ] In that regard, the tax authorities are entitled to demand, with retroactive effect, repayment of the amounts deducted in relation to each transaction whenever they find that the right to deduct has been exercised abusively (Fini H, paragraph 33). [ . . . ] 56   Joined Cases C-10/97 to C-22/97 Ministero delle Finanze v IN.CO.GE. ’90 Srl et al [1998] ECR I-6307, para [21]. 57   See Case C-309/06 Marks & Spencer plc v Her Majesty’s Commissioners of Customs and Excise [2008] ECR I-2283, para [64].

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  453 However, they must also subtract therefrom any tax charged on an output transaction for which the taxable person was artificially liable under a scheme for reduction of the tax burden and, if appropriate, they must reimburse any excess. [ . . . ] Similarly, it must allow a taxable person who, in the absence of transactions constituting an abusive practice, would have benefited from the first transaction not constituting such a practice, to deduct, under the deduction rules of the Sixth Directive, the VAT on that input transaction. [...] It follows that the answer [ . . . ] must be that, where an abusive practice has been found to exist, the transactions involved must be redefined so as to re-establish the situation that would have prevailed in the absence of the transactions constituting that abusive practice.58

In essence, the Court requires the domestic authorities to apply the abuse concept developed by it, since the rules of the VAT Directives do not give any room to the domestic legislator to decide the question itself who should benefit or suffer from VAT rules. As far as the case is concerned, in which the referring court had to deal with the issue, the court complies with EU law if it applies this concept directly. However, as the Court’s case law requires the concept to be expressly implemented into domestic law, it is necessary to create a legal basis for the application of this concept in these cases. In countries where the concept developed by the Court is identical with unwritten interpretation principles, an explicit provision will not be needed. However, in countries where such principles do not exist, it might be necessary to introduce an explicit statutory provision. It has been submitted in academic writing that it should make a difference whether the Court decides a case at the level of justification or at the level of proportionality. If the ECJ had concluded that the UK CFC provisions were not fit for their avowed purpose and therefore could not be objectively justified, the provisions would be precluded from being applied. Since the Court held it was not disproportionate to apply these provisions in the case of ‘wholly artificial arrangements’, domestic courts seem to be free to apply these provisions to ‘wholly artificial arrangements’.59 However, in my view, this does not make a difference. The level of justification and the level of proportionality are exchangeable. The Court could either hold that that there are two different cross-border situations: A (no wholly artificial arrangement) and B (wholly artificial arrangement), and in situation B there is a justification for a worse treatment, while in situation A there is none, or the Court could hold that a worse treatment is justified insofar as the arrangement is wholly artificial. There is no difference in substance between presenting the issue at the justification or the proportionality level. In both examples, EU law neither prevents the domestic legislator from treating wholly artificial arrangements in cross-border situations as worse than other domestic transactions, nor forces the domestic legislator to do so. From a domestic point of view there could be valid reasons to reduce the scope of a provision which provides worse treatment for cross-border situations to wholly artificial arrangements. There is no need to go beyond what EU law requires. However, some courts seem to take the view that there are valid reasons not to apply such a provision at all and let the legislator make the law. One could never be sure whether reducing the scope of such a provision to wholly artificial arrangements would really have been the second best solution in the eyes of the legislator, had he anticipated that CFC regulations may only cover wholly artificial arrangements. The legislator could also have chosen to limit   Halifax (n 22 above), paras [90–98].   S Whitehead, ‘Practical Implications Arising from the European Court’s Recent Decisions Concerning CFC Legislation and Dividend Taxation’ (2007) 16 EC Tax Review 176, 182. 58 59

454  Michael Lang its CFC legislation even further or not to apply any such CFC provision or to combat aggressive tax planning by other means. Thus, it could be seen as ‘judicial self-restraint’ not to apply the CFC rules at all, in order to avoid the implication that the law is designed by judges. Instead, it may force the legislator to take action himself. In any case, how a domestic judge implements a judgment of the Court is not a question of EU law but of domestic law. The England and Wales High Court (Chancery Division) was obviously guided by the idea of judicial self-restraint when it decided that the UK CFC rules, as an immediate consequence of ECJ’s judgment in Cadbury Schweppes, may not be applied at all in Vodafone 2.60 Judge Evans-Lombe gave the following explanation: It seems to me that all UK taxpayers, including Vodafone, were and are entitled to be told by legislation, of which the meaning is plain, what the tax consequences for them will be if they decide to incorporate a CFC in a Member State. As it stands, unamended, Section 748(3), although its meaning, on the basis of conventional methods of construction, is plain, when viewed against the background of Article 43, is wholly misleading as to its actual effect brought about by the operation of Section 2 of the European Communities Act 1972. The CFC legislation and the motive test are of potentially wide application throughout the UK business world. To adapt the speech of Lord Hope in that case, the nature of the defect [in Section 748(3)] is such that a single solution is required that can reasonably be applied to all taxpayers. That can only be done by Parliament, or possibly by appropriate executive steps as was suggested by the House of Lords in the Fleming case. [ . . . ] In my judgment the CFC legislation, which depends on Section 747 and Section 748 for its effectiveness, must be disapplied so that, pending such amending legislation or executive action, no charge can be imposed on a company such as Vodafone under the CFC legislation. It follows that HMRC’s enquiry into Vodafone’s tax return for the Accounting Period has no legitimate purpose and should be closed.

However, there is also case law which illustrates that the Court sometimes is concerned with how an EU law obligation is implemented into domestic law and considers this to be a question of EU law as well.61 The Court seems to demand that the domestic legislator create a proper legal basis in order to implement the requirements which it has imposed. This could be seen as contradictory if one assumes that the mode of implementation is up to domestic law. However, the Court’s judgment in Biehl II has to be understood to be exceptional. In this case, the ECJ seemed to be concerned that the domestic legislator ignored its judgment. Therefore, the Court did not accept the explanation provided by the domestic legislator that equity measures were applied in order to ensure the application of the freedoms. In practice, if a domestic legislator is required to draft a provision which implements the Court’s case law, this exercise can be quite difficult since the content of the rules developed by Court are by no means clear. There are other examples, such as the implementation of the Scorpio62 judgment, which illustrate that the legislator might be tempted to simplify this task by copying the phrases of the relevant judgment.63 However, such a technique does not necessarily simplify life for the taxpayer, since all the ambiguity of the judgment has to be borne directly by him. Furthermore, by deliberately using a phrase stemming   Vodafone 2 v Her Majesty’s Revenue and Customs Commissioners [2008] EWHC 1569 (Ch).   Case C-151/94 Commission v Luxembourg (Biehl II)) [1995] ECR I-3699, para [18]. 62   Case C-290/04 KP Scorpio Konzertproduktionen GmbH v Finanzamt Hamburg-Eimsbüttel [2006] ECR I-9461. 63   See Sec 99 par 2 number 2 of the Austrian Income Tax Act (‘mitteilen’). 60 61

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  455 from a judgment of the Court, the interpretation of this phrase might become a question of interpretation of EU law as well. Thus, domestic courts might become enabled or even obliged to refer such an issue to the Court of Justice. A closer look at the Sixth VAT Directive, which had to be interpreted in Halifax, illustrates that the question whether there is an EU law obligation to implement certain rules is even more complex than indicated before. The VAT Directive contains some provisions which give room for domestic tax policy decisions. For example, Article 28(2)(a) of the Directive allows pre-existing exemptions to be maintained. In such a situation, as long as the EU law requirements are not in question, the interpretation of the relevant domestic provisions is in the exclusive competence of the domestic court and is not influenced by EU law. Therefore, there is no need to apply the abuse concept of the Court’s case law.64 C.  No Obligation to Implement the Court’s Abuse Concept There are, however, other legal situations where there is no obligation for Member States to implement anti-abuse concepts developed by the Court of Justice. The Merger Directive may serve as an example. In Kofoed,65 the referring Danish court asked whether the tax authorities may react to a possible abuse of rights, even though the national legislature has not enacted specific measures to transpose Article 11 of Directive 90/434.66 The ECJ replied as follows: Under Article 11(1)(a) of Directive 90/434, by way of exception and in specific cases, Member States may refuse to apply or withdraw the benefit of all or any part of the provisions of that directive, inter alia, where the exchange of shares has tax evasion or tax avoidance as its principal objective or as one of its principal objectives. That same provision also provides that the fact that the operation is not carried out for valid commercial reasons, such as the restructuring or rationalisation of the activities of the companies participating in the operation, may constitute a presumption that the operation has such an objective (see, to that effect, Leur-Bloem, paragraphs 38 and 39). [ . . . ] Thus, Article 11(1)(a) of Directive 90/434 reflects the general Community law principle that abuse of rights is prohibited. Individuals must not improperly or fraudulently take advantage of provisions of Community law. The application of Community legislation cannot be extended to cover abusive practices, that is to say, transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law (see, to that effect, Case C‑212/97 Centros [1999] ECR I-1459, paragraph 24; Case C‑255/02 Halifax and Others [2006] ECR I-1609, paragraphs 68 and 69; Case C‑456/04 Agip Petroli [2006] ECR I-3395, paragraphs 19 and 20; and Case C‑196/04 Cadbury Schweppes and Cadbury Schweppes Overseas [2006] ECR I-7995, paragraph 35). [ . . . ] As indicated by the Advocate General in point 59 of her Opinion, it is true that, in the main proceedings, there is some evidence which might justify application of Article 11(1)(a) of Directive 90/434. [ . . . ] However, it is necessary, as a preliminary issue, to determine whether, in the absence of a specific transposition provision transposing Article 11(1)(a) of Directive 90/434 into Danish law, that provision may nevertheless apply in the case in the main proceedings. [ . . . ]   McCarthy (n 50 above) 168.   Case C-321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I-5795. 66   Council Directive 90/434/EEC of 23 July 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States (Merger Directive) [1990] OJ L225/1–5. 64 65

456  Michael Lang In that regard, it should be borne in mind that, according to Articles 10 EC and 249 EC [now in essence Articles 4(3) TEU and 288 TFEU], each of the Member States to which a directive is addressed is obliged to adopt, within the framework of its national legal system, all the measures necessary to ensure that the directive is fully effective, in accordance with the objective that it pursues [ . . . ]. Moreover, the principle of legal certainty precludes directives from being able to create obligations for individuals by themselves. Directives cannot therefore be relied upon per se by the Member State as against individuals [ . . . ]. However, the Court observes, first, that, according to the actual wording of the third paragraph of Article 249 EC, Member States may choose the form and methods for implementing directives which best ensure the result to be achieved by those directives (see, to that effect, Commission v Italy, paragraph 51). [ . . . ] Accordingly, provided that the legal situation arising from the national transposition measures is sufficiently precise and clear and that the persons concerned are put in a position to know the full extent of their rights and obligations, transposition of a directive into national law does not necessarily require legislative action in each Member State. Likewise, as noted by the Advocate General in point 62 of her Opinion, the transposition of a directive may, depending on its content, be achieved through a general legal context, so that a formal and express re-enactment of the provisions of the directive in specific national provisions is not necessary [ . . . ]. The Court notes, second, that all authorities of a Member State, in applying national law, are required to interpret it as far as possible in the light of the wording and purpose of the Community directives in order to achieve the result pursued by those directives. Moreover, although it is true that the requirement of a directive-compliant interpretation cannot reach the point where a directive, by itself and without national implementing legislation, may create obligations for individuals or determine or aggravate the liability in criminal law of persons who act in contravention of its provisions, a Member State may nevertheless, in principle, impose a directive-compliant interpretation of national law on individuals (see, to that effect, Kolpinghuis Nijmegen, paragraphs 12 to 14, and Arcaro, paragraphs 41 and 42). [ . . . ] As noted by the Advocate General in point 63 of her Opinion, in the main proceedings it is therefore for the national court to ascertain whether there is, in Danish law, a provision or general principle prohibiting abuse of rights or other provisions on tax evasion or tax avoidance which might be interpreted in accordance with Article 11(1)(a) of Directive 90/434 and thereby justify taxation of the exchange of shares in question (see also Case 8/81 Becker [1982] ECR 53, paragraph 34). [ . . . ] If so, it will be for the national court to determine whether the conditions for the application of those national provisions are satisfied in the main proceedings.67

According to the ECJ’s reasoning in Kofoed, the absence of a Danish provision which denies the benefits of the Merger Directive to certain abusive practices would not constitute an infringement of EU law. This view is convincing: Article 11(1)(a) of the Directive does not require the Member States to exclude abusive practices from benefits the Merger Directive would provide for. According to the provision, a ‘Member State may refuse to apply or withdraw the benefit of all or any part of the provisions of Titles II, III and IV where it appears that the merger, division, transfer of assets or exchange of shares [ . . . ] has as its principal objective or as one of its principal objectives tax evasion or tax avoidance.’ There is no obligation for Member States to implement that exception. Such situations, as mentioned in Article 11(1)(a) of the Directive, are outside the scope of the Directive. According to the system of the Directive, no Member State is prevented from being more generous than is required by the Directive. Therefore,   Kofoed (n 65), paras [37–47].

67

Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective  457 there is no breach of EU law if a Member State refrains from implementing the exception for abusive situations.68 However, if a Member State exercises its right under the Directive to exclude abusive practices from the benefits granted by the Directive, it has to pay attention to the interpretation of Article 11(1)(a) provided by the Court. That does not mean that referring to the case law of the Court is the only way to implement this provision in domestic law. Member States may exclude only some of the abusive practices covered by Article 11(1)(a) of the Directive. However, they must not go beyond this provision. The same is true under the freedoms. In Cadbury Schweppes, the ECJ made it clear that Member States may only cover ‘wholly artificial arrangements’ under their CFC regimes. The phrase ‘wholly artificial arrangements’ has to be understood in the way it has been interpreted by the Court. However, no Member State is obliged to have a CFC regime at all. It is perfectly in line with EU law not to allocate foreign passive income channelled through a legal entity established in a low tax jurisdiction within the EU to the domestic shareholder, even if such an arrangement would qualify as ‘wholly artificial’. Neither does a domestic provision constitute an infringement of EU law if its content remains below the threshold of ‘wholly artificial arrangements’. However, such a provision has to be interpreted in conformity with EU law, and particularly the Court’s case law, as its scope could otherwise reach beyond ‘wholly artificial arrangements’. One way to interpret a CFC provision in accordance with EU law is to interpret this provision beyond the scope of ‘wholly artificial arrangements’ as a presumption which can be rebutted. From Test Claimants in the Thin Cap Group Litigation we learned that such presumptions do not necessarily constitute an infringement of the freedoms as long as ‘the taxpayer is given an opportunity, without being subject to undue administrative constraints, to provide evidence of any commercial justification that there may have been for that arrangement’.69 So under EU law, domestic CFC rules may survive, if they can be understood as just shifting the burden of proof to the taxpayer in situations beyond ‘wholly artificial arrangements’ and only covering ‘wholly artificial arrangements’. However, the question whether the methods of interpretation permit alteration of CFC rules in such a dramatic way, or whether those CFC rules must first be left to one side and then changed, is a question of domestic law and not of EU law. EU law is neutral as far as the techniques of its implementation into domestic law are concerned. If domestic law has to be ignored insofar as it contradicts European law, it is also a question of domestic law whether CFC rules still may be applied on ‘wholly artificial arrangements’ or whether CFC rules may not be applied at all, since no explicit legal basis exists for the application of such rules merely for ‘wholly artificial arrangements’.

68  According to Kromann Reumert (‘Recent Danish Tax Developments 2008’, 26) the Danish Ministry of Finance announced that it would not seek a ruling on this issue from the domestic court that had referred the case to the ECJ. The Ministry settled the case in favour of Mr Kofoed. In its announcement, the Ministry stated that the decision to settle the case was based on the fact that the National Tax Tribunal in a ruling from 1999 had stated that Article 11(1)(a) of the Merger Directive could not be relied upon by the Danish tax authorities as long as Article 11(1)(a) of the Directive had not been transposed into Danish law. In the opinion of the Danish ministry of taxation, this ruling prevented the Ministry from seeking a ruling against Mr Kofoed. 69   Thin Cap (n 2 above), para [82].

458  Michael Lang IV. Conclusion

It is not at all clear whether it is justified to refer to the Cadbury Schweppes line of case law as far as wholly artificial arrangements are concerned. The abuse concept developed by the Court of Justice concentrates on the object and purpose of the law. It is not required under EU law to extend a benefit which seems to be granted under the wording of a certain provision to situations where the object and purpose of the provision does not permit its application. However, this is not at all surprising. The Court implicitly has taken the position that the wording of a provision is not the end of the interpretation process, but only the starting point. First of all, determining the wording of a provision almost always requires taking into account the history of the provision, the other rules in which the provision is embedded, and the object and purpose of the provision. Therefore it is difficult to see whether there can be a clear meaning at all of the wording as such. Secondly, in order to interpret a provision, one may be required to set aside the wording if the object and purpose of the provision, or other aspects which are relevant in the interpretation process, lead in a different direction. Since the Court follows these interpretation rules regularly, the Cadbury Schweppes case law does not seem to be special. However, it is difficult to explain why the Court also developed a subjective criterion in this situation; it does not seem convincing that the object and purpose of the law be taken into account if it can be shown that taxpayers have the intention to circumvent the law if such subjective criteria is not applied to other situations as well. However, if one accepts that there is a specific Cadbury Schweppes case law, according to which, certain situations are qualified as ‘wholly artificial arrangements’, the consequences may differ from the context of EU law. If, as it is the case under the VAT Directive, Member States are required to refuse certain benefits in situations which are considered to be abusive, they have to implement the case law of the Court. However, if Member States may only deny benefits without being obliged to do so, there is no need to implement the case law of the Court in this context. Nothing prevents the Member States from being more generous to taxpayers than is required under the Merger Directive or the freedoms.

31 Cadbury Schweppes: Breach, Abuse Justification and Why They Are Different Julian Ghosh

T

his chapter is not a reply to Michael Lang’s thoughtful and thorough chapter.1 Rather, it tackles a question that has not been adequately analysed, at least to this writer’s satisfaction. The question is this: what is the difference between a finding by the Court of Justice that the particular provision of a Member State’s domestic tax law does not breach the fundamental freedoms at all and a finding that the provision does indeed breach the freedoms but the breach is justified? As is made clear below, the answer to the question is critical. However, the notion of a finding of no breach on the one hand and a justified breach on the other are often confused, not least by the Court itself. This has led, in turn, not only to confusion of language but also to the Court of Justice and the courts in the United Kingdom reaching incorrect conclusions as a matter of law. Cadbury Schweppes2 provides a useful (indeed, stark) illustration of the different conceptual bases for a finding of no breach and justified breach, while the first of its (UK) progeny, Vodafone 2,3 is an example of how conceptual confusion can lead to severe practical misconceptions. I.  The difference between no breach and a justified breach

Where a party to proceedings complains that a particular provision of a Member State’s domestic tax law (for example, in Cadbury Schweppes, the UK Controlled Foreign Companies (CFC) provisions, which attributed, broadly, the profits of subsidiaries established in low-tax jurisdictions4 to a controlling UK parent company) contravene the fundamental freedoms in the Treaty,5 the Court can reach one of three conclusions: (i) there is simply no breach of any of the fundamental freedoms; (ii) one or more of the freedoms are breached but the breach is cured by a permitted justification; (iii) the prima facie breach is not cured by any permitted justification.

  M Lang, ‘Cadbury Schweppes’ Line of Case Law from the Member States’ Perspective’, ch 30 above.   Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995. 3   Vodafone 2 v HMRC (No 2) [2008] STC 2391 (HC). 4  Here, Ireland, which taxed the profits of the two relevant subsidiaries at ten per cent. 5  In Cadbury Schweppes (n 2 above) the applicant complained that the Controlled Foreign Companies provisions contravened each of Arts 43, 49 and 56 EC [now Articles 49, 56 and 63 TFEU]. 1 2

460  Julian Ghosh The difference between conclusion (i) (no breach) and conclusion (ii) (breach but justified) is crucial. Any action which complains that a domestic provision of a Member State infringes the freedoms and thus breaches the applicant’s EU rights involves an enquiry into the definitional scope of the freedoms. Insofar as the fundamental freedoms have simply not been offended, EU rights are not engaged at all. In other words, the definitional scope of the respective freedoms is conclusive to the case. If, on the other hand, EU rights have been infringed by reason of one or more of the fundamental freedoms being interfered with, the breach caused by such interference is only cured by an explanation that a conflicting Treaty-compatible aim (the focus of the conference prompting this volume was the legitimate tackling of ‘abuse’ which seeks to distort the application of EU provisions) was being pursued. Moreover (and critically), proportionality is engaged. It is as well to recollect what the two notions of justification and proportionality entail. The express justifications for a breach of the freedoms in the Treaty, in particular the ‘public policy’ proviso, which are common to all of the provisions which define and confer the fundamental freedoms are not considered in this chapter (since the ‘public policy’ proviso does not permit a plea of loss of revenue).6 The implied category of derogations—that is, the so-called ‘mandatory requirements’—are an expressly openended7 implied category of derogations which Member States may make to qualify the fundamental freedoms. However, the category is restricted to requirements which pursue an objective which is compatible with the Treaty since it is only then that the provision (and the objective it pursues) can take precedence over the fundamental freedoms.8 In other words, a breach of fundamental freedom can only be legitimately explained by the pursuit of another (conflicting) Treaty objective.9 Proportionality holds the balance of the relationship of the fundamental freedoms and the permitted other Treaty objective which permits their infringement. So the provision must, in the language of the Court, be variously ‘necessary’,10 or ‘indispensable’11 to achieve its desired Treaty-compatible objective. Orthodoxy is that a provision is not ‘necessary’ or ‘indispensable’ (and therefore, not proportionate) if it is conceivable that a less restrictive regime could have attained the other Treaty-compatible objective (and hence breach the fundamental freedom in a less restrictive way).12 So if it is possible to have achieved the other Treaty-compatible objective less restrictively (that is, in a manner which effected less of an infringement on the fundamental freedoms), the provision is not proportional.13 So far, so obvious. But this potted summary of the mandatory requirements and proportionality reveals a critical and rarely discussed role of proportionality, which,     6   R (on the application of Professional Contractors Group Limited) v IRC [2002] STC 165, para [82], citing Case C-398/95 Syndesmos ton en Elladi Touristikon kai Taxidiotikon Grafeion v Ypourgos Ergasias [1997] ECR I-3091: ‘[A] Member State cannot rely on its own economic interest to flout the principles on which the common market is founded’.     7   Case C-120/78 Rewe-Zentral AG v Bundesmonopolverwaltung für Branntwein [1979] ECR 649, para [8].     8   Joined Cases C-60/84 and C-61/84 Cinéthèque SA et al v Fédération nationale des cinémas français [1985] ECR 2605, para [9].     9   Case 362/88 GB-INNO-BM v Confédération du commerce luxembourgeois [1990] ECR I-667, paras [15] and [19]; Case C-250/95 Futura Participations SA and Singer v Administration des contributions [1997] ECR I-2471, para [26]. 10   Case C-204/90 Hanns-Martin Bachmann v Belgian State [1992] ECR I-249, para [27]. 11   Case C-205/84 Commission v Germany [1986] ECR 3755, para [52]. 12  Case C-118/96 Jessica Safir v Skattemyndigheten i Dalarnas Län, formerly Skattemyndigheten i Kopparbergs Län [1998] ECR I-1897, para [33]. 13   Case C-76/90 Manfred Säger v Dennemeyer & Co Ltd [1991] ECR I-4221, paras [16] and [19]; Case C-19/92 Dieter Kraus v Land Baden-Württemberg [1993] ECR I-1663, para [37].

Breach, Abuse Justification and Why They Are Different  461 at least at the level of the compatibility of Member State domestic tax provisions with the Treaty, establishes a very clear hierarchy between the freedoms on the one hand and the competing Treaty-compatible aim on the other. By demanding that the freedoms are infringed as little as possible in attaining that other Treaty-compatible aim (by restricting the cure for their breach to measures which are ‘necessary’, ‘indispensable’, etc) proportionality reinforces a strong free market über-capitalist agenda for the Treaty. A finding of no breach is founded on an analysis which ascertains the definitional scope of the freedoms. If the purported interference of the freedoms is in fact outside their field of application, no more analysis is relevant or necessary. The conceptual analysis needed for a finding of a justified breach involves an additional scrutiny of the competition of different Treaty objectives and a resolution of any conflict which arises. It should now be readily apparent as to why an analysis of no breach is not only conceptually radically different from an analysis of a justified breach. It should also be apparent that an analysis which views the proper question as to whether there is breach or not can yield a very different result to an analysis which finds breach and focuses on whether that assumed breach is justified or not. Take Futura,14 where the ECJ held that the requirement to give a set of Luxembourg accounts, imposed on a Luxembourg permanent establishment of a company resident in France, breached the French resident company’s freedom of establishment. The Court observed that: In practice if [the non-resident company with the Luxembourg permanent establishment] . . . must keep, in addition to its own accounts which must comply with the tax accounting rules applicable in the Member State in which it has its seat, separate accounts for its branch’s activities complying with the tax accounting rules applicable in the State in which its branch is established . . . the imposition of such a condition which specifically affects companies having their seat in another Member State is, in principle, prohibited by Article 52 of the [EEC] Treaty [now Article 49 TFEU].15

The reason that the Court found a prima facie breach was the following: it scrutinised the interaction between the compliance rules of the French company’s home State and the State in which it had its permanent establishment (Luxembourg) and found that the cumulative burden of having to keep two sets of accounts infringed what is now Article 49 TFEU. Thus the Court found a prima facie breach which was potentially justified by the notion of ‘effective fiscal supervision’ (compliance requirements).16 But the principle of proportionality was applied by the Court as follows: [T]he competent authorities of a Member State may always request the competent authorities of another Member State to provide them with all the information enabling them to ascertain, in relation to the legislation which they have to apply, the correct amount of revenue tax payable by a taxpayer having his residence in that other Member State.17

It followed, according to the Court, that the requirement to keep accounts in Luxembourg was not ‘essential’ to the effective supervision conducted by the host State (Luxembourg) and was hence not proportional.18 Indeed, it follows that any domestic legislation relating to information-gathering on the part of the host State for compliance purposes was   Kraus (n 13 above).   Futura (n 9 above), para [26]. 16  In fact this justification has been accepted by the Court since Cassis de Dijon (n 7 above), see para [8]. 17   Futura (n 9 above), para [41]. 18   Futura (n 9 above), para [40]. 14 15

462  Julian Ghosh disproportionate.19 It may legitimately be observed that the application of proportionality by the Court in Futura to what seems to be an unexceptionable requirement to keep a set of accounts in the State in which a non-resident company has a permanent establishment reduced the notion of effective fiscal supervision to a justification which is more apparent than real. But it will be recollected that the reason that there was a prima facie breach at all in Futura was because the Court examined the interaction of the compliance requirements in the non-Luxembourg resident company’s home State (France) and the State in which it had its permanent establishment (Luxembourg) and was able to ascertain a cumulative burden (and hence, a prima facie breach of what is now Article 49 TFEU). But if the Court had restricted its scrutiny to the compliance requirements in Luxembourg, there would arguably have been no breach at all. The requirement to keep accounts was not discriminatory (it applied to all companies, resident and non-resident in Luxembourg, which were established in Luxembourg, whether by way of a primary establishment or a secondary establishment). Neither could it be realistically said to impede market access, since a requirement to keep accounts is a market participation condition, which is only engaged once the taxpayer is established in Luxembourg. Indeed, it is a curiosity that if the action had been raised by the Commission in a direct action against Luxembourg, there would have been no home State to scrutinise in order to establish a cumulative burden. And if this is right, there would have been no finding of breach and no application of proportionality to reduce the justification of effective fiscal supervision to something which is virtually illusory. Put another way, the approach of the ECJ in Futura fixed a Member State (Luxembourg) with the burden of effecting tax compliance in a cross-border circumstance by communicating with the home State authorities of a particular taxpayer established within it. This in turn raises all of the attendant problems of language barriers and translation of the taxpayer home State substantive and formal rules to a form which satisfies the compliance-effecting Member State’s compliance requirements. These problems arise despite the absence of any evidence that the keeping of accounts in Luxembourg was unduly burdensome to the particular taxpayer (Singer) which requirement, at first sight, does not appear to impede the Single Market in any realistic or intelligible sense. Indeed, the confusion of what ought to be a finding of no breach with a finding of justified breach can make the legal process verge on the unjusticiable. Take the so-called ‘Sunday trading cases’:20 the Court treated rules which restricted Sunday trading as prima facie obstacles to intra-Community trade which required justification.21 However the Court considered the rules to be justified on the basis that the retail opening hours ‘reflected certain political and economic choices to accord with national or regional socio-cultural characteristics’ and considered them to be proportionate.22 The better view is that the restrictions on Sunday trading did not represent a breach of the freedoms (free movement of goods, as it happens) at all. The restriction on Sunday trading could legitimately be observed as being neither discriminatory (it applied to all traders, national and non  Case C-136/00 Rolf Dieter Danner [2002] ECR I-8147, para [49].   Case 145/88 Torfaen Borough Council v B & Q plc [1989] ECR I-3851; Case C-312/89 Union départementale des syndicats CGT de l’Aisne v SIDEF Conforama, Société Arts et Meubles and Société Jima [1991] ECR I-997; Case C-169/91 Council of the City of Stoke-on-Trent and Norwich City Council v B & Q plc [1992] ECR I-6635. 21   Torfaen Borough Council (n 20 above), para [13]; SIDEF Conforama (n 20 above), paras [9–11]; Stoke-onTrent City Council (n 20 above), para [15]. 22   Torfaen Borough Council (n 20 above), para [14]; SIDEF Conforama (n 20 above), para [11]. 19 20

Breach, Abuse Justification and Why They Are Different  463 national in each host State). Neither can they be described as any sort of impediment to market access, since, precisely because the restricted retail opening hours applied to all traders within each host State, the market was simply defined by reference to those opening hours. There was no additional impediment to market access for any new trader. Thus the correct finding would have been no breach.23 A finding of non-breach would have been conceptually simple, elegant (and correct). The finding of justified breach by reference to ‘certain political and economic choices [which accorded with] national or regional socio-cultural characteristics’ engaged a justification which is not Treaty-compatible (the objective of Articles 2 and 3 of the EC Treaty [now in essence Articles 3 TEU and 3-6 TFEU] is to ensure that such national characteristics do not impede inter-State trade; they cannot be used as a defence to a breach of the freedoms) and begs the question as to how such national or regional socio-cultural characteristics are to be established. Even more perplexing is how proportionality is to be engaged (in what sense are the retail opening hours, restrictive on Sundays, ‘necessary’ or ‘indispensable’ in preserving these national or regional socio-cultural characteristics?). This is an example of how a case can indeed verge on becoming unjusticiable by reference to illegitimate justifications where a finding of no breach is confused with a finding of justified breach. II.  Cadbury Schweppes: A Case Study

The approach of the Court in Cadbury Schweppes throws the conceptual question of the distinction between no breach and justified breach into sharp relief. In Cadbury Schweppes, as observed above, the complaint was that the attribution of profits (but not losses) of the Irish resident subsidiaries of the UK resident parent (Cadbury Schweppes plc) contravened Articles 43,24 4925 and 5626 [now Articles 49, 56 and 63 TFEU]. The ECJ, unsurprisingly, held that the mere selection of Ireland by reference to its lower tax rate (ten per cent) by the UK company (the corporation tax rate in the UK was thirty per cent at the relevant time) did not constitute any form of ‘abuse’ and there was a prima facie breach of Article 43,27 subject to further scrutiny (by the national court) of a socalled ‘motive defence’ which was claimed by the UK (but was ultimately found not to28) attack wholly artificial structures. Crucially, however, the Court qualified its finding by observing that ‘wholly artificial structures’ would not be protected by Article 43, and in particular, that Article 43 was only engaged if the subsidiaries had: (i) premises (an ‘actual establishment’) in the host State (Ireland);29 (ii) staff and equipment in the host State;30 23  Which the Court eventually recognised in subsequent case law after Keck and Mithouard: Joined Cases C-267/91 and 268/91 Criminal proceedings against Bernard Keck and Daniel Mithouard [1993] ECR I-6097, paras [15–17]. 24   The freedom of the UK plc to establish itself in Ireland. 25   The right of the Irish subsidiaries, which were finance and treasury companies, to provide services to the UK. 26  An impediment to the capitalisation of the subsidiaries by the UK plc. 27  See Cadbury Schweppes (n 2 above), paras [43–46], [50]. Only Article 43 was addressed. The other Articles were ignored. 28   Vodafone 2 (n 3 above). 29   Cadbury Schweppes (n 2 above), paras [66, 67]. 30   Cadbury Schweppes (n 2 above), para [67].

464  Julian Ghosh (iii)  conducted ‘genuine economic activities’ in the host State (in other words, actually performed the services to which the profits purportedly arising in Ireland were properly attributable, so that the subsidiaries could not be described as ‘fictitious’, or mere ‘letter-box companies’)31. The subsidiaries must ‘reflect economic reality’.32 A simple and intelligible reading of the Court’s observations would be to view the requirements as simply examples of specific tests or guarantees that Article 43 was indeed engaged, that Article 43, at least in the context of a claim that profits were properly taxed in the jurisdiction of alleged establishment, required a guarantee of substance. A parent claiming to have established itself through a secondary establishment in the form of a subsidiary in a host State must really have established itself in that host State. In other words, a mere brass plate company would not, of itself, engage what was then Article 43 EC because to form a subsidiary as a mere brass plate company would not engage the freedom of establishment. After all, the freedom of establishment has been defined as ‘the actual pursuit of an economic activity through a fixed establishment [in the host State] for an indefinite period’.33 The various tests of having sufficient staff, premises and equipment and the reference to ‘genuine economic activities’ in contra-distinction to ‘fictitious’ and ‘letter-box’ companies simply express that test in more specific terms. Put another way, the Court was not implying that establishment without one or more of staff, premises or ‘conducting genuine economic activities’ was in some manner ‘abusive’ which enabled a disapplication of the freedoms which were prima facie engaged. Rather, the Court was simply refining the definitional scope of the freedom of establishment in the context of the case which involved a UK tax resident parent asserting Community rights in relation to its Irish tax resident subsidiaries. The Court is best seen as merely observing that if the subsidiaries failed one or more of these tests it could be said against them that they had not engaged Article 43 EC at all (and thus had no protection against the application of the Controlled Foreign Companies provisions). If the subsidiaries had no substance, they could not properly be said to have engaged Community rights (certainly not in relation to the freedom of establishment). The failure of the tests as to substance was a simple application of the definitional scope and limits of the freedom of establishment, not a disapplication of the right to establishment intra-Community. This was not, however, the language of the Court. The Court expressed its analysis of the substance tests as one relating to justification, here a justified restriction on the freedom of establishment ‘where it specifically relates to wholly artificial arrangements aimed at circumventing the application of the legislation of the Member State concerned’.34 Thus, the Court’s language was very firmly couched in an analysis of achieving a legitimate Treaty aim (countering abuse designed solely to circumvent national legislation). But this seems to confuse the notion of ‘abuse’ in the sense that a claim is simply made illegitimately (that a brass plate company with no substance engages the rights of establishment in what is now Article 49 TFEU) with a very different notion of ‘abuse’, where the rights under the freedoms are indeed engaged (in the sense that the text of the freedoms cannot be construed35 to exclude their protection for a transaction in a context which distorts and perverts the freedoms’   Cadbury Schweppes (n 2 above), para [68].   Cadbury Schweppes (n 2 above), para [65]. 33   The formulation is well established: see Cadbury Schweppes (n 2 above), para [54]. 34   Cadbury Schweppes (n 2 above), para [51]. 35  By reference to linguistic analysis or to general principles of EU Law. 31 32

Breach, Abuse Justification and Why They Are Different  465 application by reference to their purpose). This second sense of ‘abuse’ is the Halifax36 sense of abuse which has been explored by other chapters presented in this volume. There is no point in my repeating the various analyses of this second Halifax notion of ‘abuse’, since it is set out in the other chapters in this collection. But it is instructive to recollect that this Halifax sense of ‘abuse’ is only engaged when the process of construction of the text of the freedoms has been exhausted and the offending transaction still falls within their definitional scope. Put another way, the Halifax sense of abuse will generally speaking be a Member State defence to a charge of breach of the freedoms of last resort where the rights conferred by the freedoms are conferred in circumstances in which the draftsmen of the Treaty would not have conferred them (including circumstances in which rights are claimed without engaging their co-relative duties). But the prospective ‘abuse’ in Cadbury Schweppes was not the Halifax sense of ‘abuse’ at all. As has been observed, the Court expressly stated that the selection of a jurisdiction to access its tax rate (because that rate is appreciably lower than the home State rate) was no sort of abuse at all (see paragraph [49]). What the Court objected to was the prospect of the Irish resident subsidiaries claiming to have their profits taxed in Ireland when the economic reality was that those profits were attributable to activities which were truly carried on in the UK. That is why the Court was troubled by ‘the creation of wholly artificial arrangements, which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory [that is, the UK]’37 (emphasis added) and ‘practices which have no purpose other than to escape the tax normally due on the profits generated by activities carried on in national territory’38 (emphasis added). Abuse, in the eyes of the Court, was not the swapping of one tax rate for another via the establishment of a subsidiary. Abuse was the claiming of a tax rate by a parent for its subsidiary when the relevant profits were properly attributable to the home State of the parent (because the substance tests revealed the subsidiaries not to be established in the host State at all, meaning that the profits belonged to and were properly taxed by the home State of the parent). This exercise is nothing more than the refinement of the scope of the right of establishment. If the Irish subsidiaries failed the substance tests they were simply not established in Ireland and the UK Controlled Foreign Company provisions could properly be applied to the UK parent (although see further on this point below). ‘Abuse’ is simply a pejorative label for the observation that Article 49 TFEU does not protect brass plate companies. It might be thought that in the context of Cadbury Schweppes, the no breach/justified breach analysis makes a distinction without a difference. After all, the Court observed that it was a matter of evidence before the national court as to whether the subsidiaries were truly established in Ireland or not (see paragraph [70]; incidentally the language here makes it clear that the ‘abuse’ here goes to the definitional scope of what now is Article 49 TFEU). So why does it matter whether that evidence goes to the question of breach or justification? In fact it matters very much indeed. Quite apart from the necessity to deal with the application of proportionality if the question is one of justified breach,39 36   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 37   Cadbury Schweppes (n 2 above), para [55]. 38   Cadbury Schweppes (n 2 above), para [59]. 39  The question of proportionality was inadequately dealt with by the Court (see Cadbury Schweppes (n 2 above), paras [60–72]). In particular, the question as to why it is proportionate to tax profits but not relieve losses of a subsidiary within the Controlled Foreign Company regime if Community rights are indeed prima facie engaged was nowhere addressed.

466  Julian Ghosh the confusion between no breach and justified breach had led the English High Court to reach an unintelligible decision as to the application of Cadbury Schweppes to the UK Controlled Foreign Company legislation. While the practical consequences of the High Court decision was reversed by the Court of Appeal, the conflation of ‘abuse’ as a definitional factor in framing the scope of the freedoms and ‘abuse’ as a justification for a Member State in limiting the application of the freedoms remains (unsurprisingly, since the Court of Appeal merely followed the language of the ECJ in Cadbury Schweppes). The English High Court in Vodafone 2 had held that a UK resident parent company could invoke Article 43 EC (regarding certain compliance obligations) in relation to a Luxembourg resident subsidiary with a third country (Switzerland) branch without the need to adduce any evidence at all. The UK Crown had argued, perfectly unexceptionably, that the decision of the ECJ in Cadbury Schweppes required the UK parent company to lead evidence that there was no abuse in the Cadbury Schweppes sense before the parent could rely on Article 43. For, the UK Crown argued, if there was no substance in the required sense (there in Luxembourg) there were no Community rights under Article 43.40 The High Court, however, held that because the UK Controlled Foreign Company provisions impeded establishment within the Community41 a ‘single solution’ was required for all UK resident companies who relied on Article 43, which, in turn, required that all such companies could invoke Article 43 without needing to lead any evidence as to abuse.42 This approach is illegitimate on several levels. Firstly and at the highest level of abstraction, it effectively means that the Controlled Foreign Companies provisions cannot be enforced at all (ever) on the basis that the legislation is non-Treaty compliant as a matter of principle (because the motive defence does not simply attack wholly artificial structures). In other words, the English High Court has held that the UK Controlled Foreign Companies provisions fall contra mundum on the footing that a prima facie breach has occurred and cannot be justified. That is something which the Court has never said (and indeed cannot say) since it has no jurisdiction to strike down legislation of Member States but rather to declare whether they comply with the terms of the Treaty or not. A subtle defence of the High Court’s conclusion might have been made on the basis that the Court of Justice has on occasion purported to have Member State legislation ‘set aside’ by reference to general principles43 but the High Court did not provide any such explanation of its ‘single solution’ in its decision. One can add that there was no discussion of these or any other relevant ECJ authorities in relation to the ‘single solution’ adopted by the High Court. And this approach has in any event had doubt cast upon it by the Court.44 Secondly, the elevation of the question of breach of the freedom of establishment to a question of principle ignores the formulation of the very right itself by the Court as requiring the   Cadbury Schweppes (n 2 above), para [87].   The High Court, in the main part of its judgment up to para [86], rightly found that not only did the ‘motive test’ which the UK had claimed attacked only ‘wholly artificial arrangements’ in Cadbury Schweppes itself go far beyond what the UK had claimed in that case and was non-Treaty compliant, but that the relevant provisions could not be construed to make them Treaty compliant. 42   Cadbury Schweppes (n 2 above), para [89]. 43   See Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9981 and now Case C-555/07 Seda Kücükdeveci v Swedex GmbH & Co KG, ECJ Judgment of 19 January 2010, nyr; the general principle in play here would have been the principle of non-discrimination, which would have had its own conceptual difficulties, for which see J Ghosh, Principles of the Internal Market and Direct Taxation (Oxford, Key Haven 2007), Ch 4 (discrimination is not a relevant concept in cases where market access impediment is the ground of breach). 44   Case C-427/06 Birgit Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH [2008] ECR I-7245. 40 41

Breach, Abuse Justification and Why They Are Different  467 penetration of the host State economy on a stable and indefinite basis (see above). This test is necessarily fact sensitive and demands a scrutiny of the relevant evidence. And finally, the notion of a ‘single solution’ which ignores relevant evidence is in direct contradiction to the observation of the Court in Cadbury Schweppes itself (paragraph [70]) that the question of whether the breach of Article 43 was justified by reference to the substance tests was a matter of evidence for the national court to scrutinise. The paragraph [70] observation holds good, of course, if the question had been (correctly) framed as one of breach or no breach, rather than whether there was a justified breach. Had the ECJ in Cadbury Schweppes made it clear that its observations on ‘abuse’ were confined to an effective definition of the scope of the rights conferred by Article 43 (so that Article 43 only conferred protection on the UK resident parent if the substance tests were met in relation to the subsidiaries), the fact-sensitive nature of the required analysis would have been clear peradventure, although it is difficult to see quite how the observations in paragraph [70] of the Court’s judgment in Cadbury Schweppes could have failed to alert the High Court to the factual enquiry which the Court expected. The misconception of the ‘single solution’ approach might well have been avoided. The Court of Appeal45 was (properly) sceptical of the High Court’s ‘single solution’ approach.46 The Chancellor47 observed that the application of the freedoms did indeed require scrutiny of the facts and circumstances of those who sought to rely on the freedoms to protect their Community rights.48 That is, the Court of Appeal recognised that the freedoms could only be pleaded to protect Community rights by the class of persons ‘so circumstanced that the offending [domestic] provisions [which infringed the Community rights in issue] must not be invoked against them, either in particular cases or at all’.49 In other words, the Court of Appeal was perfectly alive to the unexceptionable proposition (as was, it observed, the House of Lords) that the application of the freedoms depended on the ‘circumstances’ of those pleading them as to whether the freedoms disapplied the ‘offending’ domestic provisions or not. Thus the High Court single solution was replaced as an approach by a more conceptually satisfying case-by-case, circumstance-bycircumstance approach to identify the class of persons appropriately ‘circumstanced’ so as to have engaged EU rights which are protected from infringement by domestic provisions. And in the context of ‘abuse’, the class of persons who may plead the application of the freedoms (including the freedom of establishment), to prevent the operation of domestic provisions, who are sufficiently ‘circumstanced’ to do so, will not contain those persons engaged in ‘abuse’. The contra mundum disapplication of the prospectively offending domestic provisions has (albeit obiter) been rejected as a consequence of breach.   Vodafone 2 v HMRC (No 2) [2009] STC 1480 (CA), [2009] EWCA Civ 446.   The Court of Appeal had decided that the CFC regime could be sympathetically construed so as to accommodate Community rights, including the freedom of establishment as articulated by the ECJ in Cadbury Schweppes. Such a sympathetic construction allowed for the Court’s caveat of a justification based on abuse (see paras [28–60]). Thus the Court of Appeal’s observations on disapplication were obiter. However the Chancellor was clear that ‘[he] would need a good deal of persuading that it was appropriate simply to disallow [that is, disapply] the CFC in favour of [the taxpayer pleading for such disapplication] in the relevant accounting period . . .’, in the light of the approach of the House of Lords in Autologic Holdings plc v IRC [2005] STC 1357 (HL), para [17]; and Joined Cases Michael Fleming (trading as Bodycraft) v HMRC and Condé Nast Publications Ltd v HMRC [2008] 1 WLR 195 (HL), para [49]: see paras [62–66], especially para [66]. 47  With whom the other members of the Court of Appeal agreed. 48   Fleming/Condé Nast (n 46 above), para [62], citing Autologic (n 46 above), para [17] and Fleming/Condé Nast (n 46 above), para [49]. 49   Citing Lord Walker in Fleming/Condé Nast (n 46 above), para [49]. 45 46

468  Julian Ghosh To be sure, the notion of ‘abuse’ as a justification, rather than a definitional component of the scope of the freedoms firmly remains. This is clear from the observation of the Court of Appeal that the ECJ’s notion of abuse in Cadbury Schweppes was a ‘limitation on the freedom of establishment constituted by the justification found by the ECJ to be permissible . . . ’.50 Given the language of the ECJ in Cadbury Schweppes, this is wholly unsurprising. One supposes that one should be grateful that the confusion resulting from the Court’s approach to ‘abuse’ which gave rise to the High Court’s ‘single solution’ has at least been put to rest. What remains, however, is the spectre of proportionality holding the balance as to when and what type of ‘abuse’ legitimately permits the freedom of establishment to be infringed, which as observed above,51 has already caused Member States unexpected and, it is submitted, over-burdensome problems in relation to provisions which do not, at least at first sight, impede the Single Market at all. What emerges from Cadbury Schweppes is not only that ‘abuse’ means different things in different contexts but also that confusion of language in relation to ‘abuse’ can disguise a more profound confusion between a finding of whether there is or is not breach and a finding of justified breach. That confusion is conceptually unsatisfactory and gives rise to very practical misconceptions at the level of national court proceedings.

  Fleming/Condé Nast (n 46 above), para [65].  In the context of cases such as Futura (n 9 above): see above.

50 51

32 A Single Principle of Abuse in European Union Law: A Methodological Approach to Rejecting a Different Concept of Abuse in Personal Taxation1 Ana Paula Dourado I.  How the interpretative principle of abuse of law is applicable to taxation and personal taxation

M

y initial claim in this chapter is that there is one single principle of abuse of law in EU law (ie abuse of EU law), although its application to different fields of law, or taking into account the degree of vagueness of the EU principle or rule abused, may imply construction of second-level principles or of different guiding criteria applicable on a case-by-case basis, and adjustments resulting from a balanced application of several EU law principles leading to opposite results (eg the principles of equality, legal certainty and cohesion).2 In its broadest meaning, there is abuse of EU law when there is an accrual of advantages in a manner that conflicts with the purposes and aims of European law provisions.3 However, taking advantage of an EU right in order to benefit from a more favourable domestic legal system is not a misuse of EU law (ICI,4 Emsland-Stärke,5 Opinion of

1  The manuscript was finalised in October 2008. It was brought in line with the Treaty on the Functioning of the European Union, but subsequent literature and case-law of the Court of Justice were not taken into consideration. 2  Claiming in a different or even opposite sense: W Schön, ‘Abuse of Rights and European Tax Law’ in J Avery Jones, P Harris and D Oliver (eds), Comparative Perspectives on Revenue Law: Essays in Honour of John Tiley (Cambridge, Cambridge University Press 2008) 96; F Vanistendael, ‘Halifax and Cadbury Schweppes: One Single European Theory of Abuse in Tax Law?’ (2006) 4 EC Tax Review 195; R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax?’ (2008) 45 Common Market Law Review 395. 3  See AG Poiares Maduro in Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609, para [63]. On whether there is a common notion of abuse to the Member States, see P Harris, ‘Abus de Droit in the Field of Value Added Taxation’ (2003) 2 British Tax Review 131. 4  Case C-264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I-4695, para [26]. 5  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569.

470  Ana Paula Dourado Advocate-General Geelhoed in Akrich,6 Chen,7 Centros,8 Gemeente Leusden and Holin Group,9 Cadbury Schweppes10).11 The principle of abuse is an interpretative principle12 and in EU tax law it is an instrument with different purposes. It aims at determining whether there is abuse of EU law and at reciprocally limiting the exercise of rights conferred by European law, either directly (there is abuse of EU law ‘when Community law provisions are relied upon in order to gain advantages in a manner that conflicts with the purposes and aims of those provisions’13) or indirectly (there is abuse of EU law ‘when Community law provisions are abusively invoked in order to evade national law’14). It is therefore a method of interpreting other EU law principles and rules that directly or indirectly grant rights connected to taxes, operating in a similar (but not identical) way to a General Anti-Abuse Rule (GAAR), and which aims at achieving the principle of equality in EU tax law. Although the approach of the Court of Justice regarding the fight against abusive practices is always formal (whether there is abuse; whether there is a domestic restriction; whether it is justified by public interest reasons; whether it is proportional:15 cf Halifax, Part Service,16 Cadbury Schweppes, Thin Capitalisation Group Litigation17 or de Lasteyrie du Saillant18 and N;19 and also Marks & Spencer,20 Rewe,21 Oy AA22), the underlying reason for that fight is equality. In this way, the principle of abuse aims at defining the core and boundaries of the EU legal right. The Court’s case law on the concept of abuse is progressively related to the

    6   AG Geelhoed in Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9607, para [96].     7  Case C-200/02 Kunqian Catherine Zhu, Man Lavette Chen v Secretary of State for the Home Department [2004] ECR I-9925, paras [36–40].     8  Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459, paras [24, 27].     9   Joined Cases C-487/01 Gemeente Leusden v Staatssecretaris van Financiën and C-7/02 Holin Groep BV cs v Staatssecretaris van Financiën [2005] ECR I-5337, para [79]. 10  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995, paras [57, 64, 65]. 11   Although the case law before Emsland-Stärke and Halifax accepted a much broader concept of abuse, connected to the so-called ‘u-turn transactions’, see Case 33/74 Johannes Henricus Maria Van Binsbergen v Bestuur van de Bedijfsvereniging voor de Metaalnijverheid [1974] ECR 1299; Case C-211/91 Commission of the European Communities v Kingdom of Belgium [1992] ECR I-6773; Case C-148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I-487; and Case C-23/93 TV 10 SA v Commissariaat voor de Media [1994] ECR I-4795. For more details, see de la Feria (n 2 above) 395. 12   AG Tesauro in Case C-367/96 Alexandros Kefalas et al v Elliniko Dimosio (Greek State) Organismos Ikonomikis Anasinkrotisis Epikhirision AE (OAE) [1998] ECR I-2843, paras [18–27]; AG Poiares Maduro in Halifax (n 3 above), para [62]. 13   Halifax (n 3 above), para [63]. 14   Halifax (n 3 above), para [63]. 15  On the proportionality of anti-abuse measures in direct taxation case law, see A Zalasinski, ‘Proportionality of Anti-Avoidance and Anti-Abuse Measures in the ECJ’s Direct Tax Case Law’ (2007) 35 Intertax 310. 16  Case C-425/06 Ministero dell’Economia e delle Finanze, formerly Ministero delle Finanze, v Part Service Srl, company in liquidation, formerly Italservice Srl [2008] ECR I-897, para [46]. 17  Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107. 18  Case C-9/02 Hughes de Lasteyrie du Saillant v Ministère de l’Économie, des Finances et de l’Industrie [2004] ECR I-2409. 19  Case C-470/04 N v Inspecteur van de Belastingdiesnt Oost/kantoor Almelo [2006] ECR I-7409. 20  Case C-446/03 Marks & Spencer plc v David Halsey (Her Majesty’s Inspector of Taxes) [2005] ECR I-10837, para [57]. 21  Case C-347/04 Rewe Zentralfinanz eG v Finanzamt Köln-Mitte [2007] ECR I-2647, paras [51–52]. 22  Case C-231/05 Oy AA [2007] ECR I-6373, paras [58–60].

A Single Principle of Abuse in European Union Law  471 aim of establishing the Internal Market.23 According to the Court, that concept should not be misused by Member States in order to create distortions of the Internal Market.24 In the case of non-harmonised matters, it establishes the core and boundaries of the fundamental freedoms and it is therefore a limit to the exercise of these freedoms (see Cadbury Schweppes and de Lasteyrie du Saillant and N). In the case of harmonised matters, it operates in exactly the same way, but since the determinacy of the possibly abused rule is larger in comparison to the vagueness of the fundamental freedoms principles, it is easier for the Court to define the core and boundaries of the possibly abused harmonised rules and to reciprocally come to a conclusion on whether or not there has been abuse in harmonised matters.25 But determinacy/ indeterminacy is a quantitative issue and harmonised tax matters also contain vague concepts (see the issues discussed in Halifax,26 Denkavit,27 and Leur-Bloem28). To consider the example of Halifax, under Article 5(1) of the Sixth VAT Directive: ‘“[s]upply of goods” shall mean the transfer of the right to dispose of tangible property as owner’. Article 6(1) defines ‘[s]upply of services’ as ‘any transaction which does not constitute a supply of goods within the meaning of Article 5’. From the perspective of a tax lawyer, what the Advocate-General proposed and the Court did in Halifax corresponds to a ‘substance over form’ interpretation, which is, after all, an interpretation according to the purpose of the provisions—the supply of goods and services.29 The same is applicable to the interpretation of ‘valid economic reasons’ within Article 11 of the Mergers Directive (cf Leur-Bloem30) or to the meaning of ‘leasing arrangement’ in Part Service.31 In any case, the greater the vagueness of the EU principle or rule, the more cautious the Court is in safeguarding the effectiveness of European law.32 Moreover, the Court has to search for typical situations and criteria similar to the cases under analysis in order to reduce the vagueness of principles and rules, and in order to achieve results compatible with the rule of law. What has been called ‘typifying’ by legal theory is a way to describe interpretation both in civil and common law countries. Interpretation implies a circular movement: departing from the individual case, to the typical one foreseen in the rule, and then back to the individual case. In this way, criteria

23  See for example, Leur-Bloem (n 28 below), para [45]. On abuse as ‘a limitation of or an exception to the freedom’ see AG Lenz in TV 10 (n 11 above), para [25]. 24  Schön (n 2 above) 82; Leur-Bloem (n 28 below), para [45]. 25   M Lang and S Heidenbauer, ‘Wholly Artificial Arrangements’ in L Hinnekens and P Hinnekens (eds), A Vision of Taxes Within and Outside European Borders: Festschrift in Honour of Prof Dr Frans Vanistendael (The Hague, Kluwer Law International 2008) 609. The authors make a departure from that distinction but conclude that the type of legal instrument is not so important after all. Object and purpose of the rule are the decisive element, according to them. 26   Halifax (n 3 above). For a critical review of the case, see J Peacock QC, ‘The Law Ends Where Abuse Begins’ (2001) 4 EC Tax Journal 142. 27   Joined Cases C-283/94 Denkavit Internationaal BV, C-291/94 Vitic Amsterdam BV and C-292/94 Voormeer BV v Bundesamt für Finanzen [1996] ECR I-5063. 28   Case 28/95 Leur-Bloem v Inspecteur der Belastngdienst/Ondernemingen Amsterdam 2 [1997] ECR I-4161, para [41]. 29   Authors tend to consider that the Court treats abuse of harmonised matters and abuse of the fundamental freedoms in a different way: Schön (n 2 above) 81; see the discussion in de la Feria (n 2 above) 425, 430. 30   Leur-Bloem (n 28 above), paras [37, 46]. 31   Leur-Bloem (n 28 above), paras [8, 46]. 32   In this sense, Schön (n 2 above) 80–81.

472  Ana Paula Dourado can be found to progressively reduce the vagueness of law,33 including of course the vagueness of the principles of the fundamental freedoms. In this respect, the wholly artificial arrangement test operates in EU tax law as the classical business purpose test (although the first test is broader as it covers Articles 18 and 21 TFEU – ex Articles 12 and 18 EC): it operates through typifying criteria— recommended objective elements aimed at guiding national courts in applying the abuse concept and at allowing, in the same way, the reduction of discretionary application and the according reduction of legal uncertainty.34 But the same is applicable to some antiabuse rules in harmonised matters: the ‘valid economic reasons’ test of Article 11 of the Mergers Directive (cf Leur-Bloem35) needs to be progressively typified as it is a vague formula. In the eternal tension that law faces, in either considering the single case (by using general principles, or by being vague, leading to legal indeterminacy) or the typical case (the average typical case or the frequent typical case) through a typified method that drastically reduces or eliminates discretionary application (using legal fictions or irrebuttable presumptions),36 the principle of abuse of EU law clearly favours the first-mentioned purpose, in contrast to certainty, as it implies a case-by-case analysis.37 The fact that the Court has accepted abuse of EU law as an interpretative principle in tax matters may be regarded with some suspicion in tax law, since legal certainty, prohibition of analogy, and fear from administrative and judicial discretion have been accompanied, in some Member States, by methods of interpretation that follow the literal element very closely.38 Moreover, tax law imports most of its concepts from other legal fields, eg from private law, and that raises the issue regarding the dependent versus autonomous meaning of the imported concepts in tax law (eg provision of services, dependent worker, transfer of residence).39 In order to fulfil constitutional requirements concerning the competence to enact tax law and related demands of legal determinacy, in recent decades in many Member States control of abuse of tax law has normally been dealt with by a General Anti-Abuse Rule (hereinafter, also GAAR). The GAAR awards legitimate control of abuse and the vagueness of the GAAR is progressively reduced by case law. In other words, a General Anti-Abuse Rule gives legitimacy to the tax administration and the tax courts to use the principle of abuse as an interpretative tool and progressively define the scope and borders of tax rules. In this sense, a GAAR is a formal tool to use in the principle of abuse. Thus, it is not surprising that, in some Member States, application of the Halifax decision by national courts requires application of the (domestic) GAAR,40 and that the 33  See AP Dourado, O Princípio da legalidade fiscal: Tipicidade, conceitos jurídicos indeterminados e margem de livre apreciação (Lisbon, Almedina 2007) 556 and the literature therein; A Kaufmann, Grundprobleme der Rechtsphilosophie (Munich, Beck 1994) 112–33; F Bydlinski, Juristische Methodenlehre und Rechtsbegriff, 2nd edn (New York, Springer-Verlag Kg 1991) 543. 34  See for example, Cadbury Schweppes (n 10 above), para [61]; Thin Cap (n 17 above), para [82]. 35   Leur-Bloem (n 28 above), para [41]. 36   Dourado (n 33 above) 612, 643. 37  Tax law literature is normally impatient with the Court’s (to a certain extent, unavoidable) methodology: see for example, R de la Feria, ‘The European Court of Justice’s Solution to Aggressive VAT Planning—Further Towards Legal Uncertainty?’ (2006) 1 EC Tax Review 27. 38  See K Vogel, ‘Steuerumgehung nach innerstaatlichem Recht und nach Abkommensrecht’ (1985) 4 Steuer und Wirtschaft 371–72. Cf F Zimmer, ‘Form and Substance in Tax Law, General Report’ in International Fiscal Association, Cahiers de Droit Fiscal International (The Hague, Kluwer 2002) 37–38, 50. 39   Zimmer (n 38 above) 25–28, 56–57. 40  See Schön (n 2 above) 78.

A Single Principle of Abuse in European Union Law  473 Court has recognised application of domestic anti-abuse concepts as long as that does not prejudice the ‘full effect and uniform application of the Community law provisions allegedly relied upon in an abusive manner’.41 Besides, if domestic law does not contain an anti-abuse concept, EU law does not step in automatically, unless it is possible to solve the problem by interpretation. A GAAR goes further than the interpretative principle of abuse, as beyond applying it, it also gives legitimacy to the tax administration and courts requalifying the transaction,42 according to discretionary assessments. How far a GAAR plays a distinct role from the abuse principle depends on the accepted methods of interpretation of tax law issues, and that varies within the different legal systems. In the context of personal taxation, abuse of law operates exactly in the same way: whether the rule referred to the Court is a domestic one or part of an EU directive, abuse of law is an interpretative principle of EU law and as such it defines the meaning and scope of the EU rules conferring a tax right. Both the Court of Justice and national courts, when interpreting the compatibility of potentially restrictive national rules with EU law, are to apply the principle of abuse in EU law. The Court of Justice and the national courts apply other relevant principles of EU law, such as the principle of equality, certainty and cohesion, and include any principles of domestic law that are accepted by EU law, and this may in the future imply different results according to the field of law under analysis. Playing on the field of legal pluralism, the Court accepts international tax law principles and domestic tax law principles, as long as they are not incompatible with it. However, I will discuss the issue as to whether the Court attributes a more reduced scope to the exercise of the fundamental freedoms by an individual, and conversely, a broader scope to the abuse concept, than when corporations or similar entities are involved. II. The ‘wholly artificial arrangement’ test as a second-level principle to determine abuse

If we compare potential discriminatory measures in cases referred to the Court involving free movement of services, such as Safir,43 Danner,44 and Skandia,45 with cases involving freedom of establishment such as de Lasteyrie du Saillant, and N and one case involving free movement of citizens (Schempp46), we can find references to the ‘fiscal vacuum’ argument in the first three cases. In contrast, in de Lasteyrie du Saillant47 and N, the Court searches for possible abusive behaviour and furthermore in the context of the cohesion argument (preserving the allocation of the power to tax between Member

41   AG Poiares Maduro in Halifax (n 3 above), para [65]; Case C-206/94 Brennet AG v Vittorio Paletta [1996] ECR I-2357, para [25]; Kefalas (n 12 above), paras [21–22]; Case C-373/97 Dionisios Diamantis and Elliniko Dimosio (Greek State) [2000] ECR I-1705, paras [34–35]; Centros (n 8 above), paras [24–25]. 42  See in this sense, Lang and Heidenbauer (n 25 above) 605. 43  Case C-118/96 Safir v Skattemyndigheten i Dalarnas Län [1998] ECR I-1897. 44  Case C-136/00 Rolf Dieter Danner [2002] ECR I-8147. 45  Case C-422/01 Skandia, Ramstedt v Rikskatteverket [2003] ECR I-6817. 46  Case C-403/03 Egon Schempp v Finanzamt München V [2005] ECR I-6421. 47   De Lasteyrie (n 18 above), paras [24–26, 54].

474  Ana Paula Dourado States) (N48) and in Schempp49 non-deductibility of maintenance expenses paid to a nonresident is not even considered discriminatory. It is possible that the fiscal vacuum or the cohesion argument underlies the reasoning in Schempp, or at least justifies the different treatment granted by domestic legislation, but no reference to that argument is made, since, according to the Court, ‘the payment of maintenance to a recipient resident in Germany cannot be compared to the payment of maintenance to a recipient resident in Austria’.50 Moreover, in Van Hilten51 (a case concerning free movement of workers and not free movement of capital, according to the Court), no reference was made to potential abuse behaviour that would justify the Netherlands’ restrictive regime under analysis (if it had been considered restrictive by the Court). Taking into account existing case law, it could be tempting to say that the ‘wholly artificial arrangement’ test is only applicable to the freedom of establishment (cf ICI,52 Lankhorst-Hohorst,53 de Lasteyrie du Saillant,54 Marks & Spencer,55 Rewe,56 Cadbury Schweppes,57 Thin Cap58) and free movement of workers (cf Emsland-Stärke59). I propose, on the contrary, that there is no difference of analysis according to the different freedoms and that the ‘wholly artificial arrangement’ test is applicable to verify the exercise of any freedom and is applicable not only in the context of personal taxation, but to any taxes60 and in all fields of law. In fact, the test regarding the artificiality of the situation (or effective relocation of goods, services, persons and capital) may be used and has been used in non-specific tax matters (see General Milk Products,61 Emsland-Stärke,62 Ninni-Orasche63). Let us test my claim in the field of personal taxation. Taking into account the source and the residence elements as the currently relevant connecting elements in direct taxation (and nationality in the case of property taxes), tax abuse of fundamental freedoms operates by manipulation of one of those elements. It can also occur by manipulation of the ownership of income (EU national or not), qualification of income (interest versus dividend) and attribution of income (profits/losses). Let us take the example of free movement of services. In direct tax law, free movement of services precludes any discrimination according to the residence of the beneficiary and the provider of the service (free movement of services concerns the right of EU nationals to provide and receive services as long as two Member States are involved: see Bent Vestergaard64).   N (n 19 above), paras [42, 47].   Schempp (n 46 above), para [35]. 50   Schempp (n 46 above), para [35]. 51  Case C-513/03 Heirs of MEA van Hilten- van der Heijden v Inspecteur van de Belastingdienst /Particulieren/ Ondernemingen buitenland te Heerlen [2006] ECR I-1957. 52   ICI (n 4 above), para [26]. 53  Case C-324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I-11779, para [37]. 54   De Lasteyrie (n 18 above), para [50]. 55   Marks & Spencer (n 20 above), para [57]. 56   Rewe (n 21 above), paras [51–52]. 57   Cadbury Schweppes (n 10 above), para [51]. 58   Thin Cap (n 17 above). 59   Emsland-Stärke (n 5 above), paras [52–54]. 60  See Part Service (n 16 above), paras [17, 53]. 61  Case C-8/92 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas [1993] ECR I-779, para [21]. 62   Emsland-Stärke (n 5 above), para [53]. 63  Case C-413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I-13187, paras [34–36]. 64   Akrich (n 6 above), paras [55–57]. Case C-55/98 Skatteministeriet v Bent Vestergaard [1999] ECR I-7641. 48 49

A Single Principle of Abuse in European Union Law  475 Abuse of free movement of services could occur if an individual taxpayer claimed application of a tax regime—such as deduction of expenses in the State of residence regarding payment of services provided by a resident of another Member State—and in fact either manipulated the freedom to reside in a Member State65 or the provider had manipulated this and also therefore the source of the income (if that source is determined, as it frequently is, by the place of residence of the paying agent). Besides, the location of a permanent establishment (it can also be a paying agent, of course) can be manipulated. Moreover, abuse can occur if an entity taxed under corporate tax law in a Member State claims to be the beneficial owner of the income and of the free movement, but instead is only an agent or nominee and the taxpayer is an individual resident outside the EU. And finally, in the presence of associated persons, attribution of income and expenses can be abusively manipulated. The same is applicable to the movement of capital between two Member States or between a Member State and a third State—manipulation of the connecting elements residence or source (where again this often corresponds to the place of residence of the paying agent), of the attribution of income, or of the owner of the income, can lead to an abuse test. Thus, in the context of personal taxation (and income taxation in general), a wholly artificial arrangement test is relevant not only with regard to the freedom of establishment, but also when any of the other freedoms comes into play, although this acquires a special meaning according to types of cases (exit taxes, transfer of residence, controlled foreign companies (CFC), thin capitalisation, discriminatory taxation of services, dividends, interest, property), operating through the creation of objective criteria, as secondary EU legislation has already exemplified. Independent of discussing whether or not some or most of the concepts chosen are the best objective criteria to prevent tax abuse, examples of those concepts used in the directives are: the paying agent or the beneficial owner, in respect of the Savings Directive;66 the effective residence of the taxpayer in a Member State and taxpayers subject to taxation (Article 2(1)(b) and (c) of the Parent-Subsidiary Directive);67 the two- year holding requirement under Article 3(2) of the Parent-Subsidiary Directive; and the ‘valid economic reasons’ in Article 11(1)(a) of the Merger Directive.68 Since I claim that the wholly artificial arrangement test is a universally applicable test to check whether there is abuse of EU law, the question may be raised as to the reasons why it was not applicable in Safir, Skandia, Danner, Schempp and Van Hilten. Whereas in N, one of the issues raised was whether Mr N had actually exercised his right to establish himself in another Member State, and therefore whether he had actually transferred his residence or whether he had artificially done so in order to avoid the Netherlands taxes on capital gains, in the above-mentioned cases, no issue regarding artificial residence or source of income was raised. 65  See, on the meaning and scope of freedom of establishment by an EU citizen, AG Kokott in N (n 19 above), paras [45–57]. 66  Council Directive 2003/48/EC of 3 June 2003 on taxation of savings income in the form of interest payments (Savings Directive) [2003] OJ L157/38. 67  Council Directive 2003/123/EC of 22 December 2003 amending Directive 90/435/EEC on the common system of taxation applicable in the case of parent companies and subsidiaries of different Member States (Parent-Subsidiary Directive) [2004] OJ L007/41–44. 68  Council Directive 90/434/EEC of 23 July 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States (Merger Directive) [1990] OJ L225/1–5.

476  Ana Paula Dourado Let us take Safir as an example: according to §50 of the Swedish Premium Tax Law, the tax administration may, at the request of the policyholder, grant an exemption from payment of tax or reduce the tax by half if the company with which the insurance was taken out is subject, in the State in which it is established, to revenue tax comparable to that payable by insurance companies in Sweden.69 Since there was no manipulation of connecting elements, no artificial arrangement or business purpose test was conducted here. The Swedish government did not argue on the basis of abuse, but claimed instead a risk of fiscal vacuum (double non-taxation) in order to try to justify the different treatment. III. Balancing different legal principles in different fields of law

Since I claim that there is only one principle of abuse in EU law—and that can be justified by the absence of a hierarchy of domestic rules from the perspective of their compatibility with EU law—the issue may be raised whether that principle does not acquire special meanings according to different fields of law (eg private law versus public law or company law versus tax law).70 In fact, the constitutional prohibition of retroactivity for tax rules and for criminal offence rules does not have the same meaning/scope; the constitutional requirements of legal determinacy of a tax rule, of a rule that restricts a fundamental right, and of a rule that defines criminal conduct are not the same. Conditions for being subject to a legal regime are not necessarily the same in company law and in tax law, either71 and therefore the objective criteria for verifying whether there is an artificial arrangement may vary.72 However, this does not imply either a different concept of abuse nor inapplicability of the artificial arrangement test,73 as also follows from Court’s case law: see for example Emsland-Stärke,74 Ninni-Orasche,75 Akrich,76 Part Service,77 and the Opinion of AdvocateGeneral Poiares Maduro in Cartesio.78 It is therefore sensible to admit that the scope of free movement and of abuse and the concrete application of the ‘artificial arrangement test’ may vary according to the field of law under analysis, since each of them is ruled by different principles accepted by EU law,79 and in a concrete situation, articulation of different relevant principles implies a specific balance and may lead to different results.   Safir (n 43 above), para [11].  Schön (n 2 above) 85; W Schön, ‘Playing Different Games? Regulatory Competition in Tax and Company Law Compared’ (2005) 42 Common Market Law Review 342, 345. 71  See Schön (n 2 above) 84–85. 72   Cf in a similar sense, Lang and Heidenbauer (n 25 above) 608. 73  See KE Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 424, 449. In the opposite sense, de la Feria (n 2 above) esp 417. 74   Emsland-Stärke (n 5 above), paras [51–54]. 75   Ninni-Orasche (n 63 above), paras [28–31]. 76   Akrich (n 6 above), para [55]: ‘effective and genuine activity’. See more references to cases on free movement of workers in P Craig and G de Búrca, EU Law, Text, Cases, and Materials, 4th edn (Oxford, Oxford University Press 2008) 752–58. 77   Part Service (n 16 above), para [53]. 78   AG Poiares Maduro in Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641, para [29]. 79  See for example, Schön (n 2 above) 342–45. 69 70

A Single Principle of Abuse in European Union Law  477 Even if, in its broad meaning, the principle of abuse is indifferently applicable when a domestic rule restricts the freedom of establishment of a company for reasons that are not related to the tax burden (Centros,80 Cartesio81), and when it does, it does so in order to prevent tax abuse, application of the artificial arrangement test may vary. It must be stressed that the Court’s approach will not be consistent if the principle of abuse ceases to apply when tax treaties come into play, and if anti-abuse clauses within a tax treaty applicable between Member States will not be subject to the same tests (which is the case with limitation of benefits (LOB) clauses, according to the Court in ACT Group Litigation,82 although the issue raised involved third States). Some difficult issues may arise if protocols signed by the contracting States expressly allow application of domestic anti-abuse clauses, as it is not clear whether the Court is to analyse them as domestic rules or as tax treaty rules. Lastly, when third States are involved, the Court is more tolerant when assessing compatibility of anti-abuse rules with free movement of capital (cf Thin Cap, A83 and the Orange European Small Cap84) but, contrary to what happens in respect of tax treaties, the Court is not inconsistent in its methodological approach, as this is the same as that mentioned above (abuse and anti-abuse are handled as justifications for restrictions). IV. Fiscal Vacuum as part of the concept of abuse

Taking the Court’s reasoning into account, it seems that ‘fiscal vacuum’ is also admitted as an argument to interpret EU law and it is legitimate to ask whether it is being considered as part of a broad concept of abuse.85 However, fiscal vacuum is neither connected to the business purpose, nor to the valid economic reasons86 or to the wholly artificial arrangement tests. According to paragraph [33] of Safir: Other systems which are more transparent and are also capable of filling the fiscal vacuum referred to by the Swedish Government, whilst being less restrictive of the freedom to provide services, are conceivable, in particular a system for charging tax on the yield on life assurance capital, calculated according to a standard method and applicable in the same way to all insurance policies, whether taken out with companies established in the Member States concerned or with companies established in another Member State.

The non-deductibility of maintenance amounts paid by a resident to a non-resident, as occurs in the Schempp case, could possibly be justified due to a risk of a fiscal vacuum. Since neither in Danner, Skandia and Safir, nor in Schempp, is an abuse test mentioned, the question is whether the Court—or a particular Chamber of the Court—is not accepting the cohesion argument or the fiscal vacuum argument as a sort of irrebuttable presumption that may restrict rights granted by EU law, such as the fundamental freedoms.

 See Centros (n 8 above), para [24] and the case law quoted in respect of several law fields.   Centros (n 8 above), para [27]. 82  Case C-374/04 Test Claimants in Class IV of the ACT Group Litigation [2006] ECR I-11673. 83  Case C-101/05 Skatteverket v A [2007] ECR I-11531. 84  Case C-194/06 Staatssecretaris van Financiën v Orange European Smallcap Fund NV [2008] ECR I-3747. 85  See AG Poiares Maduro in Halifax (n 3 above). 86   Cf Danner (n 44 above), para [56], Skandia (n 45 above), para [53] and Safir (n 43 above), para [33]. 80 81

478  Ana Paula Dourado I would rather argue that the fiscal vacuum argument is not part of a broad concept of abuse, but contributes to defining the meaning and scope of the rights granted by EU law within a cohesion perspective. V.  GAAR and specific anti-abuse rules and their modus operandi concerning the individual case: the example of a case regarding personal taxation

Either a GAAR or a specific anti-abuse rule implies requalifying the legal transaction. As mentioned above, that requalification is often not simply achieved in tax law by judicial interpretation in the absence of one of those rules. In other words, anti-abuse rules (either a GAAR or a specific rule) go further than the principle of abuse, as beyond applying it, they also provide legitimacy to requalifying the transaction by the tax administration and courts, with there being no need to pass a new law. Discriminatory regimes, such as exit taxes, share an anti-abuse purpose, even if they follow other purposes, as follows from de Lasteyrie du Saillant (paragraphs [29, 64]) and N (paragraphs [42–47]), and some of these regimes belong in a very broad sense to specific anti-abuse rules. The role played by specific anti-abuse rules could presumably be played by a GAAR, which could, for example, be applied to tax capital gains when individuals have transferred their residence, or to tax non-distributed dividends by a CFC, or to tax dividends of interest paid to an associated enterprise if the amounts paid do not respect the arm’slength principle. That role could sometimes be played even if only by teleological interpretation (ie by the principle of abuse87), such as taxing someone as a resident, if the tax administration and the courts conclude the person’s permanent home, centre of vital interests, or habitual abode occurs in that State. Specific anti-abuse rules, namely if they contain irrebuttable presumptions or legal fictions, contribute to achieving legal certainty as they reduce or even eliminate administrative and judicial discretion, whereas a GAAR and the principle of abuse do this to a much lesser extent (GAAR and the principle of abuse have to be progressively defined by courts). The difference between abuse as a principle defining the core and limits of a right attributed by primary or secondary EU law and a GAAR, on the one hand, and a specific anti-abuse rule or a discriminatory measure following an anti-abuse purpose, on the other hand (such as an exit tax or a legal fiction of residence, like the one in Van Hilten or in N), lies in their relation towards the underlying case or facts. Whereas the principle of abuse as well as a GAAR considers the individual case, specific anti-abuse clauses consider the typical case, and often preclude analysis on a case-by-case basis. ‘Mass tax administration’ has placed tax administrations on ‘emergency status’ since the second half of the twentieth century, and as a consequence, legal fictions and presumptions have spread in domestic tax legislations (and tax treaties), especially since the 1980s. These have been tolerated in OECD Member States from a constitutional 87   Although the principle of abuse is not necessarily used in teleological interpretation: see Case C-63/04 Centralan Property Ltd v Commissioners of Customs & Excise [2005] ECR I-11087.

A Single Principle of Abuse in European Union Law  479 point of view with the argument that they fulfil a ‘second best’ principle of equality (the ‘achievable equality’).88 In cross-border situations, legal fictions and presumptions with anti-abuse purposes have been recommended by the OECD. All EU directives in direct tax law contain specific anti-abuse clauses and some of them use this technique, as already mentioned, which, from the point of view of EU law, can be justified by a second best principle of equality. Taking into account the previous pages, the basic assumption is that specific domestic anti-abuse rules have to be tested, exactly like any potentially discriminatory or restrictive rule, against the principle of abuse of EU law. As they operate in the interpretation sphere of EU law and do not aim at qualifying a legal transaction as a tax offence, and as they requalify the legal transaction, the Court tests whether those clauses qualify the facts within the framework of the principle of abuse of EU law—ie whether there is an abuse of EU law. Even if there is an abuse of domestic law, the relevant assessment is, I insist, whether there is an abuse of EU law. In order to check how the suggested methodology operates or should coherently operate let me take as an example the facts underlying the N case. The first test that should be carried out by the Court is whether Mr N actually exercised his right of establishment or actually established himself in another Member State. If the answer is positive, then there is a domestic regime that restricts the freedom of establishment and that has to be justified. In this situation, the Court tests, in second place, whether the restriction resulting from the application of an anti-abuse rule is proportional to fighting abuse of EU law. VI.  Is there a broader concept of abuse in personal taxation?

We may find, however, that the first test has to be complemented. The complementary test will take place if abuse is to be analysed under the cohesion perspective, or, in other words, if it is an issue of fair distribution of tax revenue between the Member States involved.89 If the answer to the complementary test is positive, a third test has to be taken, concerning how to achieve that fair distribution of tax revenue without abuse and in a proportional way, even if the person has actually exercised his right of establishment. The complementary test, taking the N case as an example, leads us to ask whether: 1. The Court has a broader concept of abuse when individuals are involved, since in Cadbury Schweppes, it simply required the wholly artificial arrangement test, that is, a test on the actual exercise of the right of establishment. Thus, the wholly artificial arrangement test seems to make the cohesion perspective irrelevant. 2. Or, whether the complementary test is not indeed a refined wholly artificial arrangement test in the case of an exit tax, taking into account that the competence to tax capital gains at the time they are realised belongs to the States of residence involved, from the time of acquisition of the assets until the time of their sale. I claim that the second answer is the right one and that there is no broader concept of abuse when individuals are involved.  See Dourado (n 33 above) 537, 571, 612, 643.  See N (n 19 above), paras [41–48].

88 89

480  Ana Paula Dourado VII. The principle of abuse and legal certainty

In Halifax, the Court made a reference to legal certainty and seems to oppose it to the aim of preventing tax abuse: Preventing possible tax evasion, avoidance and abuse is an objective recognised and encouraged by the 6th Directive [ . . . ]. However, as the Court has held on several occasions, Community legislation must be certain and its application foreseeable by those subject to it [ . . . ]. That requirement of legal certainty must be observed all the more strictly in the case of rules liable to entail financial consequences, in order that those concerned may know precisely the extent of the obligation which they impose on them.90

This excerpt from the Court seems to reveal some reluctance in attributing an interpretative role to the principle of abuse, and it may perhaps reflect dissenting opinions within the Court on the use of anti-abuse as a principle. Halifax also seems difficult to reconcile with Centros,91 Überseering92 and Inspire Art,93 because in the latter three cases, the Court regards the choice of the applicable law as the core of the freedom of establishment. On the other hand, if the Court gave a predominant role to legal certainty in the context of abuse, it would accept specific anti-abuse clauses with irrebuttable presumptions, at least in some cases. In respect of anti-abuse clauses and irrebuttable presumptions, I suggest that the Court adopts a consistent test, accepting irrebuttable presumptions as long as they correspond to ‘internationally recognised principles’ that do not jeopardise the fundamental freedoms. In fact, the Court accepts a thin capitalisation discriminatory regime if it passes the test based on the ‘arm’s-length principle’.94 And the arm’s-length principle is itself a legal fiction or an irrebuttable presumption, since a company is not allowed to demonstrate that its transactions should not be assessed according to the arm’s-length principle. Whether domestic restrictive anti-abuse rules such as legal fictions on the meaning of residence or a minimum holding period of assets in order to benefit from an exemption regime (as long as they are not abusive) should not be accepted by the Court can be a matter of discussion. However, that scrutiny will have to be made by the Court in any case, as the scope of the rights conferred by EU law, the anti-abuse aim and the proportionality of the measure always have to be tested against EU law. Another issue is whether the Court does not exaggerate in its proportionality assessments. According to the Court in the N case, paragraph [38], the ‘tax declaration required at the time of transferring the residence outside the Netherlands is an additional formality likely to further hinder the departure of the person concerned’. Curiously enough, when within a Member State there are different tax domiciles implying different tax burdens, benefiting from more favourable tax regimes often implies additional compliance costs, such as filling in returns similar to the ones required by the Netherlands in the N case.   Halifax (n 3 above), paras [68–69].  See a critical review of the case by E Werlauff, ‘The Consequences of the Centros Decision: Ends and Means in the Protection of Public Interests’ (2000) European Taxation 542. 92  Case C-208/00 Überseering BV v Nordic Construction Company Baumanagement GmbH [2002] ECR I-9919. 93  Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155. 94   Thin Cap (n 17 above), paras [83–86]; on the issue, see AP Dourado and R de la Feria, ‘Thin Capitalization and Outbound Investment: Thin Capitalization Rules in the Context of the CCCTB’ in M Lang, P Pistone, J Schuch and C Staringer (eds), Common Consolidated Corporate Tax Base (Vienna, Linde 2008) 817. 90 91

A Single Principle of Abuse in European Union Law  481 What is more, principles of tax law that are common to the Member States, which are or should be accepted as principles of the EU, such as the principle of practicability and equality (or second best equality) have been totally disregarded by the Court so far, but in a second generation of case law, they may well be taken into account.95 Secondary EU law ensures that domestic measures do not create distortions to crossborder investment in the EU, but directives in direct taxation also include legal fictions and irrebuttable presumptions with anti-abuse aims that would be considered incompatible with the Treaty if they were domestic rules.96 I suggest that specific anti-abuse clauses with irrebuttable presumptions in EU secondary law are legitimate because they are directly aimed at preventing abuse of EU law (harmonised tax law) and were not drafted to prevent abuse of domestic law (and they are not domestic law). VIII.  Abuse as justification

In the Court’s reasoning the abuse test (the artificiality of the arrangement) has recently been dealt with at the level of justifications and proportionality and not when applying the relevant freedom.97 If abuse and anti-abuse clauses appear at the level of justifications, this implies that they are automatically considered as restricting the fundamental freedoms, as the reasoning of the Court in Cadbury Schweppes (paragraphs [36–59]) very clearly demonstrates. In this context, the principle of abuse is no longer an interpretation principle aimed at defining the scope of a right granted by EU law. In its analysis, the Court seems to overlap steps of analysis that should be considered at different levels. It does so by considering that any formal exercise of an EU fundamental freedom is within the meaning and scope of the freedom. Instead of using the principle of abuse as an interpretative principle, in order to interpret the scope of the freedom, it presumes that taking advantage of the EU fundamental freedoms, independently of the aim and purpose of the movement, is within the scope of the freedom, and that any anti-abuse rule restricts the relevant freedom (see again Cadbury Schweppes, paragraphs [37–46]). My argument is based on examples such as the following. According to the Court: [N]ationals of a Member State cannot attempt, under cover of the rights created by the Treaty, improperly to circumvent their national legislation. They must not improperly or fraudulently take advantage of provisions of Community law [ . . . ].98 However, the fact that a Community national, whether a natural or a legal person, sought to profit from tax advantages in force in a Member State other than his State of residence cannot in itself deprive him of the right to rely on the provisions of the Treaty [ . . . ].99 95  On the importance of comparative law in the case law of the Court, see K Lenaerts, ‘Interlocking Legal Orders in the European Union and Comparative Law’ (2003) 52 International and Comparative Law Quarterly 873; M Poiares Maduro, ‘Interpreting European Law—Judicial Adjudication in a Context of Constitutional Pluralism’ IE Law School Working Paper WPLS08-02, 4, available at ssrn.com/abstract=1134503. 96  See Schön, ‘Abuse of Rights’ (n 2 above) 81–82; and Denkavit et al (n 27 above), paras [23–36]. 97  Case C-364/01 Barbier v Inspecteur van de Belastingdienst Particulieren/Ondernemingen Buitenland [2003] ECR I-15013, para [71]; Cadbury Schweppes (n 10 above), para [36]; see Lang and Heidenbauer (n 25 above) 607. 98  Case 115/78 J Knoors v Staatssecretaris van Economische Zaken [1979] ECR 399, para [25]; Case C-61/89 Marc Gaston Bouchoucha [1990] ECR I-3551, para [14]; Centros (n 8 above), para [24]. 99   Barbier (n 97 above), para [71].

482  Ana Paula Dourado As to freedom of establishment, the Court has already held that the fact that the company was established in a Member State for the purpose of benefiting from more favourable legislation does not in itself suffice to constitute abuse of that freedom [ . . . ]100.101

But then, instead of applying the abuse test, the Court considers that anti-avoidance rules are restrictive or discriminatory. This has not always been so102 and it seems that beyond adopting a formal concept of freedom of establishment, it is easier for the Court to test the proportionality of the antiabuse rule if this is automatically considered to be a restrictive rule and its justification is then verified.103 However, one previous step of analysis is clearly missing: the one concerning whether there has been a real exercise of a fundamental freedom or of a right granted by EU law. If this issue were considered by the Court, the answer would either lead to the acceptance of the anti-abuse measure as long as it was proportional to the aims of defining the scope of the right if there had been abusive behaviour; or the answer would lead to the rejection of the anti-abuse rule if the behaviour of the taxpayer was within the scope of the right granted by EU law. Since I argue that abuse defines the core and boundaries of a right granted by EU law, whether there has been abuse has to be tested first, and if there has been no abuse, the anti-abuse rule cannot be justified. iX. Foreseeing the next stages in the court’s case law: a path to the typical features of the case?

It follows on from the preceding statements that an irrebuttable presumption as part of a specific domestic anti-abuse clause cannot be accepted, because the anti-abuse measure has to be tested concerning its compatibility with EU law. The Court has consistently required application of the abuse reasoning on a case-by-case basis (ICI, LankhorstHohorst, Cadbury Schweppes, Thin Cap, de Lasteyrie du Saillant, N, Part Service, Lidl Belgium104). As a rule, and since, as I claimed above, abuse aims at defining the core and boundaries of the EU legal right, this methodology of the Court seems unavoidable in a first stage, as it obviously grants higher efficacy to the fundamental principles, even if it precludes, to some extent, certainty. However, the Court accepts a thin capitalisation discriminatory regime if it passes the test based on the ‘arm’s-length principle’ and this principle is itself a legal fiction. I therefore suggested that the Court adopt a consistent test, accepting irrebuttable presumptions as long as they correspond to ‘internationally recognised principles’ that do not jeopardise the fundamental freedoms. Thus, it is predictable that in the aforementioned case-by-case tests, the taking into account of every single feature of the case will at some point begin to decrease and that will happen in a progressive way.   Centros (n 8 above), para [27]; Inspire Art (n 93 above), para [96].   Cadbury Schweppes (n 10 above), paras [35–37]. 102  See, on the evolution of the ECJ case law, V Edwards and Paul Farmer, ‘Abuse of Law: What Is The Value Added of The Tax Dimension?’ in Hinnekens and Hinnekens, A Vision of Taxes (n 25 above) 302; Lang and Heidenbauer (n 25 above) 607. 103  See Lang and Heidenbauer (n 25 above) 607–08. 104  Case C-414/06 Lidl Belgium GmbH & Co KG v Finanzamt Heilbronn [2008] ECR I-3601. 100 101

A Single Principle of Abuse in European Union Law  483 This methodology will belong to a second stage of the Court’s case law: when referrals to the Court on abuse and its case law on the issue will have already dealt with a wide range of (tax) abuse issues, the Court will cross reference its previously created objective criteria (such as the ones recommended in Cadbury Schweppes or in de Lasteyrie du Saillant). That cross reference will still require a casuistic analysis of some elements, but will not imply interpretation of the free movement itself as a principle, and instead an analysis of the case under the previously found objective criteria (ie under the defined typical cases) will be carried out. The individual case will be compared to the typical case and some peculiarities of the individual case will be disregarded. This will not only result from a necessity to save work and time—since most of the tests are to be applied by the referring national courts—but from the nature of legal reasoning and its urge to achieving legal certainty and equal solutions to equal or similar cases.

33 Is There a Role for a European Principle Prohibiting Abuse of Law in the Field of Personal Taxation? A Comment Edwin Simpson

I

n her contribution to this collection, Professor Ana Paula Dourado argues that a principle of European law prohibiting abuse of law should apply in the context of personal tax law in very much the same way as it does elsewhere. I say ‘in very much the same way’ because her position is qualified by the view that ‘the concrete application of the “artificial arrangement test” may vary according to the field of law under analysis’.1 The intention in this brief comment is to explore the reasons that there might be for any such variation in the context of personal taxation; and to consider whether there could even be grounds to anticipate no general European principle of abuse in the field of personal taxation at all, such matters being instead better left for domestic law to resolve. I. The development of a general abuse principle

A.  Its Use as a ‘Shield’ by Member States Reasoning based on abuse of law has been deployed by the Court of Justice in two very different ways. The first, and rather more securely established way, is where a Member State seeks to defend a part of its national legal system on the basis that, although it involves an apparent restriction of some sort on the fundamental freedoms, that restriction ought nevertheless to be permitted, because it is in some sense justified. It is well established that the possible justifications in such cases extend beyond specific provisions in the Treaty to include circumstances where the restriction is justified as a means of achieving one of the ‘overriding objectives’ recognised by the Court—pressing matters of public importance and so forth. The Court has accepted a number of ‘concessions’ of this type that are peculiar to tax law. First, and most obviously, it has been bound to accept (at any rate, to some degree) the discrimination that is involved in taxing source and resident income differently. The national laws of most countries, and the treaties concluded between them, aim at territorial partition of the tax base, normally on the basis of first, source taxation—a country can 1   AP Dourado, ‘A Single Principle of Abuse in European Community Law: A Methodological Approach to Rejecting a Different Concept of Abuse in Personal Taxation’, ch 32 above. See Part III of the chapter.

486  Edwin Simpson tax the income of non-residents that is earned within its territory; and secondly, on the basis of residence—that states are entitled to tax residents on their worldwide income (but through treaties, credits and so forth, should seek to avoid double taxation).2 The Court has therefore had little choice but to accept that residents and non-residents cannot straightforwardly be compared in order to demonstrate that Member States are improperly discriminating between them. The differences in treatment that follow from long-established international convention—the fiscal principle of territoriality—can be permitted provided that they are justified by a relevant difference of fact, and as long as non-residents are not simply treated less favourably than residents.3 Secondly, there is the justification of ‘fiscal cohesion’, associated with the Bachmann decision in 1992;4 ie the need urged by Member States to maintain the fiscal coherence of their national tax systems (and sometimes the ‘supervision’ of those systems, if that is different) even though in particular cases this may appear to operate in a discriminatory way. The modern extent of this ‘fiscal cohesion’ doctrine is itself of course controversial,5 but it is apparent that the two styles of reasoning just discussed (ie restrictions justified on the basis of the long-standing acceptance of the fiscal principle of territoriality as well as on the basis of ‘fiscal cohesion’) themselves closely prefigure the first style of abuse reasoning (ie where a Member State seeks to defend discriminatory national provisions on the basis that they are aimed at tax avoidance, or abuse, or combating a tax advantage, or whatever it may be appropriate to call it in the particular domestic context). It has been clear for some time from the cases that this was a legitimate aim, a position that has been confirmed by Cadbury Schweppes;6 but we now know that such provisions will only be effective if they are aimed solely at genuinely abusive transactions—which it seems may be a tough test for domestic provisions to pass. But the significance in terms of the identification of any principle of European law combating or prohibiting abuse in general is that these analytically-related developments are no more than precursors to any more interventionist principle, as opposed to indicators of its current existence. On their own, they certainly cannot be said to amount to a principle ‘prohibiting abuse of law’ in the field of taxation. All that they show is a (possibly reluctant) recognition by the Court, in the tax context, that there will have to be some concessions to existing patterns of fiscal territoriality, fiscal cohesion, and the combating of avoidance or abuse by domestic legal systems, even where fundamental freedoms are infringed. One might perhaps helpfully describe this style of abuse reasoning by the Court as a ‘shield’, ie as a means of protecting domestic tax provisions which might otherwise be found to conflict with the fundamental freedoms.

2   Such treaties (which, of course, pre-date the European Union) and the processes which led to their common forms are themselves instructive precursors of today’s debates as to the appropriate allocation of tax bases between Member States within the EU. 3   See for example, Case C-250/95 Futura Participations SA and Singer v Administrations des Contributions [1997] ECR I-2471 (Luxembourg entitled to restrict the carrying forward of losses by a non-resident company to Luxembourg losses only, even though a resident company could have carried forward worldwide losses). 4  Case C-204/90 Hanns-Martin Bachmann v Belgium [1992] ECR I-249. The direct link in Belgian law between the deductibility of life insurance contributions and the taxation of the eventual benefits justified a national rule that contributions could be deducted only if paid within Belgium. 5   See Case C-319/02 Petri Manninem [2004] ECR I-7477; and Case C-446/03 Marks & Spencer plc v David Halsey (Her Majesty’s Inspector of Taxes) [2005] ECR I-10837. 6  Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995.

A European Principle Prohibiting Abuse of Law in Personal Taxation?  487 B.  Should a Principle Prohibiting Abuse Be Used as a ‘Sword’? It is a separate question whether a supra-national legal system of the European kind (ie a system that does not purport to be all-embracing, but rather, and to differing degrees in different legal fields, to respect notions of deference, subsidiarity, margin of appreciation and so forth) ought to move from such defensive examples of deference, to develop a general combative principle of its own, operating as a ‘sword’ against tax avoidance or abuse. If it were to do so, the underlying European logic would appear to have to be that, if transactions have no genuine economic purpose and are carried out purely to obtain a beneficial tax treatment that should not be available in such circumstances, then they do not assist the operation of the Single Market, and so ought not to be entitled to the protection of the rules regulating that market. It was in that context that value-added tax (VAT) avoidance, for example, in the context of the Halifax case,7 was urged by Her Majesty’s Revenue and Customs (HMRC) not to be an economic activity. As VAT law is directed only to activities carried on with a business purpose, there is a coherent internal logic to the suggestion that ‘abusive’ steps can be ignored. But even here, where an EU resource is at issue, we now know that the test is again8 a strict one: merely carrying out transactions for the sole purpose of obtaining a tax advantage, without any other economic objective, is not enough; rather ‘wholly artificial arrangements’ must exist, and a clear intent to obtain benefits in circumstances contrary to the purposes of the provisions relied on. It is clear that a considerable transition has occurred between the first group of examples of abuse reasoning as ‘shield’ above; and its apparent use as a ‘sword’ in Halifax. A move beyond a tiered, deference principle, pursuant to which the Court allows Member States in their domestic taxing capacity to derogate from the freedoms in certain circumstances, to the quite different context in Halifax, where the Court is defending a European tax base by directly ‘reviewing’ the conduct of its citizens, and seeking to do so in a consistent way across all jurisdictions. There may well be good reasons for the Court wishing to develop such a general anti-avoidance or abuse principle in the context of European taxes (the reasons will, after all, be the very same ones that have led domestic tax systems to develop such principles), but it will constitute a considerable further move still to extend any such doctrine beyond the confines of the European tax base itself. II. Lessons from the history of domestic law

A.  The Difficulties Indicated by the UK Experience In the domestic context, arguments as to whether General Anti-Avoidance Rules (so-called GAARs) are better legislated, or left to be developed by the judges (along the lines of the so-called Ramsay principle9 in the UK), are well rehearsed. But domestic systems are perhaps less likely to offer useful guidance on the question of whether any such principle 7  Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609. 8   See text following n 6 above. 9  Ie the principle associated with the decision of the House of Lords in WT Ramsay Ltd v Inland Revenue Commissioners (IRC) [1982] AC 300.

488  Edwin Simpson should extend beyond a ‘sovereign’ tax base into subsidiary ones. The legal historians who have contributed to this collection have struggled to find precedent in the common law at all for any abuse doctrine (indeed they find only its firm rejection at the end of the nineteenth century). But the UK courts do, in fact, have recent experience to offer of a very similar notion, concerning the development of the so-called Ramsay principle in the field of UK tax law in the early 1980s. The current status of that doctrine is itself highly controversial (and the guidance that it can offer to the Court will be thought by many to be little more than not to proceed in the same way), but it is an example of a domestic court developing a principle to combat abuse, of its own motion, and without specific statutory authority to do so. Of course it was not framed in terms of abuse of law, but rather in terms of a species of ‘fiscal nullity’ that deprived of effect a preordained series of transactions into which steps had been inserted which had no purpose other than the avoidance of tax. But this, on examination, may turn out to be a distinction without much of a difference. The UK doctrine, in its most fully developed form (ie during the period of the decisions of the House of Lords in Furniss v Dawson10 in 1984 and in Craven v White11 in 1988) required two indications of unacceptable activity: first, a series of transactions intended to operate as a whole; and secondly, steps inserted into that series with no purpose other than the avoidance of tax. The modern EU doctrine requires a ‘wholly artificial arrangement’ (which might well be identified by a series of transactions with otherwise unnecessary steps inserted into it) and ‘a clear intent to obtain benefits in circumstances contrary to the purposes of the provisions relied on’ (which again, the two elements of the Ramsay doctrine would indicate). The similarity is at least enough to suggest that the EU doctrine, if it is applied widely in the tax field, may well face many of the same difficulties that have already been considered by the UK courts in a domestic context. Those difficulties were many, but can perhaps usefully be considered as falling into four categories. The first concerned the constitutional legitimacy of a judicial (rather than legislated) doctrine, and is a difficulty which will clearly raise different issues in the European context. Secondly, the Ramsay principle struggled to identify satisfactorily (and without simply providing precise guidance to taxpayers as to how to prevent the principle from applying to their transactions at all) the necessary element of ‘preordainedness’ for a series of transactions to be subject to the doctrine. The EU formulation will be saved to some extent from this difficulty, because it is not phrased in terms of ‘a series of transactions’; but then it may face a related complexity determining what is to count as an ‘arrangement’ for its purposes. Thirdly, and most catastrophically in the UK cases, the Ramsay principle faced recurrent difficulties concerning the manner in which transactions subject to the doctrine were then actually to be taxed. What does it mean to ‘ignore’ steps in a transaction, which have been found genuinely to have taken place?12 Does it require the transaction to be restructured for tax purposes only; and, if so, how precisely is that to be done?13 This is a difficulty which the EU doctrine will certainly face, and indeed is prefigured by the optimistically straightforward statement in Halifax that, once abuse   Furniss (Inspector of Taxes) v Dawson [1984] AC 474.   Craven (Inspector of Taxes) v White [1989] AC 398. 12  In the sense that they are not ‘shams’ as that concept is understood in English law, and so are not devoid of any legal effect at all: see generally, Snook v London & West Riding Investments Ltd [1967] 2 QB 786, CA. 13  For an example where reconstruction was entirely straightforward, see Hatton v IRC [1992] STC 140, Chadwick J; for an example where it proved impossible, see Countess Fitzwilliam v IRC [1993] 1 WLR 1189, HL. 10 11

A European Principle Prohibiting Abuse of Law in Personal Taxation?  489 has been identified, the transactions involved ‘must be redefined so as to re-establish the situation that would have prevailed in the absence of the transactions constituting that abusive practice’.14 One can only hope that it will prove so straightforward! The difficulty—indeed in some cases, impossibility—of such restructuring, following the removal of fiscally annulled steps, led to the gradual retreat in the UK context from its then perhaps overly-precise formulation of a general anti-avoidance doctrine. There may well be lessons for the Court to learn from that experience;15 and, significantly, the direction of the retreat may be found to be of particular interest. In denying a freestanding judge-made principle of the common law capable of operating as a general overlay upon the statutory taxing provisions,16 UK judges have instead emphasised that the same work can be done (and may be more legitimately done in constitutional terms) if the principle is simply regarded as one of statutory interpretation, coupled with the obvious need to construe or characterise (but not to recharacterise) the transaction concerned. On this modern view of the UK approach,17 all the work is to be done in the particular context of a specific statutory provision (rather than there being any general, overlaid anti-avoidance principle). And the work is to be done (simply!) by analysing, in a purposive way, what it is that the legislation is designed to tax, and whether what the taxpayer has done falls within or outside that analysis. The taxpayer’s purposes will be relevant where the legislation makes them relevant, but not where it does not do so (and so subjective intentions may be significant in the context of some statutory provisions but not in others). The UK experience has therefore been that a freestanding, statute-like but judge-made, anti-avoidance rule has matured instead into no more than grown-up—ie purposive— statutory interpretation. But EU law, of course, already has very grown-up techniques of purposive construction; which may mean, if the difficulties that the UK faced with the Ramsay principle turn out to be replicated as an abuse principle is applied to tax matters generally by the Court, that a similar retreat (from a precise rule-like formulation, towards a more general principle of statutory and transaction construction) can be expected.18 The truth is that the problems experienced by the UK courts really demonstrate simply that it is vitally important to get the primary, statutory tax rules right in the first place.   Halifax (n 7 above), para [94].  The UK cases have also considered the issue of whether an essentially tax doctrine ought to apply in other areas of law where statutory consequences are manipulated: see Circuit Systems Ltd v Zuken-Redac (UK) Ltd; Norglen Ltd v Reeds Rains Prudential Ltd [1997] 3 WLR 1177, HL, in the context of possibly champertous schemes to secure legal aid; and the well-known line of cases associated with Street v Mountford [1985] AC 809, HL, concerning the manipulation of the distinction between a lease and a licence to avoid the consequences of the Rent Acts. 16  Lord Hoffmann, in MacNiven v Westmoreland Investments Ltd [2001] UKHL 6, para [29], used a very European example to make his point, stating that the detailed formulation of the Ramsay doctrine urged upon him by Counsel ‘cannot be called a principle of construction except in the sense of some paramount provision subject to which everything else must be read, like section 2(2) of the European Communities Act 1972. But the courts have no constitutional authority to impose such an overlay upon the tax legislation and, as I hope to demonstrate, they have not attempted to do so’. 17  See particularly, in addition to MacNiven (n 16 above), Barclays Mercantile Business Finance Ltd v Mawson (Inspector of Taxes) [2004] UKHL 51; and IRC v Scottish Provident Institution [2004] UKHL 52. For recent applications of the modern approach see Astall v Revenue and Customs Commissioners [2009] EWCA Civ 1010; and Mayes v Revenue and Customs Commissioners [2009] EWHC 2443 (Ch). 18   There are indications of some appreciation of this point already, at any rate at the Advocate-General level: see AG Kokott in Case C-63/04 Centralan Property Ltd v Commissioners of Customs and Excise [2005] ECR I-11087, para [61] stating that in her view, a purposive interpretation of the Common VAT System Directive (CVSD) provisions would ‘preclude these artificial transactions from giving rise to a tax exemption which would run counter to the objectives of the directive and would have to be remedied by course to unwritten principles such as the prohibition on abuse of rights’. 14 15

490  Edwin Simpson The judges, of course, are not in a position directly to achieve this for themselves. But any anti-avoidance, or prohibition of abuse, principle that they design ought to send that message clearly to legislators as well as act as a warning to citizens. As Judith Freedman has convincingly argued in the context of UK tax law,19 the message sent out by everincreasing, overly technical legal rules almost invites avoiding or abusive strategies. One undoubted advantage of the modernised UK approach is that it makes clear to those drafting legislation that their language can, if used carefully, communicate to the courts the approach to be taken to avoiding or abusive strategies in specific fiscal contexts. This may be an ambitious (and decidedly long-term)20 programme; but it is ultimately the right one. What is most needed from any judicial doctrine of this kind in the tax arena is an approach that encourages—indeed demands—that the primary, statutory tax rules be general and clear. B.  The Application of an Abuse Doctrine to Personal Tax in Particular The UK experience therefore suggests that there may be real difficulties ahead in applying a general ‘prohibition of abuse’ principle (at any rate, as presently formulated by the Court) to issues generally of tax law. But there is a further question of whether any EU doctrine, capable of being used in the manner I have described above as a ‘sword’, ought to extend beyond the protection of the EU’s own resources. Would it assist domestic systems to have the back-up of a European abuse principle, alongside their own, usually highly sophisticated, domestic provisions, even in areas where the Union’s own tax base is not threatened—such as in the realm of personal taxation? It is far from clear that Cadbury Schweppes21 is itself an indication of such an extension to the use of abuse reasoning as a ‘sword’. The Court held that the UK’s ‘controlled foreign company’ legislation will be incompatible with the EC Treaty if it can apply to situations other than wholly artificial arrangements, which appears rather to be only a further instance of its use as a ‘shield’ (albeit perhaps unsuccessfully) by a Member State. The progression so far has therefore been as follows. The ‘sword’ principle has been applied in the VAT context in Halifax, ie in the context of a (broadly) harmonised tax, which forms part of the Union’s own resources. In the context of corporate taxes, which on the contrary have only been marginally harmonised, and do not form part of the Union’s own resources, its application is very much less clear. The Thin Cap case22 also looks once again like a ‘shield’ case. Member States are to be permitted to infringe the freedoms so long as only genuinely abusive practices are being targeted. If a Member State has domestic tax provisions which go beyond that, then it is that Member State’s ‘abuse’ of the fundamental freedoms which needs to be curbed 19   J Freedman, ‘The Tax Avoidance Culture: Who is Responsible? Governmental Influences and Corporate Social Responsibility’ in J Holder and C O’Cinneide (eds), Current Legal Problems (Oxford, Oxford University Press 2006) 359. 20  In the short term, there is room for the cynic to argue that the modern UK doctrine will be a decidedly narrow one in the context of personal taxation, if it requires a clear purpose of the legislation to be identified and then shown to be undermined. 21   Cadbury Schweppes (n 6 above). 22  Case C-524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I-2107, para [72]: ‘According to established case law, a national measure restricting freedom of establishment may be justified where it specifically targets wholly artificial arrangements designed to circumvent the legislation of the Member State concerned.’

A European Principle Prohibiting Abuse of Law in Personal Taxation?  491 (although there is no need for abuse language to be used to describe such reasoning, because it is already fully captured in its usual description as an unjustified restriction on the exercise of the fundamental freedoms). It should also be observed that the move evident from Cadbury Schweppes and the Thin Cap case has the effect of narrowing the permitted scope of operation of domestic anti-avoidance provisions by requiring that they only target ‘the creation of wholly artificial transactions which do not reflect economic reality’.23 Far from broadening the combating of so-called ‘abuse’ within the EU these decisions in fact restrict it, by limiting the permitted sphere of operation of domestic anti-avoidance rules in areas of primarily domestic competence. The final group of cases to which one might look for any sign of an abuse doctrine capable of being used as a ‘sword’ in the personal tax cases would seem to be those cases, such as N24 and de Lasteyrie,25 where the Court has considered questions of purely personal taxation. But again one finds on examination that these are ‘shield’ cases, all be they cases where the ‘shield’ has not protected the Member State. They say nothing about an abuse principle acting as a ‘sword’ against citizens, but speak only about one that acts against Member States unjustifiedly restricting individual exercise of the freedoms. There accordingly seems to be rather slender existing evidence for the extension of a European principle of prohibition of abuse of law into the field of personal taxation. But should such a principle be extended into this territory? Professor Dourado argues that there is one principle, and that it should be so extended—albeit that the actual application of the ‘artificial arrangement test’ may vary from one field to another. But how and why might it vary? What are the policy issues at stake here? We know from the cases that particular considerations make the application of abuse reasoning more complex where nationality, citizenship or workers’ rights are concerned.26 There appear to be substantial and well-acknowledged difficulties in the operation of any abuse principle in these contexts of ‘real’ people. This prompts the following conundrum. If indirect taxation (Halifax) and (possibly) corporate taxation (Cadbury Schweppes) are situations where a general abuse principle may apply, and workers’ rights and citizenship are instances where such a principle is very much more restricted, then where does that leave the ‘interdisciplinary’ subject (so to speak) of personal taxation—ie of the nonharmonised taxation of real individuals by domestic law? Is it a situation where ‘abuse’ of the freedoms must be stopped, or where full exercise of the freedoms is to be encouraged? We know that in the context of social benefits for individuals (educational grants and so forth) forum shopping is to be allowed. But circumventing transactions (u-turn transactions) are not to favour corporations: Emsland-Stärke.27 Where in this spectrum does the taxation of individuals lie? Professor Dourado argues that there need be no broader concept of abuse in the context   Thin Cap (n 22 above), para [74].  Case C-470/04 N v Inspecteur van de Belastingdienst Oost/kantoor Almelo [2006] I-7409: exit charge regime in the Netherlands disapplied. 25  Case C-9/02 Hughes de Lasteyrie du Saillant v Ministère de l’Économie, des Finances et de l’Industrie [2004] I-2409: a deemed disposal for French capital gains tax purposes of assets when an individual moved his residence from France disapplied. 26   See for example, Case C-413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I-13187; Case C-109/01 Secretary of State for the Home Department v Hacene Akrich ECR I-9607; Case C-200/02 Kunqian Catherine Zhu, Man Lavette Chen v Secretary of State for the Home Department [2004] ECR I-9925; and Case C-147/03 Commission v Republic of Austria [2005] ECR I-5969. 27  Case C-110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I-11569. 23 24

492  Edwin Simpson of personal taxation,28 but might one not expect a narrower one? That is to say, would it not be reasonable to expect that individuals should be protected rather more in their exercise of the fundamental freedoms than artificial corporations? Might it not be said that the organisation of their affairs by individuals, inherently less able to move around and reinvent themselves as they are, require rather less scrutiny for abuse than the actions of corporations, and that freedom of movement for real persons ought to remain at the heart of the European enterprise? Being subject to taxation is, after all, only the other side of the coin from being entitled to benefits and so forth. If one is to be entitled to move freely to receive the one, in pursuance of the goal of a Single Market, why should one not also be free to minimise the other, and so increase the pressure on Member States to have neutral and efficient systems of taxation, as well as comparable systems of social benefits? Addressed in this way, the real issues here are revealed as ones concerning market power and political influence, and the complex spectrum that now exists between fully public bodies and fully private individuals. The proper order of ‘public’ (or perhaps one should say ‘market’) influence is probably: (i) international corporations and capital; (ii) Member State governments; and finally (iii) private individuals—a hierarchy that the European enterprise must somehow supervise and coordinate. That relative lack of brute economic power in the hands of individual citizens29 may explain a corresponding relative lack of need to police abuse by individuals of their freedoms and rights under European law in the tax context. There is a greater need to remedy abuse by corporations certainly, and indeed by Member State governments if they seek over-zealously to protect their tax bases from the harsh realities of free movement in what, of course, are increasingly difficult fiscal times. These may all be reasons why there is a less pressing need to extend any abuse doctrine into the sphere of personal taxation, and at their heart is the fact that it would bring such a principle so starkly into contest with the core exercise of the fundamental freedoms by ‘real’ people. Perhaps we had rather instead accept that individual citizens of Europe are to be allowed to exercise the fundamental freedoms—even in ways which lessen their personal liability to domestic taxation.30 Those freedoms can be restricted by domestic anti-avoidance or abuse legislation—so long as it is not too widely framed—but it may be unnecessary, and counterproductive in terms of both certainty and the costs of compliance, to have a further, general European anti-abuse doctrine operating as a ‘sword’ alongside those domestic provisions in the context of real persons. These are, of course, full-blown, and fully controversial, political questions. But they return one perhaps to the real issue at stake here, at any rate in the field of taxation generally, and so of personal taxation too. That is the fact that, whilst there may be some virtue in an anti-abuse rule while issues concerning the proper and consistent allocation of the tax base between Member States and the EU are sorted out, the real work remains to be done in deciding what those rules allocating the tax base ought actually to be, and in stating them in a way that is sufficiently general and clear for judges, whether at the domestic or EU level, to be able to apply them straightforwardly. It is a balanced allocation of the taxing power that is, as ever, required. If that is achieved, whether by harmonisation or through competition, the need for anti-abuse principles will be very much reduced.   See part IV of Professor Dourado’s chapter.  Of course there are (many) individual exceptions, but the general point remains sound.   See generally, J Snell, ‘And Then There Were Two: Products and Citizens in Community Law’ in T Tridimas and P Nebbia (eds), European Law for the Twenty-First Century: Rethinking the New Legal Order, Vol 2 (Oxford, Hart Publishing 2004) 49, 62: ‘The Court has consistently favoured the movement of European citizens over other forms of free movement.’ 28 29 30

34 State Aid, Taxation and Abuse of Law Timothy Lyons*

A

t first sight it may seem surprising to find a review of aspects of State aid law in a book on abuse of law. After all, abuse of law has not featured largely in judicial decisions in the State aid field. It was dealt with briefly in one of AdvocateGeneral Jacobs’ Opinions a few years ago.1 It is also mentioned in some relatively recent references to the Court of Justice.2 These are, however, appearances of limited significance. Furthermore, there are some notable differences between the law of State aid and those areas of law in which abuse of law has generated debate. I highlight four below. Firstly, the law of the fundamental freedoms and the right to input tax in value-added tax (VAT), in which abuse of law has featured, relate primarily to the conduct of economic operators. They are given the rights which may be abused. The law of State aid, however, seeks primarily to control the behaviour of Member States, although economic operators may be concerned with matters such as the recovery or receipt of aid. Secondly, disputes over the application of the fundamental freedoms and the right to input tax are matters to be determined primarily by litigation before national courts and the Courts of the Union. In the State aid regime, however, the Commission is given the primary role, albeit one which it fulfils under the supervision of the EU Courts.3 Thirdly, the fundamental freedoms and the provisions of law conferring the right to deduct input tax are, of necessity, positive in their stance. In contrast to this, the law of State aid is, primarily, negative in approach. Its purpose is to prohibit.4

 *  The paper on which this chapter is based was written for delivery on 4 October 2008. Some developments up to 31 March 2010 have been taken into account in the text. I am grateful to Conor Quigley QC for taking up the role of discussant in relation to my paper and to Professor Takis Tridimas for reading a draft of this chapter. Any errors are, of course, my own. 1  See Case C-91/01 Italy v Commission [2004] ECR I-4355. See paras [47, 48] of the Opinion of AG Jacobs and para [39] of the judgment. A misuse of the rules arose because a company, claiming to benefit from aid intended for small and medium-sized companies, did not suffer from the normal disadvantages of such an entity because it was not independent but, in effect, belonged to a larger group. Therefore, it satisfied, in a formal way only, the criteria for a small and medium-sized enterprise (SME) in the Community Guidelines on State aid for small and medium-sized enterprises [1996] OJ C213/4. 2  See Case C-78/08 Amministrazione delle Finanze, Agenzia delle Entrate v Paint Graphos Scarl, Case C-79/08 Adige Carni Scrl in liquidation v Amministrazione delle Finanze, Agenzia delle Entrat, and Case C-80/08 Ministero delle Finanze v Michele Franchetto. Opinion delivered 8 July 2010 too late for consideration here. 3  For a recent statement of the breadth of the Commission’s role in making an economic appraisal of whether or not an advantage has been obtained which could not have been obtained on the market and the effect of that on judicial supervision, see: Case T-163/05 Bundesverband deutscher Banken eV v European Commission, judgment of 3 March 2010, para [38] nyr. 4  It may be noted that economic operators are not given the right to avoid payment of tax on the grounds that other businesses are exempt from it and the exemption constitutes a State aid. See for example, Case C-368/04 Transalpine Olleitung in Osterreich GmbH et al v Finanzlandesdirektion für Tirol et al [2006] ECR I-9957,

494  Timothy Lyons Fourthly, the articles of the TFEU conferring the benefits of the fundamental freedoms on economic operators state both the benefits of the freedoms and their limitations in broad terms. The State aid regime on the other hand establishes procedures under which State aid schemes are analysed on a case-by-case basis by the Commission. All these matters help to explain why the concept of abuse, as applied in Halifax5 and Cadbury Schweppes6 does not appear extensively in State aid matters heard before EU Courts. Yet to reject the invitation to consider abuse of law in the context of State aid would be to lose an opportunity to reflect upon a number of important issues. That opportunity may be particularly valuable given that State aid law specifically acknowledges its relationship to general principles of law and, at one point, refers plainly to ‘a general principle of Community law’.7 Indeed, the invitation is timely because the law relating to State aid has been the subject of considerable discussion following the publication by the Commission in 2005 of the State Aid Action Plan. This envisaged a ‘comprehensive reform’ of State aid policy during the period 2005 to 20098 which has now been implemented. I look firstly and briefly at the nature of general principles of EU law and the concept of abuse within it. Some general understanding of the basic terms of the debate is clearly essential.9 Then, secondly, I comment generally on State aid, taxation and the prevention of abuse. Thirdly, I consider the nature of the specific role of general principles of EU law, including an anti-abuse principle, within State aid law. Fourthly and finally, I conclude with some cautionary observations on the dangers of using general principles, including a general anti-abuse principle, to decide cases. I do not consider competition law or policy generally, nor any aspect of the rules of the World Trade Organisation. I. GENERAL PRINCIPLES AND ABUSE: WHAT ARE THEY?

The title of this chapter, and of the conference from which this chapter stems, offers extensive opportunity for debate. In a university inhabited by the ghost of Professor HLA Hart one should be cautious about developing a narrow approach to matters of definition.10 Nevertheless, I venture a few observations on three matters under the headings of: (a) general principles, (b) abuse and (c) abuse by institutions. para [51]; Joined Cases C-393/04 Air Liquide Industries Belgium SA v Ville de Seraing and C-41/05 Air Liquide Industries Belgium SA v Province de Liège [2006] ECR I-5293, para [48]; and Case C-390/98 HJ Banks & Co Ltd v The Coal Authority and Secretary of State for Trade and Industry [2001] ECR I-6117, para [80].     5   Case C-255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I-1609.     6   Case C-196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I-7995.     7  See Council Regulation (EC) No 659/1999 of 22 March 1999 laying down detailed rules for the application of Article 93 of the EC Treaty [1999] OJ L83/1, Art 14(1). Art 107(2) TFEU [ex Art 87(2) EC], in stating what State aid is compatible with the Internal Market, refers to a specific general principle in requiring nondiscrimination as between products on the ground of origin. Compare the reference in the TFEU, Art 340(2) [ex Art 288 EC] to ‘the general principles common to the laws of the Member States’.     8  See Commission (EC), ‘State Aid Action Plan: Less and Better Targeted State Aid: A Roadmap for State Aid Reform 2005–2009’ COM (2005) 107 final, 7 June 2005, para [1].     9  Readers are referred also to the opening chapters of this book. 10  See for example, on the difficulties of defining legal terms, HLA Hart, ‘Definition and Theory in Jurisprudence’ in Essays in Jurisprudence and Philosophy (Oxford, Clarendon Press 1988) 21.

State Aid, Taxation and Abuse of Law  495 A.  General Principles In identifying the constituent elements of law, Professor Dworkin identified policies, principles and rules.11 Without adopting fully his definitions of these, his three-fold analysis may be helpful. A policy, says Professor Dworkin, is ‘a goal to be reached, generally an improvement in some economic, political, or social feature of the community’.12 In EU law these may be thought of as the fundamental tasks and activities of the EU pursued by the exercise of its exclusive or shared competences. They were referred to in Articles 2 and 3 of the EC Treaty and now appear in provisions such as the Treaty on European Union (TEU), Article 3 and TFEU Articles 3 and 4. With State aid law in mind, these would include the establishment of an Internal Market and of the rules necessary for its functioning.13 Moving on to a ‘principle’, this Dworkin described as ‘a standard that is to be observed, not because it will advance or secure an economic, political, or social situation deemed desirable, but because it is a requirement of justice or fairness or some other dimension of morality’.14 In EU law we may see principles as describing the basic standards and requirements to be observed in the exercise of the EU’s competences whether exclusive or shared. Into this category clearly fall principles such as equality, proportionality and the protection of legitimate expectations. On this basis, a requirement that EU law, EU rights and the EU’s fundamental elements, such as the Internal Market, are not to be abused may well be said to be a principle of EU law. The final Dworkinian category of rules may be seen as containing the great majority of the provisions of EU law. Amongst other things they implement general principles and provide for the specific application of principles to the situation in question. As an example of this process, even within the TFEU itself, one may highlight Article 18 of the Treaty [ex Article 12 EC] which contains a general prohibition of discrimination on grounds of nationality and which is implemented in different areas of EU law by specific provisions of the Treaty.15 The specific application of general principles will, of course, be subject to different rules in different contexts.16 In the context of this chapter, the understanding of a principle as set out above is important. It suggests that one should consider principles as having a broad role in a legal system influencing rules and their application. The role of general principles is not  See R Dworkin, Taking Rights Seriously (London, Duckworth 1978) particularly 22–28.   Dworkin (n 11 above) 22. 13  See the EC Treaty Arts 2 and 3(1)(c) and (g), the TEU Art 3(3) and the TFEU Arts 4(2)(a) and 3(1)(b). State aid law also supports subsidiary aims of the Union such as combating harmful tax competition in accordance with the Code of Conduct set out in the Annex to the Conclusions of the ECOFIN Council Meeting on 1 Dec 1997 concerning taxation policy [1998] OJ C2/1. As para [1] of the Notice on the application of the State aid rules to measures relating to business taxation [1988] OJ C384/3 says, the State aid provisions of the Treaty ‘contribute through their own mechanism to the objective of tackling harmful tax competition’, but not all measures which are State aid will constitute harmful tax competition: see Article J of the Code. The authors of the Code of Conduct plainly considered it important to combat abuse and had regard, in Article L, to national anti-abuse measures. 14  See Dworkin (n 11 above) 22. 15  See for example, Case 305/87 Commission v Greece [1989] ECR 1461, para [12]. 16  See also Part I.C below (and the pages following) where differences between rules governing the application of an anti-abuse principle to the European Commission and those governing its application to economic operators are noted. 11 12

496  Timothy Lyons confined to their role in dispute resolution. In such a role, general principles may have a ‘gap-filling function’, an interpretative function or provide grounds for review.17 The role of influencing the content of EU rules is arguably more important. It would be possible, at least in theory, for a branch of law to have a collection of specific rules which express the content of a general principle so completely that there was no gap to be filled or dispute of interpretation to be resolved. In such a situation it would be wrong to assert that general principles were absent from that branch of law. There would simply be no dispute-resolving function for them to perform. The fact, therefore, that there are few State aid cases concerned with abuse of law is not necessarily an indication of the role which the principle plays in State aid matters. B.  Abuse and Abuse of Law Abuse, it may be said, is behaviour which runs counter to the purposes of the EU and its laws. Those purposes are, amongst other things, the creation of an Internal Market and of a system ensuring that competition within it is not distorted. Abuse is linked to the specific objectives of the EU and its legislation in a way which the general notion of ‘fairness’ is not.18 Abusive behaviour is not, therefore, merely conduct which attracts disapproval from a particular moral or political standpoint. Turning to consider abuse in the context of competition law, Article 102 of the TFEU refers to the abuse of a dominant position within the Internal Market or a substantial part of it. Such abuse is, of course, stated to be incompatible with the Internal Market. Examples of the nature of abuse for these purposes are provided by Article 102 itself [ex Article 82 EC] which refers to matters such as the imposition of unfair trading conditions and the application of dissimilar conditions to equivalent transactions, so placing parties at a competitive disadvantage. It is not the law or a right but the dominant position and, consequently, the Internal Market which is abused. The State aid regime is, of course, also concerned to create and preserve the Internal Market and prevent its abuse. In doing this, its focus is on the power of Member States. Member States are not the only authorities to be subject to an anti-abuse principle. As we shall see, the General Court [previously the Court of First Instance] has specifically applied what it called ‘abuse of right’ to the Commission. If the anti-abuse principle is a general principle there is no reason to exclude the Commission (and Member States) from its scope. It is notable that those areas of law in which the anti-abuse principle has developed strongly, namely the fundamental freedoms and VAT, are also, like the State aid regime, inextricably linked to the protection of the Internal Market. In considering abuse of the

17  Professor Takis Tridimas has identified the gap-filling function of general principles in The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press 2006) 17. He rightly notes that this function is not the only function of general principles. He says they have an interpretative function (see 19), provide grounds for review (31) and give rise to a claim for damages when breached (35). The gap-filling function may have far-reaching, even constitutional, effects (see Part IV.A below). 18   Professor Tridimas has noted: ‘The meaning of fairness is so vague that it lacks objective determination.’ See Tridimas (n 17 above) 28. ‘Unfairness’ in English public law may in certain cases ground a remedy for a taxpayer: see R v IRC ex p Unilever plc [1996] STC 681. Sir Thomas Bingham MR, as he then was, said that to reject the claims to loss relief of Unilever plc would be ‘so unfair as to amount to an abuse of power’, see 691.

State Aid, Taxation and Abuse of Law  497 fundamental freedoms in Centros19 the nature of the Single Market was important in the Court’s reasoning. It said that: The right to form a company in accordance with the law of a Member State and to set up branches in other Member States is inherent in the exercise, in a single market, of the freedom of establishment guaranteed by the Treaty.20

VAT is also linked to the fundamental Market-oriented aims of the EU. The first Recital to the First Council Directive21 referred to the main objective of the Treaty being: ‘to establish within the framework of an economic union, a common market within which there is healthy competition and whose characteristics are similar to those of a domestic market’. The second Recital referred to the consequent need to have national legislation ‘such as will not distort conditions of competition’. The Sixth VAT Directive22 recorded, in the third Recital, that ‘further progress should be made in the effective removal of restrictions on the movement of persons, goods, services and capital and the integration of national economies’. In the fourth Recital it referred to a Common Market ‘resembling a real internal market’. So far as deductions are concerned, Recital twelve stated that the rules should be harmonised to the extent that they governed the amount collected and it stated the aim of a having deductible proportion calculated in a similar manner in Member States. The VAT Directive of 2006 demonstrated the same concern with the fundamental aims of the Community and the Internal Market.23 In the light of the concern of the VAT legislation with the fundamental aims of the EU and its Internal Market, it can be seen that abuse of the right to reclaim input tax is an abuse of an Internal Market permitting fair competition. The need to bear in mind the general purposes of the common system of VAT in the context of abuse was recognised in the Opinion of Advocate-General Poiares Maduro in Halifax. In his view, the prohibition of abuse should be limited to cases where the relevant economic activity has no other objective explanation than to create a claim against the tax authorities. He went on to observe that, in those circumstances, ‘recognition of the right would conflict with the purposes and results envisaged by the relevant provisions of the common system of VAT’.24 Given the obvious link between the Internal Market on the one hand and abuse of the fundamental freedoms and the right of deduction of input tax on the other, it may be that, as was suggested earlier, one can speak not merely of a principle preventing abuse of   Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459.   Centros (n 19 above), para [27]. This paragraph was specifically noted in Case C-167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I-10155, para [138]. 21   Council Directive 67/226/EEC of 11 April 1967 on the harmonisation of legislation of Member States concerning turnover taxes [1967] OJ L71/1301. 22   Council Directive 77/388/EEC of 17 May 1977 on the harmonisation of the laws of Member States relating to turnover taxes—common system of value-added tax: uniform basis of assessment [1977] OJ L145/1. 23  See Council Directive 2006/112/EC of 28 November 2006 on the common system of value-added tax [2006] OJ L347/1. It states, in Recital 4, for example: The attainment of the objective of establishing an internal market presupposes the application in Member States of legislation on turnover taxes that does not distort conditions of competition or hinder the free movement of goods and services. It is therefore necessary to achieve such harmonisation of legislation on turnover taxes by means of a system of value added tax (VAT), such as will eliminate, as far as possible, factors which may distort conditions of competition, whether at national or Community level. 24   AG Poiares Maduro in Halifax (n 5 above), para [86]. 19 20

498  Timothy Lyons right or of law, but of a more general principle preventing abuse of the Internal Market and the other aims, activities and laws of the EU. As Advocate-General Poiares Maduro has said, the Court frequently uses the term ‘abuse’ and not ‘abuse of right’ or ‘abuse of law’25 and he himself has referred to the ‘notion of abuse’.26 That notion may, of course, be referred to by terms other than ‘abuse’.27 In view of the breadth of the concept of abuse in EU law it is unsurprising that it has a significant counterpart in the political realm which is particularly apparent in the context of State aid. Advocate-General Poiares Maduro has observed: It cannot be denied that selective State measures expose the Community to a major risk: the risk of ‘abuse’ of the political system, which consists in giving an advantage to a limited group and making the other members of the community bear the cost, even though the latter have no direct influence over the decision to grant the advantage.28

A principle of abuse which seeks to ensure that the actions of economic operators and authorities are consistent with the basic purposes of the EU performs a valuable role, therefore, not merely in the EU’s legal system but also in the EU’s political system. The narrower the formulation of the anti-abuse principle, the less scope it has for limiting abuse of the political system. With that in mind we may move on to consider the antiabuse principle as it operates to control not the conduct of economic operators but of institutions and, specifically, the Commission. C.  Abuse by Institutions We have already observed that there would seem to be no fundamental reason why a general anti-abuse principle should not apply to institutions, such as the Commission and Member States, as well as to economic operators. That is particularly significant when one is considering abuse in the context of a State aid regime, overseen by the Commission and designed to control Member States. According to Article 108 of the TFEU [ex Article 88 EC], the Commission has the role of holding aid to be incompatible with the Internal Market where this is appropriate, operating always under the supervision of the EU Courts.29 The Commission is to keep under constant review all systems of existing aid in cooperation with Member States. It may decide that aid granted by a Member State is not compatible with the Internal Market, or is misused, and refer to the Court of Justice noncompliance with its decisions. The Commission is to be informed of any plans to grant or alter aid so that the compatibility of the aid with the Internal Market may be considered. Inevitably, the Commission must keep a close eye on Member States’ tax systems, although as the Court says, direct taxation falls within the competence of Member States.30  See AG Poiares Maduro in Halifax (n 5 above), para [69].  See AG Poiares Maduro in Case C-210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I-9641, para [29]. 27  See further, for example, the reference to ‘misuse’ of aid in Part III below. 28   AG Poiares Maduro in C-276/02 Spain v Commission [2004] ECR I-8091, para [25]. 29  See Case C-354/90 Fédération Nationale du Commerce Extérieur des Produits Alimentaires and Syndicat National des Négociants et Transformateurs de Saumon v French Republic [1991] ECR I-5505, para [14]; and C-143/99 Adria-Wien Pipeline GmbH and Wietersdorfer & Peggauer Zementwerke GmbH v Finanzlandesdirektion für Kärnten [2001] ECR I-8365, para [29]. 30  See for example, Joined Cases T-211/04 and T-215/04 Government of Gibraltar and United Kingdom v Commission [2008] ECR II-3745, para [146]; and outside the State aid field: Case C-303/07 Proceedings brought by Aberdeen Property Fininvest Alpha Oy [2009] ECR I 5145 para [24]. 25 26

State Aid, Taxation and Abuse of Law  499 Of course, there are other actors within the State aid system. The Council has the right, in exceptional circumstances, to decide that aid which a Member State is granting, or intends to grant, is compatible with the Internal Market. The Member State has a duty to cooperate with the Commission pursuant to Article 4(2) of the TEU [ex Article 10 EC] and as set out in Article 108 TFEU [ex Article 88(1) EC]. It must abolish existing aid which the Commission finds incompatible with the Internal Market and, subject to certain caveats, must recover such illegal aid as it has granted. Given the importance of institutional action in the State aid system, it is appropriate to note that the conditions for the application of the anti-abuse principle need not be the same for institutions as for economic operators. In Halifax the Court spoke of a ‘formal’ application of the relevant conditions leading to the accrual of an advantage contrary to the purpose of the relevant provisions and the essential aim of the transactions being to obtain the advantage.31 In Cadbury Schweppes the Court of Justice again referred to the ‘formal’ observance of conditions and the failure to achieve the objectives of freedom of establishment together with the intention of obtaining the advantage.32 Provisions seeking to counter abusive practices must, it said, prevent ‘conduct involving the creation of wholly artificial arrangements which do not reflect economic reality with a view to escaping the tax normally due on the profits generated by activities carried out on national territory’.33 The requirements for the application of the anti-abuse principle in Halifax and Cadbury Schweppes are to be regarded as appropriate given that the principle was being applied to the conduct of economic operators. They should not be regarded as confining the application of the principle to economic operators. On occasion, the impression may have been given that the principle is so confined. In Ampliscientifica Srl, for example, the Court of Justice said: [T]he principle prohibiting the abuse of rights is intended to ensure, particularly in the field of VAT, that Community legislation is not extended to cover abusive practices by economic operators, that is to say transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law.34

The reference to economic operators should be understood, however, as a reference to the nature of the person in issue in the case in question rather than any kind of limitation on the principle. Citymo35 makes clear that the principle is not limited to economic operators but encompasses the Commission. In Citymo the CFI referred to a number of ‘abuse of right’ cases from van Binsbergen to Halifax as establishing a rule that ‘individuals cannot seek to misuse Community measures’.36 It then applied that rule not to an individual but to the Commission. It went on to find that ‘by informing the applicant belatedly of its’ decision to break off the pre-contract negotiations, the Commission breached the principle of good faith to a sufficiently serious degree and abused its right not to contract’.37  See Halifax (n 5 above), paras [74–75].  See Cadbury Schweppes (n 6 above), para [64].  See Cadbury Schweppes (n 6 above), para [55]. 34   Case C-162/07 Ampliscientifica Srl, Amplifin SpA v Ministero dell’Economia e delle Finanze, Agenzia delle Entrate [2008] ECR I-4019, para [27]. See also, for example, the reference to ‘abusive practices by economic operators’ in Case C-251/06 Firma ING AUER—Die Bausoftware GmbH v Finanzamt Freistadt Rohrbach Urfahr [2007] ECR I-9689, para [41]. 35   Case T-271/04 Citymo SA v Commission [2007] ECR II-1375. 36   Citymo (n 35 above), paras [107] and [127–37]. 37   Citymo (n 35 above), para [137]. 31 32 33

500  Timothy Lyons An award of €20,000 was made against the Commission, plus interest, for the applicant’s loss of opportunity to let to a third party a building which it had been negotiating to let to the Commission. The Commission was also found to have infringed the applicant’s legitimate expectations. It is noteworthy that the abuse of right was considered together with the principle of good faith. The conduct of the Commission was considered to be a breach of both principles. It seems that a finding of abuse of right was not necessary to found the Court’s judgment, but it is inevitable that a general anti-abuse principle will be engaged in many situations in which it does not work alone. Given that the misuse of the right arose during the course of the fulfilment of the Commission’s duty it may be thought that the case may be analysed more naturally as an abuse of power or a breach of the Commission’s duty of good administration rather than an abuse of right. Abuse of power is, after all, a complaint which it has been possible to direct at the Commission for a long time, as the State aid case Italy v Commission demonstrates.38 Whether appearing in the guise of abuse of right or abuse of power, however, it is plain that an anti-abuse principle was relevant to the case. Furthermore, it was quite reasonable for the anti-abuse principle to be applied without any reference to matters such as ‘formal observance’ of conditions. As Advocate-General Poiares Maduro said in his Opinion in Halifax: ‘the absence of a unitary test for the operation in every field of Community law of the principle of the prohibition of abuse must be regarded as perfectly natural in Community law, as it is in any national legal system’.39 Having established that an anti-abuse principle is not confined to economic operators but may operate in relation to the institutional actors in the Union and, therefore, in the State aid regime, and may be applied pursuant to different rules according to the prevailing circumstances, we move on to consider how the specific rules governing the State aid regime are influenced by the need to prevent abuse. II. STATE AID, TAX AND THE PREVENTION OF ABUSE

To the extent that State aid rules are designed to prevent abuse of the Internal Market, they may be seen as the expression of an anti-abuse principle. Furthermore, the rules governing State aid in general and their application to fiscal State aid in particular, although they cannot be considered in any detail here, demonstrate a clear intention to prevent abuse. As is well known, Article 107(1) of the TFEU [ex Article 87(1) EC] sets out the definition of State aid.40 Article 107(2) [ex Article 87(2) TEC] then specifies certain aid which is compatible with the Internal Market, such as aid having a social character. Article 107(3) [ex Article 87(3) EC] goes on to set out certain categories of aid which may be considered compatible with the Internal Market, for example, aid to promote the economic development of areas where the standard of living is abnormally low.   Case 173/73 Italy v Commission [1974] ECR 709.  See Halifax (n 5 above), para [83].  It states: Save as otherwise provided in the Treaties, any aid granted by a Member State or through State resources in any form whatsoever which distorts or threatens to distort competition by favouring certain undertakings or the production of certain goods shall, insofar as it affects trade between Member States, be incompatible with the internal market. 38 39 40

State Aid, Taxation and Abuse of Law  501 Article 107(1) sets out four conditions, each of which must be satisfied if illegal State aid is to be found to exist. Lord Woolf identified them in Lunn Poly as follows: Ultimately the court has to ask itself the global question whether the matters complained of constitute (i) an aid, (ii) granted by a Member State or through State resources in any form whatsoever which (iii) distorts or threatens to distort competition and (iv) which affects trade between Member States.41

As may be expected, the Courts of the EU also follow a fourfold division of the issues. The Court of First Instance, as it was then called, in a case in which it found illegal State aid consisting of, amongst other things, tax exemptions, noted that: The principle that State aid is prohibited, laid down in Article 87(1) EC, [now Article 107(1) TFEU] includes the following requirements. First, there must be an intervention by the State or through State resources. Second, that intervention must confer an advantage on its recipient. Third, it must be liable to affect trade between Member States. Fourth, it must distort or threaten to distort competition.42

Each of the requirements of Article 107 must be satisfied if the existence of State aid is to be established.43 It is essential, if the State aid regime is to fulfil its purpose of preventing the abuse of the Internal Market, that the definition of aid is widely drawn. The width of the definition (‘any aid granted by a Member State or through state resources in any form whatsoever’) may be seen, broadly speaking, as the expression of an anti-abuse principle. It is clear that the definition in Article 107 TFEU encompasses fiscal aid44 whether national, provincial or municipal. Not only is the State aid illegal, the tax may be invalid.45 The concentration on the effects of the aid, rather than the form of it and whether it is fiscal or not, is also what one would expect from a system of rules designed to combat a distortion, or abuse, of the Internal Market. The Commission must establish whether   R v Commissioners of Customs and Excise ex p Lunn Poly Ltd et al [1999] STC 350, 358.   Case T-442/03 SIC –Sociedade Independente de Comunicação, SA v Commission [2008] ECR II-1161, para [44]. See for a more recent statement of the four issues: Case T-369/06 Holland Malt BV v Commission, CFI Judgment of 9 September 2009, nyr, para [35]. The Commission also formulated four questions. They are: ‘Does a selective advantage exist? Are state resources involved? Is Community trade affected? Is the measure justified by the nature of the tax system?’ See: ‘Report on the implementation of the Commission notice on the application of the state aid rules to measures relating to direct business taxation’ (Report) C(2004) 434, para [7]. 43  See Case C-482/99 France v Commission [2002] ECR I‑4397, para [68] and the authorities cited there. The requirements are subject to development by EU Courts, eg, the imputability of aid to the State was considered in Case T-351/02 Deutsche Bahn v Commission [2006] ECR II-1047. Deutsche Bahn complained that the German exemption of excise duty in respect of aviation fuel was a State aid. The German law, however, followed an exemption in Council Directive 92/81/EEC of 19 October 1992 on the harmonization of the structures of excise duties on mineral oils [1992] OJ L316/12. The exemption was not imputable to the State and was not State aid (see paras [101–103]). See further J-D Braun and J Kühling, ‘Article 87 EC and the Community Courts: From Revolution to Evolution’ (2008) 45 Common Market Law Review 465, 466. 44  As the court confirmed in Italy v Commission (n 38 above). Italy unsuccessfully argued that, by encroaching into the field of fiscal aid, the Commission was guilty of an abuse of power: see para [13]. For recent confirmation of the position and of the width of the definition of State aid see for example, Joined Cases T‑227/01 to T‑229/01, T‑265/01, T‑266/01 and T‑270/01 Territorio Histórico de Álava—Diputación Foral de Álava et al v Commission of the European Communities, [2009] ECR II-3029. The Commission Notice on the application of the State aid rules to measures relating to direct business taxation [1998] OJ C 384/3 states: ‘it is irrelevant whether the measure is a tax measure’ (para [8]). The Commission is considering the extension of the Notice to indirect taxation, see the State Aid Action Plan (n 8 above), para [64]. 45   Joined Cases C-266/04 to C-270/04, C-276/04 and C-321/04 to C-325/04 Distribution Casino France SAS, formerly Nazairdis SAS et al v Organic [2005] ECR I-9481, para [35]. See, for a discussion of taxation and State aid: C Quigley, European State Aid Law and Policy, 2nd edn (Oxford, Hart Publishing 2009) Ch 3. 41 42

502  Timothy Lyons or not the measures in question are liable to affect trade between Member States and to distort it.46 It is the effects of the aid with which Article 107 TFEU is concerned, not its causes or objectives.47 The concern with the effects of aid springs from the same well as the concern to counter artificiality in other cases concerned with abuse. Professor Dr Frans Vanistendael has, with respect, rightly noted that both Halifax and Cadbury Schweppes were concerned with economic reality concluding that ‘both decisions have used a reasoning to refute tax avoidance that is very similar if not identical: the economic reality of the transaction’.48 In the context of State aid, the concern with economic effects and the need to have regard to commercial reality is demonstrated with particular clarity in relation to the requirement of ‘selectivity’. The Court of Justice has said that what is now Article 107(1) TFEU requires a determination of: whether, under a particular statutory scheme, a State measure is such as to favour certain undertakings or the production of certain goods in comparison with others which, in the light of the objective pursued by the system in question, are in a comparable legal and factual situation . . . If that is the case, the measure concerned satisfies the condition of selectivity which is a constituent of the concept of State aid under that provision.49

In determining whether the condition of selectivity is satisfied, a ‘private operator test’ may be employed. This test may take two different forms depending upon whether it is applied as a ‘private investor test’ or a ‘private creditor test’.50 The latter is relevant, for example, to the issue of whether or not a favourable tax treatment for companies in commercial difficulty constitutes State aid.51 Having seen that the State aid rules are widely drawn to ensure that abuse is prohibited, and that the rules are concerned with economic reality and economic effects, we can move on to look specifically at fiscal aid. A.  Fiscal Aid: Specific Considerations The concept of fiscal State aid is of particular contemporary significance. It would appear that there were few cases concerning fiscal State aid prior to 2001,52 although as 46   Case C-148/04 Unicredito Italiano SpA v Agenzia delle Entrate, Ufficio Genova 1 [2005] ECR I-11137, para [55]. 47  See, for a recent re-statement of this, Case C-487/06 P British Aggregates Association v Commission of the European Communities [2008] ECR I-10505, para [85]. 48  F Vanistendael, ‘Halifax and Cadbury Schweppes: One Single European Theory of Abuse in Tax Law?’ (2006) 15 EC Tax Review 192, 195. 49   Case C-308/01 GIL Insurance Ltd et al v Commissioners of Customs & Excise [2004] ECR I-4777, para [68]. The Court relied upon Case C-126/01 Ministre de l’économie, des finances et de l’industrie v GEMO SA [2003] ECR I-13769, para [35], and Case C-409/00 Spain v Commission [2003] ECR I-1487, para [47] in support of the final sentence; and on Adria-Wien Pipeline (n 29 above), para [41] and Spain v Commission (this note above), para [47] for the earlier part of the statement. See further, on the concept of ‘selectivity’: Government of Gibraltar (n 30 above), paras [143–44]; British Aggregates (n 47 above), paras [81–85]; and Case T-211/05 Italy v Commission, [2009] ECR II-2777. 50  For some discussion of this see AG Poiares Maduro in Spain v Commission (n 28 above), paras [20–25]. See also Bundesverband deutscher Banken (n 3 above). 51  The possible loss of tax revenue in a statutory system establishing a regime governing an insolvent undertaking and the general body of creditors does not constitute State aid, but State aid will exist if there is specific and selective advantage: see for example, Case C-480/98 Spain v Commission [2000] ECR I-8717. 52   As the Commission noted at para [1] of the Report referred to (n 42 above).

State Aid, Taxation and Abuse of Law  503 long ago as 1994 the Court of Justice confirmed that tax exemptions could constitute State aid53 and, as we have seen, Italy’s contention that fiscal aid was outside the State aid regime failed in the first half of the 1970s.54 In recent years fiscal State aid has been of some considerable significance both in the EU Courts55 and in the UK courts.56 The role of national courts is, of course, of very significant.57 Fiscal State aid has also been the subject of considerable academic discussion.58 Not all cases of State aid come before the Courts of course. Plainly, many important cases on fiscal State aid have been dealt with by the Commission.59 State aid has, of course, become of great significance in the current 53   Case C-387/92 Banco de Crédito Industrial SA, now Banco Exterior de España SA v Ayuntamiento de Valencia [1994] ECR I-877, para [14]. 54  See Italy v Commission (n 38 above). 55  See for examples in 2009: Territorio Histórico de Álava (n 44 above), Italy v Commission (n 49 above), Case T-189/03 ASM Brescia SpA v Commission [2009] ECR II-1831; in 2008: Joined Cases C-341/06 P and C-342/06 P Chronopost SA and La Poste v Union française de l’express (UFEX) et al [2008] ECR I-477; Government of Gibraltar (n 30 above); British Aggregates (n 47 above); in 2007: Case C-441/06 Commission v France [2007] ECR I-8887; in 2006: Joined Cases C-485/03 to 490/03 Commission v Spain [2006] ECR I-11887; Air Liquide (n 4 above), Case C-88/03 Portugal v Commission [2006] ECR I-7115; and in 2005: Unicredito Italiano (n 46 above). 56  See for example, R v Attorney General ex p Imperial Chemical Industries Ltd [1985] CMLR 588 (Div Ct), [1987] 1 CMLR 72 (CA); R v Commissioners of Customs & Excise ex p Lunn Poly [1998] STC 649 (HC), [1999] STC 350 (CA) (Lunn Poly was followed by the VAT and Duties Tribunal in CGNU, Airtours et al v Commissioners of Customs & Excise IPT 00007, 8 October 2001, unreported); R ex p Professional Contractors’ Group et al v IRC [2001] STC 629 (HC), [2002] STC 165 (CA); R ex p British Aggregates Association et al v Commissioners of Customs & Excise [2002] EuLR 394 (HC) (see further for the decision of the ECJ, n 47 above); GIL Insurance Ltd v Commissioners of Customs & Excise [2000] STC 204 (HC) (see further for the decision of the ECJ, n 49 above); University of Sussex v Commissioners of Customs & Excise [2001] STC 1495 (HC). State aid was raised in Page trading as Empowerment Enterprises Ltd v Commissioners of Customs & Excise [2004] UKVAT V18820 but not considered. 57  See for example, Case C-199/06 Centre d’exportation du livre français (CELF), Ministre de la culture et de la communication v Société internationale de diffusion et d’édition (SIDE) [2008] ECR I-469, which considers the role of domestic courts in relation to illegal State aid. The Court of Justice noted that ‘national courts do no more than preserve, until the final decision of the Commission, the rights of individuals faced with a possible breach by State authorities of the prohibition laid down by Article 88(3) EC [now Art 108(3) TFEU]’ and referred to Case C‑354/90 Fédération nationale du commerce extérieur des produits alimentaires and Syndicat national des négociants et transformateurs de saumon v France [1991] ECR I‑5505. See the Study on the Enforcement of State Aid Law at National Level (March 2006) requested by the Commission and coordinated by Thomas Jestaedt, Jones Day, Jacques Derenne, Lovells, Tom Ottervanger, Allen & Overy, Section 2.4, ‘Assessment of the existing system’ 467, available at ec.europa.eu/competition/state_aid/studies_reports/studies_reports.html. The study was updated in 2009. It notes that ‘the most frequent cases dealt with tax measures’. See ec.europa. eu/competition/state_aid/studies_reports/enforcement_study_2009.pdf, 2. On the importance of the role of the courts in State aid law see for example, Commissioner Neelie Kroes’ comments in the foreword to the Study: Regrettably, however, the powers of national courts are still not being used to the full. Companies are often hesitant to launch proceedings at national level to put an end to the illegal granting of aid to their competitors, or to assert their rights to claim compensation. In some cases, national courts themselves seem uncertain about the powers that they have in the State aid arena, and the locus standi of competitors. 58   A full bibliography cannot be given here. The reader may wish to refer to Quigley (n 45 above). See also C Quigley, ‘General Taxation and State Aid’ in A Biondi, P Eeckhout and J Flynn (eds), The Law of State Aid in the European Union (Oxford, Oxford University Press 2004) 207; and K Bacon, ‘State Aids in the English Courts: Definition and Other Problems’ in Biondi et al (this note above) 337. 59  See for example, Commission Decision of 28 October 2009 on the tax amortization of financial goodwill for foreign shareholding acquisitions No C45/2007 (ex NN51/2007, ex CP9/2007) implemented by Spain and Commission Decision of 18 July 2007 on State Aid No C 37/2005 (ex NN 11/2004) implemented by Greece Tax Exempt Reserve Fund. The Commission decisions on coordination centres are discussed, for example, in the Report (n 42 above), para [10]. As to Gibraltar see, amongst other matters, Commission Decision of 30 March 2004 on the aid scheme which the United Kingdom is planning to implement as regards the Government of Gibraltar Corporation Tax Reform [2005] OJ L85/1 and Commission Decision of 30 March 2004 on the aid scheme implemented by the United Kingdom in favour of Gibraltar Qualifying Companies [2005] OJ L29/24. See too Government of Gibraltar (n 30 above).

504  Timothy Lyons economic crisis. Although little assistance has been granted in relation to taxation, it is noteworthy that the Commission has acknowledged the possibility of a general extension of tax payment deadlines.60 Whilst there may be a contemporary interest in fiscal State aid, it has long been a concern of the Commission. As is well known, in 1998 the Commission published a Notice on the application of the State aid rules to direct business taxation.61 The Report on the implementation of the Notice, published in 2004,62 observed that the Notice ‘did not signal any change in its approach to aid in the form of tax measures’.63 The Notice itself states, amongst other things, that the advantage which is essential for State aid to exist must relieve recipients of charges normally borne from their budgets.64 As the Court of Justice has put it: The concept of aid is more general than that of a subsidy. It embraces not only positive benefits, but also measures which, in various forms, mitigate the charges which are normally included in the budget of an undertaking and which, without therefore being subsidies in the strict meaning of the word, are similar in character and have the same effect.65

As the Notice indicates, an advantage may be conferred in a variety of ways: by a reduction in the tax base, a total or partial reduction in the amount of tax, or a deferment, rescheduling, or special rescheduling of the tax debt.66 It is, of course, plain that a tax exemption may constitute State aid.67 The concern to prohibit abuse is present then, not just in the width of the general definition of State aid and its concern with economic reality and economic effects, but also in the width of the understanding of prohibited fiscal aid in particular. We now move on to consider briefly the justifications for State activity. i.  Justifications for State Activity We noted earlier, that a measure may fall outside the scope of prohibited State aid for a number of reasons, but the justification of a measure by reference to the nature or general scheme of the tax system is also important.68 Where such a justification applies the necessary selectivity is absent. One case which demonstrated this was GIL Insurance Ltd in which the Court of Justice relied upon the general scheme of a tax system to justify differential rates of insurance premium tax.69 In doing so, the Court will have encouraged tax authorities that wish to counter what they perceive as abusive conduct (using the term in its widest possible sense) by noting that: [The higher rate of tax] was introduced to counteract the practice of taking advantage of the difference between the standard rate of IPT and that of VAT by manipulating the prices of rental or sale of appliances and of the associated insurance. Such conduct had given rise to a loss of 60   Communication from the Commission—Temporary Community framework for State aid measures to support access to finance in the current financial and economic crisis [2009] OJ C83/1, 83/3. 61  See the Notice (n 44 above) for the full reference. 62  See the Report (n 42 above) for the full reference. 63  See the Report (n 42 above), para [54]. 64  See the Notice (n 44 above), para [9]. 65   Adria-Wien Pipeline (n 29 above), para [38]. 66  See the Notice (n 44 above), para [9]. 67  See for example, Air Liquide (n 4 above), paras [29–30]. 68  See for example, Italy v Commission (n 38 above), para [15]; and paras [13–16] of the Notice (n 44 above). 69   GIL Insurance (n 49 above).

State Aid, Taxation and Abuse of Law  505 income in terms of VAT receipts and to shifts in the conditions of competition in the domestic appliance sector.70

A number of difficulties appear to have arisen in relation to the concept of selectivity and the identification of a general scheme, including a general scheme of taxation.71 It should be noted, however, that a desire to combat abuse and distortions in the conditions of competition informs the definition of the general scheme of taxation just as it informs other aspects of State aid law.72 Of course, what has been said so far on the influence of an anti-abuse principle on State aid law is very general. It is time to consider whether or not an anti-abuse principle could have a specific and more visible role in State aid law. III.  A SPECIFIC ROLE FOR GENERAL PRINCIPLES IN STATE AID LAW

Economic operators may want a right to refuse to pay a tax because a competitor receives State aid by virtue of an exemption from it, but it is clear that no such right exists.73 Furthermore, a Member State has no right to respond to the grant of State aid by limiting the fundamental freedoms.74 Yet there are some important rights provided in the context of State aid law. Even if the anti-abuse principle were to be regarded as a relatively narrow abuse of right principle it would still have a role in relation to these matters. State aid law refers to a general principle of EU law, as we have already noted, in Council Regulation (EC) No 659/1999,75 which lays down detailed rules for the application of what is now Article 107 TFEU [ex Article 87 EC]. Article 14 of the Regulation deals with the recovery of illegal aid and states that, where a negative decision is taken by the Commission in a case of unlawful aid, the Commission shall decide that the Member State shall take all necessary measures to recover the aid from the beneficiary. It then provides this important qualification in the final sentence of Article 14(1): ‘The Commission shall not require recovery of the aid if this would be contrary to a general principle of Community law.’ A clearer link between general principles of EU law and State aid could not be made. It has been said that: ‘The provision serves in effect as a mandate to the ECJ to colonise this field fully.’76 The remark was made in the context of a discussion of legitimate expectations and there is, no doubt, much scope for that principle in cases concerned with recovery of illegal aid. Unicredito Italiano77 is an example of a case in which the principle of legitimate expectations and legal certainty was considered78 along with the general principle of proportionality.79 In fact, none of these principles assisted the applicant in   GIL Insurance (n 49 above), para [74].  See for example, the comments made in B Kurcz and D Vallindas, ‘Can General Measures Be . . . Selective? Some Thoughts on the Interpretation of a State Aid Definition’ [2008] 45 Common Market Law Review 159. 72  Member States, as well as the Commission, may be involved in the fight against abuse: see for example, Case T-289/03 British United Provident Association Ltd (BUPA) et al v Commission [2008] ECR II-81. Ireland could properly seek to counter ‘abuse’ of services in the general economic interest (private medical insurance) in respect of which it made payments which did not constitute State aid: see, in particular, para [199]. 73  See Air Liquide (n 4 above), para [43]. 74  See AG Léger in Cadbury Schweppes (n 6 above), para [58]. 75   Council Regulation (EC) No 659/1999 (n 7 above). 76   Tridimas (n 17 above) 296. 77   Unicredito Italiano (n 46 above). 78   Unicredito Italiano (n 46 above), paras [104–12]. 79   Unicredito Italiano (n 46 above), paras [113–19]. 70 71

506  Timothy Lyons the case, although the Court noted, having regard to Article 14(1) (referred to above), that ‘a national measure providing for repayment of aid in compliance with a Commission decision is unlawful where that decision is contrary to a rule of Community law’.80 One may also note that the general principles of effectiveness and equivalence must be complied with in protecting the rights of individuals against a disregard, by Member States, of the prohibition on granting aid before it has been approved by the Commission.81 Furthermore, the general principles of EU law have to be respected by the Commission when it decides not to raise any objection to a scheme of aid as well as when illegal aid is to be recovered. One example of a case in which a decision of the Commission not to object to a scheme of aid was analysed in the context of the general principle of equal treatment is Nuova Agricast.82 So far as the anti-abuse principle is concerned, it may be that an action for recovery could be met in appropriate circumstances with the defence that the Commission and the Member State were abusing their right to recovery. If the negative right not to contract may be abused, so too may the positive right to recover. While the recovery of illegal aid is one area in which the operation of general principles—including a general antiabuse principle—is specifically encouraged, qualification for the aid in the first place is also an area where an anti-abuse principle could be relevant. That much is shown by the recent references made to the Court of Justice83 and in Italy v Commission84 noted in the introduction to this chapter. Another area in which a general anti-abuse principle may be relevant for Member States and economic operators concerns the misuse of aid. Article 108(2) TFEU [ex Article 88(2) EC] contains specific provisions dealing with this. It provides, amongst other things, that the Commission may decide that aid ‘is being misused’. Where it does so it shall decide that the State concerned shall abolish or alter the aid. What constitutes the misuse of aid is defined in Council Regulation (EC) No 659/1999, Article 1(g).85 It is said to be: ‘aid which is used by the beneficiary in contravention of a decision taken pursuant to Article 4(3) or Article 7(3) or (4) of this Regulation’. All three provisions mentioned in Article 1(g) refer to situations in which the compatibility of aid with the Common Market, as it is there referred to, is relevant. ‘Misuse’, therefore, is defined by reference to the aim of the Community to establish a Common Market. The equivalence between ‘abuse’ and ‘misuse’ is relatively clear. The fact that there is, what may be regarded as a specific expression of the anti-abuse principle may mean, however, that the principle in its general form, though relevant, will not add significantly to the reasoning in any specific any case on the topic. It may be noted here that the fact that ‘misuse’ is employed rather than ‘abuse’ is a linguistic choice without legal significance. In Akrich, for example, having considered abuse of law in that case, Advocate-General Geelhoed concluded that ‘there can be no question of a misuse of Community law’.86 One commentator has noted ‘terminological discrepancies’ and the use of a number of words, such as, ‘avoidance’, ‘evasion’,   Unicredito Italiano (n 46 above), para [123].   Transalpine Olleitung (n 4 above), para [45]. 82   Case C-390/06 Nuova Agricast Srl v Ministero delle Attività Produttive [2008] ECR I-2577. 83  See above (n 2). 84   Italy v Commission (n 1 above). 85   Council Regulation (EC) No 659/1999 (n 7 above). 86   AG Geelhoed in Case C-109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I-9609, paras [181–84]. 80 81

State Aid, Taxation and Abuse of Law  507 ‘circumven­tion’ and ‘fraud’.87 This variety of terms is not an indication of inexactitude but of the scope of the concept of abuse. If an anti-abuse principle may be relevant when the Commission is deciding on misuse of aid, it may also be relevant to the Commission’s actions more generally. It is plain that its decision-making responsibilities in the State aid field are subject to general provisions such as Article 263(4) TFEU [ex Article 230(4) EC] which allows a person to challenge a decision of ‘direct and individual concern’88 and to Article 296 TFEU [ex Article 253 EC], which, amongst other things, provides that decisions adopted by the Commission are to state the reasons on which they are based.89 It seems unlikely that the presently existing grounds for mounting a challenge to a Commission decision would, in the vast majority of cases, be much extended by the availability of a general anti-abuse principle. As was noted earlier, the principle of abuse of power has always been applicable to the Commission’s conduct in the State aid sphere as elsewhere.90 Nevertheless, the possibility of a specific role for an anti-abuse principle in relation to the activities of the Commission in the State aid regime, as well as in relation to economic operators and Member States, seems clear. IV.  CONCLUDING REMARKS

The State aid regime provides a context in which to consider abuse as a general principle of EU law which is, plainly, considerably different from that offered by the law governing the fundamental freedoms and VAT and the right to deduction of input tax. There are also significant similarities between State aid law, the fundamental freedoms and the law of VAT. Fundamentally they are all concerned with the creation, preservation and operation of the Internal Market, in which there must be no distortion of competition. By looking at the anti-abuse principle in the light of these contrasts and similarities a number of considerations may become apparent. I highlight three here. First of all, there are good grounds for contending that the anti-abuse principle operates to protect not merely EU rights, or EU law, but the creations of the Union, in particular, the Internal Market itself. The need to protect the Market requires that the anti-abuse principle be fundamentally concerned with economic reality, something on which State aid law concentrates. Secondly, tax is not beyond the reach of the general anti-abuse principle. That is demonstrated by Halifax91 and Cadbury Schweppes92 and, it may be said, by the inclusion of fiscal aid within the concept of State aid. Thirdly, whilst the anti-abuse principle may be applied to economic operators within the State aid regime, its application is not confined to them. It is capable of operating to 87  See R de la Feria, ‘Prohibition of Abuse of Community Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395, 396. 88  For a recent case in which Art 230 [now Art 263 TFEU] was relevant see Case C-521/06 P Athinaïki Techniki AE v Commission of the European Communities [2008] ECR I-5829. 89   A decision by the Commission that aid does not distort the Internal Market must, therefore, contain the reasons for that conclusion. For a State aid case in which Art 253 EC, as it then was, was relevant see Nuova Agricast (n 82 above). 90   The issue of whether or not the Commission had abused its powers was in issue in Italy v Commission (n 38 above). 91   Halifax (n 5 above). 92   Cadbury Schweppes (n 6 above).

508  Timothy Lyons govern the behaviour of other actors, including the European Commission. The specific rules governing the application of the general anti-abuse principle may, of course, differ according to the context in which it operates. In Citymo93 there was no need to satisfy rules established, in Halifax and in Cadbury Schweppes, in relation to economic operators when applying the anti-abuse principle to the Commission. A.  The Judicial Use of an Anti-Abuse Principle It is one thing to suggest that there are good grounds for contending that a long-standing general anti-abuse principle exists in EU law and is relevant to the field of State aid.94 It is quite another to recommend to the Courts the use of the general principle as a tool of dispute resolution. The use of general principles by the Courts of the EU can entail the exercise of judicial discretion to a considerable degree, unless a legislature provides specific rules for their application. This is not surprising if it is remembered that, as Dworkin has said, principles deal with the requirements of ‘justice, fairness or some other dimension of morality’.95 Professor Melchior Wathelet, formerly a judge in the Court of Justice, of course, has expressed some concern that the predictability and objectivity of the law may suffer from the use of the general principle of proportionality in cases in which the fundamental freedoms are applied to tax law.96 In my view, similar concerns may well be justified over the enthusiastic use of a general anti-abuse principle.97 The principle should be used judicially, as Advocate-General Poiares Maduro has indicated, with ‘considerable restraint’.98 Rights are not lightly to be qualified by the EU Courts. But extensive use of the anti-abuse principle in dispute resolution will not affect just the specific interests of economic operators and, on occasions, national and Union institutions. It may well harm the Union’s legal and political order more generally—the very things a general anti-abuse principle is intended to protect.

  Citymo (n 35 above).   Any suggestion that an anti-abuse principle is new is unconvincing. It has been said that: ‘The Court of Justice has been alluding to abuse and abusive practises in its rulings for more than thirty years.’ See de la Feria (n 87 above) 395. 95   Dworkin (n 11 above) 22. 96  M Wathelet, ‘Marks & Spencer plc v Halsey: Lessons to Be Drawn’ (2006) 2 British Tax Review 128, 132. The Court of Justice is clearly conscious of the need for self-restraint in applying an anti-abuse principle and of the possible role of the legislature. In, for example, Centros (n 19 above), it recognised that the Council had the power to harmonise company law if that was thought desirable (see para [28]). Prof Melchior Wathelet was a member of the Court which heard Centros. 97   The potentially dramatic impact of the judicial use of the anti-abuse principle is well shown by Cussens & Ors v Brosnan [2008] IEHC 169 in which the Irish High Court (Charleton J) rejected the submission that the principle was inconsistent with the Irish courts’ tradition of not intervening where the legislature has failed to ‘plug a hole’ in advance: see para [29]. 98  See AG Poiares Maduro in Cartesio (n 26 above), para [29]. 93 94

35 Comments on State Aid, Taxation and Abuse of Law CONOR QUIGLEY

T

his chapter offers comments on Timothy Lyon’s contribution to this volume: ‘State Aid, Taxation and Abuse of Law’.1 I. State aid in the context of EC law

Article 107(1) TFEU [ex Article 87(1) EC] essentially prohibits State aid as being incompatible with the Internal Market, subject to various exceptions. It is situated in the section of the Treaty dealing with competition policy, which is itself a subsection of the part dealing with common rules on competition, taxation and approximation of laws. As such, this Article has been described as a lex specialis under the aegis of the general rules laid down by Article 116 TFEU [ex Article 96 EC] which aim to remove distortions of competition.2 The State aid provisions differ from other competition law rules in Articles 101-102 TFEU [ex Articles 81–82 EC] on the basis that they are addressed to the Member States, rather than to the behaviour of individuals. This has effects in the context of rights and responsibilities. In particular, it is the responsibility of the Member States not to introduce aid without previously seeking authorisation from the Commission. Member States must give a full notification of the aid measure. If the aid is unlawfully implemented, Member States must take the necessary measures to effect recovery. The individual beneficiary of aid has little say in these procedures. The only specific responsibility placed on the recipient is not to misuse the aid. Even then, the extent of the recipient’s legal obligation is uncertain. Rather, where aid is misused, meaning that it is used by the recipient for purposes other than those for which it was approved,3 the Commission should order the Member State that granted the aid to recover it from the beneficiary.4 It follows that, in examining whether there exists in relation to State aid a principle of abuse of law, the principal focus is on the role of the Member States. However, it must first be noted that State aid law is principally concerned with controlling the activities of   T Lyons, ‘State Aid, Taxation and Abuse of Law’, ch 34 above.  AG Geelhoed in Case C-308/01 GIL Insurance Ltd v Commissioners of Customs and Excise [2004] ECR I-4777, para [67]; Case C-174/02 Streekgeweest Westelijk Noord-Brabant v Staatsecretaris van Financiën [2005] ECR I-85, para [24]. 3   Council Regulation (EC) No 659/1999 of 22 March 1999 laying down detailed rules for the application of Article 93 of the EC Treaty [1999] OJ L83/1, Art 1(g). 4  Art 108(2) TFEU [ex Art 88(2) EC]; Council Regulation (EC) No 659/1999 (n 3 above), Art 16. 1 2

510  Conor Quigley Member States, rather than granting them rights. In that respect, the State aid provisions differ fundamentally from the provisions of the TFEU that allow the Member States to restrict freedom of movement. Any decision to grant State aid remains solely within the discretion of the Member States. There is no requirement under EU law that State aid be granted. Thus, each Member State is free to determine whether it wishes to intervene in the economy by means of public subsidy to undertakings. The State aid provisions merely determine whether any particular aid measure is permissible. Specifically, Article 107(1) TFEU determines whether a given measure is to be categorised as State aid, and Articles 107(2)–(3) TFEU [ex Article 87(2)–(3) EC] determines whether that aid may be declared compatible with the Internal Market. Thus, the control exercised by the Commission is essentially negative, in that it prohibits incompatible State aid, rather than positive, in that the Commission cannot require Member States to grant aid. By virtue of that negative character, it is difficult to relate the action of Member States to a notion of abuse. If the Member State infringes the State aid rules, it is not obvious that it is abusing those rules or that its conduct amounts to an abuse of law. The Member States’ conduct is merely subject to supervisory control and, if the rules are infringed, the Member State may be deemed to have acted unlawfully. Nevertheless, from another perspective, it could be argued that all infringements of State aid law amount to an abuse of rights, in that the Member States have a sovereign right to intervene in the economy, subject to limitations, and that any exercise of that right in breach of those limitations amounts to an abuse. II. State aid and taxation

This limitation is particularly important in the sphere of taxation. Taxation remains essentially a matter within the scope of national sovereignty.5 On that basis, for the most part Member States are free to design their own tax system, particularly as regards income and corporation taxes. Specifically, Member States may determine whether they wish to grant tax exemptions or derogations in respect of particular undertakings or activities. However, tax measures that favour certain undertakings or the production of certain goods may fall within the scope of Article 107(1) TFEU. Indeed, the Commission estimates that up to forty per cent of all State aid measures are granted through the tax system.6 Moreover, the Commission positively encourages the use of the tax system as a means of granting subsidies. Tax exemptions may be viewed as economically more efficient than financial grants, since they generally presuppose that the aid is funded out of the recipient’s profits. General measures of economic policy are normally outside of the scope of application of Article 107(1) TFEU and are not considered as measures involving State aid.7 It follows 5   Joined Cases T-211/04 & T-215/04 Government of Gibraltar and United Kingdom v Commission [2008] ECR II-3745, para [146]. 6   European Commission, ‘State Aid Scoreboard: Autumn 2008 Update’ COM(2008) 751 final 49. 7  AG Saggio in Joined Cases C-400/97 & C-402/97 Administración del Estado v Juntas Generales de Guipúzcoa [2000] ECR I-1073, para [32]. See W Schön, ‘Taxation and State Aid in the European Union’ (1999) 36 Common Market Law Review 911; J Bourgeois, ‘State Aids, Taxation Measures and Specificity’ in M Dony and A de Walsche (eds), Mélanges en hommage à Michel Waelbroeck (Brussels, Bruylant 1999) 765; C Quigley, ‘General Taxation and State Aid’ in A Biondi, P Eeckhout and J Flynn (eds), The Law of State Aid in the European Union (Oxford, Oxford University Press 2004) 207.

Comments on State Aid, Taxation and Abuse of Law  511 that measures setting the general rates of corporate taxation do not constitute State aid, since they are generally applicable throughout the Member State to all undertakings operating within the State. Similarly, any uniform reduction in the general rate will not involve aid being granted within the meaning of Article 107(1) TFEU, since all undertakings are equally affected so that the criterion of selectivity is not fulfilled.8 However, where a measure does not apply to all economic operators, it cannot be considered to be a general measure of tax or economic policy.9 In order to bring all tax-based State aid within the scope of Article 107(1) TFEU, the Commission, supported by the Court of Justice, has given a broad definition of aid. The Court has consistently stated that this provision does not distinguish between measures of State intervention by reference to their causes or aims but defines them in relation to their effects.10 Thus, the fiscal aim of a measure cannot suffice to shield it from the application of Article 107(1) TFEU.11 It is, however, rather more accurate to say that measures that fulfil the criteria of State aid do not fall outside of the scope of this provision merely because they might seek to achieve some other social or economic objective.12 In any event, the mere fact that a measure may have fiscal, social or other objectives is irrelevant. State aid may also be granted indirectly through the tax system. For example, a tax deduction granted in respect of investments made by the taxpayer in firms in particular regions may be classified as the indirect grant of aid to the companies in which the investment is made. The ECJ held, in Germany v Commission, that a tax relief granted to individuals who invested in companies situated in Berlin or the new Länder constituted aid in favour of those companies. The advantage indirectly conferred on the companies was the renunciation of tax revenue which would normally have been received by the State, inasmuch as it was this renunciation which enabled the investors to take up holdings in those companies on conditions which were in tax terms more advantageous.13 There is no advantage for the purposes of establishing State aid where a difference in treatment is justified by reasons relating to the logic of the system.14 Tax measures which differentiate between undertakings and which might appear, therefore, to be selective will not constitute State aid where the differentiation arises from the nature or the overall     8  See generally, the Commission Notice on the application of the State aid rules to measures relating to direct business taxation [1998] OJ C384/3.     9   Case C-66/02 Italy v Commission [2005] ECR I-10901, para [99]; Case C-148/04 Unicredito Italiano SpA v Agenzia delle Entrate, Ufficio Genova 1 [2005] ECR I-11137, para [49]; Case C-222/04 Ministero dell’Economia e delle Finanze v Cassa di Risparmio di Firenze SpA [2006] ECR I-289, para [135]. 10   Case 173/73 Italy v Commission [1974] ECR 709, para [13]; Case C-241/94 France v Commission [1996] ECR I-4551, para [20]; Case C-75/97 Belgium v Commission [1999] ECR I-3671, para [25]; Case C-480/98 Spain v Commission [2000] ECR I-8717, para [16]; Case C-382/99 Netherlands v Commission [2002] ECR I-5163, para [61]; Case C-159/01 Netherlands v Commission [2004] ECR I-4461, para [51]; Case C-172/03 Wolfgang Heiser v Finanzamt Innsbruck [2005] ECR I-1627, para [46]; Case C-487/06 P British Aggregates Association v Commission [2008] ECR I-10505, para [85]. 11   Case 173/73 Italy v Commission (n 10 above), para [13]; Case C-159/01 Netherlands v Commission (n 10 above), para [51]; Case T-210/02 British Aggregates Association v Commission [2006] ECR II-2789, para [106]. 12  Case C-487/06 P British Aggregates Association v Commission (n 10 above), para [84]; Joined Cases T-127/99, T-129/99 & T-148/99 Territorio Histórico de Álava—Diputación Foral de Álava v Commission [2002] ECR II-1275, para [168]; Joined Cases T-269/99, T-271/99 & T-272/99 Territorio Histórico de Guipúzcoa— Diputación Foral de Guipúzcoa v Commission [2002] ECR II-4217, para [63]; Case T-20/03 Kahla/Thüringen Porzellan GmbH v Commission [2008] ECR II-2305, para [197]; Joined Cases T-254/00, T-270/00 & T-277/00 Hotel Cipriani SpA v Commission [2008] ECR II-3269, para [195]. 13   Case C-156/98 Germany v Commission [2000] ECR I-6857, para [26]. 14   Case C-353/95 P Tiercé Ladbroke v Commission [1997] ECR I-7007, paras [33–35]; Case C-53/00 Ferring SA v Agence centrale des organismes de sécurité sociale (ACOSS) [2001] ECR I-9067, para [17].

512  Conor Quigley structure of the system of charges of which they are a part.15 This, however, is a relatively difficult hurdle to overcome. In any event, it is an objective test, so that the Member State cannot unilaterally determine which differences will be justified by reference to the nature and structure of the system. It follows that the Member States have no scope in this respect for abusing any rights as such, but are merely subject to the normal supervisory control by the Commission and the courts. One instance where the notion of abuse of rights might be applicable is where the State’s action is intended to counteract tax avoidance. In Lunn Poly, the English Court of Appeal held that differential rates of travel insurance tax, whereby higher rates were applied to insurance sold through travel agents, constituted State aid in favour of direct insurance companies which were liable to account for the tax only at a lower rate. It was accepted that differential rates would not entail State aid if they were justified on grounds of potential tax avoidance by the persons subject to the higher rate, but on the available evidence no such justification was shown.16 The same issue was referred subsequently to the ECJ in GIL v Commissioners of Customs and Excise, where the ECJ observed that Insurance Premium Tax (IPT) was originally introduced as a general taxation measure applicable in principle to all insurance contracts at a single rate. The higher rate of IPT and Value-Added Tax (VAT) formed part of an inseparable whole, which was the aim pursued in introducing the higher rate. That rate was introduced to counteract the practice of taking advantage of the difference between the standard rate of IPT and that of VAT by manipulating the prices of rental or sale of appliances and of the associated insurance. Such conduct had given rise to a loss of income in terms of VAT receipts and to shifts in the conditions of competition in the domestic appliance sector. In view of its purpose and effect, the higher rate of IPT had the appearance of a regulatory charge intended specifically as a deterrent to the conclusion of connected insurance contracts. The introduction of a higher rate of IPT on certain contracts was not a tax scheme favouring a specified sector, since it was a system of taxation of insurance premiums intended to compensate for the fact that insurance transactions were not subject to VAT. Thus, even on the assumption that the introduction of the higher rate involved an advantage for operators offering contracts subject to the standard rate, it was justified by the nature and scheme of the tax system of insurance.17 III. Approval of aid by the Council

Article 108(2) TFEU [ex Article 88(2) EC] provides that, on application by a Member State, the Council may decide that aid which that State is granting or intends to grant 15   Case 173/73 Italy v Commission (n 10 above), para [15]; Case C-75/97 Belgium v Commission (n 10 above), para [33]; Case C-143/99 Adria-Wien Pipeline GmbH v Finanzlandesdirektion für Kärnten [2001] ECR I-8365, para [42]; Case C-351/98 Spain v Commission [2002] ECR I-8031, para [42]; Case C-409/00 Spain v Commission [2003] ECR I-1487, para [52]; Case C-355/00 Freskot v Greece [2003] ECR I-5263, para [86]; Case C-159/01 Netherlands v Commission (n 10 above), para [42]; Joined Cases C-128/03 & C-129/03 AEM SpA v Autorità per l’energia elettrica e per il gas [2005] ECR I-2861, para [39]; Case C-88/03 Portugal v Commission [2006] ECR I-7115, para [52]; Case C-487/06 P British Aggregates Association v Commission (n 10 above), para [56]; Case T-210/02 British Aggregates Association v Commission (n 11 above), para [117]; Case T-146/03 Asociación de Empresarios de Estaciones de Servicio de la Comunidad Autónoma de Madrid and Federación Catalana de Estaciones de Servicio v Commission [2006] ECR II-98, para [119]; Joined Cases T-211/04 & T-215/04 Government of Gibraltar and United Kingdom v Commission (n 5 above), para [144]. 16   R v Commissioners of Customs and Excise ex p Lunn Poly [1999] EuLR 653. 17   GIL Insurance Ltd (n 2 above), paras [70–78].

Comments on State Aid, Taxation and Abuse of Law  513 shall be considered to be compatible with the Internal Market, in derogation from the provisions of Article 107 TFEU, if such a decision is justified by exceptional circumstances. This applies both to new aid which the Commission may consider is not compatible with the Internal Market and to existing aid which the Commission may consider is no longer compatible with the Internal Market. The Council is, therefore, the institution of last resort in deciding whether exceptional circumstances exist,18 subject to review by the General Court [formerly the Court of First Instance] and the Court of Justice.19 The notion of exceptional circumstances involves the idea of something extraordinary, unforeseen and not permanent. It might apply to facts or situations which may relate to one sector in particular or to the economy in general but which, assessed in the context of a specific Member State or a specific sector, show that there has been a change on such a scale as compared to what was previously considered normal or at least not extraordinary, that corrective measures are needed for which there is no provision in the existing rules governing the sector in question.20 Given that the Council is called upon to carry out an assessment of a complex economic situation, it is entitled to a broad measure of discretion as to the nature and scope of the measures to be taken.21 One might have thought that reliance on Article 108(2) TFEU [ex Article 88(2) EC] by the Council would entail a significant divergence from the Commission’s control under Article 107 TFEU. The Commission retains power under Article 107(2)–(3) TFEU to declare certain types of aid compatible with the Internal Market. By contrast, the Council has the power to act in derogation from Article 107 TFEU, which is not open to the Commission. That division of competence, which is objectively ascertainable and subject to judicial review, should entail that there is no overlapping of jurisdictions. However, the Court of Justice has taken a rather more restrictive approach and has held, in effect, that the Council may not intervene under Article 108(2) TFEU once the Commission has decided that aid is incompatible with the Internal Market. Any such intervention by the Council would amount to an abuse of its right of action. Article 108(2) TFEU provides that if, as regards the aid in question, the Commission has already initiated the formal investigation procedure, the fact that the Member State concerned has made its application to the Council has the effect of suspending that procedure until the Council has made its attitude known. If, however, the Council has not made its attitude known within three months of the application being made, the Commission must give its decision on the case.22 In Commission v Germany, AdvocateGeneral Mayras took the view that, since the effect of applying to the Council is to suspend the procedure before the Commission, it is not open to the Member State to apply to the Council after the Commission has taken a decision which has brought that procedure to an end.23 This conclusion was subsequently confirmed by the ECJ in two cases brought by the Commission against the Council, on the ground that, since the  AG Cosmas in Case C-122/94 Commission v Council [1996] ECR I-881, para [64].  For example, in Case 253/84 Groupement agricole d’exploitation en commun (GAEC) de la Ségaude v Council and Commission [1987] ECR 123, AG Slynn advised that Council Decision 84/361/EEC was ultra vires because it had not been shown that exceptional circumstances existed. 20  AG Cosmas in Commission v Council (n 18 above), para [85]. 21   Commission v Council (n 18 above), para [19]. In the face of Commission opposition, the Council adopted Decisions 2002/361–363/EC on 3 May 2002 authorising the grant of aid by the Netherlands, Italy and France in favour of road transport undertakings [2002] OJ L131/12–14. 22  Art 108(2) TFEU. 23  AG Mayras in Case 70/72 Commission v Germany [1973] ECR 813, 835. 18 19

514  Conor Quigley Council’s power under what was then Article 88(2) EC was clearly exceptional, where the Commission had already taken a decision that aid was incompatible with the Common Market, the Council was no longer authorised to adopt a decision in relation to the same aid measure.24 In the first of these cases, the Commission decision had declared certain unlawful aid that had been paid to Portuguese pig farmers to be incompatible with the Common Market and had ordered its recovery. The Council’s decision was in fact concerned with aid that was intended to compensate the farmers for having to repay the unlawful aid. However, the ECJ held that to allow a Member State to grant new aid, in an amount equivalent to that of the unlawful aid and which was intended to neutralise the impact of the repayments which the beneficiaries were obliged to make pursuant to the Commission’s recovery decision, would clearly amount to undermining the effectiveness of the Commission’s decision.25 In the second case, the Council sought to stop the Commission from immediately terminating the tax treatment of Belgian coordination centres. Again, however, the Council’s right of action was therefore precluded once the Commission had declared the aid incompatible with the Common Market.26 IV.  Commission decisions approving State aid

State aid may be declared by the Commission to be compatible with the Internal Market on any of the grounds set out in Article 107(2)–(3) TFEU. Leaving aside the agriculture and transport sectors, the vast bulk of aid measures fall within one or other of the main categories of aid: regional aid, small and medium-sized enterprises (SME) aid, training aid, rescue and restructuring aid; aid for research, development and innovation (R&D&I) and aid for environmental protection purposes. Most aid is awarded on the basis of schemes that are notified by the Member States and approved in advance by the Commission. Occasionally, such approval requires that individual measures within the scope of those schemes must be separately notified. In addition, ad hoc grants of aid on an individual basis must be notified in advance. Where the Commission finds that the aid is compatible with the Internal Market, it takes a decision accordingly. If it is of the view that the measure raises serious difficulties, it must open the formal investigation procedure under Article 108(2) TFEU. This procedure must result in a final decision, which might allow the measure, with or without conditions, or which might prohibit the measure from being implemented. Any decision of the Commission approving aid under Article 107(3) TFEU [ex Article 87(3) EC] but which was adopted on the basis of false or incomplete information may be revoked.27 Council Regulation (EC) No 659/1999, Article 9 provides that the Commission, after having given the Member State concerned the opportunity to submit its comments, may subsequently revoke a decision which finds that a measure does not constitute aid or that aid is compatible with the Internal Market where the decision was based on incorrect information provided during the investigation procedure which was a determining factor 24   Case C-110/02 Commission v Council [2004] ECR I-6333, para [33]; Case C-399/03 Commission v Council [2006] ECR I-5629, para [24]. 25   Case C-110/02 Commission v Council (n 24 above), paras [43–45]. 26   Case C-399/03 Commission v Council (n 24 above), paras [27–28]. 27   Joined Cases 42/59 & 49/59 SNUPAT v High Authority [1961] ECR 53, 87.

Comments on State Aid, Taxation and Abuse of Law  515 for the decision.28 However, before revoking a decision and taking a new decision, the Commission must open the formal investigation procedure. Aid which has been granted may be treated in much the same way as unlawful aid and may be required to be suspended pending the Commission’s examination. Aid which is subsequently declared incompatible with the Internal Market must be repaid.29

V.  Unlawful aid and misuse of aid

Where a Member State introduces aid without first seeking and obtaining Commission approval, the aid will be regarded as unlawful. In those circumstances, the aid is subject to a recovery order.30 It is then the responsibility of the Member State concerned to take the steps necessary to effect recovery. The Commission has issued a notice in which it sets out the requirements imposed on the Member States in relation to the recovery of unlawful aid.31 Recovery of unlawful aid must be effected without delay and in accordance with the procedures under the national law of the Member State concerned, provided that they allow the immediate and effective execution of the Commission’s decision.32 According to the Commission, this implies that the authorities responsible should carefully consider the full range of recovery instruments available under national law and select the procedure most likely to secure the immediate execution of the decision.33 If national procedural rules do not enable the enforcement of recovery, the Member State must take steps to put the necessary mechanisms in place.34 The recovery obligation is subject to the qualification that recovery of aid may not be ordered by the Commission if this would be contrary to a general principle of EU law.35 Reliance on the rule that recovery of unlawful aid may not be ordered if this would be contrary to a general principle of EU law mostly involves invoking the principle of the protection of legitimate expectations. The right to rely on the principle of the protection of legitimate expectations, which constitutes one of the fundamental principles of the EU law, extends to any individual who is in a situation in which it is clear that the Union’s authorities have, by giving him precise assurances, led him to entertain legitimate expectations. Regardless of the form in which it is communicated, information that is precise, unconditional and consistent and comes from an authorised and reliable source

28  See for example, Commission Decision 2001/856/EC of 4 October 2000 concerning State aid to Verlipack, Belgium (Verlipack) [2001] OJ L320/28, which was upheld in Case C-457/00 Belgium v Commission [2003] ECR I-6931. 29   Council Regulation (EC) No 659/1999 (n 3 above), Art 9. 30   Council Regulation (EC) No 659/1999 (n 3 above), Art 14. 31   Notice from the Commission—Towards an effective implementation of Commission decisions ordering Member States to recover unlawful and incompatible State aid [2007] OJ C272/4. 32   Council Regulation (EC) No 659/1999 (n 3 above), Art 14(3); Case 94/87 Commission v Germany [1989] ECR I-175, para [12]; Case C-24/95 Land Rheinland-Pfalz v Alcan Deutschland [1997] ECR I-1591, para [24]; Case C-404/97 Commission v Portugal [2000] ECR I-4897, para [55]; Case C-378/98 Commission v Belgium [2001] ECR I-5107, para [51]; Case C-382/99 Netherlands v Commission (n 10 above), para [90]; Case C-209/00 Commission v Germany [2002] ECR I-11695, para [32]; Case C-232/05 Commission v France [2006] ECR I-10071, para [49]; Case C-419/06 Commission v Greece [2008] ECR I-27, para [59]. 33   Notice on the recovery of unlawful aid (n 31 above), para [52]. 34  AG Sharpston in Case C-214/07 Commission v France [2008] ECR I-8357, para [76]. 35   Council Regulation (EC) No 659/1999 (n 3 above), Art 14(1).

516  Conor Quigley constitutes such assurance. However, a person may not plead infringement of the principle unless he has been given precise assurances by the authorities.36 The ability of the recipient to rely on this exception to the general rule requiring recovery is limited. Since it is the logical consequence of a finding that aid is unlawful, recovery of State aid which has been unlawfully granted, for the purpose of restoring the previously existing situation, cannot in principle be regarded as disproportionate.37 Moreover, a decision ordering the recovery of unlawful aid cannot constitute a penalty, even if it is implemented long after the aid in question was granted.38 A recipient cannot invoke legitimate expectation by merely stating that he was unaware that the measure constitutes State aid. The Court of Justice has held that a diligent businessman should normally be able to determine whether the proper procedure for having the aid notified and approved has been followed.39 This applies equally to small businesses.40 However, it is not sufficient for the recipient merely to seek assurances from the Member State. Even if the Member State has mistakenly indicated that the aid is covered by a Commission approval, this will not give rise to a legitimate expectation on the part of the recipient or affect the right of the Commission to order recovery.41 A similar approach applies in the case of misuse of aid. In that event, the recipient of aid which has been approved by the Commission misuses the aid for purposes other than those for which it was intended. For example, the Commission found there to have been a misuse of aid where shipbuilding aid was allowed as development aid for Indonesia but the vessel was used outside of Indonesia.42 Aid is misused where the Commission authorises investment aid but the beneficiary deflects the aid to cover operating costs.43 Where aid has been approved under the restructuring aid guidelines, the Commission will regard any failure to implement a restructuring plan or to fulfil the other obligations as misuse of the aid.44 In each of these cases, the recipient may be regarded as abusing the right that has been granted to him. The right to receive the aid arises under national law. In order to demonstrate that aid granted under an authorised aid scheme has been misused, it must be established that the aid was used in a manner contrary to that scheme as approved 36   Joined Cases T-66/96 & T-221/97 Mellett v Court of Justice [1998] ECR SC II-1305, paras [104, 107]; Kahla/Thüringen (n 12 above), para [146]. 37   Case C-142/87 Belgium v Commission [1990] ECR I-959, para [66]; Joined Cases C-278–280/92 Spain v Commission [1994] ECR I-4103, para [75]; Case C-75/97 Belgium v Commission (n 10 above), para [68]; Case C-404/87 Commission v Portugal (n 32 above), para [54]; Case C-419/06 Commission v Greece (n 32 above), para [55]; Case C-66/02 Italy v Commission (n 9 above), para [113]; Unicredito Italiano (n 9 above), para [113]; Case T-288/97 Regione Autonoma Friuli Venezia Giulia v Commission [2001] ECR II-1169, para [105]; Case T-198/01 Technische Glaswerke Ilmenau GmbH v Commission [2004] ECR II-2717, para [133]; Hotel Cipriani (n 12 above), para [389]. 38   Case T-55/99 Confederación Española de Transporte de Mercancías (CETM) v Commission [2000] ECR II-3207, para [164]; Case T-369/00 Départment du Loiret v Commission [2003] ECR II-1789, para [57]; Case T-366/00 Scott SA v Commission [2007] ECR II-797, para [94]. 39   Case C-169/95 Spain v Commission [1997] ECR I-135, para [51]; Land Rheinland-Pfalz (n 31 above), para [25]; Case C-334/99 Germany v Commission [2003] ECR I-1139, para [42]; Joined Cases C-346/03 & C-529/03 Atzeni v Regione autonoma della Sardegna [2006] ECR I-1875, para [64]. 40   Case T-109/01 Fleuren Compost BV v Commission [2004] ECR II-127, para [140]. 41   Fleuren Compost (n 40 above), paras [143–44]. 42   Commission Decision 1999/142/EC of 25 February 1998 on development aid granted by Germany for the construction of a dredger sold to Indonesia (Indonesian shipping aid) [1999] OJ L46/52. 43   Commission Decision 1999/580/ECSC of 11 November 1998 concerning aid granted by Germany to ESF Elbestahlwerk Feralpi GmbH, Riesa, Saxony (ESF Elbestahlwerk Feralpi GmbH) [1999] OJ L220/28. 44   Communication from the Commission—Community guidelines on state aid for rescuing and restructuring firms in difficulty [2004] OJ C244/02, para [47].

Comments on State Aid, Taxation and Abuse of Law  517 by the Commission—that is, in breach of the national rules governing that scheme or supplementary conditions which have been accepted by the Member State as part of approval of the scheme by the Commission.45 Aid that has been misused may be subject to a recovery order in the same way as unlawful aid.46 However, the difference in substance between unlawful aid and misuse of aid reflects the nature of the infringement. Aid is unlawful where the Member State puts it into effect without authorisation in breach of Article 108(3) TFEU, ie in breach of an EU law requirement imposed on the State. Aid is misused where it is used in breach of the national provisions, as approved by the Commission, which may involve a breach of a national law requirement imposed on the recipient. In other words, the recipient is given the right to receive aid in accordance with the national measures that grant the aid. By abusing that right, the aid becomes misused. That abuse gives rise to consequences under EU law, regardless of whether national law also would intervene. VI. Damages in respect of unlawful aid and misuse of aid

The Commission, in its decisions finding that aid has been granted unlawfully, has no power to award compensation to competitors who may have suffered damage as a result of the aid. However, the Court has recognised that the grant of unlawful aid may cause the Member States to have to make reparation for damage caused by the unlawful nature of the aid.47 Equally, an action for damages may be commenced against a Member State that fails to comply with a recovery order.48 A mere breach of Article 108(3) TFEU is not of itself sufficient to confer a right to damages in all cases of unlawful State aid. There must also be a distortion of competition giving rise to damage. In Banks v Coal Authority, the ECJ, having held that repayment of unlawful aid could not, in the circumstances, be recovered, thought, nevertheless, that this was without prejudice to any action for compensation which competitors might bring, if the conditions were met, for compensation for any damage caused to them by the distortion of competition.49 Thus, whereas the unlawful action is the infringement of the standstill clause in Article 108(3) TFEU, an award of damages is not automatically consequent on the infringement of the direct effect of that provision.50 Distortion of competition is not, however, a cause of action in itself, but must be proved as a consequence of the State’s unlawful action. This gives rise to difficulties in the case of distortions of competition caused by misuse of aid. In that case, there is no infringement by the Member State concerned of its duty to notify the Commission under Article 108(3) TFEU. On the contrary, that requirement will, of necessity, have been fulfilled. The recipient is not under any duty of notification, whether under Article 108(3) TFEU or otherwise. Even if he has deviated from requirements in the decision authorising the   Case T-318/00 Freistaat Thüringen v Commission [2005] ECR II-4179, para [114].   Council Regulation (EC) No 659/1999 (n 3 above), Arts 14(1), 16. 47  Case C-199/06 Centre d’exportation du livre français (CELF), Ministre de la Culture et de la Communication v Société internationale de diffusion et d’édition (SIDE) [2008] ECR I-469, para [55]; Case C-334/07 P Commission v Freistaat Sachsen [2008] ECR I-9465, para [54]. 48   Commission notice on the enforcement of State aid law by national courts [2009] OJ C85/1, para [69]. 49   Case C-390/98 HJ Banks & Co Ltd v Coal Authority [2001] ECR I-6117, para [80]. 50  AG Jacobs in Case C-368/04 Transalpine Ölleitung in Österreich v Finanzlandesdirektion für Tirol [2006] ECR I-9957, para [86]. 45 46

518  Conor Quigley aid, that decision is always addressed to the Member State concerned, and not to the recipient, so the recipient cannot be regarded as having infringed any binding obligation imposed on him by the decision. Unless he has infringed national law governing the grant of the aid (which might or might not be the case), he will not have acted unlawfully, since the mere misuse of aid does not of itself entail an unlawful act. Such a conclusion is clearly unsatisfactory from the point of view of competitors who have suffered damage as a result of the distortion of competition arising out of the recipient’s behaviour. This could be addressed by elevating the principle of abuse of law to a substantive level, or by marrying the notion of abuse with the consequent distortion of competition to an actionable cause in its own right. No Court of Justice case has yet addressed this issue. In Betws Anthracite v DSK, the English High Court held that where a recipient misused aid, and thereby causes a distortion of competition to the detriment of its competitors, there was no cause of action in English or EU law against the recipient.51

  Betws Anthracite Ltd v DSK Anthrazit Ibbenburen GmbH [2003] EWHC 2403.

51

36 The Prohibition of Abuse of Law: An Emerging General Principle of EU Law Stefan Vogenauer

I

s there a general principle of European Union law, according to which ‘abuse of law’ is prohibited? This is the question we set out to answer in this book. It might be argued that the Court of Justice settled the issue in its Kofoed decision of 2007 where it referred to the general Community law principle that abuse of rights is prohibited. Individuals must not improperly or fraudulently take advantage of provisions of Community law. The application of Community legislation cannot be extended to cover abusive practices, that is to say, transactions carried out not in the context of normal commercial operations, but solely for the purpose of wrongfully obtaining advantages provided for by Community law.1

However, if anything, the preceding chapters will have shown that the precise scope, content and limits of the prohibition of ‘abuse of law’ or ‘abuse of rights’ are far from being settled. Nor is there agreement on whether the prohibition does indeed amount to a general principle of EU law. This chapter will deal with both issues in turn. I. The Prohibition of Abuse of Law in the Law of the European Union

A. Sources The prohibition of abuse of law is a straightforward instance of judge-made law.2 It has been developed in the case law of the Court of Justice since 1974.3 The idea of restraining abusive practices emerged in the context of free movement of services and has subsequently been invoked in many other areas of EU law:4 in its Centros ruling of 1999, the Court referred to a long list of cases examining purported abuses of the   Case C–321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I–5795, para [38].   Cf J-P Maublanc, ‘Réévaluation de l’abus de droit à la lumière de la jurisprudence communautaire sur les pratiques abusives’ (2007) Revue du Marché commun et de l’Union européenne 267, 269 (‘construction prétorienne communautaire’). 3   Case 33/74 Van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299, para [13] is generally seen as the starting point although the word ‘abuse’ was not used in the decision. 4  For a comprehensive overview, see R de la Feria, ‘Prohibition of Abuse of (Community) Law: The Creation of a New General Principle of EC Law through Tax’ (2008) 45 Common Market Law Review 395. Cf also A Niemann, Der gemeinschaftsrechtliche allgemeine Missbrauchsvorbehalt nach der Rechtsprechung des EuGH (PhD thesis, Heidelberg, forthcoming). 1 2

522  Stefan Vogenauer other fundamental freedoms and assessing potentially abusive practices in the fields of social security, common agricultural policy and company law.5 Since then the Court has extended the prohibition to further areas, notably tax law.6 Centros was also the first of a series of landmark decisions that have gradually refined the content of the prohibition to the present day.7 By contrast, legislation and legal scholarship have played a very limited role in the development of the prohibition. From 1990 onwards, the legislative organs of the Union began to enact prohibitions of specific kinds of abuse in particular areas of EU law, such as the taxation of exchanges of shares between companies from different Member States,8 the use of EU budget resources,9 the cross-border transmission of broadcasting services,10 international civil procedure,11 the right to reside in another Member State,12 the measures to ensure the enforcement of intellectual property rights13 and value added   Case C–212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I–1459, para [24].   Case C–255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I–1609. 7  Notably Case C–110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I–11569; Halifax (n 6 above); Case C–196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I–7995; Case C–425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I–897. 8  Article 11(1)(a) of Council Directive 90/434/EEC of 23 July 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States [1990] OJ L225/1: ‘A Member State may refuse to apply or withdraw the benefit of all or any part of the provisions of Titles II, III and IV [of the directive] where it appears that the merger, division, transfer of assets or exchange of shares: (a) has as its principal objective or as one of its principal objectives tax evasion or tax avoidance; the fact that one of the operations referred to in Article 1 is not carried out for valid commercial reasons such as the restructuring or rationalization of the activities of the companies participating in the operation may constitute a presumption that the operation has tax evasion or tax avoidance as its principal objective or as one of its principal objectives’. 9  Art 4(3) of Regulation 2988/95 concerning the protection of the European Communities financial interests [1995] OJ L312/1: ‘Acts which are established to have as their purpose the obtaining of an advantage contrary to the objectives of the Community law applicable in the case by artificially creating the conditions required for obtaining that advantage shall result, as the case shall be, either in failure to obtain the advantage or in its withdrawal.’ 10  Recital 14 of the Preamble to Council Directive 89/552/EEC of 3 October 1989 on the coordination of certain provisions laid down by law, regulation or administrative action in Member States concerning the pursuit of television broadcasting activities [1989] OJ L298/23, as amended by Directive 97/36/EC of the European Parliament and of the Council of 30 June 1997 [1997] OJ L202/60: ‘if the choice of establishment was made with a view to evading the legislation that would have applied to the organisation’; Art 3 of the same Directive, as amended by Directive 2007/65/EC of the European Parliament and of the Council of 11 December 2007 [2007] OJ L332/27. Cf D Doukas, ‘Free Movement of Broadcasting Services and Abuse of Law’, ch 6 above, 83–86. 11  Article 6(2) of Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2001] OJ L12/1; see already Article 6(2) of the Brussels Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters of 27 September 1968 [1972] OJ L299/32: ‘A person domiciled in a Member State may also be sued: . . . (2) as a third party in an action on a warranty or guarantee or in any other third party proceedings, in the court seised of the original proceedings, unless these were instituted solely with the object of removing him from the jurisdiction of the court which would be competent in his case’. Cf A Metzger, ‘Abuse of Law in EU Private Law: A (Re-)Construction from Fragments’, ch 16 above, 244; G Cuniberti, ‘The Discreet Influence of Abuse of Law in International Civil Procedure’, ch 19 above, 285. 12  Article 35 of Directive 2004/38/EC on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States has the heading ‘Abuse of rights’. Its first sentence reads: ‘Member States may adopt the necessary measures to refuse, terminate or withdraw any right conferred by this Directive in the case of abuse of rights or fraud, such as marriages of convenience.’ Cf K Ziegler, ‘“Abuse of Law” in the Context of the Free Movement of Workers’, ch 21 above, 295. 13  Article 3(2) of Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights [2004] OJ L195/16: ‘Those measures, procedures and remedies shall also be effective, proportionate and dissuasive and shall be applied in such a manner as to avoid the creation of barriers to legitimate trade and to provide for safeguards against their abuse.’ Cf Metzger (n 11 above) 242–43. 5 6

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  523 tax.14 These instances of secondary legislation are sporadic and narrow. The Court sees them as ‘codifications’ or ‘reflections’ of the broader prohibition that it has developed in its case law.15 The Community legislator is content to operate against the background of the judicial doctrine16 and accepts that the respective legislative measures will be ‘interpreted in the light of the relevant case law of the Court of Justice’.17 A much broader provision that explicitly ‘prohibits’ the abuse of EU fundamental rights can be found in Article 54 of the Charter of Fundamental Rights of the European Union, adopted in 2000.18 However, for reasons set out in Professor Arnull’s chapter, this provision has only had limited impact so far.19 With the exception of Neville Brown’s pioneering article of 1994,20 in-depth scholarly debate on the emerging doctrine of abuse did not begin until the turn of the century.21 It was mostly confined to an analysis of the evolving case law, sometimes coupled with criticism and suggestions for improvement and further development of the Court’s approach. As is often the case, EU legal scholarship reacted to the evolving case law, rather than pro-actively setting out the parameters for legal development. Thus, whilst it might still be desirable to design a more rational doctrine from first principles,22 at the present stage of development of EU law taking the existing case law as a starting point of the analysis is unavoidable.23

14  Article 395 of Council Directive 2006/112/EC of 28 November 2006 on the common system of value-added tax [2006] OJ L347/1. 15  AG Alber in Emsland-Stärke (n 7 above), para [63]: ‘Article 4(3) of Regulation No 2988/95 concerning the protection of the European Communities financial interests, which codifies a general principle of Community law’; Kofoed (n 1 above), para [38]: ‘Article 11(1)(a) of Directive 90/434 reflects the general Community law principle that abuse of rights is prohibited’. 16   Commission (EC), ‘The Application of Anti-Abuse Measures in the Area of Direct Taxation’ (Communication) COM (2007) 785 final 3, 10 December 2007. 17  Art 1(2)(b) of Annex II to the Agreement between the EC and the Swiss Confederation in the audiovisual field, establishing the terms and conditions for the participation of the Swiss Confederation in the Community programmes MEDIA Plus and MEDIA Training – Final Act – Declarations [2006] OJ L90/23, cited in n 199 on p 86 above. 18  ‘Prohibition of Abuse of Rights. Nothing in this Charter shall be interpreted as implying any right to engage in any activity or to perform any act aimed at the destruction of any of the rights and freedoms recognised in this Charter or at their limitation to a greater extent than is provided for herein.’ The provision is closely modeled on Article 17 of the European Convention on Human Rights. 19  A Arnull, ‘What is a General Principle of EU Law?’, ch 2 above, 22. See also Ziegler (n 12 above) 313; C Costello, ‘Citizenship of the Union: Above Abuse?’, ch 23 above, 325. 20   LN Brown, ‘Is there a General Principle of Abuse of Rights in European Community Law?’ in D Curtin and T Heukels (eds), Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers, Vol II (Dordrecht, Martinus Nijhoff 1994) 511. See also the inaugural lecture of PJ Wattel, Misbruik van Europees recht (Amsterdam, Moret Ernst & Young 1993). 21  Notably A Kjellgren, ‘On the Border of Abuse: The Jurisprudence of the European Court of Justice on Circumvention, Fraud and Abuses of Community Law’ in M Andenas and W-H Roth (eds), Services and Free Movement in EU Law (Oxford, Oxford University Press 2002) 245, first published in [2000] European Business Law Review 179; W Schön, ‘Der „Rechtsmissbrauch“ im Europäischen Gesellschaftsrecht’ in R Wank et al (eds), Festschrift für Herbert Wiedemann (Munich, CH Beck 2002) 1271; H Fleischer, ‘Der Rechtsmißbrauch zwischen Gemeineuropäischem Privatrecht und Gemeinschaftsprivatrecht’ [2003] Juristenzeitung 865; F Lagondet, ‘L’abus de droit dans la jurisprudence communautaire’ (2003) 95 Journal des tribunaux – Droit européen 8. 22  As advocated by H Eidenmüller, ‘Abuse of Law in the Context of European Insolvency Law’, ch 10 above, 141–42. 23   Lagondet (n 21 above) 8.

524  Stefan Vogenauer B. Terminology Discussion of the prohibition of abuse of law necessarily suffers from the terminological inconsistency and imprecision prevailing in this area of EU law.24 The prohibition is variously said to concern the ‘abuse of law’,25 the ‘misuse of law’,26 the ‘abuse of rights’,27 ‘abusive practices’,28 ‘abusive reliance’29 or simply ‘abuse’.30 It is often referred to in the same breath as notions such as ‘circumvention’,31 ‘avoidance’,32 ‘evasion’,33 ‘fraud’,34 ‘sham’,35 ‘improperly taking advantage’36 or ‘wholly artificial arrangements’,37 and it is not always clear whether these are meant to be synonyms for ‘abuse’, related concepts or doctrines that go beyond abuse and must therefore be distinguished. Much of the confusion seems to stem from the multi-lingual nature of the EU. The legal systems of the Member States feature a multitude of different, partially overlapping doctrines of ‘abuse’, ‘fraud’, ‘evasion’ or ‘avoidance’ and their legal terminologies attach different meanings to such notions. Frequently there is even disagreement at the national level as to what exactly these terms mean and how they ought to be distinguished from one another.38 If they are translated from one official language of the EU into another it is impossible to attain terminological precision.39 It may therefore be said that the muddled 24   K Engsig Sørensen, ‘Abuse of Rights in Community Law: A Principle of Substance or Merely Rhetoric?’ (2006) 43 Common Market Law Review 423, 431; de la Feria (n 4 above) 395–96; P Schammo, ‘Arbitrage and Abuse of Rights in the EC Legal System’ (2008) 14 European Law Journal 351, 358. 25  AG Poiares Maduro in Halifax (n 6 above), para [71]. Cf AG Mazák in Case C–103/09 HMRC v Weald Leasing Limited, nyr, para [10]: ‘abuse of law principle’. 26  AG Geelhoed in Case C–109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I–9607, paras [96-105, 169-185]; AG Kokott in Kofoed (n 1 above), para [67]. 27  AG Darmon in Case 81/87 R v HM Treasury and Commissioners of Inland Revenue ex p Daily Mail and General Trust plc [1988] ECR 5483, para [9]; AG Léger in Cadbury Schweppes (n 7 above), para [118]; Case C–162/07 Ampliscientifica Srl and Amplifin SpA v Ministero dell’Economia e delle Finanze and Agenzia delle Entrate [2008] ECR I–4019, para [32]: ‘the principle prohibiting the abuse of rights’; Case C–277/09 Comissioners for Her Majesty’s Revenue & Customs v RBS Deutschland Holdings GmbH, nyr, para [52]. 28   Halifax (n 6 above), para [70]; Cadbury Schweppes (n 7 above), para [55]; Weald Leasing (n 25 above), para [26]; RBS Deutschland (n 27 above), para [47]. 29  AG Stix-Hackl in Case C–452/04 Fidium Finanz AG v Bundesanstalt für Finanzdienstleistungsaufsicht [2006] ECR I–9521, paras [80, 81]. 30   Emsland-Stärke (n 7 above), para [52]; Akrich (n 26 above), para [55]; Halifax (n 6 above), para [61]. 31   Case 229/83 Association des Centres distributeurs Edouard Leclerc et al v SARL ‘Au Blé Vert’ et al [1985] ECR 1, para [27]; Akrich (n 26 above), para [57]. 32   Van Binsbergen (n 3 above), para [13]. 33   Case 115/78 J Knoors v Staatssecretaris van Economische Zaken [1979] ECR 399, para [25]. 34  Case C–373/97 Dionysios Diamantis v Elliniko Dimosio (Greek State) and Organismos Ikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [2000] ECR I–1705, para [33]; Case C–367/96 Alexandros Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I–2843, para [20]; Centros (n 5 above), para [24]; Case C–285/95 Suat Kol v Land Berlin [1997] ECR I–3069, para [28]; Kofoed (n 1 above), para [38]; Case C–370/90 R v Immigration Appeal Tribunal and Surinder Singh ex p Secretary of State for the Home Department [1992] ECR I–4265, para [24]; Case C–303/08 Metin Bozkurt v Land Baden-Württemberg, judgment of 22 December 2010, nyr, para [47]. 35   S Weatherill, ‘Fitting “Abuse of Rights” into EU Law Governing the Free Movement of Goods and Services’, ch 5 above, 56. 36   Centros (n 5 above), para [24] (cited at n 49 below); Kofoed (n 1 above), para [38]. 37   Cadbury Schweppes (n 7 above), para [55]; Case C–524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I–2107, para [74]. 38  See, eg, F Terré, Introduction générale au droit, 8th edn (Paris, Dalloz 2009) nos 496–99; S Sieker, Umgehungsgeschäfte: Typische Strukturen und Mechanismen ihrer Bekämpfung (Tübingen, Mohr Siebeck 2001). 39  For an example of how some of these terms are used in the English and French language versions of the Sixth VAT Directive, see P Harris, ‘Abus de droit in the Field of Value Added Taxation’ [2003] British Tax Review 131, 146–47.

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  525 terminology of EU law simply reflects the ‘semantic richness’40 prevailing in the national legal systems. However, greater linguistic consistency of EU law in this area is desirable, if only to reduce the potential for misunderstanding.41 It is legitimate to develop a European notion of abuse without recourse to the established legal terminologies of the Member States: as a general rule, legal concepts of EU law have an autonomous, EU-specific meaning which will normally differ from the linguistic usage prevailing in one or more of the national legal systems.42 The expression ‘prohibition of abuse of EU law’, first suggested by Advocate-General Poiares Maduro in Halifax,43 is such a concept and cannot be easily confused with established national doctrines.44 The notion is certainly not used universally, as can be seen from a cursory glance at the title chapters of this book, but it is slowly gaining ground in European legal discourse, at least as far as it is conducted in the English language.45 Most importantly, for reasons given further below, it seems to be the most appropriate label for the legal phenomenon discussed in this book.46 Therefore I will normally refer to ‘abuse of law’ whilst using the other terms only where required by the specific context or for the purpose of simplicity. C. Applications Various attempts have been made at classifying the various cases that are dealt with when reference is made to ‘abuse of law’ or ‘abuse of rights’ in EU law.47 The Court and most commentators seem to agree that two categories of cases are particularly important,48 namely to prevent . . . nationals from attempting, under cover of the rights created by the Treaty, improperly to circumvent their national legislation or to prevent individuals from improperly or fraudulently taking advantage of provisions of Community law.49   Maublanc (n 2 above) 267, referring to various French terms.   See, for the area of tax law, J Freedman, ‘The Anatomy of Tax Avoidance Counteraction: Abuse of Law in a Tax Context at Member State and European Union Level’, ch 25 above, 377. 42   Case 29/76 LTU v Eurocontrol [1976] ECR 1541, para [3]; Case 283/81 Srl CILFIT and Lanificio di Gavardo SpA v Ministry of Health [1982] ECR 3415, para [19]. 43  AG Poiares Maduro in Halifax (n 6 above), para [71] (then still as ‘prohibition of abuse of Community law’). 44  This also holds true for its other language versions, see below (at n 339). 45   Cf de la Feria (n 4 above). 46   See below (at n 339). 47   Cf Kjellgren (n 21 above) 247; W Schön, ‘Abuse of Rights and European Tax Law’ in J Avery-Jones et al (eds), Comparative Perspectives on Revenue Law: Essays in Honour of John Tiley (Cambridge, Cambridge University Press 2008) 75, 76–77; J Englisch, ‘Verbot des Rechtsmissbrauchs – ein allgemeiner Rechtsgrundsatz des Gemeinschaftsrechts?’ [2009] Steuer und Wirtschaft 3. 48  AG Stix-Hackl in Fidium Finanz (n 29 above), para [96] (see nn 55, 70, 89 below); Schön (n 21 above) 1272–73; Fleischer (n 21 above) 869; D Anderson, ‘Abuse of Rights’ [2006] Judicial Review 348, 350; LM Baudenbacher, ‘Überlegungen zum Verbot des Rechtsmissbrauchs im Europäischen Gemeinschaftsrecht’ [2008] Zeitschrift für Rechtsvergleichung, Internationales Privatrecht und Europarecht 205, 206-212; Schammo (n 24 above) 358; S Vrellis, ‘”Abus” et “fraude” dans la jurisprudence de la Cour de Justice des Communautées Européennes’ in Liber Amicorum Hélène Gaudemet-Tallon (Paris, Dalloz 2008) 633, 634; T Tridimas, ‘Abuse of Rights in EU Law: Some Reflections with Particular Reference to Financial Law’, ch 12 above, 171; P Koutrakos, ‘The Emsland-Stärke Abuse of Law Test in the Law of Agriculture and Free Movement of Goods’, ch 14 above, 204; D Weber, ‘Abuse of Law in the Context of Indirect Taxation: Why We Need the Subjective Intention Test, When is Combating Abuse an Obligation and Other Comments’, ch 27 above, 400. 49   Centros (n 5 above), para [24]. 40 41

526  Stefan Vogenauer There are, in other words, two main contexts in which the notion of abuse has been analysed by the Court. First, when Community law provisions are abusively invoked in order to evade national law. Second, when Community law provisions are abusively relied upon in order to gain advantages in a manner that conflicts with the purposes and aims of those provisions.50

These two categories may be classified as, first, ‘abusive rule avoidance’, ie abuse of an EU law rule to bring a person outside the scope of a domestic rule, and, secondly, ‘abusive rule appropriation’, ie abuse of an EU law rule to bring a person within the scope of that rule. In this section of the chapter I will examine each of them in turn before setting out their common features. In doing so, I will look at instances where, according to the Court, the prohibition of abuse of law was potentially applicable, regardless of whether the rule avoidance or the rule appropriation were actually ‘abusive’ in the case at hand. The requirements for such a ‘finding of an abuse’51 will be discussed in the following section of the chapter.52 1.  Rule Avoidance In the first category of cases a legal or natural person seeks to rely on a rule of EU law in order to avoid the application of a national provision which has, from the perspective of the person concerned, unfavourable consequences attached to it. This ‘avoidance’,53 ‘evasion’,54 ‘circumvention’,55 ‘escape’56 or ‘flight from’57 the disadvantageous legal regime of one Member State is typically combined with a ‘flight into’ the legal regime of another Member State which imposes a lesser regulatory burden on the activities pursued by the relevant person: it might offer lower tax rates,58 less strict financial supervision,59 less

 AG Poiares Maduro in Halifax (n 6 above), para [63].   Emsland-Stärke (n 7 above), para [52].   See below, Part I.D. 53   Van Binsbergen (n 3 above), para [13]: ‘activities . . . for the purpose of avoiding’ national rules which would otherwise be applicable to the person concerned. A similar phrase is used in Case C–23/93 TV10 SA v Commissariaat voor de Media [1994] ECR I–4795, para [23]. 54   Knoors (n 33 above), para [25]: ‘attempting wrongly to evade the application of their national legislation’; Case C–148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I–487, para [13]: ‘improperly evade the obligations deriving from national legislation’; Surinder Singh (n 34 above), para [24]: ‘the facilities created by the Treaty cannot have the effect of allowing the persons who benefit from them to evade the application of national legislation’; AG Poiares Maduro in Halifax (n 6 above), para [63] (cited at n 50 above). 55  AG Darmon in Daily Mail (n 27 above), para [9]; Centros (n 5 above), para [24] (cited at n 49 above); AG Stix-Hackl in Fidium Finanz (n 29 above), para [96]: ‘circumvention of a provision which imposes an obligation’ (‘Umgehung einer Norm, mit der eine Verpflichtung auferlegt wird’; ‘contournement d’une norme qui impose une obligation’); Kjellgren (n 21 above) 247: ‘evasion of national rules . . . often referred to as circumvention’; Schön (n 21 above) 1275: ‘Umgehung’. Schammo (n 24 above) 358 uses the French notion that is normally employed to designate ‘circumventions’ or ‘Umgehungen’: ‘fraude à la loi behaviour at national level’. 56   Cadbury Schweppes (n 7 above), para [55] and Thin Cap (n 37 above), para [74]: ‘wholly artificial arrange­ ments which do not reflect economic reality, with a view to escaping the tax normally due’. Cf Lagondet (n 21 above) 8: ‘échapper à un droit national jugé défavorable’; Engsig Sørensen (n 24 above) 448: ‘trying to escape from national legislation’; Weatherill (n 35 above) 56. 57  German legal writers classify such cases as ‘flight from a provision’ (‘Normenflucht’). Cf Fleischer (n 21 above) 869. 58   Daily Mail (n 27 above); Cadbury Schweppes (n 7 above). 59   Fidium Finanz (n 29 above). 50 51 52

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  527 stringent employment rights60 or immigration laws,61 lower minimum capital requirements for the formation of a company,62 etc. Many cases in this category involve so-called ‘u-turn constructions’63 whereby persons or goods based in one Member State (home State) cross the border to another Member State (host State) and then immediately return to the home State or carry out activities there. Prima facie, injecting a cross-border element enables them to rely on one of the fundamental freedoms guaranteed by the Treaty and to avoid the application of the more stringent home State rules. The Leclerc case is a typical example. It concerned books produced in France and thus subject to the French rules of price-fixing for books. These books ‘were exported with the sole purpose of re-importation in order to circumvent legislation of the type at issue’.64 In TV 10, a television broadcasting company was largely managed by Dutch nationals and had its target audience in the Netherlands. Incorporating it under Luxembourg law and locating its seat in the same country was meant to avoid the legislation applicable to domestic broadcasters in the Netherlands.65 The well-known Centros case also concerned a u-turn arrangement: a Danish couple established a company in the United Kingdom that never conducted business in that country and subsequently set up a branch in Denmark. They did not deny that they had done so ‘for the purpose of avoiding Danish legislation requiring that a minimum amount of share capital be paid’.66 Not all cases of rule avoidance involve a u-turn. In Kefalas, for example, a number of shareholders sought to avoid the application of a Greek statute that empowered a public authority to order an increase in the capital of an undertaking in difficulties. In doing so, they relied on a directly applicable provision of a directive which gave them the right to decide upon any capital increase in a general meeting, and thus to oppose a capital increase by way of administrative act.67 2.  Rule Appropriation a)  Advantage Flowing From a Provision of EU Law  The second category of cases covers legal or natural persons seeking to rely on a rule of EU law in order to trigger the application of that rule and enjoy, from the perspective of the persons concerned, the favourable consequences attached to it. The activities of the persons have the purpose of gaining an advantage derived from EU law,68 such as monetary refunds. English legal 60   Case C–341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet, Svenska Byggnadsarbetare­ förbundets avdelning 1, Byggettan and Svenska Elektrikerförbundet [2007] ECR I–11767. Cf C Barnard, ‘The Notion of Abuse and the Freedom to Provide Services: A Labour Lawyer’s Perspective’, ch 7 above, 100–03. 61   Akrich (n 26 above). 62   Centros (n 5 above). 63   Cf A Looijestij-Clearie, ‘Centros Ltd – A Complete U-Turn in the Right of Establishment for Companies’ (2000) 49 International and Comparative Law Quarterly 621, 638; Kjellgren (n 21 above) 246–47; Engsig Sørensen (n 24 above) 426; Doukas (n 10 above) 67. Weatherill (n 35 above) 56 speaks of ‘sham cross-border transactions’. 64   Leclerc (n 31 above), para [27]. 65   TV10 (n 53 above). Cf Doukas (n 10 above) 68–71. 66   Centros (n 5 above), para [18]. In a similar vein, Case C–167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I–10155, para [98]. 67   Kefalas (n 34 above). 68  AG Poiares Maduro in Halifax (n 6 above), para [63] (cited at n 50 above). Cf Centros (n 5 above), para [24] (cited at n 49 above): ‘taking advantage of provisions of Community law’.

528  Stefan Vogenauer language – as opposed to those of France69 and Germany70 – has no specific term to denote such a ‘flight into’ EU law by attempting to bring one’s activities within the scope of a EU law provision. The notion of ‘rule appropriation’ seems to encapsulate the phenomenon best. Alternatively, one might speak of ‘rule-capture’ or ‘rule-seeking’, thus echoing the economic concept of ‘rent-seeking’ which denotes the process of seeking to gain larger profits by manipulating public policy or economic conditions, for example by means of securing beneficial subsidies or tariffs. Similar behaviour is likely to occur in cases in the second category. Again, some of these involve u-turn arrangements. In Emsland-Stärke, a German company exported agricultural products to Switzerland and was granted an export refund under the regulation setting out the relevant conditions. Immediately after their release for use in Switzerland the products were brought back to Germany unaltered and by the same means of transport. They were then released for use in Germany.71 Other cases have a different structure. General Milk Products, for example, did not concern a u-turn situation but rather what might be called a y-type scenario: a German company imported several consignments of cheddar cheese from New Zealand and subsequently re-exported some of them to other Member States. It then relied on a European regulation to claim monetary compensation.72 In Cremer, the granting of an export refund for compound feeding-stuffs under yet another regulation depended on the composition of the product. By adding an insignificant amount of a certain ingredient the exporters were able to ensure that their product had the relevant composition to become one of the products ‘coming under the Community rules’ which stipulated the conditions for a refund.73 Cases of rule appropriation are not only confined to persons seeking the benefit of a rule of secondary legislation. They may also concern instances where the advantage sought derives from primary law. This can be seen from the decisions that are sometimes referred to under the rubric of ‘abuse of procedure’.74 These are cases where a national court seeks to exercise its procedural right to make a preliminary reference under Article 267 TFEU but the underlying dispute is of an ‘artificial nature’75 or the referring court asks ‘hypothetical questions’ which are not relevant to the solution of the substantive action in the underlying proceedings.76

69  French law would use the doctrine of fraude à la loi, cf Terré (n 38 above) no 499 (see further below at n 323), or speak of a ‘captation’ of the provision, cf Lagondet (n 21 above) 9. 70  German law would refer to Normerschleichung, cf Fleischer (n 21 above) 869. This concept was used by Advocate-General Stix-Hackl in Fidium Finanz (n 29 above), para [96] (‘Erschleichung eines nicht vorgesehenen Vorteils’) which only imperfectly translates into the English and French language versions (’fraudulent acquisition of an unforeseen advantage’; ‘obtention frauduleuse d’un avantage non prévu’): in cases of Erschleichung there is no ‘fraud’ in the technical sense and the advantage is better characterised as ‘unintended’ by the lawmaker, rather than as ‘unforeseen’. 71   Emsland-Stärke (n 7 above). 72   Case C–8/92 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas [1993] ECR I–779. 73   Case 125/76 Enterprise Peter Cremer v Bundesanstalt für landwirtschaftliche Marktordnung [1977] ECR 1593, para [12]. 74  Brown (n 20 above) 520-21; Lagondet (n 21 above) 11. The Court does not use the language of ‘abuse’ in these cases. 75   Case 104/79 Foglia v Novello [1980] ECR 745, para [10]. 76   Case C-343/90 Lourenço Dias v Director da Alfandega do Porto [1992] ECR I-4673, para [17]; Case C-83/91 Wienand Meilicke v ADV/ORGA F A Meyer AG [1992] ECR I–4871, para [25].

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  529 b) Advantage Flowing From a Provision of National Law  Legal or natural persons may also seek to bring themselves within the scope of application of a national rule.77 The rule sought may be one implementing a provision of a directive, as was the case in Halifax where the benefit sought was a tax deduction under a provision of the VAT Act 1994 which was based on the Sixth VAT Directive.78 Alternatively, there may be an attempt to rely on an EU provision in order to trigger the application of a genuinely domestic rule that is entirely outside the scope of EU law. The French worker Lair, for example, invoked the freedom of movement in order to benefit from the German student assistance legislation.79 In Chen, a Chinese citizen moved to Northern Ireland to give birth to her child, so that the child could acquire Irish citizenship by virtue of Irish law and both she and the child would secure the right to reside in the United Kingdom.80 3.  Common Features Originally the Court developed two separate strands of case law for instances of rule avoidance and instances of rule appropriation. The former emerged from the Van Binsbergen decision in 1974.81 The latter had the 1977 Cremer case as its point of departure.82 It was only in the late 1990s that the Court began to regard them as an integrated whole.83 In his Opinion in Kefalas, Advocate-General Tesauro pointed out that the need to ensure that rights conferred by Community law ‘are not exercised in a manner which is abusive, excessive or distorted’ had been ‘repeatedly recognised in the Court’s case law’. He first mentioned ‘the established case-law according to which “the facilities created by the Treaty cannot have the effect of allowing the persons who benefit from them to evade the application of national legislation . . .”’ and cited to the Van Binsbergen line of decisions. This he related to the Cremer line of cases where the right to rely on Community law had been denied because ‘the conduct of the person relying upon a right conferred by a provision of Community law can be demonstrated to be a fraudulent evasion of statutory law’.84 The Court condensed the Advocate-General’s line of reasoning to the laconic statement that ‘[a]ccording to the case-law of the Court, Community law cannot be relied on for abusive and fraudulent ends’. It then cited decisions from both strands of case law without distinguishing between them.85 A few months later, in Centros, it arrived at the formula set out at the beginning of this section which amalgamates instances of rule avoidance and cases of rule appropriation under the generic term ‘abuse’.86 In Emsland-Stärke, a case of rule appropriation decided in the year after the Centros decision, the Court for the first time formulated abstract criteria for assessing whether particular conduct amounts to abuse.87 The test was phrased in such general terms that   Cf Engsig Sørensen (n 24 above) 426–27, 448–49 (‘Attempt to benefit unduly from national laws’).   Halifax (n 6 above). 79   Case 39/86 Sylvie Lair v Universität Hannover [1988] ECR 3161. 80   Case C–200/02 Kunqian Catherine Zhu, Man Lavette Chen v Secretary of State for the Home Department [2004] ECR I–9925. 81   Van Binsbergen (n 3 above). 82   Cremer (n 73 above). 83   Cf Schön (n 21 above) 1273. 84  AG Tesauro in Kefalas (n 34 above), para [24]. 85   Kefalas (n 34 above), para [20]. 86   Centros (n 5 above), para [24] (cited at n 49 above), referring to cases from both lines of case law. 87   Emsland-Stärke (n 7 above). 77 78

530  Stefan Vogenauer it could also be applied to instances of rule avoidance.88 And such a broad operation of the test is indeed entirely appropriate: avoidance of a provision, which has unfavourable consequences attached to it, implies by definition the attainment of an advantage.89 Moreover, in the specific context of EU law, the abusive avoidance of a national rule is based on the abusive reliance on a fundamental freedom of EU law. The person engaging in abusive practices therefore ultimately aims to obtain an advantage from a rule of EU law, just as in the straightforward cases of rule appropriation.90 As a result, similar requirements for establishing the existence of abuse apply to both groups of cases.91 I will discuss these requirements in the following section. D. Test In Emsland-Stärke, the Court specified the requirements for prohibiting abusive behaviour much more clearly than it had done in its previous case law.92 Subsequent decisions, notably Halifax, have further refined this test.93 Whilst the case law is still in flux and there is as yet no universally accepted test for the application of the ‘prohibition of abuse of law’ in EU law, it seems that the following proposition broadly reflects the position of the Court and would be shared by most commentators: A given rule of law will not be applied where (1) a particular set of facts is clearly and unambigu­ ously covered by the wording of the rule but (2) the result of applying the rule would be contrary to the purpose of that rule and (3) the person’s reliance on the rule is abusive.

As restated here, the prohibition of abuse has three requirements. First, formal observance: the natural or legal person invoking a particular rule must meet all of its requirements by complying strictly with its letter. Secondly, frustration of purpose: the application of the rule to the conduct of the person would lead to a result that runs contrary to the purpose of the rule. Thirdly, abusive reliance: in order to benefit from the application of the rule the person has arranged its affairs in an artificial manner. If these requirements are met the legal consequence is triggered: the rule is not applied to the facts and the person invoking it cannot derive rights from it.

 AG Stix-Hackl in Fidium Finanz (n 29 above), para [95].   Cf AG Stix-Hackl in Fidium Finanz (n 29 above), para [96]: ‘wohnt der Umgehung einer Norm, mit der eine Verpflichtung auferlegt wird, zugleich auch die Erschleichung eines nicht vorgesehenen Vorteils inne’, unfortunately somewhat distorted in the English language version (see n 70 above). 90   In a similar vein, Schammo (n 24 above) 358: ‘. . . commonly the aim of the person is not merely to avoid domestic provisions but to benefit of the more permissive (efficient?) provisions of another state by relying on Community rules. Such behaviour can trigger two claims before the ECJ: (i) one of abuse/avoidance of national provisions; and (ii) one of abuse of EC law (because the reliance on Community rules, for the purposes of avoiding national provisions, is treated as potential “abuse” of EC law)’; Tridimas (n 48 above) 171; Koutrakos (n 48 above) 205: if a trader seeks to escape from the application of a national rule ‘the abusive behaviour would entail bringing within the scope of EU law an activity which would otherwise be subject to national law’; Weber (n 48 above) 401. 91   Contra Englisch (n 47 above) 5, 22. 92   Emsland-Stärke (n 7 above), paras [52–53], discussed by Koutrakos (n 48 above) 206–12. 93   Halifax (n 6 above), paras [74–75]. 88 89

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  531 1.  Formal Observance The first requirement for a finding of abuse is the formal observance of the legal rule in question.94 The person who invokes a particular rule must meet all of its requirements by complying strictly with its letter, ie his activities are clearly and unambiguously covered by the wording of the provision.95 Prima facie, therefore, the provisions concerned ‘might appear available by virtue of their literal meaning’.96 I do not wish to argue that any legal provision, particularly in the multilingual context of EU law, ever has a ‘clear and unambiguous’ meaning. However, every legal rule has a ‘core’ set of circumstances that are clearly covered by its wording.97 For example, in the Cremer case of 1977 a regulation provided for the grant of an export refund for compound feeding-stuffs ‘containing’ certain products. The regulation did not stipulate that these products had to be present to a certain degree, ie whether they had to amount to a minimum percentage of the feeding-stuffs. A trader who exported feeding-stuffs containing only a tiny proportion of one of the relevant products, tapioca flour, did comply squarely with the letter of this rule.98 Similarly, the facts of Centros were clearly covered by the wording of Articles 52 and 58 of the EC Treaty [now 49 and 54 TFEU].99 These provisions grant individuals the right to establish a company in any Member State which may then set up a branch in another Member State. This is what the Danish couple did when they established a company in the United Kingdom (which never conducted business in that country) and through it subsequently set up a branch in Denmark. These examples show that the rule which is formally observed may be a rule of primary or secondary EU law. As far as primary law is concerned, it is conceptually conceivable that someone might rely on the formal observance of an uncodified general principle of EU law.100 A lawyer may, for example, technically meet the requirements that trigger the right to legal privilege.101 With regard to secondary legislation, the formal observance of a rule may pertain to provisions in both regulations102 and directly applicable directives.103 94   Emsland-Stärke (n 7 above), para [52]: ‘despite formal observance of the conditions laid down by the Community rules’. Cf ibid, para [46]: ‘all the formal conditions for the grant of . . . export refunds laid down by Regulation No 2730/79 were fulfilled’; Halifax (n 6 above), paras [74, 86]: ‘notwithstanding formal application of the conditions laid down by the relevant provisions of the Sixth Directive and [of] the national legislation transposing it’; AG Poiares Maduro in Halifax (n 6 above), para [73]: ‘actions . . . formally complying with a legal provision’; Case C–456/04 Agip Petroli SpA v Capitaneria di porto di Siracusa et al [2006] ECR I–3395, para [23]: ‘notwithstanding that technically the conditions laid down by . . . the regulation apply’; D Simon and A Rigaux, ‘La technique de la consécration d’un nouveau principe général du droit communautaire: l’exemple de l’abus de droit’ in Mélanges en hommage à Guy Isaac: 50 ans de droit communautaire (Toulouse, Presses de l’Université des sciences sociales de Toulouse 2004) 566, 568: ‘de pratiques qui se conforment formellement à la règle’. 95  AG Mazák in Weald Leasing (n 25 above), para [11]: ‘notwithstanding the fact that a trader has formally complied with the letter of the VAT Legislation’. 96  AG Poiares Maduro in Halifax (n 6 above), para [91]. See also the references of AG Poiares Maduro in paras [74, 76, 79] to ‘a formalistic application of [provisions of Community Law] based solely on their plain meaning’, ‘relying on a literal meaning’ and ‘the literal meaning of the legal provision’. Cf Maublanc (n 2 above) 268: ‘des opérations qui respectent formellement, littéralement, les conditions prévues’. 97   P Heck, ‘Gesetzesauslegung und Interessenjurisprudenz’ (1914) 112 Archiv für die civilistische Praxis 1, 46–47, 173; HLA Hart, The Concept of Law (Oxford, Oxford University Press 1961) 119. 98  AG Reischl in Cremer (n 73 above), at 1611 (sub 3.). 99   Centros (n 5 above). 100  As far as I can see, there is no case law on this point. 101   Case 155/79 AM & S Europe Limited v Commission [1982] ECR 1575. 102   Cremer (n 73 above). 103   Diamantis (n 34 above).

532  Stefan Vogenauer Furthermore, the prohibition of abuse of law may also be relevant if the formal obser­ vance relates to a national rule, whether it implements a directive,104 is adopted in full compliance with a derogation permitted under a directive105 or is simply invoked in any other case having an EU law dimension.106 The ‘prohibition of abuse of law’ in EU law therefore goes beyond a mere ‘prohibition of abuse of EU law’. In most cases the rule that is observed will grant subjective entitlements (‘rights’) to private persons, such as the freedom to provide services or the right to claim an export refund. However, as Professor Eidenmüller’s contribution shows, it is at least conceivable that public bodies, such as national legislators, courts and authorities engage in abusive practices whilst observing the letter of EU law.107 His case is supported by the abovementioned instances of ‘abuse of procedure’ where national courts arguably abused the preliminary reference procedure by referring non-genuine disputes to the Court of Justice.108 2.  Frustration of Purpose The prohibition of abuse of law further requires that the application of the rule to the facts of the case would be ‘contrary to the purpose’,109 the ‘spirit’,110 the ‘aims and results’111 or the ‘objectives’112 of the rule in question.113 The objectives pursued by the rule ‘would be frustrated’ if it were applied.114 The prohibition thus aims ‘to prevent the formal and mechanical application of legal rules from leading to unacceptable results with respect to the objectives of the legal system’.115 104   Weald Leasing (n 25 above), para [42]. Cf Halifax (n 6 above), paras [74, 86] (see n 94 above); RBS Deutschland (n 27 above), para [49]. 105  AG Mazák in Weald Leasing (n 25 above), para [24]. 106   Cf de la Feria (n 4 above) 429–30; Fleischer (n 21 above) 869; Engsig Sørensen (n 24 above) 435–36; Maublanc (n 2 above) 267. 107  Eidenmüller (n 22 above) 143, 152–54, with examples from insolvency law. Contra Costello (n 19 above) 323. 108   See above (at nn 74–76). 109   Halifax (n 6 above), paras [74, 86]; Part Service (n 7 above), para [42]; Weald Leasing (n 25 above), para [29]; cf Emsland-Stärke (n 7 above), para [52] and Case C–515/03 Eichsfelder Schlachtbetrieb GmbH v Hauptzollamt Hamburg-Jonas [2005] ECR I–7355, para [39]: ‘the purpose of those rules has not been achieved’. 110  AG Tizzano in Chen (n 80 above), para [114]: ‘it must be ascertained whether the person concerned, by invoking the Community provision which grants the right in question, is betraying its spirit and scope’. 111  AG Poiares Maduro in Halifax (n 6 above), paras [68, 89, 91]; see also ibid, para [79]: ‘aims and objectives’. 112   Case C–441/93 Panagis Pafitis et al v Trapeza Kentrikis Ellados AE et al [1996] ECR I–1347, para [68]: ‘an improper advantage, manifestly contrary to the objective of that provision’ (repeated in Kefalas (n 34 above), para [28] and Diamantis (n 34 above), para [33]); AG Poiares Maduro in Halifax (n 6 above), paras [65] (‘most importantly, it must be made in conformity with the the purpose and objectives of the provision of Community law’) and [73, 92]; Part Service (n 7 above), para [58]: ‘verify, first, whether the result sought is a tax advantage, the granting of which would be contrary to one or more of the objectives of the Sixth Directive’. 113  AG Alber in Emsland-Stärke (n 7 above), para [71]: ‘That purpose could be frustrated by a re-importation of the goods’; AG Poiares Maduro in Halifax (n 6 above), para [88]: ‘the purpose and objectives of the Community rules allegedly being abused are compared with the purpose and results achieved by the activity at issue’; AG Tizzano in Chen (n 80 above), para [115]: ‘whether or not there has been a distortion of the purposes and objectives of the Community provision which grants the right in question’. 114  AG Poiares Maduro in Halifax (n 6 above), para [91]: ‘the aims and results pursued by the legal provisions formally giving rise to the tax advantage invoked would be frustrated if that right were conferred’; Agip Petroli (n 94 above), para [23]: ‘frustrating the aim of . . . the regulation’. The notion of ‘frustration of purpose’ in this context is not to be confused with the eponymous doctrine of English contract law (cf Krell v Henry [1903] 2 KB 740, CA). 115  AG Poiares Maduro in Halifax (n 6 above), para [74] n 68, referring to scholarly discussion of the French doctrine of abus de droit.

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  533 In the above-mentioned Cremer case, for example, the Court emphasised ‘the objectives of the system of export refunds’ as expressed in the preamble of the regulation in question. It appeared ‘clearly from these recitals that the objective of the refund on export . . . is to compensate for the effect on the prices’. This objective was not achieved if there was no ‘actual presence in the compound feeding-stuff, in significant proportions,’ of the relevant products.116 Many of the early cases concerning abuse glossed over the requirement of frustration of purpose.117 More recently it has taken centre stage.118 It was crucial in cases where a finding of abuse was rejected. The most prominent of these is the decision in Centros. Here, the Court explicitly acknowledged that it is in principle legitimate to prevent individuals from abusively relying on provisions of Community law.119 Yet, it emphasised that in such cases one ‘must nevertheless assess such conduct in the light of the objectives pursued by those provisions’.120 It held that in the case at issue ‘the provisions of the Treaty on freedom of establishment are intended specifically to enable companies formed in accordance with the law of a Member State and having their registered office . . . within the Community to pursue activities in other Member States through an agency, branch or subsidiary’.121 As a result, to establish a company in a Member State with less restrictive rules of company law and to set up branches in other Member States ‘cannot, in itself, constitute an abuse of the right of establishment. The right to form a company in accordance with the law of a Member State and to set up branches in another Member State is inherent in the exercise, in a single market, of the freedom of establishment guaranteed by the Treaty’.122 The reasoning of the Court in Centros may not be entirely convincing. It is difficult to argue that the objective of freedom of establishment is promoted by the incorporation of a company in another Member State if there is no genuine economic integration in the host state at all.123 What matters for our purpose, however, is that the Court followed the Opinion of Advocate-General La Pergola who had suggested that the situation before the Court was, ‘whether one likes it or not, . . . the logical consequence of the rights guaranteed under the Treaty’ and was ‘consistent with the objective behind the inclusion of the freedom of establishment in the Treaty’. Thus it could not be held that the founders of the company ‘took “an improper advantage, manifestly contrary to the objective” pursued by Articles 52 et seq. of the Treaty’.124

  Cremer (n 73 above), paras [13] and [14].   V Edwards and P Farmer, ‘The Concept of Abuse in the Freedom of Establishment of Companies: A Case of Double Standards?’ in A Arnull, P Eeckhout and T Tridimas (eds), Continuity and Change in EU Law – Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press 2008) 205, 215. 118   See, eg, Diamantis (n 34 above), paras [33, 34], Kefalas (n 34 above), paras [22, 28], Emsland-Stärke (n 7 above), para [52] with the Opinion of AG Alber, para [69]; Halifax (n 6 above), para [74] and AG Poiares Maduro in Halifax (n 6 above), paras [65, 68]; Weald Leasing (n 25 above), para [32]. Cf Schön (n 47 above) 79. 119   Centros (n 5 above), para [24]. 120   Centros (n 5 above), para [25]. 121   Centros (n 5 above), para [26] (my emphasis). 122   Centros (n 5 above), para [27]. Cf G Ringe, ‘Sparking Regulatory Competition in European Company Law: The Impact of the Centros Line of Case Law and its Concept of “Abuse of Law”’, ch 8 above, 108–10. 123  The point is convincingly made by Edwards and Farmer (n 117 above) 218 and J Vella, ‘Sparking Regulatory Competition in European Company Law: A Response’, ch 9 above, 129, 131–32. 124  AG La Pergola in Centros (n 5 above), para [20], citing Kefalas (n 34 above), para [28] (my emphasis). Cf the interpretation of this aspect of Centros by AG Stix-Hackl in Fidium Finanz (n 29 above), paras [91–92]. 116 117

534  Stefan Vogenauer The Court subsequently adopted the same reasoning in another company law case, Inspire Art,125 and extended it to the area of corporate taxation in Cadbury Schweppes. In that case, the Advocate-General invited the Court to give ‘precedence to the purpose of the right of establishment’, a freedom that ‘is . . . intended’ to enable a company established in one Member State to set up a secondary establishment in any other Member State.126 Accordingly, the Court held that the fact that someone establishes a company in another Member State in order to benefit from the tax advantages in that State ‘cannot in itself deprive him of the right to rely on the provisions of the Treaty’.127 In Akrich, a case on free movement of workers, the Advocate-General also relied on Centros-style reasoning in order to reject a finding of abuse.128 In other words, there is no abuse of law if the application of EU law provisions ‘is compatible with the purposes and objectives pursued by the relevant provisions’.129 In the field of indirect taxation it may be the case that applying a rule would not be contrary to its purpose because the VAT system has anomalies and inconsistencies and the taxable person simply makes use of such anomalies: ‘not every “use” made of that possibility by a taxable person may automatically be deemed to be constitutive of an “abuse” ’.130 3.  Abusive reliance Finally, a finding of abuse requires that the reliance on the rule by the person invoking it can be classified as abusive.131 This requirement is often couched in terms describing the person’s conduct, motives or aims as ‘improper’,132 ‘not bona fide’,133 ‘opportunistic’,134 ‘wrongful’135 or ‘fraudulent’.136 The Court still uses such ‘evocative’137 phrases although its language has gradually shifted towards more neutral terminology since Emsland-Stärke. What matters under the test set forth in this decision is whether the person aims ‘to obtain an advantage from the Community rules by creating artificially the conditions laid down

125   Inspire Art (n 66 above), paras [137–38]: ‘while in this case Inspire Art was formed under the company law of a Member State . . . for the purpose in particular of evading the application of Netherlands company law, which was considered to be more severe, the fact remains that the provisions of the Treaty on freedom of establishment are intended specifically to enable companies . . .’. 126  AG Léger in Cadbury Schweppes (n 7 above), paras [46 and 41]; see also the reference to the ‘objective’ in para [43] of the Opinion. 127   Cadbury Schweppes (n 7 above), para [36]. The Court did, however, revisit the issue of abuse at the level of justification, see below, Part I.H.1, and Vella (n 123 above) 129–33. 128  AG Geelhoed in Akrich (n 26 above), paras [100, 178–81]. The Court reached the same outcome but it is not clear whether it did so for the same reasons: Akrich (n 26 above), paras [55–57]. 129  AG Poiares Maduro in Halifax (n 6 above), para [95]. 130  AG Mazák in RBS Deutschland (n 27 above), para [76]. The Court did not consider this point. 131  AG Poiares Maduro in Halifax (n 6 above), paras [65] (‘the provision of Community law . . . relied upon in an abusive way’) and [79] (‘the provision abusively relied upon’). 132   Veronica (n 54 above), para [13]; Pafitis (n 112 above), para [68]; Kefalas (n 34 above), para [28]; Diamantis (n 34 above), para [33] (cited in n 112 above); Centros (n 5 above), para [24] (cited at n 49 above); Kofoed (n 1 above), para [38]. Cf Vella (n 123 above) 128. 133   General Milk Products (n 72 above), para [21]; Emsland-Stärke (n 7 above), para [51]. 134  AG Poiares Maduro in Halifax (n 6 above), para [76]: ‘opportunistic behaviour by taxable persons relying on the literal meaning of its provisions to improperly gain tax advantages against the tax authorities’. 135   General Milk Products (n 72 above), para [21]; Emsland-Stärke (n 7 above), para [51]; Bozkurt (n 34 above), para [49]. 136   Kofoed (n 1 above), para [38]; cf Diamantis (n 34 above), para [33]: ‘abusive or fraudulent ends’. 137   Maublanc (n 2 above) 267.

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  535 for obtaining it’.138 Thus the touchstone is the artificiality of the conduct139 which turns otherwise unobjectionable behaviour into an ‘abusive practice’.140 As with frustration of purpose, the early decisions concerning abuse of law were not explicit about the requirement of artificiality. In the 1977 Cremer case, the Court simply mentioned the ‘manipulation’ on the part of the exporters that, according to the customs authorities, amounted to an ‘abusive practice’: they had added an insignificant proportion (two percent) of tapioca flour to the compound feeding-stuffs.141 Presumably this had been done with the sole purpose of acquiring the entitlement to claim the export refund but the Court did not discuss this issue. A decade later, in Lair, the Court raised the bar for the establishment of ‘abuse’ by requiring ‘objective evidence’ that the activities were conducted ‘for the sole purpose of enjoying’ the beneficial consequences of a rule.142 Emsland-Stärke then added the language of artificiality, as mentioned in the previous paragraph. To this day the Court has refrained from developing hard and fast rules to determine in the abstract whether or not particular conduct is artificial. The assessment therefore ultimately depends on (i) the specific area of law concerned and (ii), as the Court has frequently emphasised, a careful analysis of the circumstances of the case.143 Economic operators engage in ‘abusive practices’ if their transactions are ‘carried out not in the context of normal commercial operations but solely for the purpose of wrongfully obtaining advantages provided for by Community law’.144 This is the case if there are ‘economic activities for which there is objectively no other explanation than the creation of the right claimed’.145 The absence of such an explanation, or ‘justification’,146 indicates artificial conduct which, in turn, makes it abusive to rely on the rule invoked.147 In the export refund cases, an ‘abuse, that is to say, a purely formal dispatch from Community territory with the sole purpose of benefiting from export refunds’ can be 138   Emsland-Stärke (n 7 above), para [53]. Cf ibid, para [56]: ‘the conditions required to obtain the advantage derived from the Community rules were created artificially’; Agip Petroli (n 94 above), para [23]: ‘there must also be objective evidence to show that the essential aim of the international voyage in ballast is to avoid the application of Article 3(2) of the regulation’. 139   Halifax (n 6 above), para [81]: ‘take account of the purely artificial nature of those transactions’, and AG Poiares Maduro, para [70]: ‘a finding that the situation giving rise to the application of a certain Community rule was purely artificial’; Agip Petroli (n 94 above), para [22]: ‘have set up artificially an international voyage in ballast’. Cf de la Feria (n 4 above) 429; AG Geelhoed in Akrich (n 26 above), para [172]: the parties had ‘created a loophole’. 140   See references in n 28 above. 141   Cremer (n 73 above), paras [12, 14]. 142   Lair (n 79 above), para [43]: ‘where it may be established on the basis of objective evidence that a worker has entered a Member State for the sole purpose of enjoying, after a very short period of occupational activity, the benefit of the student assistance system in that State’. Cf Case C–413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I–13187, para [34]. 143  AG Darmon in Daily Mail (n 27 above), para [9]: ‘in a specific case and having regard to the circumstances’; AG Mazák in Weald Leasing (n 25 above), para [10] n 36: ‘the referring court must examine and weigh all the applicable contractual terms and relevant circumstances’. For the need for a ‘case by case’ analysis, see Centros (n 5 above), para [25]; Bozkurt (n 34 above), para [47]. Cf Lagondet (n 21 above) 11; Engsig Sørensen (n 24 above) 453–54. 144   Halifax (n 6 above), para [69] (not confined to VAT cases but concerning ‘economic operators’ in general); Ampliscientifica (n 27 above), para [27]; Weald Leasing (n 25 above), para [26]; AG Mazák in RBS Deutschland (n 27 above), para [70]. Apparently the concept of ‘normal commercial operations’ was first used by AG Darmon in Leclerc (n 31 above), para [17]. 145  AG Poiares Maduro in Halifax (n 6 above), para [91]. 146  AG Poiares Maduro in Halifax (n 6 above), para [87]. 147  AG Poiares Maduro in Halifax (n 6 above), para [73]: actions which ‘amount to abusive exploitation of the possibilities left open by that provision’.

536  Stefan Vogenauer suggested by ‘the specific circumstances of the operation at issue’.148 One ‘of the factual elements which can be taken into account . . . to establish the artificial nature of the operation concerned’ is the fact that the undertakings concerned had ‘personal and commercial links’, ie there is ‘evidence of collusion between the Community exporter receiving the refunds and the importer of the goods in the non-member country’.149 In the field of indirect taxation, it must ‘be apparent from a number of objective factors that the essential aim of the transactions concerned is to obtain a tax advantage’.150 It is therefore necessary ‘to determine the real substance and significance of the transactions concerned. In so doing, [the court] may take account of the purely artificial nature of those transactions and the links of a legal, economic and/or personal nature between the operators involved in the scheme’.151 The ‘contractual terms of the . . . transaction at issue’ are equally important. For example, it would be a factor pointing towards an assessment of artificiality if in a leasing transaction ‘the rentals were set at levels which were unusually low or did not reflect any economic reality’.152 In sum, ‘wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due’ will be held to be abusive.153 Conversely, ‘the prohibition of abuse is not relevant where the economic activity carried out may have some explanation other than the mere attainment of tax advantages’.154 It ‘cannot be regarded as constituting an abuse of rights’ if the transactions between two parties that ‘were legally unconnected’ were ‘carried out in the context of normal commercial operations’ and ‘contain nothing to suggest an artificial arrangement that does not reflect economic reality and the sole aim of which is to obtain a tax advantage’ – in short when there was ‘genuine economic activity’.155 Factors pointing to this include contractual terms that ‘correspond to arm’s length terms’156 and the pursuit of ‘economic objectives arising from, for example, marketing, organisation or guarantee considerations’157 or ‘cost efficiencies’.158 A transaction may be a normal commercial operation even if the economic operator does not usually engage in that particular type of transaction. What matters is the particular purpose of the relevant transaction.159 In the area of free movement of workers, the artificial creation of conditions entitling a worker to student assistance benefits may be indicated by the fact that the previous employment was of ‘short duration’ only. However, this ‘cannot, in itself, exclude that employment from the scope of Article 48 of the [EC] Treaty’ [now Article 45 TFEU].   Emsland-Stärke (n 7 above), para [50].   Emsland-Stärke (n 7 above), paras [58, 53].   Halifax (n 6 above), para [75]. 151   Halifax (n 6 above), para [81]. Repeated in Part Service (n 7 above), para [62]; AG Mazák in RBS Deutschland (n 27 above), para [79]. 152   Weald Leasing (n 25 above), para [39]. 153   Cadbury Schweppes (n 7 above), para [55]; Thin Cap (n 37 above), para [74]. In these cases, the question whether there was an abusive practice was only asked at the level of justification of a national restriction (see below, Part I.H.1). 154   Halifax (n 6 above), para [75]. Cf AG Poiares Maduro in Halifax (n 6 above), paras [87] (‘determine whether the activity at issue has some autonomous basis which, if tax considerations are left aside, is capable of endowing it with some economic justification in the circumstances of the case’) and [96]. 155   RBS Deutschland (n 27 above), paras [50–52], taking up Ampliscientifica (n 27 above), para [28]. Cf AG Mazák in RBS Deutschland (n 27 above), para [71]. 156   Weald Leasing (n 25 above), para [45]. 157   Part Service (n 7 above), para [62]. 158  AG Mazák in RBS Deutschland (n 27 above), para [79]. 159   Weald Leasing (n 25 above), paras [43–45]. Cf AG Mazák in Weald Leasing (n 25 above), para [32]. 148 149 150

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  537 What matters is rather whether the employment performed was ‘purely marginal and ancillary’.160 Nor is it abusive conduct per se if a third-country citizen marries an EU citizen, they then move together to a Member State that is not the home State of the EU citizen, and after a short time they jointly return to the spouse’s home State – even if all of this is done with the avowed aim of the third-country citizen benefitting from EU law in order to evade the restrictive immigration legislation of the spouse’s home State and to acquire the right of permanent residence there. ‘Conversely, there would be an abuse if the facilities afforded by Community law in favour of migrant workers and their spouses were invoked in the context of marriages of convenience entered into in order to circumvent the provisions relating to entry and residence of nationals of non-Member States’.161 Only if there is a ‘factor suggesting that . . . the marriage was a sham contracted with the sole aim of enjoying abusively advantages provided for by the law’ can an assessment be made that the spouse has artificially created the conditions laid down for obtaining the right to abode.162 All these considerations confirm that ‘the dividing line between abuse and use for a purpose not contemplated by the legislator is hard to define’.163 Be that as it may, whether the reliance of a person will be found to be ‘abusive’ does not only depend on the specific area of law concerned and the particular facts of the case at hand. There are four issues of a more general nature which merit further discussion. a)  Abusive Practice Alone not Sufficient  First, abusive conduct as such is not sufficient to trigger the prohibition of abuse of law. As has been shown above, it is always necessary that the application of the respective rule to the situation is contrary to its purpose. In Centros, for example, the circumstances certainly indicated – and the Danish couple did not deny – that they had acted ‘for the purpose of avoiding Danish legislation requiring that a minimum amount of share capital be paid’.164 However, there was no finding of abuse because their conduct was not inconsistent with the objectives underlying the freedom of establishment.165 In a more recent VAT case, the referring court had stated that the essential aim of the relevant transactions was to obtain a tax advantage. The Court of Justice warned that ‘before it can be concluded that there was an abusive practice, it must also be the case that, notwithstanding formal application of the conditions laid down in the relevant provisions . . . that tax advantage is contrary to the purpose of those provisions’.166 It is indeed perfectly conceivable that there are ‘instances where the sole purpose of the activity might be to diminish tax liability but where the purpose is actually a result of a choice between different tax regimes that the Community legislature intended to leave open’. In these cases there is no contradiction between recognising the claim made by the taxable person and the aims and results pursued by the provision invoked, so there is no abuse of law.167   Ninni-Orasche (n 142 above), paras [25–26].   Akrich (n 26 above), para [57].   Bozkurt (n 34 above), para [50]. 163  AG Geelhoed in Akrich (n 26 above), para [173]. 164   Centros (n 5 above), para [18]. In a similar vein, Inspire Art (n 66 above), para [98]; AG Geelhoed in Akrich (n 26 above), paras [100, 172, 178–84]. 165   See above (after n 119). 166   Weald Leasing (n 25 above), para [32]. However, in paras [33–45] the Court mixed up the second and third limbs of the test and discussed factors indicating abusive reliance in the context of frustration of purpose. 167  AG Poiares Maduro in Halifax (n 6 above), para [88]. 160 161 162

538  Stefan Vogenauer b)  Objective Determination of the Purpose of the Activities  Secondly, there has been discussion of whether the person relying on the rule must have had the subjective ‘intention’ to obtain an advantage from it by artificially creating the conditions laid down for obtaining it.168 This was the phrase used in the Emsland-Stärke decision where the Court made the first attempt to formulate an overarching test for the prohibition of abuse.169 Requiring such a ‘subjective element’ is very much in line with the Romanistic tradition of requiring an ‘intention to inflict harm’ (animus nocendi, intention de nuire) in order to establish an abus de droit. However, as is known from the jurisdictions concerned, it is normally difficult to prove that the person relying on a rule did indeed have such an intention. Most of these jurisdictions have therefore ‘objectivised’ their subjective element and acknowledge that it is sufficient to show that the exercise of the right has a harmful effect or is, according to objective standards, unreasonable.170 This is even the case in Italy171 where Article 833 of the Civil Code explicitly requires that the activities in question ‘have no other purpose than that of harming and causing annoyance’.172 The Court in Emsland-Stärke implicitly embarked on a similar ‘objectivisation’ when it conceded that the ‘subjective element’ could be established by evidence of objective criteria.173 The decision in Halifax took the analysis one step further. The Advocate-General had forcefully argued that what mattered was the purpose of the activities in question. That purpose – which must not be confused with the subjective intention of the participants in those activities – is to be objectively determined on the basis of the absence of any other economic justification for the activity than that of creating a tax advantage.174

The Court followed this suggestion. It abandoned the requirement of a ‘subjective element’ and made it clear that it rather looked for ‘the essential aim [résultat, Ziel] of the transactions’, as ‘apparent from a number of objective factors’.175 In the later case law of the Court the ‘subjective element consisting in the intention to obtain [an] advantage’ has occasionally resurfaced.176 Nevertheless, there is now a broad consensus that the prohibition of abuse does not necessitate an inquiry into the inner motivation of the persons invoking the rule at issue. An abusive practice will be established on the basis of objective factors which provide evidence of genuine commercial reasons for the transaction.177 This amounts to a significant relaxation of the abuse test: 168  For further discussion, see Schön (n 21 above) 1284–87; Baudenbacher (n 48 above) 215–16; Schammo (n 24 above) 367–69. 169   Emsland-Stärke (n 7 above), para [53]. 170  HC Gutteridge, ‘Abuse of Rights’ (1935) 5 Cambridge Law Journal 22, 26; A Gambaro, ‘Abuse of rights in civil law tradition’ (1995) 3 European Review of Private Law 561. 171   See the references in G Cian and A Trabucchi, Commentario breve al Codice civile, 8th edn (Milano, CEDAM 2007) Art 833, para I.4. 172  Translation by M Beltramo, The Italian Civil Code and Complementary Legislation (Dobbs Ferry/NY, Oceana 1991-93). 173   Emsland-Stärke (n 7 above), para [53]. 174  AG Poiares Maduro in Halifax (n 6 above), para [87]; cf para [71]: ‘the artificial character of the situation to be assessed in the light of a set of objective circumstances’. 175   Halifax (n 6 above), paras [75, 86] (cited at n 150 above). 176   Cadbury Schweppes (n 7 above), para [64]. 177  AG Geelhoed in Akrich (n 26 above), paras [102, 174]; Akrich (n 26 above), paras [55–56]; AG Léger in Cadbury Schweppes (n 7 above), paras [118, 119]; Agip Petroli (n 94 above), para [23]; AG Mazák in Weald Leasing (n 25 above), para [33]; Engsig Sørensen (n 24 above) 457; Edwards and Farmer (n 117 above) 221–24; Baudenbacher (n 48 above) 216; AK Lenaerts, ‘The General Principle of the Prohibition of Abuse of Rights:

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  539 such objective factors are frequently present whereas an abusive intention will rarely be successfully proved. The objectivisation of the test will therefore increase the number of cases where the courts will be able to make a finding of abuse. c) Sole Purpose or Essential Aim?  The third question relating to the requirement of abusive reliance is whether obtaining an advantage from EU rules must have been ‘the sole aim pursued’ by the person, to the exclusion of other economic objectives, or whether it is sufficient that this was merely the ‘essential’ or ‘principal’ of two or more purposes the person pursued. The earlier decisions of the Court on export refunds seemed at least implicitly to support the first view. They required that the transactions be realised ‘only in order wrongfully to benefit from the grant of monetary compensatory amounts’178 or ‘with the sole purpose of benefiting from export refunds’.179 Equally, there was an ‘abuse’ of the freedom of movement where ‘a worker has entered a Member State for the sole purpose of enjoying, after a very short period of occupational activity, the benefit of the student assistance system in that State’.180 A thorough analysis of the issue was first undertaken by Advocate-General Poiares Maduro in the Halifax case. According to him, in the area of VAT a finding of abuse presupposed that the ‘sole purpose’ of the transaction was to obtain tax advantages. It was necessary that the ‘right invoked derives from economic activities for which there is objectively no other explanation than the creation of the right claimed’. Any other view ‘would introduce a high degree of uncertainty’ and ‘affect economic activities which clearly deserve protection, provided that they are, at least to some extent, accounted for by ordinary business aims’.181 The Court was less clear. It summarised the previous case law on export refunds as referring to transactions carried out ‘solely for the purpose of wrongfully obtaining advantages provided for by Community law’ and stated that, according to the referring court, the relevant transactions had the sole purpose of obtaining a tax advantage.182 However, when it rephrased the abstract abuse test established in Emsland-Stärke it merely required that obtaining such an advantage was the ‘essential aim of the transactions concerned’.183 It used a similar phrase when it gave further guidance to the referring court.184 The subsequent case law is confused. The language of ‘sole purpose’ continued to be used185 and was sometimes employed interchangeably with the ‘essential aim’ formula.186 In the Part Service case, the Court was explicitly asked to clarify its Halifax judgment. The Second Chamber held that the ‘minimum threshold for classifying a practice as abusive’ is reached if obtaining a tax advantage has been the sole purpose. But there can also ‘be a finding of an abusive practice when the accrual of a tax advantage constitutes the principal aim of the transaction or transactions at issue’,187 despite the ‘possible A Critical Position on Its Role in a Codified European Contract Law’ (2010) 18 European Review of Private Law 1121, 1136; Eidenmüller (n 22 above) 143; Koutrakos (n 48 above) 210. Contra Weber (n 48 above) 398–99. 178   General Milk Products (n 72 above), para [21]; Emsland-Stärke (n 7 above), para [51] (emphasis added). 179   Emsland-Stärke (n 7 above), para [50] (emphasis added). 180   Lair (n 79 above), para [43] (emphasis added). 181  AG Poiares Maduro in Halifax (n 6 above), paras [88–91]. 182   Halifax (n 6 above), paras [69, 82]. 183   Halifax (n 6 above), para [75] (emphasis added). 184   Halifax (n 6 above), para [81]: ‘must essentially seek to obtain a tax advantage’. 185   Thin Cap (n 37 above), para [77]; Cadbury Schweppes (n 7 above), para [63]; Kofoed (n 1 above), para [38]. 186   Agip Petroli (n 94 above), paras [20] (‘only with the aim’) and [23] (‘essential aim’). 187   Part Service (n 7 above), paras [40–45] (emphasis added).

540  Stefan Vogenauer existence, in addition, of economic objectives arising from, for example, marketing, organ­isation or guarantee considerations’.188 Quite astonishingly, the Third Chamber has since continued to require that the transactions were ‘set up with the sole aim of obtain­ ing a tax advantage’189 and has even used this formula and ‘essential aim’ as apparently equivalent tests.190 It may be the case that in the decisions following Part Service the standard of purpose did not matter because a finding of abuse was in any event not to be realistically expected. However, the distinction is important.191 On the one hand, requiring that the attainment of an advantage from EU law must have been the sole aim pursued is a much higher hurdle than merely requiring that it was the ‘essential’ or ‘principal’ objective. Persons frequently pursue a mixture of purposes in their activities. Findings of abuse will therefore be much rarer if the first standard is applied. On the other hand, the first standard is much more conducive to legal certainty. If the ‘essential aim’ test is used it will be necessary to assess whether the aim of gaining an advantage was so predominant, by comparison to the other purposes pursued, that reliance on the right granting the advantage must be classified as abusive. This requires a much greater value judgment than applying the ‘sole purpose’ standard and will render the law less predictable. Perhaps there is no room for a unitary test on this issue. Whilst the standard of ‘essential aim’ might be acceptable in many areas of law, the ‘sole purpose’ test might be more appropriate in fields such as tax law which require a specific emphasis on legal certainty. d)  Need for a Value Judgment  Finally, determining whether a person has engaged in abusive practices always requires a value judgment. The notion of ‘abusive’ as such has a strongly negative moral connotation. As has been seen further above, the Court has traditionally used equally value-laden synonyms such as ‘improper’, ‘wrongful’ or ‘fraudulent’ as synonyms.192 It is this blurring of the line between law and morality that makes doctrines of abuse so unpalatable to common lawyers steeped in the tradition of legal positivism.193 The concepts of ‘artificiality’, ‘genuine transaction’ and ‘normal commercial operation’ used in the wake of Emsland-Stärke are, prima facie, less value-laden. However, they also require the courts to engage in a normative exercise. In the area of tax law, for example, the courts are in effect involved in ‘defining the realm of choices that the common VAT rules have left open to taxable persons’.194 They are asked to determine whether activity of a certain ‘kind, even if not unlawful, deserves no protection’ under EU law.195

  Part Service (n 7 above), para [62].   Ampliscientifica (n 27 above), para [28] and, in a similar vein, para [27]; RBS Deutschland (n 27 above), paras [49, 51]. See also AG Sharpston in Bozkurt (n 34 above), para [67]. 190   Weald Leasing (n 25 above), paras [26, 30, 31]. 191   See J Snell, ‘The Notion of and a General Test for Abuse of Rights: Some Normative Reflections’, ch 15 above, 227. 192   See above, Part I.B. 193   See, eg, the comments by R O’Sullivan, ‘Abuse of Rights’ (1955) 8 Current Legal Problems 61, 67–68. 194  AG Poiares Maduro in Halifax (n 6 above), para [85]. For a similar point, see Ziegler (n 12 above) 297, 314. 195  AG Poiares Maduro in Halifax (n 6 above), para [86]. 188 189

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  541 E.  Legal Consequences 1.  Non-application of the Rule If the three requirements of the prohibition of abuse are met the rule in question is not applied to the facts of the case: ‘the application of the doctrine of abuse leads to the non-application of a Community law provision and, consequently, to the conclusion that the right is not conferred’.196 In other words, ‘abuses are not covered by the Community provisions in question’.197 Alternatively, it is frequently said that the application of a beneficial rule is ‘precluded’,198 that EU law ‘cannot be relied on’ for abusive ends,199 that the persons concerned are ‘denied’ the benefit of the provision they rely on,200 that they have ‘forfeited’ their right flowing from the rule201 or that their act results in a ‘failure to obtain the advantage’.202 All these phrases have in common that they denote a process by which a ‘right is not in fact conferred, contrary to the literal meaning of the legal provision’.203 The process is therefore comparable to that of ‘reading down’ the scope of legislation, a restriction or reduction of the scope of the rule, in the context of domestic laws.204 This is obscured by another formula that has been used more or less consistently in the cases on export refunds and VAT. According to this, ‘the scope’ of a particular rule or EU law in general ‘must in no case be extended to cover abusive practices’.205 The language of ‘extension’ creates the impression that what is involved is simply refraining from enlarging the scope of application of the rule to situations which are clearly not covered by its literal meaning. However, the consequence is more severe: the rule will simply not be applied to an instance that is clearly covered by its plain meaning. Ultimately, the result of the non-application of the rule invoked depends on the category of case concerned.206 In all instances the rule of EU law which is relied upon is not applied. In cases of abusive rule appropriation this simply means that the advantage that was sought to be obtained will not be gained. In cases of abusive rule avoidance the 196  AG Poiares Maduro in Halifax (n 6 above), para [71] n 66. Cf ibid, para [79]: the provision ‘cannot be regarded as conferring the right at issue’. Cf AG Darmon in Daily Mail (n 27 above), para [9]: ‘the national court may assess whether . . . it should decide not to apply Community law’. 197   Lair (n 79 above), para [43]; Ninni-Orasche (n 142 above), para [36]. 198   Emsland-Stärke (n 7 above), para [51]: operations may ‘preclude the application of positive monetary compensatory amounts’; Halifax (n 6 above), para [85]: ‘precluding any right . . . to deduct input VAT’. 199   Kefalas (n 34 above), para [20]; Diamantis (n 34 above), para [33]; Case C–32/03 I/S Fini H v Skatteministeriet [2005] ECR I–1599, para [32]. 200   Centros (n 5 above), para [25]: ‘deny them the benefit of the provisions of Community law on which they seek to rely’; Diamantis (n 34 above), para [34]; AG Stix-Hackl in Fidium Finanz (n 29 above), para [82]; Bozkurt (n 34 above), para [47]. 201   Emsland-Stärke (n 7 above), para [59]: ‘a Community exporter can forfeit his right to payment of [an] export refund’. 202  Article 4(3) of Regulation 2988/95 (n 9 above). 203  AG Poiares Maduro in Halifax (n 6 above), para [79]. Cf ibid, para [91]: ‘not conferring the rights that might appear to be available by virtue of their literal meaning’. 204  Recently there has been some confusion regarding this issue, see below Part I.H.1. 205   Cremer (n 73 above), para [21] (‘que l’application du règlement en cause . . . ne saurait en aucun cas être étendue jusqu’à couvrir des pratiques abusives’; ‘daß der Anwendungsbereich der Verordnung . . . keinesfalls so weit ausgedehnt werden darf, daß er mißbräuchliche Praktiken . . . deckt’) (my emphasis); Emsland-Stärke (n 7 above), para [51]; Agip Petroli (n 94 above), para [20]. Cf Halifax (n 6 above), para [69]; Kofoed (n 1 above), para [38]; Ampliscientifica (n 27 above), para [27] and Weald Leasing (n 25 above), para [26]: ‘Community legislation cannot be extended to cover abusive practices’ (my emphasis). 206   See above, Part I.C.

542  Stefan Vogenauer additional consequence is that the more stringent rules of the state that were sought to be avoided will continue to apply ‘to the Member State’s own nationals who are using Community law purely and simply in order to evade the legislation’.207 2.  Restitution of the Advantage Obtained The simple non-application of the rule in question may not amount to a satisfactory response to instances of abusive rule appropriation. This is the case where the relevant authorities have already applied the rule by granting the advantages provided for by it. Here, the finding of abuse results in the ‘withdrawal’ of the advantage.208 If the advantage was monetary, such as payment of an export refund or allowing the deduction of VAT, the authorities are entitled to demand reimbursement of the refund209 or repayment of the deduction made.210 Conversely, there is an ‘obligation to repay’ on the part of the person who abusively received the advantage. According to the Court, this is ‘simply the consequence of a finding that the conditions required to obtain the advantage derived from the Community rules were created artificially, thereby rendering the refunds granted undue payments and justifying the obligation to repay them’.211 EU law will normally not stipulate detailed rules on the restitution of advantages obtained. It is therefore, as a rule, for the Member States to provide the respective mechanisms within the limits imposed by EU law.212 3.  Partially Abusive Practices The position is particularly complex if the facts of a case relate to a number of transactions only some of which amount to an abusive practice. The non-application of the rule will then be partial: it will be confined to the abusive transactions, whilst the rule will be applied to the other transactions.213 Similarly, an obligation to repay does not arise with regard to advantages gained from transactions which are not considered to be abusive.214 4.  Re-definition of the Transaction Moreover, a particular transaction that was engaged in with the purpose of artificially obtaining an overall advantage may nevertheless have created some liabilities, for example a tax liability at a lower rate or a rental payment at a lower level. Such transactions ‘must be redefined so as to re-establish the situation that would have prevailed in the absence of the transactions constituting that abusive practice’.215 In cases where the lower tax rate  AG Geelhoed in Akrich (n 26 above), para [105].   Cf, for some cases, Article 4(3) of Regulation 2988/95 (n 9 above). 209   Eichsfelder Schlachtbetrieb (n 109 above), para [42]. 210   Halifax (n 6 above), para [95]. 211   Emsland-Stärke (n 7 above), para [56]. Cf Halifax (n 6 above), para [93] (see n 214 below). 212   Halifax (n 6 above), para [91]. 213  AG Poiares Maduro in Halifax (n 6 above), para [97]: ‘it must be concluded that the relevant provisions . . . do not confer that right or confer it only partially’. 214   Halifax (n 6 above), para [93] (‘an obligation to repay, simply as a consequence of that finding, which rendered undue all or part of the deductions of input VAT’) (emphasis added) and para [95] (authorities are entitled to demand ‘repayment of the amounts deducted in relation to each transaction whenever they find that the right to deduct has been exercised abusively’). 215   Halifax (n 6 above), paras [94, 98]; Weald Leasing (n 25 above), para [48]. 207 208

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  543 has been charged the authorities must subtract this amount from the amount that has to be repaid.216 Where a lower rent has been paid it may be possible to redefine the situation ‘by varying or disapplying those contractual terms’ to the extent that is necessary for the prevention of tax evasion.217 A third person who would have been able to obtain an advantage in the absence of the transactions constituting an abusive practice will be able to claim that advantage.218 F.  Procedural Aspects Questions of abuse of law always arise within the procedural framework of the preliminary reference procedure under Article 267 TFEU (ex Article 234 EC). The main proceedings before the referring court are almost invariably conducted between a private person and a national public authority.219 In cases of abusive rule avoidance the authority has applied, or wishes to apply, a domestic rule which the person had sought to avoid by relying on an EU law rule. In cases of abusive rule appropriation the authority has refused to apply, or refuses to apply, an EU law rule which the person had relied on in order to obtain an advantage. The proceedings are then brought by the private persons who invoke their rights under the relevant EU law provisions or, in cases where an advantage has already been obtained through abusive reliance on an EU law rule, by the public authorities who claim restitution. The abuse test as set out above requires both the interpretation of EU law and its application to the facts of the case. Under the division of functions provided for by Article 267 TFEU, the Court of Justice is responsible for the former and the national courts are charged with the latter. This division also applies when the prohibition of abuse of law is invoked.220 Thus it is for the Court of Justice to elucidate the wording of the EU rule that is relied upon and to define its object and purposes. It is then for the national court, in the light of the ruling of the Court on the interpretation of EU law, to establish whether the person concerned acted in formal compliance with the wording, whether the application of the rule would frustrate its purpose221 and whether reliance on the rule would be abusive in the circumstances.222   Halifax (n 6 above), para [96].   Weald Leasing (n 25 above), paras [51, 52]. 218   Halifax (n 6 above), para [97]. 219  With the exception of cases of ‘abuse of procedure’ (see at nn 74–76 above) and abusive reliance on EU rules on jurisdiction in civil matters (see at n 420 below). 220  AG Tesauro in Kefalas (n 34 above), paras [12, 25–27]; AG Stix-Hackl in Fidium Finanz (n 29 above), paras [95, 102]; AG Mazák in RBS Deutschland (n 27 above), para [73]. 221   Emsland-Stärke (n 7 above), para [54]: ‘It is for the national court to establish the existence of those two elements’; AG Poiares Maduro in Halifax (n 6 above), para [95]: ‘the responsibility of the national courts to establish whether recognition of the right . . . is compatible with the purposes and objectives pursued by the relevant provisions’; AG Stix-Hackl in Fidium Finanz (n 29 above), para [99]. 222   General Milk Products (n 72 above), para [21]: ‘The bona fide nature of those transactions is a question of fact to be decided by the national court’; Emsland-Stärke (n 7 above), para [54] (see n 221 above); AG Poiares Maduro in Halifax (n 6 above), para [96]: ‘it will also be for the national courts to determine whether, in the cases before them, the economic activities . . . are directed towards anything other than the creation of a tax advantage’; Halifax (n 6 above), para [76]; AG Stix-Hackl in Fidium Finanz (n 29 above), para [101]; Part Service (n 7 above), para [54]: ‘It is for the national court to assess if, the contractual structure of the transaction notwithstanding, the evidence before the court discloses the characteristics of a single transaction’, cf ibid, para [63]. 216 217

544  Stefan Vogenauer In doing so the national court ‘may find it necessary to extend its analysis by seeking evidence of indications of the existence of an abusive practice’.223 As the Court made clear in Emsland-Stärke, such proof must always be adduced in accordance with the rules of evidence of national law.224 It thereby declined to follow the Commission which had contended that the test of abuse ought to comprise ‘a procedural law element relating to the burden of proof’.225 The Court rather upheld the procedural autonomy of the Member States, albeit with the usual proviso requiring the national courts to ensure an effective protection of EU rights.226 Since the rules of evidence vary considerably between Member States it may be expected that at least slightly different standards of evidence with regard to abusive practices will apply throughout the Union. Of course most, if not all, jurisdictions will agree that the burden of proof is on the party alleging abusive reliance, ie the public authority.227 However, some laws of procedure may concede that the burden is reversed by prima facie evidence of an extremely serious instance of abuse,228 whilst other procedural regimes might reject such a relaxation. In sum, it is always for ‘the national court to verify in accordance with the rules of evidence of national law . . . whether action constituting such an abusive practice has taken place in the case before it’.229 As a result, the Court cannot make a finding of abuse of law itself. However, as usual, the ‘Court, when giving a preliminary ruling, may, where appropriate, provide clarification designed to give the national court guidance’.230 And, as usual, such guidance may be so explicit that it does not leave the national court with a meaningful choice of outcomes.231 In some cases of rule avoidance the Court made it clear that it did not believe that it would be possible to find abuse on the facts.232 In others, such as TV 10, it was obvious that it regarded the activities in question as abusive.233 A similar observation can be made with regard to the cases of rule appropriation.234 The strength of the Court’s ‘guidance’ can be influenced by the referring courts. If they make it clear, for example, that they do not regard the respective transactions as artificial the Court is willing to find that there is no abuse of law.235 G. Limits Although the prohibition of abuse of law has become an established doctrine of EU law it is confined within narrow boundaries. Decisions where the Court actually indicates   Part Service (n 7 above), para [55].   Emsland-Stärke (n 7 above), para [54]; cf Eichsfelder Schlachtbetrieb (n 109 above), paras [40, 42]. 225   Emsland-Stärke (n 7 above), para [39]. 226   Emsland-Stärke (n 7 above), para [54]: ‘provided that the effectiveness of Community law is not thereby undermined’; cf Eichsfelder Schlachtbetrieb (n 109 above), para [40]. 227   Cf AG Alber in Emsland-Stärke (n 7 above), para [83]. 228  The issue was raised by the Commission in its observations in Emsland-Stärke (n 7 above): see para [39] of the judgment; cf AG Alber in Emsland-Stärke (n 7 above), para [83]. The Court had rejected such a reversal of proof based on national law in Case C-206/94 Brennet AG v Vittorio Paletta [1996] ECR I-2356 [26] (a case of fraud and therefore not discussed in this chapter: see below, Part I.H.4); cf Ziegler (n 12 above) 309-10. 229   Eichsfelder Schlachtbetrieb (n 109 above), para [40]; Halifax (n 6 above), para [76]. 230   Halifax (n 6 above), para [77]; Part Service (n 7 above), para [56]. 231  For an overview of the guidance given by the Court in the relevant cases, see Baudenbacher (n 48 above) 218. 232   Pafitis (n 112 above). 233   TV10 (n 53 above). 234   Cremer (n 73 above). 235   RBS Deutschland (n 27 above), paras [24, 51, 52]. 223 224

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  545 that it considers the conduct of one of the parties to be abusive are comparatively rare. Instead, the Court frequently reiterates that, as a general rule, abusive practices will not be tolerated before going on to say that it will be more or less impossible for the national courts to establish abuse on the facts of the case at hand.236 There are three reasons for this narrow conception of abuse.237 First, the prohibition must be applied in a way that does not infringe general principles of EU law, notably the principles of legal certainty, separation of powers and uniform application (and/or effectiveness) of EU law. Secondly, the application of the prohibition must be balanced against the promotion of market integration, one of the key goals of the Union. Thirdly, there is no room for the prohibition if more specific mechanisms for the combat of abusive mechanisms apply. 1.  General Principles of EU Law a)  Legal Certainty  As mentioned above, any finding of abuse requires the court to make a value judgment, and it has to reach this on a case-by-case basis. The outcome of such value judgments is not always predictable and might vary from one court to another. Any abuse-based reasoning can therefore be expected to diminish legal certainty and to increase judicial discretion.238 National doctrines of abuse are frequently criticised for this reason. Their salient features are, as famously opined by one common lawyer, ‘vacillation and nebulosity’. They ‘may get out of hand and result in serious inroads on individual rights, thus becoming an instrument of dangerous potency in the hands of the demagogue and the revolutionary’.239 Some of the common lawyers contributing to this volume have voiced similar concerns, albeit couched in less drastic language.240 The Court of Justice has contributed to the allaying of such fears by developing a more refined test of abuse with a set number of requirements and legal consequences. Nevertheless, the criteria employed, such as ‘artificiality’ or ‘normal commercial transaction’, are necessarily flexible and thus inherently vague. Thus Advocate-General La Pergola in Centros advised against a finding of abuse, inter alia, because any requirement that a company should ‘effectively carry on business’ was ‘difficult to apply owing to its indeterminate nature. What kind of business must the parent company conduct, for how long and on what scale . . .?’241 The issue is particularly pressing in tax law. As the Court recognised in Halifax, the ‘requirement of legal certainty must be observed all the more strictly in the case of rules liable to entail financial consequences, in order that those concerned may know precisely the extent of the obligations which they impose on them’.242 The Court was able to draw on a nuanced discussion of the issue by Advocate-General Poiares Maduro. He had acknowledged the ‘legitimate concerns about legal certainty in tax law’, whilst insisting that the principles of legal certainty and protection of legitimate expectations 236   See, eg, Kefalas (n 34 above), Centros (n 5 above). Cf AG Poiares Maduro in Halifax (n 6 above), para [65]; Schön (n 21 above) 1281; Ringe (n 122 above) 113–14. 237  For a concise overview of the issues involved, see Snell (n 191 above) 224. 238   Cf Kjellgren (n 21 above) 245. 239  Gutteridge (n 170 above) 35, 43–44, quoted with approval by O’Sullivan (n 193 above) 72. 240  Arnull (n 19 above) 22-23; S Whittaker, ‘Comments on ‘Abuse of Law’ in European Private Law’, ch 17 above. 241  AG La Pergola in Centros (n 5 above), para [17]. For a similar type of argument, see AG Geelhoed in Akrich (n 26 above), para [173] (cited at n 163 above). 242   Halifax (n 6 above), para [72].

546  Stefan Vogenauer have to be ‘balanced against other values of the legal system’, such as the combating of ‘opportunistic behaviour’ promoted by the prohibition of abuse of law.243 The result of this balancing exercise found its way into his suggestion for framing the requirements of the prohibition of abuse in the area of tax:244 the prohibition could only apply when the relevant economic activity carried out has no other objective explanation than to create [a] claim against the tax authorities and recognition of the right would conflict with the purposes and results envisaged by the relevant provisions of the common system of VAT. Economic activity of that kind, even if not unlawful, deserves no protection from the Community law principles of legal certainty and protection of legitimate expectations because its only likely purpose is that of subverting the aims of the legal system itself.245

The Court did not discuss the issue in detail but agreed with the Advocate-General’s advice to apply the prohibition of abuse to the sphere of VAT. In doing so it explicitly referred to the relevant part of his Opinion.246 The argument may be summarised as follows: the formulation of the requirements of the prohibition must strike a balance between the legal values of legal certainty and substantive justice. It is impossible to altogether eliminate the inherent uncertainty surrounding the notion of abuse but the dangers to legal certainty may be reduced by developing a test that provides the courts with guidance and that is also strict enough to make it difficult for them to easily make a finding of abuse. The principle of legal certainty therefore imposes limits on the prohibition of abuse of law: as its application involves a deviation from the letter of the law it should be applied ‘only in exceptional cases where the abuse is evident’, and this is particularly so in tax law.247 b) Separation of Powers and Rule of Law  The prohibition of abuse of law may potentially conflict with the separation of powers. Arguably, the proper remedy for abuses is not a judicial doctrine allowing the courts to disapply a rule in particular circumstances. It is rather for the legislature to make the necessary policy choices, clarify the law and expressly set out the types of conduct that are acceptable and those that ought to be regarded as abusive and therefore banned.248 This is particularly so because the application of the prohibition typically works to the detriment of the individual and benefits the national administrative authority seeking to prevent the individual from asserting a right granted under EU law. The prohibition runs therefore, as Professor Poiares Maduro reminds us in his foreword to this book, counter to the ‘presumption of freedom’: in a

243  AG Poiares Maduro in Halifax (n 6 above), para [77]: ‘tax law is frequently dominated by legitimate concerns about legal certainty, deriving, in particular, from the need to guarantee the predictability of the financial burden imposed on taxpayers . . . Legal certainty must be balanced against other values of the legal system. Tax law should not become a sort of legal “wild-west’ in which virtually every sort of opportunistic behaviour has to be tolerated’. Cf para [84]: ‘prohibition of abuse has to be balanced against the principles of legal certainty and protection of legitimate expectations that also “form part of the Community legal system” . . . From those principles it follows that taxpayers must be entitled to know in advance what their tax position will be and, for that purpose to rely on the plain meaning of the words of the VAT legislation’. 244  AG Poiares Maduro in Halifax (n 6 above), para [82]. 245  AG Poiares Maduro in Halifax (n 6 above), para [86]. Note the connection of this passage to the proposed test of ‘sole purpose’, discussed at n 181 above. 246   Halifax (n 6 above), paras [70, 75]. 247  AG Mazák in Weald Leasing (n 25 above), para [11]. 248  This type of argument is run by Eidenmüller (n 22 above) 154, Whittaker (n 240 above) 259–60 and Ziegler (n 12 above) 297, 314. See also Metzger (n 11 above) 251.

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  547 legal system based on liberal rule of law values anything that is not expressly prohibited by the law is permitted.249 Such arguments from principle are easily countered by referring to the realities of the EU legislative process. The view from inside the Commission, as presented by Jonathan Faull in this volume, clearly shows the institutional and practical constraints within which the legislative organs of the Union operate.250 Other contributors have made similar points.251 It is simply not realistic to expect the EU legislator to engage in the kind of legislative micro-management that would be required to formulate rules with the degree of precision necessary to pre-empt all abusive practices. In a decision from the late 1970s, the Court indicated that it remained open to the EU legislature ‘to remove the causes of any abuses of the law’ by arranging for further harmonisation of the area of law concerned.252 Professor Weatherill comments that such a belief in legislative solutions is typical of a period when ‘harmonisation was seen as a good deal more politically feasible and economically desirable than it is today’. However, we ‘know today that a great many obstacles to inter-State trade caused by diversity between national laws that might have been expected to be tackled by legislative harmonisation have instead been left in the hand of the judiciary’.253 The rule of law argument has also been considered by the Court. The attempt to rely on the ‘principle of lawfulness’ by the exporter in Emsland-Stärke was plainly rejected: the exporter’s obligation to repay the refund obtained was ‘not a penalty for which a clear and unambiguous legal basis would be necessary, but simply the consequence of a finding that the conditions required to obtain the advantage derived from the Community rules were created artificially’.254 Advocate-General Alber was even more robust: ‘the presumption here is that the abuse of rights as such does not merit protection. The abuse of rights causes the protection guaranteed by the legal system to a trader acting in good faith to be forfeited’.255 In Halifax, the Court reiterated that ‘a finding of abusive practice must not lead to a penalty’256 and, as has been seen above, implicitly subscribed to the Advocate-General’s view that the principle of legitimate expectations does not protect taxpayers who engage in opportunistic behaviour.257 In sum, the Court correctly assumes that it is possible to apply the prohibition of abuse of law without violating the separation of powers and the rule of law. However, the need to uphold these principles as much as possible is a powerful argument against too broad a conception of the prohibition. c)  Uniform Application and Effectiveness of EU Law  A finding of abuse by a national court will lead to the non-application of a rule of EU law in a given case. This has, to all intents and purposes, the effect of rendering the rule inoperative in that particular 249   M Poiares Maduro, Foreword, p vii above. Ziegler (n 12 above) 296, 312 also draws attention to the fact that the prohibition operates in a ‘vertical’, rather than ‘horizontal’ dimension; see also Arnull (n 19 above) 20–21. 250   J Faull, ‘“Prohibition of Abuse of Law”: A New General Principle of EU Law’, ch 20 above, 291–92. 251   Snell (n 191 above) 220–23; Schammo (n 24 above) 374–75. 252   Knoors (n 33 above), para [27]. 253  Weatherill (n 35 above) 56. 254   Emsland-Stärke (n 7 above), para [56]. 255  AG Alber in Emsland-Stärke (n 7 above), para [80]. However, he strongly relied on the fact that the behaviour must have been intentional, thus subscribing to the subjective approach that has now been superseded, see above, Part I.D.3.b). 256   Halifax (n 6 above), para [93]. 257   See above (at nn 242–46).

548  Stefan Vogenauer instance. From the perspective of the EU legal order, this is problematic. It is one of the constitutional fundamentals of the Union that national courts do not have the power to declare measures of the Union institutions invalid. Otherwise divergences between courts in the Member States as to the validity of such measures might arise and endanger the uniform application of EU law.258 If a national court applies the prohibition of abuse of law there is a similar concern: a rule of EU law might be applied in the courts of one Member State and not in the others. Invoking the prohibition might therefore ‘undermine the full effect and uniform application of Community law’. Ultimately, this would ‘go against the fundamental principle that Community law has primacy over national law’.259 The Court therefore has to strike a cautious balance.260 On the one hand, the principle of procedural autonomy and the division of functions under Article 267 TFEU oblige it to leave to the national courts the task of establishing abuses.261 On the other hand, invoking the prohibition of abuse of law ‘must not detract from the full effect and uniform application of Community law in the Member States’.262 In other words, the prohibition must be applied in a way that ensures that ‘the effectiveness of Community law is not thereby undermined’.263 In its Kefalas judgment, the Court devised the following formula in order to meet concerns about the uniform application and the effectiveness of Community law: it is not open to national courts, when assessing the exercise of a right arising from a provision of Community law, to alter the scope of that provision or to compromise the objectives pursued by it.264

The reference to the ‘objectives’ of the rule in question simply reinforces the second limb of the abuse test, the decision on which is a matter for the Court of Justice and not for the national courts: a finding of abuse is impossible if the application of the rule to the facts is not ‘contrary to the purpose’ of the rule.265 Declaring it impermissible to ‘alter the scope’ of the EU law rule means that a finding of abuse may not have the result to make the rule inapplicable across the board, ie in all cases to which it potentially applies, rather than in specific circumstances only. Such a finding must not ‘have the effect of utterly negating the provision concerned’ or to ‘nullify the provision in question entirely’. The rule of EU law must not be ‘paralysed’ or ‘rendered inoperative’ in its entirety.266 A broad-brush finding of abuse for an entire group of cases would be ‘tantamount to censuring not the abusive exercise of the right but rather any exercise of it’.267 This cannot be tolerated because it would be ‘tantamount to altering the scope’ of the respective   Case 314/85 Firma Foto-Frost v Hauptzollamt Lübeck-Ost [1987] ECR 4199, para [15].  AG Tesauro in Kefalas (n 34 above), para [15]. Cf ibid, para [17]; AG Tesauro in Pafitis (n 112 above), para [28]; D Triantafyllou, ‘Abuse of rights versus primacy?’ (1999) 36 Common Market Law Review 157; Engsig Sørensen (n 24 above) 424. In Kefalas (like in Pafitis and Diamantis) the situation was further complicated by the fact that the referring court sought to rely on a national anti-abuse provision (see below, Part I.H.6). However, the constitutional principles discussed in the text above are relevant for all cases where the prohibition of abuse is potentially applicable: cf AG Tizzano in Chen (n 80 above), para [112]. 260  AG Poiares Maduro in Halifax (n 6 above), para [65]. 261   See above, Part I.F. 262   Pafitis (n 112 above), para [68]. Cf Kefalas (n 34 above), para [22]; Diamantis (n 34 above), para [34]. 263   Emsland-Stärke (n 7 above), para [54]; Eichsfelder Schlachtbetrieb (n 109 above), para [40]; Halifax (n 6 above), para [76]. 264   Kefalas (n 34 above), para [22]. 265   See above (at n 109). See also Koutrakos (n 48 above) 210–11. 266  AG Tesauro in Pafitis (n 112 above), para [34]; AG Tesauro in Kefalas (n 34 above), paras [12, 15]. 267  AG Tesauro in Pafitis (n 112 above), para [31] (emphasis in the original). 258 259

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  549 EU law provision268. This is particularly so in cases of rule avoidance where the person concerned relies on a fundamental freedom in order to escape a provision of his domestic law.269 The national court may not apply such a domestic law if it amounts to a ‘blanket ban’ on the exercise of the freedom, ie if it requires an ‘abstract, general assessment’ of the situation, rather than ‘the existence of an actual basis for concluding there to have been abuse in the individual case’.270 The situation is different if the conduct of the person relying on a provision of EU law leads to the non-application of the rule in a particular instance, whilst still leaving room for the rule to apply in other cases. This would not be ‘tantamount to altering the scope’ of the rule.271 The rule is therefore not ‘paralysed’ in its entirety, and the full effect and uniform application of EU law are not undermined. For this reason the court insists that findings of abuse must always be made on a case-by-case-basis.272 However, even if the national courts make such an assessment of the individual circumstances, uniformity of application throughout the EU will never be fully achieved. Given the inherent vagueness of the abuse test, it is to be expected that national courts reach different outcomes in broadly similar cases.273 The existence of such divergences is an unavoidable and well-known phenomenon of EU law that is bound to occur whenever vague and ambiguous standards such as ‘proportionality’ or ‘good faith’ are used.274 It would be more worrying if certain courts were generally too keen, or at least significantly keener than those of other Member States, to establish findings of abuse and thus to disapply EU law rules. However, as Professor Koutrakos shows, the risk that national courts might ‘abuse the abuse of law test’ does not seem to have materialised.275 2.  Market Integration Keeping the prohibition of abuse of law within narrowly confined boundaries is particularly important in cases of rule avoidance where persons ‘escape’ from the application of the more stringent regulatory regime of their home State by exercising one of the fundamental freedoms.276 They thereby choose a more favourable regime, a process frequently referred to as ‘regulatory arbitrage’.277 From the perspective of EU law, this is arguably a good thing. The freedoms are meant to promote the Union’s key goal of market integration and thus to encourage the movement of persons, goods, services and capital between jurisdictions. As a general rule, it is clearly legitimate and even desirable that citizens and companies should exploit the differences between regulatory regimes for their own (and, ultimately, for the overall) benefit. As a matter of principle, such choices ought to be protected, rather than being labelled ‘abusive’ and prohibited: ‘Community law can have no complaint with such mobility; rather it is precisely the objective of   Pafitis (n 112 above), para [70].   See above, Part I.C.1. 270  AG Alber in Inspire Art (n 66 above), para [117] (emphasis in the original). Cf ibid, paras [119–120]. 271   See, eg, Diamantis (n 34 above), para [43]. 272   See above (at n 143). 273   See above (at n 241). 274  S Vogenauer, ‘Eine gemeineuropäische Methodenlehre des Rechts – Plädoyer und Programm’ [2005] Zeitschrift für Europäisches Privatrecht 234. Cf Koutrakos (n 48 above) 215–16. 275   Koutrakos (n 48 above) 216. 276   See above, Part I.C.1. 277   Schammo (n 24 above) 353–54. 268 269

550  Stefan Vogenauer Community law to promote mobility’.278 Any application of a prohibition of abuse is therefore raising an obstacle to market integration and has the capacity to undermine the internal market.279 However, as in all matters of legal integration, a balance must be struck between the interests of the EU and the sovereignty of the Member States. The Court has therefore consistently acknowledged the ‘legitimate interest which a Member State may have in preventing certain of its nationals, by means of facilities created under the Treaty, from attempting to evade the application of their national legislation’.280 In other words, EU law ‘allows – encourages – regulatory arbitrage but it does not promote it unconditionally’.281 The prohibition of abuse of law is a mechanism for resolving the tension between market integration and national sovereignty by spelling out the conditions in which regulatory arbitrage is and is not permitted. In the light of the overriding goal of market integration such a mechanism must have freedom of choice as the rule and a restriction based on abusive behaviour must be the exception.282 Thus it is not abusive conduct as such to take advantage of the most attractive regime, thereby avoiding the application of the rules of another Member State.283 Persons ‘are in principle entitled to arrange their circumstances in such a way as to come within a given set of rules, in this case Community law, and to benefit therefrom’.284 The judgment of the Court in Centros is the paradigm example: choosing a company law regime ‘cannot, in itself, constitute an abuse of the right of establishment’ but is ‘inherent in the exercise, in a single market, of the freedom of establishment guaranteed by the Treaty’.285 In a similar vein, the Court has consistently emphasised that ‘taxpayers may choose to structure their business so as to limit their tax liability’.286 The mere fact that certain transactions have been structured ‘with the effect that a tax advantage accrues, cannot in itself lead to a finding that the essential aim of the . . . transactions at issue is merely to obtain a tax advantage’.287 It is only when the objectives of the relevant provisions of EU law are frustrated and the conditions for obtaining a benefit from these provisions have been artificially created that legitimate arbitrage turns into illegitimate abuse of law. Since the prohibition of  AG Geelhoed in Akrich (n 26 above), para [179].  Arnull (n 19 above) 22. Cf D Triantafyllou, ‘L’interdiction des abus de droit en tant que principe general du droit communautaire’ (2002) 38 Cahiers de droit européen 611, 613: ‘conflit de base entre l’intégration européenne et l’“abus de droit”’; Schammo (n 24 above) 361–62: abuse of rights doctrines are ‘a potential threat to legal integration. . . . in the hands of national Courts, they potentially risk undermining integration by prompting national judges to prohibit reliance on Community law provisions’; Schön (n 47 above) 82–83; P Schammo, ‘Comments on Abuse of Rights in EU Law’, ch 13 above, 194. 280   Knoors (n 33 above), para [25]. Restated frequently, inter alia in Case C–61/89 Bouchoucha [1990] ECR I–3551, para [14] and Surinder Singh (n 34 above), para [24]. 281  Weatherill (n 35 above) 60. Cf Tridimas (n 48 above) 176. 282   Cf AG Jacobs in Joined Cases C–34/95 to C–36/95 Konsumentombudsmannen (KO) v De Agostini (Svenska) Förlag AB (C-34/95) and TV-Shop i Sverige AB [1997] ECR I–3843, para [45], citing AG Lenz in Case C–11/95 Commission v Belgium [1996] ECR I–4115 (Belgian Cable Network), para [75]: the principle of abuse of rights ‘should not be read too widely: . . . as an exception to one of the freedoms to the internal market, the scope for intervention which that principle confers on the receiving Member State must be narrowly interpreted’. 283  Engsig Sørensen (n 24 above) 444. 284  Observations of the Greek government, referred to by AG Geelhoed in Akrich (n 26 above), para [171]. 285   Centros (n 5 above), para [27] (cited at n 122 above) (emphasis added). 286   Halifax (n 6 above), para [73]; Part Service (n 7 above), para [47]; Weald Leasing (n 25 above), para [27]; AG Mazák in RBS Deutschland (n 27 above), para [77]. 287  AG Mazák in RBS Deutschland (n 27 above), para [78] (emphasis added). 278 279

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  551 abuse is an exception to the general rule that regulatory arbitrage is allowed, it must be particularly narrowly construed in cases of rule avoidance. This may explain why findings of abuse are less frequent in such instances than in cases of rule appropriation.288 3.  Specific Prohibitions of Abuse of Law in EU Law Finally, there is no room for recourse to the general prohibition of abuse of law if a more specific solution for preventing abuses is available. This may be the case if the general principle ‘has been expressed in a concrete manner’ in a provision of secondary EU law.289 The contribution of Dr Doukas shows how the implementation of the TWF Directive which deals with a number of potentially abusive situations has reduced the need to have recourse to the general prohibition.290 Invoking the general prohibition – which is by definition less clear and precise – in such instances might contravene the harmonisation objective of the provision. In the event that a specific anti-abuse provision is contained in a directive that has not been transposed as yet, the general prohibition cannot be invoked either. Doing so would undermine the rule that unimplemented directives may not be applied to the detriment of individuals.291 In the absence of a specific anti-abuse provision in secondary EU law recourse to the general principle may still be barred if it is possible to prevent the abusive conduct by employing another legal mechanism, such as proportionality or the public interest defence. As Professor Weatherill shows, it may well be that there is no need, and thus there is no room, for an independent doctrine of abuse in the area of free movement of goods and services because the problems flowing from abusive practices can be adequately tackled within the existing framework developed for the protection of these freedoms: the Treaty requires States to show sufficient reasons to justify the application of their own rules that restrict the freedom of movement. These rules might offer a sufficiently ‘subtle, flexible . . . instrument for advancing market integration while seeking to take account of legitimate national interests in regulatory protection’.292 H.  Doctrinal Foundations The requirements, legal consequences, procedural aspects and limits of the prohibition of abuse of law have been gradually clarified by the Court and legal scholars in the course of the past decade. From the early, rather intuitive invocations of ‘abuse’ in the 1970s a surprisingly well defined doctrine has emerged. However, there is still considerable uncertainty with regard to some important structural and conceptual foundations of the doctrine: what is the actual mechanism that leads to the non-application of the rule in   See below (at n 422).  AG Kokott in Kofoed (n 1 above), para [67]. For examples, see nn 8–14 above. See also the ‘in-built defence against regulatory arbitrage’ identified by Schammo (n 279 above) 195 in Article 5(4) MIFID. 290   Doukas (n 10 above) 72–86. For similar arguments, see Tridimas (n 48 above) 186, 191 (with regard to MIFID) and Cuniberti (n 11 above) 286 (on the revision of Art 6(1) of the Brussels I Regulation). 291  AG Kokott in Kofoed (n 1 above), para [67], noting that the status of this rule is in doubt after Case C–144/04 Werner Mangold v Rüdiger Helm [2005] ECR I–9981 (see the discussion by Arnull (n 19 above) 13–18). Cf K Engsig Sørensen, ‘What is a General Principle of EU Law? A Response’, ch 3 above, 29–32. 292  Weatherill (n 35 above) 49, 50–51, 58, 60–61. For a similar point in the area of free movement of workers, see Ziegler (n 12 above) 306, 307, 313–14. 288 289

552  Stefan Vogenauer question? Does the prohibition concern the abuse of ‘law’ or the abuse of ‘rights’? To what extent, if any does, the application of the prohibition differ from the interpretation of EU law? How does the prohibition relate to other doctrines, such as fraud or proportionality? And how is it connected to national anti-abuse doctrines? 1.  Pre-emption or Justification? As has been shown above, the legal consequence of a finding of abuse is the nonapplication of the EU law rule in question.293 However, the doctrinal justification for this result is not clear. The Court employs two distinct techniques, sometimes alternatively and sometimes cumulatively, and the reasons for these differences in approach are by no means obvious.294 The first technique, sometimes referred to as ‘pre-emption’,295 is to use the prohibition of abuse in order to determine the scope of application of the rule. The scope of the rule is a priori limited to non-abusive practices. Abusive practices are simply not covered: the rule does not apply to them at all; the application of the rule is pre-empted. The Court has employed this approach since its early case law. In Lair, for example, it simply held that abuses ‘are not covered by the Community provisions in question’,296 a formulation that it repeated in the 2003 Ninni-Orasche case.297 A similar understanding is suggested by the cases on export refunds and VAT that use the – arguably infelicitous – formula according to which EU law rules will not be ‘extended to cover abusive practices’.298 The second technique is to frame the prevention of abuse as a justification of restrictions of fundamental freedoms granted by the Treaty. Here, the abusive conduct is regarded as falling within the scope of application of the relevant provision of the Treaty. The prohibition of abuse then operates as an a posteriori limitation of a rule that prima facie applies. This approach can be traced back to the Centros decision where Advocate-General La Pergola relied on the established, ‘pre-emptive’ notion of abuse. Referring to Marcel Planiol’s classic statement that ‘the law ends where abuse begins’299 he claimed that ‘the problem of abuse is resolved in the last analysis by defining . . . the scope of the right conferred on the individual concerned’.300 Initially, the Court followed this approach. As has been seen above, it rejected a finding of abuse by holding that the application of Articles 52 and 58 of the EC Treaty [now 49 and 54 TFEU] to the facts of the case would not run counter to the objectives of these provisions.301 It did so when assessing whether the Treaty rights were engaged in the first instance. However, the Court also asserted that the ‘question of the application of those articles of the Treaty is different from the question whether or not a Member State may adopt measures in order to prevent attempts by certain of its   See above, Part I.E.  For an extensive discussion, see Edwards and Farmer (n 117 above) 208–15; J Ghosh, ‘Cadbury Schweppes: Breach, Abuse Justification and Why They Are Different’, ch 31 above. See also Schammo (n 24 above) 363–66; P Farmer, ‘Prohibition of Abuse of (European) Law: The Creation of a New General Principle of EC Law through Tax: A Response’, ch 1 above, 4–6. 295  Edwards and Farmer (n 117 above) 208. 296   Lair (n 79 above), para [43]. 297   Ninni-Orasche (n 142 above), para [36]. 298   Cremer (n 73 above), para [21] and the other cases cited in n 205 above. 299   M Planiol, Traité élémentaire de droit civil, Vol II (Paris, F Pichon 1900) no 87: ‘le droit cesse là ou l’abus commence’. 300  AG La Pergola in Centros (n 5 above), para [20]. 301   Centros (n 5 above), paras [24–27]. See above (at nn 119–24). 293 294

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  553 nationals to evade domestic legislation by having recourse to the possibilities offered by the Treaty’.302 It then returned to the question of evasion of national legislation when it considered possible justifications for the legislation authorising the refusal to register the company in Denmark. The Court acknowledged that Member States were not precluded from adopting measures for ‘preventing or penalising fraud . . . where it has been established that [the members of the company] are in fact attempting . . . to evade their obligations towards private or public creditors’.303 In effect it thereby ruled that the need to combat abuse is a potential justification for national restrictions of the freedom of establishment, although it did not regard the relevant Danish legislation as justified on these grounds and it did not use the language of ‘abuse’ at the stage of justification.304 In the subsequent decisions on corporate establishment in the areas of tax and company law the Court increasingly analysed abuse both in terms of pre-emption and justification305 or exclusively in terms of justification.306 In doing so it gradually merged the traditional abuse criterion of ‘artificiality’ derived from Emsland-Stärke with the socalled ‘wholly artificial arrangements doctrine’ from tax law. According to this doctrine, tax avoidance rules of the Member States restricting the freedom of movement are only justified if they apply exclusively to wholly artificial arrangements.307 Such arrangements are now regarded as ‘abusive’ for the purposes of the prohibition of abuse.308 It is difficult to say whether the gradual shift away from pre-emption and towards justification is a conscious choice or simply the result of muddled reasoning on the part of the Court. If it is the former three possible explanations exist. First, the development might confirm the point made in the previous section, ie that there is no room for the prohibition of abuse where other mechanisms to combat abusive practices are already in place.309 Such a mechanism, the wholly artificial arrangements doctrine, exists for tax avoidance by way of corporate establishment, so the Court can rely on this doctrine rather than on the general prohibition of abuse. Secondly, given that both techniques are sometimes used cumulatively, the development might be an indication that the prohibition of abuse of law has become an all-pervasive principle of law: it is not only relevant at the stage of defining the scope of application of a rule but also at the stage of justifying restrictions.310 Thirdly (and conversely), the shift might be an attempt to reduce the number of findings of abuse by developing a stricter test. In order to justify a national measure limiting the exercise of a freedom the prevention of abusive conduct will have to be proportionate and imperatively required in the general interest.311 This is a significantly more stringent requirement than the traditional approach that generally   Centros (n 5 above), para [18].   Centros (n 5 above), para [38]. 304  Edwards and Farmer (n 117 above) 211. 305   Cadbury Schweppes (n 7 above), paras [34–38] on the one hand and paras [48–68] on the other. 306   Inspire Art (n 66 above), paras [136–39]; Thin Cap (n 37 above), paras [71–80]. Cf Edwards and Farmer (n 117 above) 214. 307   Case 264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) [1998] ECR I–4695, para [26]. Cf LV Faulhaber, ‘Sovereignty, Integration and Tax Avoidance in the European Union: Striking the Proper Balance’ (2010) 48 Columbia Journal of Transnational Law 177, 204–05. 308   Cadbury Schweppes (n 7 above), para [55]; Thin Cap (n 37 above), para [74]. Cf above (at n 153) and de la Feria (n 4 above) 427–29. 309   See above, Part I.G.3. 310   Cf de la Feria (n 4 above) 438. 311   Centros (n 5 above), para [34]. Cf Edwards and Farmer (n 117 above) 215; Ziegler (n 12 above) 295–96. 302 303

554  Stefan Vogenauer presumes the ‘legitimate interest’ of Member States to prevent abuse.312 The activities of the Danish couple and their company in Centros certainly did not reach this threshold.313 Only the first of these explanations would provide a fully conclusive reason for the Court’s recent shift in approach. The third explanation is to be rejected unless it is intended to render the prohibition of abuse more or less inoperative as a result of an absurdly restrictive test. Also, the second explanation is only partly convincing once it is acknowledged that the test is more stringent at the level of justification than at the level of defining the scope of the rule. The Court would thus be well advised to clarify its doctrinal position and return to its original, ‘pre-emptive’ approach. This is particularly so because the justificatory approach cannot be employed in cases of ‘rule appropriation’ where there is no evasion of a domestic rule that would require justification.314 If the Court maintains its current two-track approach it will be impossible to maintain a coherent doctrine of prohibition of abuse that covers cases of both rule avoidance and rule appropriation.315 2.  Abuse of Law or Abuse of Rights? Further uncertainty as to the doctrinal foundations of the prohibition of abuse of law arises from the terminological inconsistencies mentioned at the outset of this chapter.316 These are particularly pronounced with regard to the object of the prohibition. What precisely is it that is abused? The Court and legal writers refer to the abuse of ‘rights’, ‘freedoms’,317 ‘provisions of EU law’ or ‘law’ in general, and they appear to treat these as interchangeable concepts. The variety of terminology is somewhat less pronounced in the Romance languages and in German where the notions of droit, diritto or Recht refer to both the ‘law’ in general (droit objectif, Recht im objektiven Sinn) and an individual’s ‘right’ against another private person or entity, ie a ‘claim right’ or ‘right to something’ (droit subjectif, subjektives Recht). Nevertheless, continental lawyers generally think in terms of an ‘abuse of rights’ if they employ the concepts of abus de droit and Rechtsmissbrauch. Unfortunately, in doing so lawyers from different civilian traditions do not refer to similar definitions, requirements, areas of application and legal consequences. There are, as has been said in a Europewide comparative study, ‘almost as many conceptions of the abuse of rights as there are member states of the Council of Europe’, and the diversity is further enhanced by the fact that, ‘within most countries, there is no unanimous agreement as to the scope of the prohibition of abuse of rights; doctrinal disputes and contradictory judgments are commonplace’.318 Blanket assertions that ‘a common concept of abuse of rights exists in the traditions of the Member States’319 lack the eye for nuances and detail. Such a

  Knoors (n 33 above), para [25] (n 280 above).   Centros (n 5 above), paras [35–38].   Cf above, Part I.C.2. 315  The reasons for having such a coherent doctrine are set out above, Part I.C.3. 316   See above, Part I.B. 317   See, eg, Cadbury Schweppes (n 7 above), para [34]: ‘abuse of freedom of establishment’. 318   J Voyame, B Cottier and B Rocha, ‘Abuse of Rights in Comparative Law’ in Council of Europe, Abuse of Rights and Equivalent Concepts: The Principle and Its Present Day Application (Strasbourg, Council of Europe 1990) 23. Cf AG Tizzana in Chen (n 80 above), para [111]. 319   Lenaerts (n 177 above) 1121. 312 313 314

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  555 common core is not discernable, unless it is in the most general terms. The basic premise is that someone exercises his or her right (usually a property right, a contractual right or a procedural right) in such a way that it may be described as ‘excessive’, ‘unjustified’, ‘without legitimate interest’, ‘wrong’ or ‘against good faith’.320 Traditionally, the notion of abuse of rights also required an ‘intention to harm another’, the ‘purpose of harming another’ or the ‘exercise of the right contrary to its social purpose’.321 One of the legal consequences of an abuse of rights is that the holder of the right is barred from exercising it although he would normally be entitled to rely upon it. If he sues on the right the other party can raise the abusive conduct as a defence. Medieval and modern Roman lawyers called this the exceptio doli generalis. The notion of ‘abuse’ is even less consistently used by continental lawyers when they deal with cases where someone attempts to ‘circumvent’, ‘avoid’ or ‘evade’ the application of a legal rule. French lawyers use a doctrine that can be traced back to another Roman law concept, that of fraus legis.322 They speak of ‘fraud against the statute’ (fraude à la loi) if a person intentionally attempts to escape the application of a mandatory rule of law which would otherwise be applicable and he has no legitimate grounds for doing so.323 The doctrine of fraude à la loi which has close equivalents in Italy324 and Spain325 also applies if the person brings himself within the scope of a provision. The reason is that he has ‘fraudulently’ caused the application of the rule, ‘disregarding the finality of the provision, whilst at the same time scrupulously respecting its letter’.326 In similar cases, German lawyers tend to refer to their doctrine of ‘circumvention of a statute’ (Gesetzesumgehung). They distinguish between two types of circumvention. In cases of the first type, persons try to avoid the application of a provision which would have negative effects for them. This is frequently called ‘flight from the rule’ (Normenflucht). In the second group of cases they try to bring themselves within the scope of the rule by way of ‘rule appropriation’ (Normerschleichung).327 It is easy to see that the prohibition of abuse of law, as developed by the Court for cases of rule avoidance and rule appropriation,328 is much more akin to domestic doctrines such as fraude à la loi and Gesetzesumgehung than to the typical ‘abuse of rights’ scenario.329 Commentators familiar with French and German law have been quick 320  For an overview, see J Gordley, ‘The Abuse of Rights in the Civil Law Tradition’, ch 4 above, and F Ranieri, ‘Abuse of Law’ in J Basedow, K Hopt and R Zimmermann (eds), Max Planck Encyclopedia of European Private Law (Oxford, Oxford University Press forthcoming). For brief overviews on the diverging national approaches in the literature dealing with the EU doctrine of abuse of rights, see Brown (n 20 above) 513–15, M Byers, ‘Abuse of Rights: An Old Principle, A New Age’ (2002) 47 McGill Law Journal 389, 392–95; Lenaerts (n 177 above) 1125–27. Cf AG Tesauro in Kefalas (n 34 above), para [22]. 321   See above (at n 170). 322  R Zimmermann, The Law of Obligations: Roman Foundations of the Civilian Tradition (Oxford, Oxford University Press 1996) 702–05. 323  H Desbois, La notion de fraude à la loi et la jurisprudence française (Paris, Dalloz 1927) 11, 15, 21; J Ghestin, G Goubeaux, M Fabre-Magnan, Traité de droit civil: Introduction générale, 4th edn (Paris, LGDJ 1994) 823; Terré (n 38 above) no 499. Cf Schammo (n 279 above) 193–94; Whittaker (n 240 above) 258. 324  U Morello, Frode alla legge (Milano, Giuffré 1969). 325  E de la Iglesia Prados, El fraude de ley (Valencia, Tirant lo Blanch 2007). 326   Maublanc (n 2 above) 269: ‘appliqué frauduleusement une règle en méconnaissance de la finalité du dispositif, tout en respectant scrupuleusement la lettre du texte’. 327   See above (nn 57, 70). 328   See above, Part I.C. 329   See the reference to ‘fraudulent evasion’ by AG Tesauro in Kefalas (n 34 above), para [24] (cited at n 84 above).

556  Stefan Vogenauer to point out this parallel,330 and it is perhaps the main reason for Professor Gordley’s assertion that the continental tradition with regard to ‘abuse of rights’ hardly contributes to the understanding of the modern prohibition in EU law.331 In fact there are only three decisions of the Court that resemble the classic case where it is alleged that there is an abus de droit or Rechtsmissbrauch on the part of an individual in a ‘horizontal’332 relationship with another private person. All of these decisions dealt with preliminary references from Greek courts. They concerned broadly similar actions brought by existing or former shareholders of a company against new shareholders.333 In one of these cases, Kefalas, the claimants contested an increase in the capital of the company on the grounds that they had not been given the opportunity to decide on the increase at the company’s general meeting. The defendants argued that the claimants’ exercise of their shareholders’ rights was abusive: the increase in capital had ultimately worked to their economic benefit because it had led to the financial recovery of a company that would otherwise have been insolvent. However, as has been shown above,334 it is easy to reframe this case as an instance of potentially abusive avoidance of a rule of EU law: the right invoked by the shareholders was conferred by a directly applicable directive. The increase in capital without a decision of the general meeting had been ordered by a public authority which had based its decision on a Greek statute concerning the restructuring of undertakings in financial difficulties. As a result, the decision turned on the question of whether the claimants’ attempt to escape the application of the Greek statute by relying on a provision of EU law had to be categorised as an abuse of that provision. The answer was in the negative. The other cases of rule avoidance to which the Court applied the prohibition of abuse follow a similar pattern. They assess the activities of persons who attempt to evade national legislation by relying on an EU law provision, usually a fundamental freedom. Since these freedoms are rights conferred by the Treaty it is easy to see why the language of ‘abuse of rights’ is intuitively used. This language also fits the cases of rule appropriation which concern persons seeking to obtain an advantage from a rule of EU law. Again, these provisions of EU law grant rights (to monetary compensation or to a reduction in tax), so the abuse of a rule of EU law coincides with an abuse of a right conferred by EU law. There is therefore no need in the law of the European Union to draw a strict terminological distinction between an abuse of law and an abuse of rights. However, given that the prohibition of abuse is a doctrine of EU law, there is a need for an autonomous notion of abuse that will not be easily confused with existing national doctrines.335 This is particularly so since English law has traditionally been hostile to the notion of ‘abuse of rights’336 and has never felt it necessary to develop a freestanding doctrine of ‘evasion’ 330   Maublanc (n 2 above) 267–69; Schammo (n 24 above) 354, 355; Schammo (n 279 above) 193–94; Vrellis (n 48 above) 635; Schön (n 21 above) 1279; Fleischer (n 21 above) 870; Tridimas (n 48 above) 171; Cuniberti (n 11 above) 284. 331  Gordley (n 320 above) 42, 45–46. 332   Cf Ziegler (n 12 above) 296, 314. 333   Pafitis (n 112 above); Kefalas (n 34 above); Diamantis (n 34 above). 334   See above (at n 84). 335   See above (at n 42). 336   P Catala and TA Weir, ‘Delict and Torts: A Study in Parallel, Part II’ (1964) 38 Tulane Law Review 221, 237; M Taggart, Private Property and Abuse of Rights in Victorian England: the Story of Edward Pickles and the Bradford Water Supply (Oxford, Oxford University Press 2002). This does not exclude the possibility that English law frequently intervenes to censure or control actions by persons relying on their strict legal rights

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  557 or ‘circumvention of the law’.337 It is therefore advisable to avoid the concepts of ‘abuse of rights’ and ‘evasion of the law’ and to use the notion of ‘abuse of law’ or ‘abuse of Community law’ that was suggested by Advocate-General Poiares Maduro in his Opinion in Halifax.338 It has, as Professor Whittaker observes, ‘no real resonance to most English lawyers’.339 The same is true for lawyers of other Member States who encounter the concept in one of its official translations, such as interdiction de faire un usage abusif des dispositions du droit communautaire, divieto di abusi del diritto comunitario and Verbot des Missbrauchs des Gemeinschaftsrechts. These notions are certainly not catchy but they do signal the arrival of a new doctrine of EU law that genuinely differs from comparable doctrines at the national level. For the reasons just given, there is no room to criticise the absence of a distinction in EU law between ‘abus de droit/Rechtsmissbrauch’ on the one hand and ‘fraude à la loi/ Gesetzesumgehung’ on the other as ‘a loss of clarity of thought’.340 Moreover, EU law can hardly be expected to resolve an issue that has confounded national legal systems for centuries. The French and German doctrines are less clearly delineated than the short account given above suggests. Legal writers offer differing accounts, and the case law on both of the doctrines remains unsettled. It is well known that the respective doctrines are sometimes overlapping or lead to the same outcome.341 Moreover, their structural similarity has been noted: in both cases the purpose of the rule or right deviates from its literal meaning or formal content.342 The legal terminology used often adds to the lack of precision. Some German lawyers speak of ‘institutional abuse of rights’ (institutioneller Rechtsmissbrauch) when referring to circumventions.343 In France, the Act of Fiscal Procedure has a specific provision dealing with ‘abus de droit’ in tax matters which requires a ‘fictitious’ or ‘fraudulent’ transaction in the narrow and technical sense.344 The Conseil d’État extended this provision to transactions that were neither fictitious nor fraudulent but nevertheless undertaken with the sole purpose of avoiding taxation. Thus the doctrines of ‘abus de droit’ and ‘fraude à la loi’ have been merged and the latter has become a sub-category of the former in French tax law.345 German tax legislation also has a provision that amalgamates both doctrines by stipulating that tax legislation ‘cannot be circumvented by way of abuse of formally admissible legal arrangements’ (‘Missbrauch

to the unfair disadvantage of others (eg, tort of nuisance, abuse of procedure, the equity of redemption in relation to mortgages and the law against clogs on the equity of redemption, relief against forfeiture of leases, the concept of quasi-partnership for the purposes of winding up companies on equitable grounds in the law of unfair prejudice under section 994 of the Companies Act 2006 – I am grateful to Mr Justice Sales for suggesting some of these examples to me; for a further example from the law of international civil procedure, see A Briggs, ‘The Rejection of Abuse in International Civil Procedure’, ch 18 above, 261, 272). What matters for our purposes is the refusal of English law to generalise from these instances and to develop a free-standing doctrine or general principle of ‘abuse’. 337  RH Graveson, ‘The Doctrine of Evasion of the Law in England and America’ (1937) 19 Journal of Comparative Legislation and International Law (Series 3) 21; S Vogenauer, Die Auslegung von Gesetzen in England und auf dem Kontinent, Vol II (Tübingen, Mohr Siebeck 2001) 813–16; Schammo (n 24 above) 357. 338  Albeit for different reasons, cf AG Poiares Maduro in Halifax (n 6 above), para [71]. 339  Whittaker (n 240 above) 258. 340  As suggested by Fleischer (n 21 above) 870. Cf Schön (n 21 above) 1279. 341  Fleischer (n 21 above) 870; Schammo (n 24 above) 355; Vrellis (n 48 above) 633–34 n 2. 342  Ghestin et al (n 323 above) 722; Baudenbacher (n 48 above) 216. 343   Schön (n 21 above) 1279; Fleischer (n 21 above) 870, Englisch (n 47 above) 5. 344  Article L 64 Livre des procédures fiscales. 345  For references, see Vogenauer (n 337 above) 270-71; Harris (n 39 above) 134.

558  Stefan Vogenauer von rechtlichen Gestaltungsmöglichkeiten’).346 The quest for the exact meaning of this provision is one of the most discussed and trickiest issues of German tax law. 3.  Abuse of Law or Interpretation of EU Law? A further unresolved doctrinal question is whether the prohibition of abuse of law as developed by the Court is a free-standing doctrine or merely a technique for the interpretation of EU law. The latter view is shared by many commentators.347 An indication could potentially be gleaned from the specific prohibitions of abuse contained in primary and secondary EU law, but these unfortunately do not adopt a uniform approach. Article 54 of the Charter of Fundamental Rights clearly frames the prohibition as an interpretative principle.348 By contrast, the 2004 Citizenship Directive349 expressly conceptualises the prohibition as carving out an exception from an existing right, rather than a factor influencing the assessment of the exact scope of the right.350 The argument in favour of the prohibition being a mere interpretative principle was made most forcefully in Advocate-General Poiares Maduro’s influential Opinion in Halifax. The reason for this view was that the decisive factor in affirming the existence of an abuse is the teleological scope of the Community rules invoked, which must be defined in order to establish whether the right claimed is, in effect, conferred by such provisions, to the extent to which it does not manifestly fall outside their scope.351

He further emphasised that the result of the application of the ‘prohibition of abuse of Community law, seen as a principle of interpretation’, is that the legal provision interpreted cannot be regarded as conferring the right at issue because the right claimed is manifestly beyond the aims and objectives pursued by the provision abusively relied upon.352

A finding of abuse resulting in the non-application of the rule does indeed resemble a purposive or teleological interpretation whereby the wording of the rule (being found to be ‘over-inclusive’ for the situation in question) is ‘read down’ in order to ‘restrict’ or ‘reduce’ the scope of the rule.353 Writings on legal methodology in the German legal tradition would describe such an outcome as a ‘teleological reduction’ (teleologische Reduktion).354 Moreover, it might be argued that the general interpretative approach prevailing in EU law represents such a strong version of purposivism that there is no need for a further   § 42(1) Abgabenordnung. Cf Freedman (n 41 above) 373.   Schön (n 21 above) 1283, 1285; Eidenmüller (n 22 above) 142; Ringe (n 122 above) 114; Costello (n 19 above) 321. However, some authors are comparatively cautious in their assessment, eg Kjellgren (n 21 above) 275–76; Anderson (n 48 above) 353; Schammo (n 24 above) 372–73; Edwards and Farmer (n 117 above) 205. See also the discussion by Freedman (n 41 above) 375–77. 348   See above (n 18). 349  Article 35 of the Directive (see n 12 above). 350   Snell (n 191 above) 223; Ziegler (n 12 above) 296. 351  AG Poiares Maduro in Halifax (n 6 above), para [69]. Cf also ibid, para [62] n 44; AG Tesauro in Kefalas (n 34 above), para [25]: ‘the question of any abuse of rights . . . ultimately turns into a question of interpretation of the Community provision at issue’; AG Saggio in Diamantis (n 34 above), para [25]. 352  AG Poiares Maduro in Halifax (n 6 above), para [79]. 353   See above (at n 204). 354  EA Kramer, Juristische Methodenlehre, 3rd edn (Bern, Stämpfli 2010) 213. 346 347

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  559 doctrine to ‘read down’ provisions of EU law. After all, the Court has rarely shied away from judicial activism. On the contrary, it has frequently been criticised for according much more weight to the object and purposes of the provisions that it interprets than the national courts of the Member States would.355 If it wants to ‘read down’ the scope of application of a provision of EU law the necessary methodological tools are at its disposal.356 A similar point was made by Advocate-General Tesauro who argued against the existence of a ‘general principle of law relative to abuse of rights’ in Kefalas. He suggested that the European legal order was ‘more promptly amenable to adaptation to the needs of society’ than other legal systems, and that this had been achieved ‘through judicial interpretation’.357 However, conceptualising the prohibition of abuse as an interpretative technique ignores the well-established third requirement of the abuse test. A finding that the application of the rule in question is contrary to its purpose is not regarded as sufficient for a finding of non-application. The reliance of the person invoking the rule must be ‘abusive’ according to the criteria set out above.358 The non-application of a rule under the prohibition of abuse is thus subject to stricter requirements than a ‘reading down’ of the relevant rule of EU law by way of interpretation. This makes sense, given that the prohibition is conceived as an exception because of its anti-integrationist tendency and its one-sided operation to the detriment of the individual.359 Therefore the two issues of interpreting a provision of EU law and establishing an abusive practice are, as was recently suggested by Advocate-General Mazàk in RBS Deutschland, ‘conceptually distinct and should accordingly be dealt with one after the other rather than together’.360 In that case the Court, having carefully sidestepped the issue in its earlier Halifax decision, seemed to tacitly subscribe to Advocate-General Mazàk’s view. In a first step, it interpreted the relevant directive and it then went on, under a separate heading, to establish whether ‘the principle of prohibiting abusive practices may influence the interpretation adopted’.361 The current state of EU law is that the prohibition of abuse of law has the character of a free-standing doctrine This does not exclude the possibility that it may evolve into an interpretative principle at a later stage. Interpretative approaches develop over time. French legal theory, for example, still draws a strict distinction between the doctrine of fraude à la loi and the interpretation of legislation. German law initially proceeded from a similar position but has gradually come to regard cases of Gesetzesumgehung as a subset of statutory interpretation. In the area of tax law, German law still maintains a specific doctrine of ‘abuse of formally admissible legal arrangements’, rather than resolving cases of tax avoidance by simple interpretation.362 Similarly, English law resorted to various Ramsay, Furniss and other anti-avoidance doctrines before its principles of statutory 355   See, eg, T Hartley, The Foundations of European Union Law, 7th edn (Oxford, Oxford University Press 2010) 71–77. 356  For examples, see Vogenauer (n 337 above) 374–76. 357  AG Tesauro in Kefalas (n 34 above), para [23]. Cf Snell (n 191 above) 223. 358   See above, Part I.C.3. 359   See above, Parts I.G.1.b) and I.G.2. 360  AG Mazàk in RBS Deutschland (n 27 above), paras [28-31]. For a similar view, Engsig Sørensen (n 24 above) 429; Engsig Sørensen (n 291 above) 25–26; cf also the reading of Case C–116/02 Erich Gasser Gmbh v MISAT srl [2003] ECR I–14692 by Cuniberti (n 11 above) 281–82. 361   RBS Deutschland (n 27 above), para [47]. 362   See above (at n 346).

560  Stefan Vogenauer interpretation had become sufficiently relaxed to reach appropriate outcomes in tax avoidance cases without the need for specific doctrines of tax law.363 4.  Abuse and Fraud As has been seen above, the Court frequently refers to the notions of ‘fraud’ and ‘abuse’ as if they were synonyms.364 This is justified to the extent that in ordinary parlance ‘fraud’ denotes exceptionally improper conduct and may therefore be seen as a particularly serious case of abuse.365 Moreover, persons may engage in fraudulent conduct for the purposes of both rule avoidance (for example, to avoid the application of a national rule imposing taxes) and rule appropriation (for example, making false statements concerning marriage to an EU citizen in order to seek the application of a rule granting a residence permit,366 providing an incorrect health certificate in order to seek the application of a rule providing for migrant workers’ sickness benefit367 or purporting to use leased premises as a restaurant whilst actually using them for purely private purposes in order to obtain the right to deduct VAT368), and it seems appropriate that they should not benefit from EU rights if fraud is proved. However, if the notion of ‘fraud’ is used in a technical sense it must be clearly distin­ guished from abuse.369 Abuse of law concerns the question of whether a rule is or is not applied to a given set of facts. Fraud is concerned with the issue of whether the facts to which the rule is subsequently applied correspond to reality. Fraudulent conduct aims to induce others, for example a national authority or a court, to believe that an untrue representation of the facts is true. The person committing fraud claims a right on the basis of objectively false premises.370 This will be the case if he is ‘knowingly providing inaccurate information to, or deliberately concealing relevant matters from, the authorities concerned’, ie if there is an ‘element of deception’.371 A clear-cut distinction between abuse and fraud is also inherent in the primary and secondary law of the EU. Article 325 TFEU obliges the Union and the Member States to ‘counter fraud and any other illegal activities affecting the financial interests of the Union’. The protection of these interests is achieved by criminal sanctions which Member States are obliged to provide for under a convention of July 1995 and by administrative sanctions which are stipulated in a regulation of December of the same year. The convention aims   MacNiven v Westmoreland Investments Ltd [2001] UKHL 6, [2003] 1 AC 311.   See above (n 136).  Kjellgren (n 21 above) 247, 249–54 (fraud ‘can be perceived as a special form of abuse’, conceding, however, that the claim to the right invoked here is generally ‘based on objectively false premises’); Engsig Sørensen (n 24 above) 431–32, conceding, however, that ‘fraud and abuse seem only partly related’. Lagondet (n 21 above) 9–10 also sees fraud as a subcategory of rule appropriation. 366   Suat Kol (n 34 above), paras [26, 28]. 367   Paletta (n 228 above), paras [25–26]: ‘the Court has consistently held that Community law cannot be relied on for the purposes of abuse and fraud’, citing Van Binsbergen (n 3 above), TV10 (n 53 above), Leclerc (n 31 above), Lair (n 79 above), General Milk Products (n 72 above). 368   Fini H (n 199 above), para [32]. 369   Schön (n 21 above) 1278; Fleischer (n 21 above) 870; Englisch (n 47 above) 20; Eidenmüller (n 22 above) 142; Ziegler (n 12 above) 296; Costello (n 19 above) 323–24. Cf AG Tesauro in Kefalas (n 34 above), para [25] n 28: ‘On a proper view, [Paletta] did not involve abuse of rights in the true sense of the term but fraud’. 370   Case C–36/96 Faik Günaydin, Hatice Günaydin, Günes Günaydin and Seda Günaydin v Freistaat Bayern [1997] ECR I–5143, para [60]: ‘if the national court establishes that the Turkish worker made the statement with the sole intention of inducing the competent authorities to issue the requisite permits on false premises’. 371  AG Elmer in Faik Günaydin (n 370 above), paras [39] and [37]. 363 364 365

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  561 to create minimum standards of criminal liability. It defines ‘fraud affecting the European Communities’ financial interests’ in the areas of both EU expenditure and revenue as ‘the use or presentation of false, incorrect or incomplete statements or documents’, ‘nondisclosure of information in violation of a specific obligation’ and the ‘misapplication’ of funds or legally obtained benefits’.372 The regulation provides for administrative measures and penalties ‘concerning irregularities with regard to Community law’. The notion of ‘irregularity’ is defined as ‘any infringement of a provision of Community law resulting from an act or omission by an economic operator, which has, or would have, the effect of prejudicing the general budget of the Communities’.373 Furthermore ‘irregular conduct’ explicitly ‘includes fraudulent actions as defined in the Convention’.374 By contrast, acts which are ‘abusive’ under the general abuse test developed by the Court set out above375 are dealt with as if they were irregularities376 – which implies that they are distinct from the notion of ‘irregularity’ and thus from ‘fraud’. As a result, the prohibition of abuse of law should not be used to sanction fraudulent conduct. The combat of fraud should follow its own rules and procedures. 5.  Further Related Doctrines There are a number of further legal doctrines, concepts and mechanisms that are frequently referred to in the context of the prohibition of abuse of law. These include proportionality,377 good faith and reasonableness,378 forfeiture,379 substance over form,380 simulation,381 sham transactions,382 contra bonos mores and public policy.383 Each of these is complex and complicated in its own right. Their precise scope and content is difficult to discern in national legal systems, and even more so in EU law. In light of this, no attempt is made here to shed light on their exact relationship to the prohibition of abuse of law. However, their loose association with the notion of abuse does tell us something about the nature of the prohibition. They all belong to what Advocate-General Tesauro once referred to as the ‘self-protection measures’ that ‘any legal order which aspires to achieve a minimum level of completion must contain . . . to ensure that the rights it confers are not exercised in a manner which is abusive, excessive or distorted’.384 Abuse of law is a similar corrective or limiting mechanism. It ‘constitutes an indispensable safety-valve for protecting the aims of all provisions of Community law against a formalistic application of them based solely on their plain meaning’.385 372  Article 1(1) of the Convention drawn up on the basis of Article K.3 of the Treaty on European Union, on the protection of the European Communities’ financial interests of 26 July 1995 [1995] OJ C 316/49. The Convention entered into force on 17 October 2002. 373  Article 1(1) and (2) of Regulation 2988/95 (n 9 above). 374  Recitals 4 and 5 of Regulation 2988/95 (n 9 above). 375   See above, Part I.D. 376  Article 4(3) of Regulation 2988/95 (cited in n 9 above), as opposed to Article 4(1) of the Regulation. 377  Brown (n 20 above) 520–22; AG La Pergola in Centros (n 5 above), para [20]: ‘a certain affinity’. 378  Whittaker (n 240 above) 259. 379   Emsland-Stärke (n 7 above), para [59]. 380  Freedman (n 41 above) 365. 381   Lenaerts (n 177 above) 1136 n 88 argues that French and Belgian law would have approached the facts of Halifax (n 6 above) on the basis of ‘simulations frauduleuses’ or ‘simulations illicites’. 382  Weatherill (n 35 above) 56; Freedman (n 41 above) 365. 383   Case C-303/08 Bozkurt (n 34 above), para [52]. 384  AG Tesauro in Kefalas (n 34 above), para [24]. 385  AG Poiares Maduro in Halifax (n 6 above), para [74].

562  Stefan Vogenauer 6.  EU Prohibition and Abuse and National Prohibitions of Abuse The prohibition of abuse of law, as developed in the case law of the Court, is a doctrine of EU law that must be applied by the domestic courts.386 However, most of the Member States also have some kind of established home-grown doctrine of abuse.387 The question is whether the courts of such States are permitted to apply such national doctrines to situations where the rights in question derive from a rule of EU law or whether they are bound to apply the EU prohibition of abuse of law. This issue came to the fore in the three cases on Greek shareholders’ rights that were referred to earlier in this chapter.388 The referring courts asked the Court of Justice whether they were permitted to apply the broad anti-abuse rule in Article 281 of the Greek Civil Code. In the first case, the Court ultimately did not have to decide the issue but it was quick to warn that, ‘in any event, the application of such a rule must not detract from the full effect and uniform application of Community law in the Member States’.389 These two factors, as has been seen before, militate against an overly broad conception of abuse in EU law.390 In the second case, the Court had to address the issue. It referred to its previous case law in which it had established that EU law cannot be relied on for abusive purposes. As a consequence, an application of national anti-abuse rules would not as such be contrary to EU law: it would only be so if it were to prejudice the effectiveness and uniform application of EU law.391 Such prejudice would have occurred in the case at hand, so as a result Article 281 of the Greek Civil Code could not be applied. However, the Court held that, as a general rule, ‘Community law does not preclude national courts from applying a provision of national law in order to assess whether a right arising from a provision of Community law is being exercised abusively’.392 In the third case, the Court added that it is the task of the national courts to determine whether the application of the national anti-abuse rule ‘is compatible’ with the requirement of safeguarding the full effect and uniform application of EU law.393 It held that on the facts of the case an application of Article 281 ‘would not alter the scope of [the] provision’ of EU law on which the claimants had relied to exercise their rights.394 As has been seen before, this means that the application of the national rule would still leave room for the application of the EU rule in other cases and thus not detract from the full effect and uniform application of EU law.395 At first sight, allowing national courts to apply domestic anti-abuse doctrines may appear to be a major concession on the part of the Court. But, as was pointed out by Advocate-General Saggio in the third case, this was ‘more apparent than real’. Of course, he pointed out, the Court had accepted that a finding of abuse could be made by applying a national anti-abuse rule rather than the Community law prohibition: ‘however, it was quick to make clear the limits that Community law imposes on the application of that national rule’. The Court had ‘laid down precise criteria . . . which keep in well defined   See above, Part I.F.   See above, Part I.H.2.   Pafitis; Kefalas; Diamantis (see at n 323 above). 389   Pafitis (n 112 above), para [68]. 390   See above, Part I.G.1.c). 391   Kefalas (n 34 above), paras [20–22]. 392   Kefalas (n 34 above), para [29]. 393   Diamantis (n 34 above), para [35]. 394   Diamantis (n 34 above), para [44]. 395   See above (at nn 266–72). 386 387 388

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  563 limits the application of the national rule’. The rule in question was simply ‘an instrument available to the national court for the purpose of ensuring the proper application of Community law’.396 In other words, national anti-abuse doctrines apply only to the extent that they conform with the abuse test of the Court. In effect, what the Court requires is an interpretation of the domestic rule in conformity with that test. Such an interpretation is also required where a directive contains anti-abuse provisions that have not been transposed into national law. It is then, for example in the area of tax law, for the domestic courts to ascertain whether there is, in national law, ‘a provision or general principle prohibiting abuse of rights or other provisions on tax evasion or tax avoidance which might be interpreted in accordance with [the directive] and thereby justify taxation of the exchange of shares in question’.397 The national anti-abuse rules are therefore clearly subordinate to the EU prohibition of abuse of law. The domestic courts may apply them but only to the extent that they reach a result that conforms with the EU prohibition.398 In the final analysis this is nothing but a consequence of the primacy of EU law.399 Of course the national anti-abuse rule need not be a statutory one. A case in point is the antisuit injunction, a procedural device of English law. The English Court of Appeal granted such an in injunction in the case of Turner v Grovit. Arguably, the defendant in that case had initiated proceedings in another Member State abusively and in bad faith. In doing so he had relied on a European jurisdictional rule. The Court of Justice did not seem to regard this scenario as being covered by the EU prohibition of abuse, so it did not question the applicability of the jurisdictional rule which, in turn, rendered the national doctrine inapplicable. ‘There may have been an abuse,’ as Professor Briggs suggests, ‘but it was not an abuse which it was in the power of the English court to restrain’.400 II. The Prohibition of Abuse of Law as a General Principle of EU Law

In recent years, much of the debate with regard to abuse of law has turned on the question whether the prohibition amounts to a general principle of EU law. This was first suggested by Professor Brown in 1994401 but explicitly rejected by Advocate-General Tesauro in the Opinion he delivered in Kefalas in 1998.402 Only a few months later, in his Opinion in Centros, Advocate-General La Pergola asserted more or less in passing that the prevention of abusive conduct was ‘among the general principles of Community law’.403 In the following year, Advocate-General Saggio echoed this view, however without any further explanation.404 From then on the debate continued, with the number of proponents405 gaining on the number of opponents,406 not least because the first view was  AG Saggio in Diamantis (n 34 above), paras [22–24].   Kofoed (n 1 above), para [46], cf ibid, paras [40, 45, 48]. 398  Weber (n 48 above) 400. 399   See the discussion above at n 259. 400  Briggs (n 336 above) 270. 401  Brown (n 20 above) 511: ‘at present only latent or inchoate in Community law, deserves proper recognition’. 402  AG Tesauro Kefalas (n 34 above), para [21]. 403  AG La Pergola in Centros (n 5 above), para [20]. 404  AG Saggio in Diamantis (n 34 above), paras [22, 28]. 405   eg Engsig Sørensen (n 24 above) 458; Simon and Rigaux (n 94 above) 579; Baudenbacher (n 48 above) 218; de la Feria (n 4 above) 438; Lenaerts (n 177 above) 1139, 1151, with comprehensive references in n 104. 406   eg Kjellgren (n 21 above) 271; Englisch (n 47 above) 5, 22. 396 397

564  Stefan Vogenauer supported in Advocate-General Poiares Maduro’s Opinion in Halifax407 and then in 2007 by the First Chamber of the Court in Kofoed.408 The contributions in this volume show that the question is still far from resolved, with supporters409 and sceptics410 evident in almost identical numbers. The debate is vigorous because it is important whether the prohibition is classified as a general principle. If we regard a number of observable legal phenomena as single instances they are just that: unconnected events that are to be dealt with independently, simply ‘individual components, each of which possesses its own significance or range of significances and its own utility in different contexts’.411 If we bring these instances together and view them as manifestations of a general principle we assume that they are – or ought to be – dealt with in a broadly similar fashion. We expect some coherence, for example, with regard to the criteria employed and the legal consequences arising in the different contexts. Most importantly, we acknowledge that the general principle will have relevance in future cases on which there is not yet any settled law. This is particularly important in a legal system such as EU law that is constantly evolving and expanding. Unfortunately, the notion of ‘general principles of law’ is notoriously elusive. There is no consensus amongst, and even within, national legal systems as to what exactly a general principle of law is, how its existence is determined and what its functions are. Fortunately, the issue is slightly less contested in EU law. Here, despite continuing differences of opinion with regard to individual aspects, there seems to be broad agreement on the most important features of general principles of law. The debate has recently been the subject of a number of important and comprehensive studies412 and it is admirably summarised in Professor Arnull’s contribution to this volume.413 For the purposes of this chapter it is therefore sufficient to focus on five features that are relevant to the discussion of abuse of law. All of them, it is submitted, point towards classifying the prohibition as a general principle of law. A. Generality First, a general principle is general, rather than specific.414 It is of fundamental importance and thus does not only apply to individual situations but, potentially, to a wide range of instances. Its impact is pervasive, ie it is relevant to all branches and at all levels of 407  AG Poiares Maduro in Halifax (n 6 above), para [64]: ‘a general principle can certainly be considered to derive from the case-law’. See also AG Sharpston in Bozkurt (n 34 above), para [58]: ‘clear beyond doubt is that it is a general principle of European Union law that abuse of rights prohibited’. 408   Kofoed (n 1 above), para [38]; possibly already in Halifax (n 6 above), para [70], but less clear. Arnull (n 19 above) 20 rightly points out that the dictum in Kofoed does not conclude the matter. 409   See, eg, Engsig Sørensen (n 291 above) 25, 27; Doukas (n 10 above) 89–90; Tridimas (n 48 above) 173. 410   See, eg, Arnull (n 19 above) 23; Whittaker (n 240 above) 260; M Dougan, ‘Some Comments on the Idea of a General Principle of Union Law Prohibiting Abuses of Law in the Field of Free Movement for Union Citizens’, ch 24 above, 355. 411  This is the approach favoured by Whittaker (n 240 above) 260. 412  T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press 2006); X Groussot, General Principles of Community Law (Groningen, Europa Law Publishing 2006); A Metzger, Extra legem, intra ius: Allgemeine Rechtsgrundsätze im Europäischen Privatrecht (Tübingen, Mohr Siebeck 2009). 413  Arnull (n 19 above). 414   Case C–101/08 Audiolux SA et al v Groupe Bruxelles Lambert SA (GBL) et al and Bertelsmann AG et al [2009] ECR I–9823, nyr, para [42]: ‘the general, comprehensive character which is . . . naturally inherent in general principles of law’.

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  565 EU law.415 The attribute of generality is frequently expressed by describing a particular principle as ‘overarching’ or ‘underlying’ a body of more specific rules. As this book has shown, the Court has considered the application of the prohibition of abuse of law in many cases cutting across a very broad spectrum of EU law.416 Moreover, there is a growing number of explicit anti-abuse provisions in secondary EU law.417 Whether it is possible to infer the existence of a ‘general principle’ from these instances is difficult to say. There is no hard and fast rule with regard to the ‘critical mass’ of individual occurrences needed in order to proceed to a generalisation. It might be argued that the number of situations where the Court and the legislative organs of the EU have employed abuse-based reasoning is still too low. Whereas the Court has frequently considered the application of the prohibition of abuse of law, it has been much less inclined to allow national courts to find abuse in a given case. In some areas of law it has consistently declined to do so, particularly in the fields of free movement of workers and citizenship, so the approach to proceedings involving natural persons is notably different to that taken in cases dealing with purely commercial situations.418 In other areas, notably contract419 and international civil procedure,420 abuse-based reasoning does not even seem to have been considered at all or has at best had an extremely limited and ‘discreet’ influence.421 There also seems to be a difference in approach between cases of rule avoidance and cases of rule appropriation, as defined above.422 Overall, the Court appears to be more willing to provide guidance that a certain type of conduct ought to be assessed as abusive in the second group. Finally, it has been suggested that the prohibition lacks the ‘general, comprehensive character’ required for its classification as a general principle because it is not, as general principles of EU law typically are, 423 a principle of administrative law with a rule of law element:424 as has been seen before, the prohibition only benefits the administration when it asserts alleged abuses by private parties.425 Against this, it must be pointed out that EU lawyers have been prepared to accept the existence of ‘general principles’ on the basis of much more slender evidence.426 Several principles have become well-established although they vary in their effects from one area to another.427 Abuse of law may not have been considered in all areas of EU law but it has featured with regard to all of the fundamental freedoms and in a broad range of other situations, involving provisions from primary and secondary law as well as from  FG Jacobs, ‘Foreword’ in Tridimas (n 412 above).   See above (at nn 4–7). 417   See above (at nn 8–14). 418   de la Feria (n 4 above) 417; Ziegler (n 12 above) 306: ‘merely verbal acceptance as a legal principle’; Costello (n 19 above) 353. 419   Metzger (n 11 above) 251. 420  Briggs (n 336 above) 262, 275–76. 421   Cuniberti (n 11 above) 279, 286. 422   See above, Part I.C. For a similar assessment, see Schön (n 47 above) 80–81; Baudenbacher (n 48 above) 217. 423  AG Tesauro in Kefalas (n 34 above), para [19]; Arnull (n 19 above) 10. 424   Dougan (n 410 above) 355–56; Arnull (n 19 above) 20–21 and 22, citing from Audiolux (n 414 above), para [42] which, however, dealt with a much narrower potential basis for a legal principle: the alleged ‘general principle of equality of shareholders’ had been based on two directives; however, according to the Court these ‘apply to well-defined situations . . . [T]hose provisions are essentially limited to regulating very specific company-law situations’. 425   See above (at n 249). 426  For an overview, cf Metzger (n 412 above) 323–59. 427  Arnull (n 19 above) 19. 415 416

566  Stefan Vogenauer national law falling within the scope of EU law.428 The Court has never explicitly held that the prohibition does not apply to a particular field as such. Quite to the contrary, it has developed an overarching abuse test that is broad enough to be used in all areas of EU law, including those where it has not yet been employed,429 and it has consistently emphasised that EU law may not be relied on for abusive ends even in cases where it has ruled out the possibility of a finding of abuse.430 That it has declined to make a positive finding in a significant number of cases is no argument against characterising the prohibition as a general principle of law: there is no doubt that proportionality is such a principle although the Court frequently refuses to hold that a given measure was disproportionate. Finally, it is not clear why a principle should not be recognised on the basis that it is skewed towards the administration. Other principles of EU law operate, by their nature, exclusively to the advantage of certain constituencies and to the disadvantage of others: for the individual and against the public authorities (right to a fair hearing), for the individual and against the Member States (state liability) or for the Member States and against the institutions of the EU (procedural autonomy). There is no reason why a general principle should never give rise to obligations against private persons,431 as long as the necessary safeguards apply and the rule of law is observed.432 B.  Dimension of Weight Secondly, a general principle is a principle, rather than a rule. As such, its legal consequences do not automatically follow once its requirements are met. A principle does not, as has been famously said, apply in an ‘all-or-nothing fashion’. It rather ‘states a reason that argues in one direction, but does not necessitate a particular decision’. It is a ‘consideration inclining’ in a certain direction which may always be outweighed by ‘other principles or policies arguing in the other direction’.433 In other words, a principle does not require its strict application. It is a mere ‘optimisation requirement’ which requires that something be realised to the greatest extent possible, given the factual and legal possibilities. What is legally possible is, inter alia, determined by opposing principles and rules.434 The prohibition of abuse of law clearly possesses this dimension of weight. It is precisely this feature that provides an explanation for the divergences in approach with regard to different areas of law and types of case that were mentioned in the previous section of this part. Where a finding of abuse is rejected there is generally a good reason for doing so. The prohibition is outweighed by other principles pointing towards a different outcome. For example, the characteristic structure of principles accounts for the observation that the Court is more likely to suggest a finding of abuse in cases of rule appropriation than   See above, Parts I.A and I.D.1.   Cf Metzger (n 11 above) 240; Briggs (n 336 above) 263.   See, eg, Kefalas (n 34 above), para [20]; Centros (n 5 above), para [24]. 431   See the discussion in Tridimas (n 412 above) 36, 47–50; X Groussot and HH Lidgard, ‘Are There General Principles of Community Law Affecting Private Law?’ in U Bernitz, J Nergelius and C Cardner (eds), General Principles of EC Law in a Process of Development (Alphen aan den Rijn, Kluwer International 2008) 155, 171, 175. 432   See above, Part I.G.1.b). 433  R Dworkin, Taking Rights Seriously (London, Duckworth 1976) 24–26. 434  R Alexy, A Theory of Constitutional Rights (Oxford, Oxford University Press 2002) 47–48. 428 429 430

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  567 in cases of rule avoidance: in both categories of cases the persons concerned ultimately aim to bring themselves within the scope of a rule of EU law in order to derive a benefit but it is only in the second group that this benefit (for example a monetary compensation) is normally derived at the expense of the EU itself. In the first group of cases it is the fiscal interests of the Member States (for example their tax revenue), rather than those of the EU, that are at stake.435 Moreover, it is only in the first group of cases that the prohibition of abuse conflicts with, and might be outweighed by, the fundamental freedoms and the Union’s key goal of market integration.436 Similarly, the prohibition will not prevail if there are strong countervailing human rights arguments, as in the cases of free movement of workers who are able to invoke the right to respect for family life,437 or if other principles are regarded as overriding, such as the principle of mutual ‘trust’ or respect of the courts of the Member States for each other’s decisions in the area of international civil procedure438 or the mutual recognition of Member States’ designations of nationality in the area of citizenship.439 The principled nature of the prohibition of abuse also accounts for the fact that the precise requirements for a finding of abuse are always dependent on the area of law concerned and the circumstances of the case.440 What amounts to an abusive practice in one context does not necessarily constitute abuse in another because the nature and the strength of countervailing principles is bound to be different in different contexts: ‘the absence of a unitary test for the operation in every field of Community law of the principle of the prohibition of abuse must be regarded as perfectly natural in Community law, as it is in any national legal system’.441 C.  Legal Nature Thirdly, a general principle of law is a legal, as opposed to a non-legal, proposition. It is not a mere collective ‘policy’, ie an economic, political, social or moral goal which, as an external standard, may or may not have an impact on the long-term development of the law. It may be classified as a guideline or an ‘optimisation requirement’ but it is nevertheless binding. Judges, practitioners, officials and other participants in the legal process must take it into account when resolving legal matters. To this extent, a general principle of law is, to use a similarly obscure and contested notion, a ‘source of law’.442 This feature of a general principle is present in the prohibition of abuse of law. The fact that the application of the prohibition requires a value judgment by the courts443 does not detract from its genuine legal character. In fact many legal propositions require the making of such judgments, and it is indeed characteristic of many general principles of EU   Cf Harris (n 39 above) 139–45.   See above, Part I.G.2. Cf Ziegler (n 12 above) 307, 309–10. 437   Akrich (n 26 above), para [58]. Cf R de la Feria, ‘Introducing the Principle of Prohibition of Abuse of Law’ pp xvii, xxi above; Ziegler (n 12 above) 307, 311–13. 438  Briggs (n 336 above) 272–74; Cuniberti (n 11 above) 287–88. 439   Chen (n 80 above), paras [36–40]; cf Costello (n 19 above) 352. 440   See above (at n 143). 441  AG Poiares Maduro in Halifax (n 6 above), para [83]. Cf Engsig Sørensen (n 291 above) 29: ‘the principle applies broadly in EU law, although slightly different approaches have been used in the different areas’. 442   S Vogenauer, ‘Sources of Law and Legal Method in Comparative Law’ in M Reimann and R Zimmermann (eds), The Oxford Handbook of Comparative Law (Oxford, Oxford University Press 2006) 869. 443   See above, Part I.D.3.d). 435 436

568  Stefan Vogenauer law, notably the principle of proportionality, that they involve a subjective evaluation.444 The prohibition establishes genuine legal requirements and stipulates legal consequences. It creates rights and duties, such as the right of an authority to claim restitution of an advantage obtained and the corresponding ‘obligation to repay’ on the part of the person who abusively relied on a rule of EU law.445 It binds the national courts that may not apply national anti-abuse doctrines to the extent that they are incompatible with the EU rule.446 It affects the legal position of the Member States: if a finding of abuse is made in cases of rule avoidance the national legislation that is sought to be avoided continues to apply to the person in question; if such a finding cannot be made the national legislation does not apply because the person is entitled to rely on a provision of EU law. The prohibition is therefore a potential constraint on the activities of the Member States.447 Yet the prohibition is not binding in a way that is typical for other established general principles of EU law. These have the function of an independent ground for the review of the legality of EU measures, so that measures infringing a general principle are declared invalid.448 The prohibition of abuse of law has never been used for this purpose and, given its particular nature, ‘it is not immediately obvious how it could do so’.449 It might be argued that a finding of abuse can achieve at least a partial annulment of secondary EU legislation because it leads to the non-application of the rule in question. However, the non-application only applies to the facts of the given case, not across the board.450 Yet again, the capacity to be invoked to strike down secondary law is by no means a necessary attribute of a general principle of EU law. It is true that particularly fundamental principles of a constitutional nature have the same status as Treaty provisions and are thus capable of overriding secondary law. However, others, for example the revocation of administrative acts granting rights to individuals,451 are on the same level as secondary law, and the Union legislator is free to deviate from these without the resulting measure being invalidated.452 D.  Linguistic Indeterminacy Fourthly, general principles of law are characterised by a high degree of linguistic indeter­ minacy or ‘open-endedness’. They use vague and open-textured phrases, rather than precise and determinate language. To some extent, this attribute follows from the first two features mentioned above. The high degree of generality of principles requires that they be framed at a high level of abstraction which makes them capable of covering a   Cf Tridimas (n 412 above) 552.   See above, Part I.E.2.   See above (at n 398). 447  For this function of general principles, see JA Usher, General Principles of EC Law (Harlow, Longman 1998) 131–32. Cf de la Feria (n 4 above) 438. 448  Tridimas (n 412 above) 31–32. 449  Farmer (n 294 above) 4. 450   See above (at nn 266-72). In a similar vein, Engsig Sørensen (n 291 above) 27. 451   Joined Cases 7/56 and 3-7/57 Algera v Common Assembly [1957-58] ECR 39. 452  T Oppermann, CD Classen and M Nettesheim, Europarecht, 4th edn (Munich, CH Beck 2009) 170; B de Witte, ‘Institutional Principles: a Special Category of General Principles of EC Law’ in U Bernitz and J Nergelius, General Principles of European Community Law (The Hague, Kluwer Law International 2000) 143. The view promoted in Audiolux (n 414 above), para [63] and relied on by Costello (n 19 above) 322–23 would require a significant narrowing of the established concept of general principles of EU law. 444 445 446

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  569 multitude of cases, including those that are as yet unforeseen at the time when the general principle is first articulated. This is facilitated by the use of loose and flexible language. At the same time, there is no need to frame principles with utmost linguistic precision since they do not aspire to spell out clear, ‘bright-line’ propositions that will apply in an ‘all-or-nothing fashion’. Once again, the prohibition of abuse of law fits in with this attribute of general principles of EU law. The inherently vague nature of the critical third limb of the abuse test (‘abusive reliance’) has been mentioned before. What is, and what is not ‘abusive’, ‘artificial’, ‘improper’ or not a ‘normal commercial operation’ is difficult to define in the abstract. Indeed, it has been seen that the linguistic indeterminacy, and the corresponding concern about a lack of legal certainty, is one of the main reasons militating towards a cautious use of the prohibition.453 E. Pedigree Fifthly, it is often said that general principles of EU law are necessarily ‘unwritten’,454 ie they are ultimately ‘judge-made’ law.455 The Court does not however, conjure them up out of thin air. It normally establishes the existence of a general principle by way of extrapolation from one or more legal propositions which are drawn from either EU law, the laws of the Member States, international law or a combination of any of these.456 A general principle may thus be derived, inductively, from one or more express provisions of existing EU law (the acquis), particularly from the Treaties and their objectives but also from the more detailed rules of secondary Union legislation which may be seen as specific manifestations of the principle. A general principle may also be inferred from the laws of the Member States, most notably from their constitutional traditions. Finally, a general principle may be drawn from general principles of public international law, primarily those that are laid down in international conventions to which Member States are parties. It is perhaps at this stage where the characterisation of the prohibition of abuse as a general principle of law is susceptible to doubt. The prohibition is certainly judgemade457 but the Court has never justified its provenance. It has not attempted to show that the prohibition follows from particular Treaty provisions, the system of the Treaties or the nature of the European legal order, as it has done with certain other general principles that have been developed autonomously, ie without reference to legal systems outside the Union legal order, such as primacy, direct effect, national procedural autonomy and institutional balance.458 Nor has the Court ever referred to the general principle of abuse of law that is known in public international law where it applies between sovereign states.459 And it never seems to have enquired whether the prohibition is one of the ‘the   See above, Part I.G.1.a).   de Witte (n 452 above) 143. 455  Groussot and Lidgard (n 431 above) 160. Cf M Herdegen, ‘General Principles of EU Law – the Methodological Challenge’ in Bernitz et al (n 431 above) 343, 352–53; Oppermann, Classen and Nettesheim (n 452 above) 167–68; Hartley (n 355 above) 141. 456  Usher (n 447 above) 12. 457   See above, Part I.A. 458  The notion of principles which are autonomous to the Union’s legal order is used by Arnull (n 19 above) 7. 459   German Interests in Polish Upper Silesia, PICJ, Series A, no 7, 1926, p 30; Free Zones, PICJ, Series A, no 30, 1930, p 12 and no 46, 1932, p 167. See also AC Kiss, L’abus de droit en droit international (Paris, LGDJ 1953); Byers (n 320 above) 431. 453 454

570  Stefan Vogenauer general principles common to the laws of the Member States’, as suggested by Article 340(2) TFEU. As has been seen above, there is no such common tradition across the EU with regard to abuse of rights, and neither is there unanimity on how to deal with the evasion or circumvention of legal rules (fraude à la loi or Gesetzesumgehung). This is not necessarily fatal: the Court has long asserted that reference to the general principles of the Member States ‘does not seek to determine the arithmetical average of national laws or to fall into line with the lowest common denominator’ but rather empowers the Court to choose the ‘most appropriate solution’.460 However, only one Advocate-General seems to have engaged in a very short analysis of the various doctrines of abuse of rights which, incidentally, only contributed to his conclusion that the existence of a corresponding general principle of Community law ought to be denied.461 The national doctrines of evasion and circumvention which are, as has been seen, much more relevant for the EU prohibition of abuse of law, were first referred to at a time when the prohibition was already established, namely in Advocate-General Poiares-Maduro’s Opinion in Halifax.462 The impression is rather that the prohibition was developed in an intuitive fashion, on the basis of a tacit understanding of the nature of rules and their application in the legal process which assumes that ‘any legal order’ needs ‘self-protection measures’ or ‘safety valves’ to ensure that ‘the rights it confers are not exercised in a manner which is abusive, excessive or distorted’ and that its rules are not applied mechanically.463 The obvious parallel of this intuitive process is the gradual development of a body of more or less coherent rules and principles for the interpretation of EU law in the first decades of the Court’s existence. This lack of proper pedigree may give rise to concern from the perspective of the separation of powers: the doctrine may be judge-made but it might be doubted whether its development is a legitimate case of judicial law-making. One possibility to dispel such doubts is to refer to the Court’s duty to ‘ensure that in the interpretation and application of the Treaties the law is observed’,464 as has been done in other controversial instances of judicial activism in the development of general principles of law.465 More important for present purposes is the general acknowledgment that even in those cases where the Court purports to derive a general principle from the laws of the Member States the finding of the ‘most appropriate solution’ invariably involves the making of policy choices that result in the making of new law and shaping principles with independent normative value in accordance with the exigencies of the Union polity.466 This is, in effect, what has happened in the case of the prohibition of abuse, albeit without as much reliance on the legal traditions of the Member States as might have been desirable. 460  AG Léger in Case C-87/01 P Commission v CEMR [2003] ECR I-4207, para [64], echoing the language of AG Lagrange in Hoogovens v High Authority [1962] ECR 253, 283-84. Cf, in the context of abuse of law, AG Tesauro in Kefalas (n 34 above), para [19]. 461  For a brief overview, see AG Tesauro in Kefalas (n 34 above), para [22]. For even briefer references to national abuse of law doctrines, see AG Tesauro in Pafitis (n 112 above), para [29]; AG Poiares Maduro in Halifax (n 6 above), para [73] n 67. 462  AG Poiares Maduro in Halifax (n 6 above), paras [74] n 68 (in passing on abuse of rights) and [77] n 72 as well as [83] n 80 (in greater detail on national doctrines of tax avoidance). 463  AG Tesauro in Kefalas (n 34 above), para [24] and AG Poiares Maduro in Halifax (n 6 above), para [74] (cited at nn 384–85 above). 464  Article 19 TEU [ex Article 220 TEC]. 465   See, eg, Joined Cases C–46/93 and C–48/93 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and R v Secretary of State for Transport ex p Factortame Ltd et al [1996] ECR I–1029, para [27]. 466   See, amongst others, Hartley (n 355 above) 141, Herdegen (n 455 above) 344, 347, 352–53; Tridimas (n 412 above) 548.

The Prohibition of Abuse of Law: An Emerging General Principle of EU Law  571

III.  Conclusion: an Emerging General Principle

According to the judge-made prohibition of abuse of EU law, a given rule of law will not be applied where (1) a particular set of facts is clearly and unambiguously covered by the wording of the rule but (2) the result of applying the rule would be contrary to the purpose of that rule and (3) the person’s reliance on the rule is abusive. The requirements, legal consequences, procedural aspects and limits of the prohibition have been gradually clarified by the Court and legal scholars in the course of the past decade. The types of cases where it potentially applies are well defined. The doctrine exhibits all the necessary attributes in order for it to be recognised as a general principle of EU law. However, this general principle is still emerging. It still awaits universal recognition. Some of its doctrinal foundations are still shaky. The conceptual framework is not yet consistent. As might be expected from an ‘embryonic principle’,467 its emergence raises normative concerns: the prohibition has the potential of endangering legal certainty, the proper division of powers between the judiciary and the legislature, the rule of law, the uniform application and effectiveness of EU law and the overriding goal of market integration. The Court has met these concerns by subjecting the prohibition to carefully crafted limits: it ‘continues to use the notion of abuse with considerable restraint – and rightly so’.468

 Faull (n 250 above) 293: ‘the embryonic principle of abuse of rights’.  AG Poiares Maduro in Case C–210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I–9641, para [29].

467 468

Table of Cases I  European Union Court of Justice of the European Union (European Court of Justice and Court of First Instance) listed chronologically by Case year/number Case 8/55 Fédération Charbonnière de Belgique v High Authority [1954–56] ECR 245......9 Joined Cases 7/56 and 3-7/57 Dineke Algera et al v Common Assembly of the European Coal and Steel Community [1957–58] ECR 39..........................................8, 566 Joined Cases 42/59 and 49/59 SNUPAT v High Authority [1961] ECR 53.........................514 Joined Cases 43/59, 45/59 and 48/59 Eva von Lachmüller, Bernard Peuvrier, Roger Ehrhardt v Commission of the European Economic Community [1960] ECR 463......................................................................................................................235, 250 Case 14/61 Hoogovens v High Authority [1962] ECR 253..............................................9, 568 Case 75/63 MKH Hoekstra (née Unger) v Bestuur der Bedrijfsvereniging voor Detailhandel en Ambachten (Administration of the Industrial Board for Retail Trades and Businesses) [1964] ECR 177...........................................................................229 Case 34/67 Firma Gebrüder Lück v Hauptzollamt Köln-Rheinau [1968] ECR 245......................................................................................................................451, 452 Case 11/70 Internationale Handelsgesellschaft v Einfuhr- und Vorratsstelle Getreide [1970] ECR 1125..................................................................................................................10 Case 70/72 Commission v Germany [1973] ECR 813.........................................................513 Case 4/73 J Nold, Kohlen- und Baustoffgrosshandlung v Commission [1974] ECR 491............................................................................................................................9, 10 Case 173/73 Italy v Commission [1974] ECR 709................500, 501, 503, 504, 507, 511, 512 Case 2/74 Jean Reyners v Belgian State [1974] ECR 631...............................................56, 402 Case 8/74 Procureur du Roi v Benoît and Gustave Dassonville [1974] ECR 837................................................................................................................................51 Case 17/74 Transocean Marine Paint v Commission [1974] ECR 1063.....................236, 238 Case 33/74 Van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid [1974] ECR 1299........................xv, xxii, 19, 51, 58, 64, 66, 67, 68, 69, 70, 72, 73, 86, 90, 93, 109, 170, 180, 184, 248, 271, 296, 361, 384, 470, 499, 519, 522, 524, 527, 558 Case 36/75 Rutili v Minister for the Interior [1975] ECR 1219.............................................13 Case 39/75 R-G Coenen et al v Sociaal-Economische Raad [1975] ECR 1547.....................69 Case 43/75 Defrenne v SABENA [1976] ECR 455.................................................................13 Case 25/76 Galeries Segoura v Société Rahim Bonakdarian [1976] ECR 1851..................235 Case 29/76 LTU v Eurocontrol [1976] ECR 1541.................................................................523 Case 35/76 Simmenthal v Ministero delle Finanze [1976] ECR 1871.................................210

574  Table of Cases Case 74/76 Iannelli and Volpi SpA v Ditta Paolo Meroni [1977] ECR 557................358, 451 Case 125/76 Enterprise Peter Cremer v Bundesanstalt für landwirtschaftliche Marktordnung [1977] ECR 1593............................................. 174, 203, 204, 208, 428, 526, 527, 529, 531, 533, 539, 542, 550 Case 30/77 R v Pierre Bouchereau [1977] ECR 1999.......................................................98, 99 Case 115/78 J Knoors v Staatssecretaris van Economische Zaken [1979] ECR 399.................................................................... 55, 56, 58, 59, 61, 67, 69, 109, 170, 176, 350, 351, 481, 522, 524, 545, 548, 552 Case 120/78 Rewe-Zentrale AG v Bundesmonopolverwaltung für Branntwein [1979] ECR 649 (Cassis de Dijon).............................................51, 52, 53, 56, 328, 460, 461 Case 240/78 Atlanta Amsterdam BV v Produktschap Voor Vee en Vlees [1979] ECR 2137............................................................................................................................355 Case 250/78 DEKA Getreideprodukte GmbH & Co KG, iL (formerly Firma Contifex Getreideprodukte GmbH & Co KG) v European Economic Community [1983] ECR 421......................................................................................................................248, 249 Case 34/79 R v Maurice Donald Henn and John Frederick Ernest Darby [1979] ECR 3795............................................................................................................................206 Case 44/79 Hauer v Land Rheinland-Pfalz [1979] ECR 3727...............................................10 Case 68/79 Hans Just I/S v Danish Ministry for Fiscal Affairs [1980] ECR 501...................21 Case 104/79 Foglia v Novello [1980] ECR 745............................................... 19, 171, 243, 526 Case 155/79 AM & S Europe Limited v Commission [1982] ECR 1575.................9, 237, 529 Case 66/80 SpA International Chemical Corporation v Amministrazione delle finanze dello Stato [1981] ECR 1191................................................................................224 Case 96/80 Jenkins v Kingsgate [1981] ECR 911....................................................................97 Case 169/80 Administration des douanes v Société anonyme Gondrand Frères and Société anonymed Garancini [1981] ECR 1931.......................................................xxv Case 244/80 Foglia v Novello [1981] ECR 3045...................................................................171 Case 8/81 Becker [1982] ECR 53.....................................................................................13, 456 Case 38/81 Effer v Kantner [1982] ECR 825........................................................................264 Case 53/81 Levin v Staatssecretaris van Justitie [1982] ECR 1035............. 231, 297, 298, 300, 301, 302, 306, 315, 340, 345 Case 61/81 Commission v United Kingdom [1982] ECR 2601...........................................238 Case 261/81 Walter Rau Lebensmittelwerke v De Smedt PVBA [1982] ECR 3961........................................................................................................................ 13, 50, 356 Case 283/81 Srl CILFIT and Lanificio di Gavardo SpA v Ministry of Health [1982] ECR 3415............................................................................................ 9, 211, 230, 523 Case 40/82 Commission v United Kingdom [1982] ECR 2793...........................................206 Case 152/82 Sandro Forcheri and his wife Marisa Forcheri, née Marino, v Belgian State and asbl Institut Supérieur de Sciences Humaines Appliquées - Ecole Ouvrière Supérieure [1983] ECR 2323...............................................................................................16 Case 199/82 Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595..............................................................................................................................21 Case 227/82 Van Bennekom [1983] ECR 3883.......................................................................50 Case 238/82 Duphar BV et al v The Netherlands State [1984] ECR 523............................209 Case 14/83 Sabine von Colson and Elisabeth Kamann v Land Nordrhein-Westfalen [1984] ECR 1891................................................................................................................215 Case 129/83 Zelger v Salinitri [1984] ECR 2397..................................................................283

Table of Cases  575 Case 229/83 Association des Centres distributeurs Edouard Leclerc et al v SARL‚ Au Blé Vert‘ et al [1985] ECR 1.............................................19, 56, 59, 60, 61, 69, 109, 113, 170, 171, 205, 350, 522, 525, 533, 558 Case 293/83 Gravier v City of Liège [1985] ECR 593............................................................16 Case 294/83 Parti écologiste ‚Les Verts‘ v European Parliament [1986] ECR 1339..............................................................................................................................12 Case 299/83 SA Saint-Herblain distribution, centre distributeur Leclerc et al v Syndicat des libraries de Loire-Océan [1985] ECR 2515.................................................205 Joined Cases C–60/84 and C–61/84 Cinéthèque SA et al v Fédération nationale des cinémas français [1985] ECR 2605...................................................................................460 Case 95/84 Boriello v Alain Darras and Dominique Tostain [1986] ECR 2253.................205 Case 152/84 MH Marshall v Southampton and South-West Hampshire Area Health Authority (Teaching) [1986] ECR 723.............................................................................221 Case 178/84 Commission v Germany [1987] ECR 1227.......................................................53 Case 181/84 R v Intervention Board ex p Man (Sugar) Ltd [1985] ECR 2889...................355 Case 205/84 Commission v Germany [1986] ECR 3755........................... 64, 67, 69, 184, 460 Case 220/84 AS Autoteile Service GmbH v Pierre Malhé [1985] ECR 2267...............245, 283 Case 222/84 Johnston v Chief Constable of the Royal Ulster Constabulary [1986] ECR 1651................................................................................................................10, 13, 356 Case 253/84 Groupement agricole d’exploitation en commun (GAEC) de la Ségaude v Council and Commission [1987] ECR 123......................................................................513 Case 66/85 Deborah Lawrie-Blum v Land Baden-Württemberg [1986] ECR 2121............................................................................................................297, 298, 302 Case 79/85 DHM Segers v Bestuur van de Bedrijfsvereniging voor Bank- en Verzekeringswezen, Groothandel en Vrije Beroepen [1986] ECR 2375...................................................................................65, 68, 88, 119, 129, 176, 177 Case 139/85 RH Kempf v Staatssecretaris van Justitie [1986] ECR 1741...........................298 Joined Cases 201/85 and 202/85 Klensch v Secrétaire d’Etat [1986] ECR 3477...................15 Case 314/85 Foto-Frost v Hauptzollamt Lübeck-Ost [1987] ECR 4199.................12, 13, 546 Case 316/85 Centre public d’aide sociale de Courcelles v Marie-Christine Lebon [1987] ECR 2811................................................................................................................348 Case 39/86 Sylvie Lair v Universität Hannover [1988] ECR 3161................ 69, 109, 296, 298, 301, 302, 303, 304, 310, 311, 328, 345, 346, 347, 349, 527, 533, 537, 539, 550, 558 Case C–62/86 AKZO Chemie BV v Commission [1991] ECR I–3359...............................226 Case 80/86 Kolpinghuis Nijmegen [1987] ECR 3969........................................ 12, 21, 22, 456 Case 120/86 Mulder v Minister van Landbouw en Visserij [1988] ECR 2321..............13, 355 Joined Cases 138/86 and 139/86 Direct Cosmetics Ltd and Laughtons Photographs Ltd v Commissioners of Customs & Excise [1988] ECR 3937................ 384, 385, 386, 387 Case 197/86 Brown v Secretary of State for Scotland [1988] ECR 3205............................................................................................ 171, 299, 301, 304, 347 Joined Cases 46/87 and 227/88 Hoechst v Commission [1989] ECR 2859..........................13 Case 81/87 R v HM Treasury and Commissioners of Inland Revenue ex p Daily Mail and General Trust plc [1988] ECR 5483............107, 135, 199, 229, 337, 522, 524, 533, 539 Case 94/87 Commission v Germany [1989] ECR I–175......................................................515 Case C–142/87 Belgium v Commission [1990] ECR I–959................................................516 Case 186/87 Cowan v Trésor Public [1989] ECR 195............................................................16

576  Table of Cases Case 189/87 Athanasios Kalfelis v Bankhaus Schröder, Münchmeyer, Hengst & Co et al [1988] ECR 5565................................................................ 244, 265, 283, 284, 285, 286 Case 305/87 Commission v Greece [1989] ECR 1461.........................................................495 Case 374/87 Orkem v Commission [1989] ECR 3283...........................................................13 Case 382/87 R Buet and Educational Business Services (EBS) v Ministère public [1989] ECR 1235..................................................................................................................53 Case 5/88 Hubert Wachauf v Bundesamt für Ernährung und Forstwirtschaft [1989] ECR 2609................................................................................................................356 Case 70/88 European Parliament v Council of the European Communities [1990] ECR I–2041..............................................................................................................12 Case 130/88 C Van de Bijl v Staatssecretaris van Economische Zaken [1989] ECR 3039........................................................................................................................67, 69 Case 145/88 Torfaen Borough Council v B & Q plc [1989] ECR I–3851................13, 52, 462 Joined Cases 306/88; C–304/90 and C–169/91 Stoke-on-Trent City Council and Norwich City Council v B & Q plc [1992] ECR I–6457.............................52, 211, 462 Case 362/88 GB-INNO-BM v Confédération du commerce luxembourgeois [1990] ECR I–667..............................................................................................................460 Case C–61/89 Bouchoucha [1990] ECR I–3551.................................................. 109, 176, 326, 350, 351, 481, 548 Case C–106/89 Marleasing SA v La Comercial Internacional de Alimentacion SA [1990] ECR I–4135........................................................................................ 5, 215, 369, 377 Case C–154/89 Commission v France [1991] ECR I–659.....................................................53 Case C–189/89 Spagl [1990] ECR I–4539...............................................................................13 Case C–213/89 R v Secretary of State for Transport ex p Factortame Ltd et al [1990] ECR I–2433........................................................................................................21, 58 Case C–221/89 R v Secretary of State for Transport ex p Factortame Ltd et al [1991] ECR I–3905...................................................................58, 70, 84, 112, 170, 228, 368 Case C–246/89 Commission v United Kingdom [1991] ECR I–4585................................112 Case C–260/89 Elliniki Radiophonia Tiléorassi AE (ERT) and Panellinia Omospondia Syllogon Prossopikou v Dimotiki Etairia Pliroforissis and Sotirios Kouvelas and Nicolaos Avdellas et al [1991] ECR I–2925........................................................................10 Case C–288/89 Stichting Collectieve Antennevoorziening Gouda v Commissariaat voor de Media [1991] ECR I–4007.......................................................................57, 66, 181 Case C–292/89 R v Immigration Appeal Tribunal ex p Antonissen [1991] ECR I–745..........................................................................................................................348 Case C–312/89 Union départementale des syndicats CGT de l’Aisne v SIDEF Conforama, Société Arts et Meubles and Société Jima [1991] ECR I–997.............211, 462 Case C–332/89 André Marchandise, Jean-Marie Chapuis and SA Trafitex [1991] ECR I–1027........................................................................................................................211 Case 344/89 Bettray v Staatssecretaris van Justitie [1989] ECR 1621............................................................................................ 297, 298, 301, 304, 310 Case C–353/89 Commission v Netherlands [1991] ECR I–4069............................57, 66, 181 Case C–357/89 VJM Raulin v Minister van Onderwijs en Wetenschapen [1992] ECR I–1027................................................................................................ 298, 302, 303, 304 Joined Cases C–1/90 and C–176/90 Aragonesa de Publicidad Exterior SA and Publivia SAE v Departamento de Sanidad y Seguridad Social de la Generalitat de Cataluña [1991] ECR I–4151............................................................................................................206

Table of Cases  577 Case C–3/90 Bernini v Minister van Onderwijs en Wetenschappen [1992] ECR I–1071........................................................................................................................298 Case C–76/90 Manfred Säger v Dennemeyer & Co Ltd [1991] ECR I–4221......................460 Case C–204/90 Hanns-Martin Bachmann v Belgium [1992] ECR I–249...................460, 486 Case C‒343/90 Lourenço Dias v Director da Alfandega do Porto [1992] ECR I‒4673........................................................................................................................526 Case C–354/90 Fédération Nationale du Commerce Extérieur des Produits Alimentaires and Syndicat National des Négociants et Transformateurs de Saumon v French Republic [1991] ECR I–5505......................................................498, 503 Case C–369/90 Mario Vicente Micheletti et al v Delegación del Gobierno en Cantabria [1992] ECR I–4239........................................................... 331, 332, 334, 335, 336 Case C–370/90 R v Immigration Appeal Tribunal and Surinder Singh ex p Secretary of State for the Home Department [1992] ECR I–4265......................... 304, 305, 317, 338, 339, 340, 347, 350, 360, 522, 524, 548 Joined Cases C–31/91 to C–44/91 SpA Alois Lageder et al v Amministrazione delle Finanze dello Stato [1993] ECR I–1761............................................................................356 Case C–83/91 Wienand Meilicke v ADV/ORGA F A Meyer AG [1992] ECR I–4871..................................................................................................................19, 526 Case C–148/91 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media [1993] ECR I–487........................... 57, 64, 66, 67, 68, 69, 70, 71, 78, 79, 80, 81, 87, 93, 109, 171, 181, 208, 248, 308, 470, 524, 532 Case C–168/91 Christos Konstantinidis v Stadt Altensteig-Standesamt and Landratsamt Calw Ordnungsamt [1993] ECR I–1191....................................................337 Case C–211/91 Commission v Belgium [1992] ECR I–6757 (Flemish Cable Network case)...............................................................................64, 71, 72, 73, 82, 208, 470 Joined Cases C–267/91 and C–268/91 Bernard Keck and Daniel Mithouard [1993] ECR I–6097...................................................................................... 52, 209, 328, 463 Case C–8/92 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas [1993] ECR I–779....................................................................... 27, 109, 174, 204, 208, 225, 226, 474, 526, 532, 537, 541, 558 Case C–19/92 Dieter Kraus v Land Baden-Württemberg [1993] ECR I–1663........................................................................................................109, 460, 461 Case C–125/92 Mulox IBC [1993] ECR I–4075...................................................................264 Joined Cases C–278–280/92 Spain v Commission [1994] ECR I–4103..............................516 Case C–315/92 Verband Sozialer Wettbewerb eV v Clinique Laboratories SNC et Estée Lauder Cosmetics GmbH [1994] ECR I–317............................................358 Case C–334/92 Teodoro Wagner Miret v Fondo de Garantía Salarial [1993] ECR I–6911................................................................................................................222, 377 Case C–387/92 Banco de Crédito Industrial SA (now Banco Exterior de España SA) v Ayuntamiento de Valencia [1994] ECR I–877..................................................................503 Case C–404/92P X v Commission [1994] ECR I–4737 ........................................................13 Case C–406/92 The owners of the cargo lately laden on board the ship Tatry v The owners of the ship Maciej Rataj (The Tatry) [1994] ECR I–5439.... 266, 267, 270, 274 Case C–23/93 TV10 SA v Commissariaat voor de Media [1994] ECR I–4795...................................................26, 57, 58, 59, 61, 64, 66, 68, 69, 70, 71, 73, 78, 79, 80, 81, 86, 87, 93, 109, 171, 177, 180, 181, 184, 185, 189, 208, 209, 226, 308, 375, 398, 470, 471, 524, 525, 542, 558

578  Table of Cases Joined Cases C–46/93 and C–48/93 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and R v Secretary of State for Transport ex p Factortame Ltd et al [1996] ECR I–1029......................................................................................21, 215, 568 Joined Cases C–133/93; C–300/93 and C–362/93 Antonio Crispoltoni v Fattoria Autonoma Tabacchi and Giuseppe Natale and Antonio Pontillo v Donatab Srl [1994] ECR I–4863..............................................................................................................21 Case C–143/93 Gebroeders van Es Douane Agenten BV v Inspecteur der Invoerrechten en Accijnzen [1996] ECR I–431........................................................................................355 Case C–384/93 Alpine Investments BV v Minister van Financiën [1995] ECR I–1141.....184 Case C–392/93 R v Her Majesty‘s Treasury ex p British Telecommunications plc [1996] ECR I–1631....................................................................................................212, 216 Case C–415/93 Union Royale Belge des Sociétés de Football Association ASBL et al v Jean-Marc Bosman et al [1995] ECR I–4921.........................................298, 318, 328 Case C–441/93 Panagis Pafitis et al v Trapeza Kentrikis Ellados AE et al [1996] ECR I–1347..................................................................21, 108, 113, 141, 172, 173, 217, 530, 532, 542, 546, 547, 554, 560, 568 Case C–470/93 Verein gegen Unwesen in Handel und Gewerbe Köln eV v Mars GmbH [1995] ECR I–1923..................................................................................................53 Case C–479/93 Andrea Francovich v Italian Republic [1995] ECR I–3843........................216 Case C–484/93 Svensson and Gustavsson v Minstre du Logement et de l’Urbanisme [1995] ECR I–3955............................................................................................................184 Case C–4/94 BLP Group plc v Commissioners of Customs & Excise [1995] ECR I–983..........................................................................................................................395 Case C–13/94 P v S and Cornwall County Council [1996] ECR I–2143..............................15 Case C–55/94 Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano [1995] ECR I–4165................................................................. 66, 160, 183, 318, 361 Case C–122/94 Commission v Council [1996] ECR I–881.................................................513 Case C–151/94 Commission v Luxembourg [1995] ECR I–3699 (Biehl II).......................454 Case C–206/94 Brennet AG v Vittorio Paletta [1996] ECR I–2357............................103, 309, 310, 473, 542, 558 Case C–222/94 Commission v UK [1996] ECR I–4025......................65, 72, 74, 75, 77, 78, 80 Case C–241/94 France v Commission [1996] ECR I–4551.................................................511 Joined Cases C–283/94 Denkavit Internationaal BV, C–291/94 Vitic Amsterdam BV and C–292/94 Voormeer BV v Bundesamt für Finanzen [1996] ECR I–5063...............471 Joined Cases C–286/94, C–340/95, C–401/95 and C–47/96 Garage Molenheide BVBA (C-286/94), Peter Schepens (C-340/95), Bureau Rik Decan-Business Research & Development NV (BRD) (C-401/95) and Sanders BVBA (C-47/96) v Belgische Staat [1997] ECR I–7281...................................................................................................386 Case C–3/95 Reisebüro Broede v Gerd Sanker [1996] ECR I–6511......................................50 Case C–11/95 Commission v Belgium [1996] ECR I–4115 (Belgian Cable Network)....................................................................... 65, 72, 73, 74, 76, 77, 78, 80, 86, 548 Case C–24/95 Rheinland-Pfalz v Alcan [1997] ECR I–1591...............................250, 515, 516 Case C–28/95 A Leur-Bloem v Inspecteur der Belastingdienst/Ondernemingen Amsterdam 2 [1997] ECR I–4161..................................................... 113, 386, 404, 471, 472 Joined Cases C–34/95 to C–36/95 Konsumentombudsmannen (KO) v De Agostini (Svenska) Förlag AB (C-34/95) and TV-Shop i Sverige AB [1997] ECR I–3843.........................................................................65, 72, 78, 79, 80, 81, 82, 86, 548

Table of Cases  579 Case C–106/95 Mainschiffahrts-Genossenschaft eG (MSG) v Les Gravières Rhénanes sarl [1997] ECR I–911 (ECJ).................................... 244, 245, 266, 270, 282, 283 Case C–168/95 Criminal Proceedings against Luciano Arcaro [1996] ECR I–4705................................................................................................................175, 456 Case C–169/95 Spain v Commission [1997] ECR I–135.....................................................516 Case C–185/95P Baustahlgewebe v Commission [1998] ECR I–8417................................355 Case C–189/95 Harry Franzén [1997] ECR I–2471...............................................................50 Joined Cases C–192/95 to C–218/95 Société Comateb et al v Directeur Général des Douanes et Droits Indirects [1997] ECR I–165.................................................................21 Case C–222/95 Parodi v Banque H Albert de Bary [1997] ECR I–3899.............................184 Case C–244/95P Moskof AE v Ethnikos Organismos Kapnou (EOK) [1997] ECR I–6441........................................................................................................................171 Case C–250/95 Futura Participations SA and Singer v Administration des contributions [1997] ECR I–2471.................................................... 460, 461, 462, 468, 486 Case C–269/95 Francesco Benincasa v Dentalkit srl [1997] ECR I–3767...................264, 276 Case C–285/95 Suat Kol v Land Berlin [1997] ECR I–3069................................. 29, 296, 300, 324, 342, 522, 558 Case C–299/95 Kremzow v Austrian State [1997] ECR I–2629.............................................10 Case C–353/95P Tiercé Ladbroke v Commission [1997] ECR I–7007...............................511 Case C–368/95 Familiapress v Bauer Verlag [1997] ECR I–3689....................................13, 53 Case C–398/95 Syndesmos ton en Elladi Touristikon kai Taxidiotikon Grafeion v Ypourgos Ergasias [1997] ECR I–3091.............................................................................460 Case C–14/96 Criminal Proceedings against Paul Denuit [1997] ECR I–2785.................................................................................65, 72, 74, 75, 76, 77, 78, 86 Case C–36/96 Faik Günaydin, Hatice Günaydin, Günes Günaydin and Seda Günaydin v Freistaat Bayern [1997] ECR I–5143...................................... 29, 324, 342, 558 Case C–56/96 VT4 Ltd v Vlaamse Gemeenschap [1997] ECR I–3143.................................................................................65, 72, 76, 77, 78, 79, 83, 86 Joined Cases T–66/96 and T–221/97 Mellett v Court of Justice [1998] ECR SC II-1305..................................................................................................................516 Case C–85/96 María Martínez Sala v Freistaat Bayern [1998] ECR I–2691.................................................................................................. 16, 299, 311, 344 Case T–111/96 ITT Promedia NV v Commission [1998] ECR II-2937.............................227 Case C–118/96 Jessica Safir v Skattemyndigheten i Dalarnas Län, formerly Skattemyndigheten i Kopparbergs LΣn [1998] ECR I–1897............ 460, 473, 475, 476, 477 Case C–249/96 Grant v South-West Trains [1998] ECR I–621.............................................15 Case C–264/96 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty‘s Inspector of Taxes) [1998] ECR I–4695..................... 4, 113, 368, 386, 387, 397, 431, 435, 436, 469, 474, 482, 551 Case C–350/96 Clean Car Autoservice GmbH v Landeshauptmann von Wien [1998 ] ECR I–2521...........................................................................................................298 Case C–367/96 Alexandros Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [1998] ECR I–2843................................................................ 18, 20, 21, 89, 108, 109, 113, 114, 141, 172, 173, 205, 216, 221, 223, 297, 309, 313, 395, 470, 473, 522, 525, 527, 530, 531, 532, 539, 541, 543, 546, 553, 554, 556, 557, 558, 559, 560, 561, 563, 564, 568

580  Table of Cases Case C–1/97 Birden v Stadtgemeinde Bremen [1998] ECR I–7747....................................298 Joined Cases C–10/97 to C–22/97 Ministero delle Finanze v IN.CO.GE. ‘90 Srl et al [1998] ECR I–6307............................................................................................................452 Case C–33/97 Colim NV v Bigg’s Continent Noord NV [1999] ECR I–3175....................358 Case C–75/97 Belgium v Commission [1999] ECR I–3671................................511, 512, 516 Case C–90/97 Swaddling v Adjudication Officer [1999] ECR I–1075................................348 Joined Cases C–147/97 and 148/97 Deutsche Post AG v Gesellschaft für Zahlungssysteme mbH and Citicorp Kartenservice GmbH [2000] ECR I–825............................................87 Case C–212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I–1459................................................. xvi, xvii, 6, 19, 22, 23, 55, 58, 59, 61, 65, 86, 88, 89, 101, 102, 107, 108, 109, 111, 112, 113, 114, 115, 117, 118, 120, 123, 125, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 141, 142, 144, 171, 176, 177, 178, 181, 186, 188, 189, 199, 208, 210, 228, 229, 306, 308, 313, 329, 330, 335, 341, 386, 399, 401, 409, 422, 424, 447, 448, 455, 470, 473, 477, 480, 481, 482, 497, 508, 520, 522, 523, 524, 525, 527, 529, 531, 532, 533, 535, 539, 543, 548, 550, 551, 552, 559, 561, 564 Case T–266/97 Vlaamse Televisie Maatschapij NV v Commission [1999] ECR II-2329.....79 Case C–240/97 Spain v European Commission [1999] ECR I–6571..................................257 Case T–288/97 Regione Autonoma Friuli Venezia Giulia v Commission [2001] ECR II-1169.......................................................................................................................516 Case C–337/97 CPM Meeusen v Hoofddirectie van de Informatie Beheer Groep [1999] ECR I–3289............................................................................................................298 Case C–373/97 Dionysios Diamantis v Elliniko Dimosio (Greek State) and Organismos Ikonomikis Anasygkrotisis Epicheiriseon AE (OAE) [2000] ECR I–1705.......................................... 89, 108, 113, 114, 141, 171, 172, 173, 473, 522, 529, 530, 531, 532, 539, 546, 547, 554, 556, 560, 561 Case C–378/97 Wijsenbeek [1999] ECR I–6207..................................................................348 Joined Cases C–400/97 and C–402/97 Administración del Estado v Juntas Generales de Guipúzcoa [2000] ECR I–1073.........................................................................................510 Case C–404/97 Commission v Portugal [2000] ECR I–4897......................................515, 516 Case C–23/98 Staatssecretaris van Financiën v J Heerma [2000] ECR I–419....................430 Case C–55/98 Skatteministeriet v Bent Vestergaard [1999] ECR I–7641....................392, 474 Case C–78/98 Preston et al v Wolverhampton Healthcare NHS Trust et al [2000] ECR I–3201........................................................................................................................356 Joined Cases C-110 to C-147/98 Gabalfrisa and Others [2000] ECR I-1577.....................452 Case C–156/98 Germany v Commission [2000] ECR I–6857.............................................511 Case C–179/98 Belgian State v Fatna Mesbah [1999] ECR 1–7955....................................332 Case C‒190/98 Volker Graf v Filzmoser Maschinenbau GmbH [2000] ECR I‒493...........xix Case C–224/98 Marie-Nathalie D’Hoop v Office national de l’emploi [2002] ECR I–6191........................................................................................ 318, 347, 348, 349, 359 Case C–254/98 Schutzverband gegen unlauteren Wettbewerb v TK-Heimdienst Sass GmbH [2000] ECR I–151..................................................................................................356 Case C–281/98 Roman Angonese v Cassa di Risparmio di Bolzano SpA [2000] ECR I–4139..................................................................................................................16, 319 Case C–351/98 Spain v Commission [2002] ECR I–8031...................................................512 Case C–376/98 Germany v Parliament and Council [2000] ECR I–8419..........................330

Table of Cases  581 Case C–378/98 Commission v Belgium [2001] ECR I–5107..............................................515 Case C–390/98 HJ Banks & Co Ltd v The Coal Authority and Secretary of State for Trade and Industry [2001] ECR I–6117...................................................................494, 517 Joined Cases C–397/98 and C–410/98 Metallgesellschaft Ltd et al (C-397/98), Hoechst AG and Hoechst (UK) Ltd (C-410/98) v Commissioners of Inland Revenue and Her Majesty’s Attorney General [2001] ECR I–1727.................................397 Case C–405/98 Konsumentombudsmannen (KO) v Gourmet International Products AB (GIP) [2001] ECR I–1795...........................................................................................206 Case C–454/98 Schmeink & Cofreth AG & Co KG v Finanzamt Borken and Manfred Strobel v Finanzamt Esslingen [2000] ECR I–6973.................................................175, 452 Case C–480/98 Spain v Commission [2000] ECR I–8717...........................................502, 511 Case T–55/99 Confederación Española de Transporte de Mercancías (CETM) v Commission [2000] ECR II-3207.....................................................................................516 Case C–63/99 Queen, The v Secretary of State for the Home Department ex p W Gloszczuk and E Gloszczuk [2001] ECR I–6369.........................................................341 Case C–108/99 Cantor Fitzgerald International [2001] ECR I–7257.................................395 Case C–110/99 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas [2000] ECR I–11569...................... xv, xvi, xvii, 4, 5, 19, 23, 27, 61, 87, 89, 102, 103, 113, 129, 171, 173, 203, 204, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 219, 220, 223, 224, 225, 227, 228, 230, 239, 240, 242, 244, 245, 247, 248, 249, 255, 258, 263, 264, 305, 322, 324, 326, 356, 360, 362, 368, 375, 381, 383, 384, 385, 389, 396, 397, 400, 401, 403, 406, 412, 417, 428, 441, 442, 443, 448, 449, 450, 452, 469, 470, 474, 476, 491, 520, 521, 522, 524, 526, 527, 528, 529, 530, 531, 532, 533, 534, 536, 537, 538, 539, 540, 541, 542, 545, 546, 551, 559 Joined Cases C–122/99P and C–125/99P D and Sweden v Council [2001] ECR I–4319.......15 Joined Cases T–127/99, T–129/99 and T–148/99 Territorio Histórico de Álava Diputación Foral de Álava v Commission [2002] ECR II-1275......................................511 Case C–143/99 Adria-Wien Pipeline GmbH and Wietersdorfer & Peggauer Zementwerke GmbH v Finanzlandesdirektion für Kärnten [2001] ECR I–8365.......... 498, 502, 504, 511 Case C–157/99 BSM Geraets-Smits v Stichting Ziekenfonds VGZ and Peerbooms v Stichting CZ Groep Zorgverzekeringen [2001] ECR I–5473...........................................318 Case C–164/99 Portugaia Construções Lda [2002] ECR I–787............................................99 Case C–184/99 Rudy Grzelczyk v Centre public d’aide sociale d’Ottignies-Louvain-laNeuve [2001] ECR I–6193..........................................299, 304, 311, 345, 346, 347, 348, 359 Case C–192/99 R v Secretary of State for the Home Department ex p Kaur [2001] ECR I–1237....................................................................................................229, 331 Case C–268/99 Jany v Staatssecretaris van Justitie [2001] ECR I–8615..............................297 Joined Cases T–269/99, T–271/99 and T–272/99 Territorio Histórico de Guipúzcoaù Diputación Foral de Guipúzcoa v Commission [2002] ECR II-4217.............................511 Case C–334/99 Germany v Commission [2003] ECR I–1139.............................................516 Case C–382/99 Netherlands v Commission [2002] ECR I–5163................................511, 515 Case C–385/99 Müller-Fauré v Onderlinge Waarborgmaatschappij OZ Zorgverzekeringen UA, and between EEM van Riet and Onderlinge Waarborgmaatschappij ZAO Zorgverzekeringen [2003] ECR I–4509....................................221, 318 Case C–413/99 Baumbast and R v Secretary of State for the Home Department [2002] ECR I–7091....................................................................................................339, 344

582  Table of Cases Case C–459/99 Mouvement contre le racisme, l’antisémitisme et la xénophobie ASBL (MRAX) v Belgian State [2002] ECR I–6591.........................................292, 309, 341 Case C–482/99 France v Commission [2002] ECR I–4397.................................................501 Case C–53/00 Ferring SA v Agence centrale des organismes de sécurité sociale (ACOSS) [2001] ECR I–9067............................................................................................511 Case C–60/00 Mary Carpenter v Secretary of State for the Home Department [2002] ECR I–6279.............................................................................. 13, 309, 318, 320, 339 Case C–94/00 Roquette Frères v Commission [2002] ECR I–9011......................................13 Case C–112/00 Eugen Schmidberger, Internationale Transporte und Planzüge v Republik Österreich [2003] ECR I–5659........................................................ 18, 53, 54, 312 Case C–136/00 Rolf Dieter Danner [2002] ECR I–8147............................. 462, 473, 475, 477 Case C–137/00 Queen, The v The Competition Commission, Secretary of State for Trade and Industry and The Director General of Fair Trading ex p Milk Marque Ltd and National Farmers’ Union [2003] ECR I–7975....................................................205 Case C–208/00 Überseering BV v Nordic Construction Company Baumanagement GmbH [2002] ECR I–9919....................................................................... 110, 115, 127, 137, 141, 145, 178, 329, 480 Case C–209/00 Commission v Germany [2002] ECR I–11695...........................................515 Case C–246/00 Commission v Netherlands [2003] ECR I–7485................................324, 351 Joined Cases T–254/00, T–270/00 and T–277/00 Hotel Cipriani SpA v Commission [2008] ECR II-3269...................................................................................................511, 516 Case C–294/00 Deutsche Paracelsus Schulen für Naturheilverfahren GmbH v Kurt Gräbner [2002] ECR I–6515...............................................................................................50 Case T–318/00 Freistaat Thüringen v Commission [2005] ECR II-4179...........................517 Case C–324/00 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt [2002] ECR I–11779............................................... 113, 176, 369, 397, 401, 431, 435, 436, 474, 482 Case C–336/00 Republik Österreich v Martin Huber [2002] ECR I–7699.........................250 Case C–355/00 Freskot v Greece [2003] ECR I–5263..........................................................512 Case T–366/00 Scott SA v Commission [2007] ECR II-797................................................516 Case T–369/00 Départment du Loiret v Commission [2003] ECR II-1789.......................516 Case C–409/00 Spain v Commission [2003] ECR I–1487...........................................502, 512 Case C–436/00 X and Y v Riksskatteverket [2002] ECR I–10829....... 330, 369, 397, 431, 436 Case C–442/00 Rodríguez Caballero [2002] ECR I–11915...................................................14 Case C–457/00 Belgium v Commission [2003] ECR I–6931..............................................515 Case C–473/00 Cofidis v Fredout [2002] ECR I–10875.......................................................238 Case C‒87/01P Commission v CEMR [2003] ECR I‒4207................................................568 Case C–91/01 Italy v Commission [2004] ECR I–4355...............................................493, 506 Case C–109/01 Secretary of State for the Home Department v Hacene Akrich [2003] ECR I–9607............................................ 22, 29, 71, 79, 171, 226, 291, 292, 302, 304, 305, 306, 307, 308, 309, 311, 315, 317, 340, 341, 343, 344, 350, 360, 361, 398, 401, 470, 474, 476, 491, 506, 522, 525, 532, 533, 535, 536, 540, 543, 548, 565 Case T–109/01 Fleuren Compost BV v Commission [2004] ECR II-127...........................516 Case C–126/01 Ministre de l’Économie, des finances et de l’industrie v GEMO SA [2003] ECR I–13769..........................................................................................................502 Case C–159/01 Netherlands v Commission [2004] ECR I–4461................................511, 512

Table of Cases  583 Case C–167/01 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd [2003] ECR I–10155................................................22, 65, 110, 111, 112, 113, 115, 120, 127, 129, 132, 137, 141, 145, 176, 178, 329, 409, 447, 448, 480, 482, 497, 525, 532, 535, 547, 551 Case T–168/01 GlaxoSmithKline Services v Commission [2006] ECR II-2969.................225 Case T–198/01 Technische Glaswerke Ilmenau GmbH v Commission [2004] ECR II-2717.......................................................................................................................516 Case C–206/01 Arsenal Football Club plc v Matthew Reed [2002] ECR I–10273...............23 Joined Cases T–227/01 to T–229/01, T–265/01, T–266/01 and T–270/01 Territorio Historico de Alava - Diputación Foral de Alava et al v Commission of the European Communities, [2009] ECR II-3029...........................................................................501, 503 Case C–308/01 GIL Insurance Ltd et al v Commissioners of Customs & Excise [2004] ECR I–4777............................................................................ 502, 504, 505, 509, 512 Case C–320/01 Busch v Klinikum Neustadt GmbH & Co Betriebs-KB [2003] ECR I–2041........................................................................................................................327 Case C–322/01 Deutscher Apothekerverband eV v 0800 DocMorris NV, Jacques Waterval [2003] ECR I–14887..........................................................................................205 Case C–364/01 Heirs of H Barbier v Inspecteur van de Belastingdienst Particulieren/ Ondernemingen buitenland te Heerlen [2003] ECR I–15013................................176, 481 Case C–413/01 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst [2003] ECR I–13187............................................22, 26, 298, 299, 302, 303, 304, 329, 346, 361, 474, 476, 491, 533, 535, 539, 550 Case C–422/01 Skandia, Ramstedt v Rikskatteverket [2003] ECR I–6817.........473, 475, 477 Case C–464/01 Johann Gruber v Bay Wa AG [2005] ECR 1–439...............................248, 259 Case C–476/01 Kapper [2004] ECR I–5205.........................................................................351 Joined Cases C–482/01 and 493/01 Orfanopoulos and Oliveri v Land BadenWürttemberg [2004] ECR I–5257............................................................................310, 356 Joined Cases C–487/01 Gemeente Leusden v Staatssecretaris van Financiën and C–7/02 Holin Groep BV cs v Staatssecretaris van Financiën [2005] ECR I–5337................................................................................................................374, 470 Case C–491/01 R v Secretary of State ex p British American Tobacco [2002] ECR I–11453........................................................................................................................21 Case C–9/02 Hughes de Lasteyrie du Saillant v Ministère de l’Économie, des Finances et de l’Industrie [2004] ECR I–2409........................ 369, 392, 397, 401, 437, 470, 471, 473, 474, 478, 482, 483, 491 Case C–14/02 ATRAL SA v Belgian State [2003] ECR I–4431..............................................50 Case C–36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I–9609...... 18, 53, 176, 312, 447 Case C–66/02 Italy v Commission [2005] ECR I–10901.............................................511, 516 Case C–110/02 Commission v Council [2004] ECR I–6333...............................................514 Case C–116/02 Erich Gasser Gmbh v MISAT srl [2003] ECR I–14692................245, 246, 262, 264, 270, 272, 273, 274, 276, 280, 281, 287, 288, 557 Case C–138/02 Brian Francis Collins v Secretary of State for Work and Pensions [2004] ECR I–2703............................................................. 318, 334, 348, 359, 360 Case C–148/02 Garcia Avello v Belgian State [2003] ECR I–11613............................331, 338 Case C–159/02 Turner v Grovit [2004] ECR I–3565.................................. 245, 246, 269, 270, 271, 272, 273, 274, 276, 287

584  Table of Cases Case C–174/02 Streekgeweest Westelijk Noord-Brabant v Staatsecretaris van Financiën [2005] ECR I–85...............................................................................................509 Case C–200/02 Kunqian Catherine Zhu, Man Lavette Chen v Secretary of State for the Home Department [2004] ECR I–9925...................xvi, xix, 22, 177, 228, 229, 315, 334, 335, 338, 339, 359, 360, 361, 470, 491, 527, 530 Case T–210/02 British Aggregates Association v Commission [2006] ECR II-2789...............................................................................................................511, 512 Case C–255/02 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise [2006] ECR I–1609.........................xv, xvi, xviii, xxii, xxiii, xxiv, 4, 5, 20, 23, 25, 27, 28, 31, 61, 88, 102, 129, 141, 171, 173, 174, 175, 181, 184, 185, 210, 219, 226, 227, 239, 240, 242, 244, 245, 247, 248, 249, 255, 258, 313, 324, 329, 366, 367, 368, 369, 371, 372, 374, 375, 376, 378, 379, 381, 383, 385, 386, 387, 388, 389, 390, 391, 392, 393, 395, 396, 397, 398, 399, 403, 404, 407, 412, 417, 418, 419, 420, 422, 423, 424, 427, 428, 429, 430, 431, 433, 441, 442, 443, 448, 449, 452, 453, 455, 465, 469, 470, 471, 472, 473, 477, 480, 487, 488, 489, 490, 491, 494, 497, 498, 499, 500, 502, 507, 508, 522, 523, 524, 525, 527, 528, 529, 530, 531, 532, 533, 534, 535, 536, 537, 538, 539, 540, 541, 542, 543, 544, 545, 546, 548, 555, 556, 557, 559, 562, 565, 568 Case C–262/02 Commission v France [2004] ECR I–6569.................................................176 Case C–276/02 Spain v Commission [2004] ECR I–8091...........................................498, 502 Case C–281/02 Andrew Owusu v NB Jackson, trading as ‘Villa Holidays Bal-Inn Villas’ et al [2005] ECR I–1383.....................................267, 268, 269, 270, 271, 272, 274, 275, 276 Case C–319/02 Petri Manninem [2004] ECR I–7477..........................................................486 Case T–351/02 Deutsche Bahn v Commission [2006] ECR II-1047...................................501 Joined Cases C–387/02, C–391/02 and C–403/02 Berlusconi and Others [2005] ECR I–3565..........................................................................................................................12 Case C–395/02 TransportService [2004] ECR I–1991.........................................................452 Case C–408/02 José Antonio da Silva Carvalho...................................................................351 Case C–419/02 BUPA Hospitals Ltd and Goldsborough Developments Ltd v Commissioners of Customs & Excise [2006] ECR I–1685.......................xxiii, 88, 210, 397 Case C–442/02 Caixa Bank v Ministère de l‘Économie, des Finances et de l‘Industrie [2004] ECR I–8961....................................................................................................183, 184 Case C–456/02 Trojani v Centre public de l‘aide sociale de Bruxelles (CPAS) [2004] ECR I–7573.................................................................................... 297, 301, 359, 361 Case T–20/03 Kahla/Thüringen Porzellan GmbH v Commission [2008] ECR II-2305...............................................................................................................511, 516 Case C–32/03 I/S Fini H v Skatteministeriet [2005] ECR I–1599................. 27, 452, 539, 558 Case C–72/03 Carbonati Apuani [2004] ECR I–8027.........................................................348 Case C–88/03 Portugal v Commission [2006] ECR I–7115........................................503, 512 Case C–105/03 Pupino [2005] ECR I–5285.................................................................222, 223 Joined Cases C–128/03 and C–129/03 AEM SpA v Autorità per l’energia elettrica e per il gas [2005] ECR I–2861..........................................................................................512 Case T–146/03 Asociación de Empresarios de Estaciones de Servicio de la Comunidad Autónoma de Madrid and Federación Catalana de Estaciones de Servicio v Commission [2006] ECR II-98...................................................................512 Case C–147/03 Commission v Austria [2005] ECR I–5969.......................... 22, 330, 349, 491

Table of Cases  585 Case C–157/03 Commission v Spain [2005] ECR I–2911...........................................292, 341 Case C–172/03 Wolfgang Heiser v Finanzamt Innsbruck [2005] ECR I–1627........................................................................................................................511 Case T–189/03 ASM Brescia SpA v Commission [2009] ECR II-1831...............................503 Case C–209/03 R (on the application of Dany Bidar) v London Borough of Ealing and Secretary of State for Education and Skills [2005] ECR I–2119.........................................................250, 298, 304, 311, 318, 347, 348, 349, 359 Case C–210/03 Queen, The (on the application of Swedish Match AB and Swedish Match UK Ltd) v Secretary of State for Health [2004] ECR I–11893.............................330 Case C–223/03 University of Huddersfield Higher Education Corporation v Commissioners of Customs & Excise [2006] ECR I–1751.......................xxiii, 26, 210, 397 Case C–286/03 Hosse v Land Salzburg [2006] ECR I–1771................................................314 Case T–289/03 British United Provident Association Ltd (BUPA) et al v Commission [2008] ECR II-81.........................................................................................505 Joined Cases C–346/03 and C–529/03 Atzeni v Regione autonoma della Sardegna [2006] ECR I–1875............................................................................................................516 Joined Cases C–354/03, C–355/03 and C–484/03 Optigen Ltd, Fulcrum Electronics Ltd and Bond House Systems Ltd v Commissioners of Customs & Excise [2006] ECR I–483.............................................................................................. xxv, 388, 428 Case C–380/03 Germany v Parliament and Council [2006] ECR I–11573........................330 Case C–399/03 Commission v Council [2006] ECR I–5629...............................................514 Case C–403/03 Egon Schempp v Finanzamt München V [2005] ECR I–6421................................................................................................ 473, 474, 475, 477 Case C–411/03 Sevic Systems AG [2005] ECR I–10805.......................................................120 Case T–442/03 SIC - Sociedade Independente de Comunicaço SA v Commission [2008] ECR II-1161...........................................................................................................501 Case C–446/03 Marks & Spencer plc v David Halsey (Her Majesty‘s Inspector of Taxes) [2005] ECR I–10837.......................................................... 435, 437, 470, 474, 486 Case C‒452/03 Chancery Division RAL (Channel Islands) Ltd et al v Commissioners of Customs & Excise [2005] ECR I‒3947....................................................................... xxiii Case C–453/03 ABNA Ltd et al v Secretary of State for Health and Food Standards Agency [2005] ECR I–10423.............................................................................................355 Case C-456/03 Commission v Italy [2005] ECR I-5335......................................................456 Joined Cases C–485/03 to 490/03 Commission v Spain [2006] ECR I–11887...................503 Case C–513/03 Heirs of MEA van Hilten- van der Heijden v Inspecteur van de Belastingdienst/Particulieren/ Ondernemingen buitenland te Heerlen [2006] ECR I–1957............................................................................................474, 475, 478 Case C–515/03 Eichsfelder Schlachtbetrieb GmbH v Hauptzollamt Hamburg-Jonas [2005] ECR I–7355.....................................................212, 213, 214, 215, 530, 540, 542, 546 Case C–1/04 Susanne Staubitz-Schreiber [2006] ECR I–701...................... 138, 141, 151, 160 Case C–63/04 Centralan Property Ltd v Commissioners of Customs & Excise [2005] ECR I–11087.........................................................................................xxiii, 478, 489 Case C–96/04 Standesamt Stadt Niebüll v Stefan Grunkin, Dorothee Regina Paul [2006] ECR I–3561....................................................................................................338 Case C–144/04 Werner Mangold v Rüdiger Helm [2005] ECR I–9981...... xxi, 10, 13, 14, 15, 16, 17, 18, 20, 30, 54, 175, 176, 306, 357, 358, 466, 549 Case C–145/04 Spain v UK [2006] ECR I–7917...................................................................333

586  Table of Cases Case C–148/04 Unicredito Italiano SpA v Agenzia delle Entrate, Ufficio Genova 1 [2005] ECR I–11137.................................................................. 502, 503, 505, 506, 511, 516 Joined Cases C–151/04 and 152/04 Claude Nadin, Nadin-Lux SA and Jean-Pascal Durré [2005] ECR I–11203...............................................................................................316 Joined Cases C–154/04 and C–155/04 Alliance for Natural Health [2005] ECR I–6451......330 Joined Cases C‒158/04 and C‒159/04 Alfa Vita Vassilopoulos AE, formerly Trofo Super‒markets AE (C‒158/04) and Carrefour Marinopoulos AE (C‒159/04) v Elliniko Dimosio, Nomarkhiaki Aftodiikisi Ioanninon ‘[2006] ECR I‒8135..................xix Case C–196/04 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue [2006] ECR I–7995............ xvi, xvii, xviii, xxiii, 4, 5, 6, 22, 23, 27, 29, 61, 111, 112, 113, 128, 129, 130, 131, 132, 133, 135, 136, 141, 176, 177, 178, 179, 180, 181, 227, 228, 326, 368, 369, 372, 374, 377, 380, 387, 388, 393, 397, 398, 399, 402, 403, 404, 407, 408, 409, 410, 411, 412, 413, 417, 421, 422, 424, 427, 431, 432, 433, 435, 436, 437, 439, 440, 441, 444, 445, 446, 447, 448, 450, 454, 455, 457, 458, 459, 463, 464, 465, 466, 467, 468, 470, 471, 472, 474, 479, 481, 482, 483, 486, 490, 491, 494, 499, 502, 505, 507, 508, 520, 522, 524, 532, 534, 536, 537, 551, 552 Joined Cases T–211/04 and T–215/04 Government of Gibraltar and United Kingdom v Commission [2008] ECR II-3745.................................. 498, 502, 503, 510, 512 Case C–212/04 Konstantinos Adeneler et al v Ellinikos Organismos Galaktos [2006] ECR I–6057..............................................................................................26, 249, 327 Case C–222/04 Ministero dell‘Economia e delle Finanze v Cassa di Risparmio di Firenze SpA [2006] ECR I–289.....................................................................................511 Case C–227/04P Maria-Luise Lindorfer v Council of the European Union [2007] ECR I–6767..........................................................................................................................15 Joined Cases C–266/04 to C–270/04, C–276/04 and C–321/04 to C–325/04 Distribution Casino France SAS (formerly Nazairdis SAS) et al v Organic [2005] ECR I–9481............................................................................................................501 Case T–271/04 Citymo SA v Commission [2007] ECR II-1375......................... xxiv, 499, 508 Case C–290/04 KP Scorpio Konzertproduktionen GmbH v Finanzamt HamburgEimsbüttel [2006] ECR I–9461.........................................................................................454 Joined Cases C–295/04 to C–298/04 Manfredi v Lloyd Adriatico Assicurazioni SpA (C-295/04), Antonio Cannito v Fondiaria Sai SpA (C-296/04) and Nicolò Tricarico (C-297/04) and Pasqualina Murgolo (C-298/04) v Assitalia SpA [2006] ECR I–6619........................................................................................................................239 Case C–300/04 MG Eman and OB Sevinger v College van burgemeester en wethouders van Den Haag [2006] ECR I–8055...............................................................333 Case C–341/04 Eurofood IFCS Ltd–Enrico Bondi v Bank of America et al [2006] ECR I–3813................................................. 138, 139, 141, 151, 159, 166, 167, 179, 444, 445 Case C–347/04 Rewe Zentralfinanz eG v Finanzamt Köln-Mitte [2007] ECR I–2647................................................................................................................470, 474 Case C–354/04P Gestoras Pro Amnistía, Juan Mari Olano Olano and Julen Zelarain Errasti v Council of the European Union [2007] ECR I–1579.................................. xxv, 11 Case C–355/04P Segi, Araitz Zubimendi Izaga and Aritza Galarraga v Council of the European Union [2007] ECR I–1657...........................................................................11 Case C–368/04 Transalpine Ölleitung in Österreich GmbH et al v Finanzlandesdirektion für Tirol et al [2006] ECR I–9957.....................................................................493, 506, 517

Table of Cases  587 Case C–372/04 Queen, The (on the application of Yvonne Watts) v Bedford Primary Care Trust and Secretary of State for Health [2006] ECR I–4325...........................299, 318 Case C–374/04 Test Claimants in Class IV of the ACT Group Litigation [2006] ECR I–11673..............................................................................................................431, 477 Joined Cases C–393/04 Air Liquide Industries Belgium SA v Ville de Seraing and C–41/05 Air Liquide Industries Belgium SA v Province de Liège [2006] ECR I–5293................................................................................................ 494, 503, 504, 505 Joined Cases C–439/04 and C–440/04 Axel Kittel v Belgian State (C-439/04) and Belgian State v Recolta Recycling SPRL (C-440/04) [2006] ECR I–6161.......................388 Case C–441/04 A-Punkt Schmuckhandels GmbH v Claudia Schmidt [2006] ECR I–2093..........................................................................................................................53 Case C–452/04 Fidium Finanz AG v Bundesanstalt für Finanzdienstleistungsaufsicht [2006] ECR I–9521........................................................88, 89, 181, 184, 185, 186, 401, 447, 522, 523, 524, 526, 528, 531, 539, 541 Case C–456/04 Agip Petroli SpA v Capitaneria di porto di Siracusa et al [2006] ECR I–3395...........................................................26, 399, 455, 529, 530, 533, 536, 537, 539 Case C–470/04 N v Inspecteur van de Belastingdienst Oost/kantoor Almelo [2006] ECR I–7409.............................................470, 471, 474, 475, 478, 479, 480, 482, 491 Case C–492/04 Lasertec Gesellschaft für Stanzformen mbH v Finanzamt Emmendingen [2007] ECR I–3775...................................................................................447 Case C–513/04 Mark Kerckhaert and Bernadette Morres v Belgian State [2006] ECR 1-10967......................................................................................................................416 Case C–524/04 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue [2007] ECR I–2107................... 4, 5, 6, 29, 315, 397, 403, 404, 432, 433, 435, 438, 440, 441, 446, 447, 448, 457, 470, 472, 474, 477, 480, 482, 490, 491, 522, 524, 534, 537, 551 Case C–1/05 Yunying Jia v Migrationsverket [2007] ECR I–0001......................293, 305, 340 Case C–13/05 Sonia Chacón Navas v Eurest Colectividades SA [2006] ECR I–6467....14, 15 Case C–16/05 R (on the application of Veli Tum and Mehmet Dari) v Secretary of State for the Home Department [2007] ECR I–7415........................xxiv, 29, 324, 341, 342 Case C–101/05 Skatteverket v A [2007] ECR I–11531.........................................................477 Case C–103/05 Reisch v Kiesel [2006] ECR I–6827.............................................................244 Case C–110/05 Commission v Italy (pending)......................................................................52 Case C–142/05 Aklagaren v Percy Mickelsson and Joakim Roos (pending)........................52 Case T–163/05 Bundesverband deutscher Banken eV v European Commission, judgment of 3 March 2010, nyr................................................................................493, 502 Case C–178/05 Commission v Greece [2007] ECR I–4185.................................397, 400, 401 Case C–192/05 Tas-Hagen and Tas v Raadskamer WUBO van de Pensioen- en Uitkeringsraad [2006] ECR I–10451................................................................................359 Case C–201/05 Test Claimants in the CFC and Dividend Group Litigation v Commissioners of Inland Revenue [2008] ECR I–2875..................................................439 Case T–211/05 Italy v Commission [2009] ECR I-2777......................................................502 Case C–212/05 Hartmann v Freistaat Bayern [2007] ECR I–6303.....................................298 Case C–227/05 Halbritter v Freistaat Bayern [2006] ECR I–49..........................................351 Case C–231/05 Oy AA [2007] ECR I–6373..........................................................................470 Case C–232/05 Commission v France [2006] ECR I–10071...............................................515 Case C–274/05 Commission v Greece [2008] ECR I–7969.................................................351

588  Table of Cases Case C–277/05 Société thermale d‘Eugénie-les-Bains v Ministère de l‘Economie, des Finances et de l‘Industrie [2007] ECR I–6415...........................................................257 Case C–279/05 Vonk Dairy Products BV v Productschap Zuivel [2007] ECR I–239.................................................................................................. 209, 214, 215, 450 Case C–284/05 European Commission v Republic of Finland, [2009] ECR I-11705.........................................................................................................................11 Case C–287/05 D P W Hendrix v Raad van Bestuur van het Uitvoeringsinstituut Werknemersverzekeringen [2007] ECR I–6909...............................................................359 Case C–291/05 Minister voor Vreemdelingenzaken en Integratie v RNG Eind [2007] ECR I–10719..............................................................................................................309, 350 Case C–298/05 Columbus Container Services BVBA & Co v Finanzamt BielefeldInnenstadt [2007] ECR I–10451....................................................... 407, 414, 415, 435, 445 Case C–303/05 Advocaten voor de Wereld VZW v Leden van de Ministerraad [2007] ECR I–3633..........................................................................................................................11 Case C–321/05 Hans Markus Kofoed v Skatteministeriet [2007] ECR I–5795............................................ xvi, xxiii, 20, 25, 28, 29, 30, 31, 171, 175, 176, 190, 369, 370, 371, 383, 388, 389, 399, 401, 407, 421, 422, 455, 456, 519, 521, 522, 532, 537, 539, 561, 562 Case C–340/05 Kremer [2006] ECR I–98.............................................................................351 Case C–341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet, Svenska Byggnadsarbetareförbundets avdelning 1, Byggettan and Svenska Elektrikerförbundet [2007] ECR I–11767.................... 21, 22, 53, 54, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 319, 327, 525 Case C–387/05 European Commission v Italian Republic, judgment of 15 December 2009, nyr...............................................................................................................................11 Joined Cases C–402/05P and C–415/05P Kadi and Al Barakaat v Council [2008] ECR I–6351....................................................................................................................10, 12 Case C–411/05 Félix Palacios de la Villa v Cortefiel Servicios SA [2007] ECR I–8531..............................................................................................................14, 15, 16 Case C–438/05 International Transport Workers’ Federation and Finnish Seamen’s Union v Viking Line ABP and OU Viking Line Eesti [2007] ECR I–10779.................................................................................... 53, 54, 94, 104, 319, 327 Joined Cases C–11/06 and C–12/06 Rhiannon Morgan v Bezirksregierung Köln (C–11/06) and Iris Bucher v Landrat des Kreises Düren (C–12/06) [2007] ECR I–9161...................359 Case C–98/06 Freeport plc v Olle Arnoldsson [2007] ECR I–8319 (ECJ)........................................................................................... 244, 256, 265, 284, 285, 286 Case T–170/06 Alrosa Co Ltd v European Commission [2007] ECR II-2601....................257 Case C–194/06 Staatssecretaris van Financiën v Orange European Smallcap Fund NV [2008] ECR I–3747............................................................................................477 Case C–199/06 Centre d’exportation du livre français (CELF), Ministre de la Culture et de la Communication v Société internationale de diffusion et d’Édition (SIDE) [2008] ECR I–469......................................................................................................503, 517 Case C–210/06 Cartesio Oktató és Szolgáltató bt [2008] ECR I–9641............xxiv, 22, 59, 60, 88, 111, 112, 120, 123, 132, 135, 136, 137, 145, 161, 178, 199, 225, 229, 329, 337, 448, 476, 477, 498, 508, 569 Case C–239/06 European Commission v Italian Republic, judgment of 15 December 2009, nyr........................................................................................................11

Table of Cases  589 Case C–244/06 Dynamic Medien Vertriebs GmbH v Avides Media AG [2008] ECR I–505......................................................................................................................53, 54 Case C–246/06 Josefa Velasco Navarro v Fondo de Garantia Salarial (Fogasa) [2008] ECR I–105................................................................................................................14 Case C–250/06 United Pan-Europe Communications Belgium SA et al v Belgian State [2007] ECR I–11135................................................................................. 65, 72, 82, 83 Case C–251/06 Firma ING AUER - Die Bausoftware GmbH v Finanzamt Freistadt Rohrbach Urfahr [2007] ECR I–9689...............................................................................499 Case C–267/06 Tadao Maruko v Versorgungsanstalt der deutschen Bühnen [2008] ECR I–1757..........................................................................................................................15 Case C–286/06 Commission v Spain [2008] ECR I–8025...................................................351 Case C–294/06 Queen, The (on the application of Payir, Akyuz, Ozturk) v Secretary of State for the Home Department [2008] ECR I–203....................................................342 Case C–309/06 Marks & Spencer plc v Her Majesty‘s Commissioners of Customs and Excise [2008] ECR I–2283..........................................................................................452 Case C–311/06 Consiglio Nazionale degli Ingegneri v Ministero della Giustizia, Marco Cavallera (CNDI) [2009] ECR I–415.............................................. xix, xx, xxiv, 401 Case C–319/06 Commission v Luxembourg [2008] ECR I–4323.............................93, 98, 99 Joined Cases C–329/06 and C–343/06 Arthur Wiedemann v Land Baden-Württemberg (C–329/06) and Peter Funk v Stadt Chemnitz (C–343/06) [2008] ECR I–4635................................................................................................................351, 352 Joined Cases C–341/06P and C–342/06P Chronopost SA and La Poste v Union française de l’express (UFEX) et al [2008] ECR I–477....................................................503 Case C–346/06 Dirk Rüffert v Land Niedersachsen [2008] ECR I–1989.......... 93, 96, 97, 100 Case C–353/06 Stefan Grunkin and Dorothee Regina Paul (Grunkin Paul) [2008] ECR I–7639........................................................................................................................338 Case T–369/06 Holland Malt BV v Commission, [2009] ECR II-3313..............................501 Case C–390/06 Nuova Agricast Srl v Ministero delle Attività Produttive [2008] ECR I–2577.........................................................................................................................506, 507 Case C–399/06P Faraj Hassan v Council of the European Union and European Commission, judgment of 3 December 2009.....................................................................11 Case C–403/06P Chafiq Ayadi v Council of the European Union,[2009] ECR I-11393.........................................................................................................................11 Case C–414/06 Lidl Belgium GmbH & Co KG v Finanzamt Heilbronn [2008] ECR I–3601........................................................................................................................482 Case C–419/06 Commission v Greece [2008] ECR I–27.............................................515, 516 Case C–425/06 Ministero dell’Economia e delle Finanze v Part Service Srl [2008] ECR I–897................................................. xxiv, 227, 371, 375, 388, 392, 393, 397, 398, 407, 417, 420, 422, 423, 425, 427, 429, 443, 470, 471, 474, 476, 482, 520, 530, 534, 537, 538, 541, 542, 548 Case C–427/06 Birgit Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH [2008] ECR I–7245.............................................................. 13, 16, 17, 357, 358, 466 Case C–441/06 Commission v France [2007] ECR I–8887.................................................503 Case C–487/06P British Aggregates Association v Commission of the European Communities [2008] ECR I–10505.......................................................... 502, 503, 511, 512 Joined Cases C–501, 513, 515 and 519/06P GlaxoSmithKline Services Unlimited et al v Commission et al, [2009] ECR I-9291..............................................................................225

590  Table of Cases Case C–520/06 Stringer and Others v Her Majesty‘s Revenue and Customs, [2009] ECR I-179.............................................................................................................................20 Case C–521/06P Athinaïki Techniki AE v Commission of the European Communities [2008] ECR I–5829............................................................................................................507 Case C–527/06 RHH Renneberg v Staatssecretaris van Financiën [2008] ECR I–7735........................................................................................................................391 Case C–1/07 Frank Weber [2008] ECR I–8571....................................................................352 Case C–25/07 Alicja Sosnowska v Dyrektor Izby Skarbowej we Wroc awiu O rodek Zamiejscowy w Walbrzychu [2008] ECR I–5129.............................................................392 Case C–84/07 Commission v Greece [2008] ECR I–171.....................................................351 Case C–105/07 NV Lammers & Van Cleeff v Belgische Staat [2008] ECR I–0173.....438, 441 Case C–151/07 Theologos-Grigoris Khatzithanasis v Ipourgos Igeias kai Kinonikis Allilengiis [2008] ECR I–9013...........................................................................................351 Case C–158/07 Jacqueline Förster v Hoofddirectie van de Informatie Beheer Groep [2008] ECR I–8507.................. 299, 302, 303, 304, 310, 311, 318, 347, 348, 349, 359 Case C–162/07 Ampliscientifica Srl and Amplifin SpA v Ministero dell’Economia e delle Finanze and Agenzia delle Entrate [2008] ECR I–4019................. 392, 393, 401, 404, 499, 522, 533, 534, 538, 539 Case C–214/07 Commission v France [2008] ECR I–8357.................................................515 Case C–219/07 Nationale Raad van Dierenkwekers en Liefhebbers VZW, Andibel VZW v Belgische Staat [2008] ECR I–4475........................................................................53 Case C‒260/07 Pedro IV Servicios SL v Total España SA [2009] ECR I‒2437..................xxiv Case C–303/07 Proceedings brought by Aberdeen Property Fininvest Alpha Oy [2009] ECR I 5145.............................................................................................................498 Case C–330/07 Jobra Vermögensverwaltungs-Gesellschaft mbH v Finanzamt Amstetten Melk Scheibbs [2008] ECR I–9099.................................................................440 Case C–334/07P Commission v Freistaat Sachsen [2008] ECR I–9465..............................517 Case C–339/07 Christopher Seagon v Deko Marty Belgium NV [2009] ECR I–767.........158 Joined Cases C–378/07 and C–380/07 Angelidaki and Others v Nomarkhiaki Aftodiikisi Rethimnis and Another, judgment of 23 April 2009.......................................12 Case C–388/07 Incorporated Trustees of the National Council on Ageing (Age Concern England) v Secretary of State for Business, Enterprise and Regulatory Reform [2009] ECR I–1569...............................................................................................299 Case C‒396/07 Mirja Juuri v Fazer Amica Oy [2008] ECR I‒8883.....xxiv Case C–465/07 Elgafaji v Staatssecretaris van Justitie [2009] ECR I–921...................291, 314 Case C–551/07 Deniz Sahin v Bundesminister für Inneres [2008] ECR I–10453..............321 Case C–555/07 Seda Kücükdeveci v Swedex GmbH & Co KG, judgment of 19 January 2010, nyr..............................................................................................11, 16, 466 Case C‒7/08 Har Vaessen Douane Service BV v Staatssecretaris van Financiën [2009] ECR I‒5581...........................................................................................................xxiv Joined Cases C–22/08 and C–23/08 Athanasios Vatsouras (C–22/08) and Josif Koupatantze (C–23/08) v Arbeitsgemeinschaft (ARGE) Nürnberg 900 [2009] ECR I–4585........................................................................................................................329 Case C–45/08 Spector Photo Group NV Chris Van Raemdonck v Commissie voor het Bank-, Financie- en Assurantiewezen, judgment of 23 December 2009, nyr.............11 Case C–46/08 Carmen Media Group Ltd v Land Schleswig-Holstein, judgment of 8 September 2010, nyr.........................................................................................................69

Table of Cases  591 Case C–78/08 Amministrazione delle Finanze, Agenzia delle Entrate v Paint Graphos Scarl....................................................................................................................................493 Case C–79/08 Adige Carni Scrl (in liquidation) v Amministrazione delle Finanze, Agenzia delle Entrat...........................................................................................................493 Case C–80/08 Ministero delle Finanze v Michele Franchetto.............................................493 Case C–101/08 Audiolux SA et al v Groupe Bruxelles Lambert SA (GBL) et al and Bertelsmann AG et al [2009] ECR I–9823............................................ 8, 322, 562, 563, 566 Case C–127/08 Blaise Baheten Metock et al v Minister for Justice, Equality and Law Reform [2008] ECR I–6241...........................................29, 31, 171, 291, 292, 293, 295, 296, 298, 304, 305, 306, 309, 311, 317, 321, 341, 360, 361 Case C–135/08 Janko Rottman v Freistaat Bayern, Opinion of 30 September 2009, judgment of 2 March 2010, nyr............................................... 229, 332, 333, 336, 352 Case C–168/08 Laszlo Hadadi (Hadady) v Csilla Marta Mesko married name Hadadi (Hadady) [2009] ECR I‒6871..................................................................... xxiv, 230 Case C–174/08 NCC Construction Danmark A/S v Skatteministeriet, [2009] ECR I-10567.............................................................................................................12 Case C–303/08 Metin Bozkurt v Land Baden-Württemberg, judgment of 22nd December 2010..........................................................300, 522, 532, 533, 535, 539, 559 Case C–352/08 Modehuis A Zwijnenburg BV, Opinion of 16 July 2009, nyr........................................................................................................ xxiii, 30, 227, 370, 379 Case C–376/08 Serrantoni Srl, sorzio stabile edili Scrl v Comune di Milano, judgment of 23 December 2009, nyr..................................................................................11 Case C‒434/08 Arnold und Johann Harms als Gesellschaft bürgerlichen Rechts v Freerk Heidinga, judgment of 20 May 2010, nyr............................................................xxiv Case C–451/08 Helmut Müller GmbH v Bundesanstalt für Immobilienaufgaben, ECJ Opinion of 17 November 2009, nyr..........................................................................227 Case T–27/09 Stella Kunststofftechnik v OHIM, [2009] ECR II-4481.................................11 Case C–103/09 HMRC v Weald Leasing Limited, Opinion of 26 October 2010, nyr...................................................................... xxiv, 371, 393, 522, 529, 530, 531, 533, 534, 535, 536, 538, 539, 540, 541, 544, 548 Case C–277/09 Comissioners for Her Majesty‘s Revenue & Customs v RBS Deutschland Holdings GmbH, Opinion of 30 September 2010, nyr........xxiv, 393, 522, 530, 532, 533, 534, 538, 541, 542, 548, 557 Case C–403/09 PPU Jasna Detiícek v Maurizio Sgueglia, judgment of 23 December 2009, nyr........................................................................................................11 Case C‒417/10 Ministero dell’Economia e delle Finanze; Agenzia delle Entrate v 3M Italia SpA [2010] OJ C288/23....................................................................................xxiv

listed alphabetically by case name Aberdeen Property Fininvest Alpha Oy (Case C–303/07) [2009] ECR I 5145...................498 ABNA Ltd et al v Secretary of State for Health and Food Standards Agency (Case C–453/03) [2005] ECR I–10423........................................................................................355 Adige Carni Scrl (in liquidation) v Amministrazione delle Finanze, Agenzia delle Entrat (Case C–79/08).......................................................................................................493 Administración del Estado v Juntas Generales de Guipúzcoa (Joined Cases C–400/97 and C–402/97) [2000] ECR I–1073..................................................................................510

592  Table of Cases Administration des douanes v Société anonmye Gondrand Frères and Société anonyme Garancini (Case 169/80) [1981] ECR 1931......................................................................xxv Adria-Wien Pipeline GmbH and Wietersdorfer & Peggauer Zementwerke GmbH v Finanzlandesdirektion für Kärnten (Case C–143/99) [2001] ECR I–8365................................................................................................ 498, 502, 504, 511 Advocaten voor de Wereld VZW v Leden van de Ministerraad (Case C–303/05) [2007] ECR I–3633..............................................................................................................11 AEM SpA v Autorità per l‘energia elettrica e per il gas (Joined Cases C–128/03 and C–129/03) [2005] ECR I–2861..........................................................................................512 Age Concern England (Incorporated Trustees of the National Council on Ageing) v Secretary of State for Business, Enterprise and Regulatory Reform (Case C–388/07) [2009] ECR I–1569............................................................................................................299 Agip Petroli SpA v Capitaneria di porto di Siracusa et al (Case C–456/04) [2006] ECR I–3395...........................................................26, 399, 455, 529, 530, 533, 536, 537, 539 Air Liquide Industries Belgium SA v Ville de Seraing and C–41/05 Air Liquide Industries Belgium SA v Province de Liège (Joined Cases C–393/04) [2006] ECR I–5293................................................................................................ 494, 503, 504, 505 Aklagaren v Percy Mickelsson and Joakim Roos (Case C–142/05) (pending).....................52 AKZO Chemie BV v Commission (Case C–62/86) [1991] ECR I–3359............................226 Alicja Sosnowska v Dyrektor Izby Skarbowej we Wroc awiu O rodek Zamiejscowy w Walbrzychu (Case C–25/07) [2008] ECR I–5129.............................................................392 Alliance for Natural Health (Joined Cases C–154/04 and C–155/04) [2005] ECR I–6451........................................................................................................................330 Alexandros Kefalas et al v Elliniko Dimosio (Greek State) and Organismos Oikonomikis Anasygkrotisis Epicheiriseon AE (OAE) (Case C–367/96) [1998] ECR I–2843.....................18, 20, 21, 89, 108, 109, 113, 114, 141, 172, 173, 205, 216, 221, 223, 297, 309, 313, 395, 470, 473, 522, 525, 527, 530, 531, 532, 539, 541, 543, 546, 553, 554, 556, 557, 558, 559, 560, 561, 563, 564, 568 Alfa Vita Vassilopoulos AE, formerly Trofo Super‒Markets AE (C‒158/04) and Carrefour Marinopoulos AE (C‒159/04) v Elliniko Dimosio, Nomarkhiaki Aftodiikisi Ioanninon (Joined Cases C‒158/04 and C‒159/04) [2006] ECR I‒8135.....xix Algera et al v Common Assembly of the European Coal and Steel Community (Joined Cases 7/56 and 3-7/57) [1957–58] ECR 39......................................................., 566 Alpine Investments BV v Minister van Financiën (Case C–384/93) [1995] ECR I–1141........................................................................................................................184 Alrosa Co Ltd v European Commission (Case T–170/06) [2007] ECR II-2601.................257 AM & S Europe Limited v Commission (Case 155/79) [1982] ECR 1575.............9, 237, 529 Amministrazione delle Finanze, Agenzia delle Entrate v Paint Graphos Scarl (Case C–78/08)..................................................................................................................493 Amministrazione delle Finanze dello Stato v SpA San Giorgio (Case 199/82) [1983] ECR 3595..............................................................................................................................21 Ampliscientifica Srl and Amplifin SpA v Ministero dell’Economia e delle Finanze and Agenzia delle Entrate (Case C–162/07) [2008] ECR I–4019................... 392, 393, 401, 404, 499, 522, 533, 534, 538, 539 André Marchandise, Jean-Marie Chapuis and SA Trafitex (Case C–332/89) [1991] ECR I–1027............................................................................................................211 Andrea Francovich v Italian Republic (Case C–479/93) [1995] ECR I–3843.....................216

Table of Cases  593 Angelidaki and Others v Nomarkhiaki Aftodiikisi Rethimnis and Another (Joined Cases C–378/07 and C–380/07), judgment of 23 April 2009...............................12 Angonese v Cassa di Risparmio di Bolzano SpA (Case C–281/98) [2000] ECR I–4139..................................................................................................................16, 319 ex p Antonissen (Case C–292/89) [1991] ECR I–745..........................................................348 A-Punkt Schmuckhandels GmbH v Claudia Schmidt (Case C–441/04) [2006] ECR I–2093..........................................................................................................................53 Aragonesa de Publicidad Exterior SA and Publivia SAE v Departamento de Sanidad y Seguridad Social de la Generalitat de Cataluña (Joined Cases C–1/90 and C–176/90) [1991] ECR I–4151............................................................................................................206 Arcaro (Case C–168/95) [1996] ECR I–4705...............................................................175, 456 Arnold und Johann Harms als Gesellschaft bürgerlichen Rechts v Freerk Heidinga (Case C‒434/08), judgment of 20 May 20120, nyr.........................................................xxiv Arsenal Football Club plc v Matthew Reed (Case C–206/01) [2002] ECR I–10273........................................................................................................................23 Arthur Wiedemann v Land Baden-Württemberg (C–329/06) and Peter Funk v Stadt Chemnitz (C–343/06) (Joined Cases C–329/06 and C–343/06) [2008] ECR I–4635................................................................................................................351, 352 ASM Brescia SpA v Commission (Case T–189/03) [2009] ECR II-1831............................503 Asociación de Empresarios de Estaciones de Servicio de la Comunidad Autónoma de Madrid and Federación Catalana de Estaciones de Servicio v Commission (Case T–146/03) [2006] ECR II-98...................................................................................512 Athanasios Vatsouras (C–22/08) and Josif Koupatantze (C–23/08) v Arbeitsgemeinschaft (ARGE) Nürnberg 900 (Joined Cases C–22/08 and C–23/08) [2009] ECR I–4585.......329 Athinaïki Techniki AE v Commission of the European Communities (Case C–521/06P) [2008] ECR I–5829............................................................................................................507 Atlanta Amsterdam BV v Produktschap Voor Vee en Vlees (Case 240/78) [1979] ECR 2137............................................................................................................................355 ATRAL SA v Belgian State (Case C–14/02) [2003] ECR I–4431...........................................50 Atzeni v Regione autonoma della Sardegna (Joined Cases C–346/03 and C–529/03) [2006] ECR I–1875............................................................................................................516 Audiolux SA et al v Groupe Bruxelles Lambert SA (GBL) et al and Bertelsmann AG et al (Case C–101/08) [2009] ECR I–9823........................................... 8, 322, 562, 563, 566 Autoteile Service GmbH v Pierre Malhé (Case 220/84) [1985] ECR 2267.................245, 283 Axel Kittel v Belgian State (C-439/04) and Belgian State v Recolta Recycling SPRL (C-440/04) (Joined Cases C–439/04 and C–440/04) [2006] ECR I–6161......................388 Ayadi v Council of the European Union (Case C–403/06P), [2009] ECR I-11393..............11 Bachmann v Belgium (Case C–204/90) [1992] ECR I–249.........................................460, 486 Banco de Crédito Industrial SA (now Banco Exterior de España SA) v Ayuntamiento de Valencia (Case C–387/92) [1994] ECR I–877..............................................................503 Baumbast and R v Secretary of State for the Home Department (Case C–413/99) [2002] ECR I–7091....................................................................................................339, 344 Baustahlgewebe v Commission (Case C–185/95P) [1998] ECR I–8417.............................355 Becker (Case 8/81) [1982] ECR 53..................................................................................13, 456 Belgian Cable Network (Case C–11/95) [1996] ECR I–4115....................... 65, 72, 73, 74, 76, 77, 78, 80, 86, 548 Belgian State v Fatna Mesbah (Case C–179/98) [1999] ECR 1–7955.................................332

594  Table of Cases Belgium v Commission (Case C–142/87) [1990] ECR I–959.............................................516 Belgium v Commission (Case C–75/97) [1999] ECR I–3671.............................511, 512, 516 Belgium v Commission (Case C–457/00) [2003] ECR I–6931...........................................515 Benincasa v Dentalkit srl (Case C–269/95) [1997] ECR I–3767.................................264, 276 Berlusconi and Others (Joined Cases C–387/02, C–391/02 and C–403/02) [2005] ECR I–3565..........................................................................................................................12 Bernini v Minister van Onderwijs en Wetenschappen (Case C–3/90) [1992] ECR I–1071........................................................................................................................298 Bettray v Staatssecretaris van Justitie (Case 344/89) [1989] ECR 1621............................................................................................ 297, 298, 301, 304, 310 Bidar v London Borough of Ealing and Secretary of State for Education and Skills (Case C–209/03) [2005] ECR I–2119.............................................................. 250, 298, 304, 311, 318, 347, 348, 349, 359 Biehl II (Case C–151/94) [1995] ECR I–3699......................................................................454 Birden v Stadtgemeinde Bremen (Case C–1/97) [1998] ECR I–7747.................................298 Birgit Bartsch v Bosch und Siemens Hausgeräte (BSH) Altersfürsorge GmbH (Case C–427/06) [2008] ECR I–7245............................................. 13, 16, 17, 357, 358, 466 BLP Group plc v Commissioners of Customs & Excise (Case C–4/94) [1995] ECR I–983..........................................................................................................................395 Boriello v Alain Darras and Dominique Tostain (Case 95/84) [1986] ECR 2253............................................................................................................................205 Bouchereau (Case 30/77) [1977] ECR 1999.....................................................................98, 99 Bouchoucha (Case C–61/89) [1990] ECR I–3551.................109, 176, 326, 350, 351, 481, 548 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and R v Secretary of State for Transport ex p Factortame Ltd et al (Joined Cases C–46/93 and C–48/93) [1996] ECR I–1029..............................................................................................21, 215, 568 Brennet AG v Vittorio Paletta (Case C–206/94) [1996] ECR I–2357................................................................................ 103, 309, 310, 473, 542, 558 British Aggregates Association v Commission (Case T–210/02) [2006] ECR II-2789...............................................................................................................511, 512 British Aggregates Association v Commission of the European Communities (Case C–487/06P) [2008] ECR I–10505................................................... 502, 503, 511, 512 ex p British American Tobacco (Case C–491/01) [2002] ECR I–11453................................21 British United Provident Association Ltd (BUPA) et al v Commission (Case T–289/03) [2008] ECR II-81...................................................................................505 Brown v Secretary of State for Scotland (Case 197/86) [1988] ECR 3205............................................................................................ 171, 299, 301, 304, 347 BSM Geraets-Smits v Stichting Ziekenfonds VGZ and Peerbooms v Stichting CZ Groep Zorgverzekeringen (Case C–157/99) [2001] ECR I–5473....................................318 Buet and Educational Business Services (EBS) v Ministère public (Case 382/87) [1989] ECR 1235..................................................................................................................53 Bundesverband deutscher Banken eV v European Commission (Case T–163/05) judgement of 3 March 2010......................................................................................493, 502 BUPA Hospitals Ltd and Goldsborough Developments Ltd v Commissioners of Customs & Excise (Case C–419/02) [2006] ECR I–1685..........................xxiii, 88, 210, 397 Busch v Klinikum Neustadt GmbH & Co Betriebs-KB (Case C–320/01) [2003] ECR I–2041........................................................................................................................327

Table of Cases  595 Cadbury Schweppes plc, Cadbury Schweppes Overseas Ltd v Commissioners of Inland Revenue (Case C–196/04) [2006] ECR I–7995.............. xvi, xvii, xviii, xxiii, 4, 5, 6, 22, 23, 27, 29, 61, 111, 112, 113, 128, 129, 130, 131, 132, 133, 135, 136, 141, 176, 177, 178, 179, 180, 181, 227, 228, 326, 368, 369, 372, 374, 377, 380, 387, 388, 393, 397, 398, 399, 402, 403, 404, 407, 408, 409, 410, 411, 412, 413, 417, 421, 422, 424, 427, 431, 432, 433, 435, 436, 437, 439, 440, 441, 444, 445, 446, 447, 448, 450, 454, 455, 457, 458, 459, 463, 464, 465, 466, 467, 468, 470, 471, 472, 474, 479, 481, 482, 483, 486, 490, 491, 494, 499, 502, 505, 507, 508, 520, 522, 524, 532, 534, 536, 537, 551, 552 Caixa Bank v Ministère de l’Économie, des Finances et de l’Industrie (Case C–442/02) [2004] ECR I–8961....................................................................................................183, 184 Cantor Fitzgerald International (Case C–108/99) [2001] ECR I–7257..............................395 Carbonati Apuani (Case C–72/03) [2004] ECR I–8027......................................................348 Carmen Media Group Ltd v Land Schleswig-Holstein (Case C–46/08), judgment of 8 September 2010, nyr.....................................................................................................69 Carpenter v Secretary of State for the Home Department (Case C–60/00) [2002] ECR I–6279.......................................................................................... 13, 309, 318, 320, 339 Cartesio Oktató és Szolgáltató bt (Case C–210/06) [2008] ECR I–9641................................ xxiv, 22, 59, 60, 88, 111, 112, 120, 123, 132, 135, 136, 137, 145, 161, 178, 199, 225, 229, 329, 337, 448, 476, 477, 498, 508, 569 Centralan Property Ltd v Commissioners of Customs & Excise (Case C–63/04) [2005] ECR I–11087.........................................................................................xxiii, 478, 489 Centre d‘exportation du livre français (CELF), Ministre de la Culture et de la Communication v Société internationale de diffusion et d‘Édition (SIDE) (Case C–199/06) [2008] ECR I–469.........................................................................503, 517 Centros Ltd v Erhvervs- og Selskabsstyrelsen (Case C–212/97) [1999] ECR I–1459............................ xvi, xvii, 6, 19, 22, 23, 55, 58, 59, 61, 65, 86, 88, 89, 101, 102, 107, 108, 109, 111, 112, 113, 114, 115, 117, 118, 120, 123, 125, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 141, 142, 144, 171, 176, 177, 178, 181, 186, 188, 189, 199, 208, 210, 228, 229, 306, 308, 313, 329, 330, 335, 341, 386, 399, 401, 409, 422, 424, 447, 448, 455, 470, 473, 477, 480, 481, 482, 497, 508, 520, 522, 523, 524, 525, 527, 529, 531, 532, 533, 535, 539, 543, 548, 550, 551, 552, 559, 561, 564 Chancery Division RAL (Channel Islands) Ltd et al v Commissioners of Customs & Excise (Case C‒452/03) [2005] ECR I‒3947.............................................................. xxiii Christopher Seagon v Deko Marty Belgium NV (Case C–339/07) [2009] ECR I–767......158 Chronopost SA and La Poste v Union française de l’express (UFEX) et al (Joined Cases C–341/06P and C–342/06P) [2008] ECR I–477....................................................503 Clean Car Autoservice GmbH v Landeshauptmann von Wien (Case C–350/96) [1998] ECR I–2521............................................................................................................298 CILFIT and Lanificio di Gavardo SpA v Ministry of Health (Case 283/81) [1982] ECR 3415........................................................................................................ 9, 211, 230, 523 Cinéthèque SA et al v Fédération nationale des cinémas français (Joined Cases C–60/84 and C–61/84) [1985] ECR 2605.........................................................................460 Citymo SA v Commission (Case T–271/04) [2007] ECR II-1375..................... xxiv, 499, 508 Coenen et al v Sociaal-Economische Raad (Case 39/75) [1975] ECR 1547.........................69 Cofidis v Fredout (Case C–473/00) [2002] ECR I–10875....................................................238

596  Table of Cases Colim NV v Bigg’s Continent Noord NV (Case C–33/97) [1999] ECR I–3175.................358 Collins v Secretary of State for Work and Pensions (Case C–138/02) [2004] ECR I–2703........................................................................................ 318, 334, 348, 359, 360 Columbus Container Services BVBA & Co v Finanzamt Bielefeld-Innenstadt (Case C–298/05) [2007] ECR I–10451............................................. 407, 414, 415, 435, 445 Comissioners for Her Majesty’s Revenue & Customs v RBS Deutschland Holdings GmbH (Case C–277/09), Opinion of 30 September 2010, nyr..............xxiv, 393, 522, 530, 532, 533, 534, 538, 541, 542, 548, 557 Commission v Austria (Case C–147/03) [2005] ECR I–5969....................... 22, 330, 349, 491 Commission v Belgium (Case C–378/98) [2001] ECR I–5107...........................................515 Commission v CEMR (Case C‒87/01P) [2003] ECR I‒4207.............................................568 Commission v Council (Case C–122/94) [1996] ECR I–881..............................................513 Commission v Council (Case C–110/02) [2004] ECR I–6333............................................514 Commission v Council (Case C–399/03) [2006] ECR I–5629............................................514 Commission v France (Case C–154/89) [1991] ECR I–659..................................................53 Commission v France (Case C–262/02) [2004] ECR I–6569..............................................176 Commission v France (Case C–232/05) [2006] ECR I–10071............................................515 Commission v France (Case C–441/06) [2007] ECR I–8887..............................................503 Commission v France (Case C–214/07) [2008] ECR I–8357..............................................515 Commission v Freistaat Sachsen (Case C–334/07P) [2008] ECR I–9465...........................517 Commission v Germany (Case 70/72) [1973] ECR 813......................................................513 Commission v Germany (Case 178/84) [1987] ECR 1227....................................................53 Commission v Germany (Case 205/84) [1986] ECR 3755........................ 64, 67, 69, 184, 460 Commission v Germany (Case 94/87) [1989] ECR I–175...................................................515 Commission v Germany (Case C–209/00) [2002] ECR I–11695........................................515 Commission v Greece (Case 305/87) [1989] ECR 1461......................................................495 Commission v Greece (Case C–178/05) [2007] ECR I–4185..............................397, 400, 401 Commission v Greece (Case C–274/05) [2008] ECR I–7969..............................................351 Commission v Greece (Case C–419/06) [2008] ECR I–27..........................................515, 516 Commission v Greece (Case C–84/07) [2008] ECR I–171..................................................351 Commission v Italy (Case C-456/03) [2005] ECR I-5335….456 Commission v Italy (Case C–110/05) (pending)...................................................................52 Commission v Luxembourg (Case C–319/06) [2008] ECR I–4323..........................93, 98, 99 Commission v Netherlands (Case C–353/89) [1991] ECR I–4069.........................57, 66, 181 Commission v Netherlands (Case C–246/00) [2003] ECR I–7485.............................324, 351 Commission v Portugal (Case C–404/97) [2000] ECR I–4897...................................515, 516 Commission v Spain (Case C–157/03) [2005] ECR I–2911........................................292, 341 Commission v Spain (Joined Cases C–485/03 to 490/03) [2006] ECR I–11887................503 Commission v Spain (Case C–286/06) [2008] ECR I–8025................................................351 Commission v United Kingdom (Case 61/81) [1982] ECR 2601........................................238 Commission v United Kingdom (Case 40/82) [1982] ECR 2793........................................206 Commission v United Kingdom Case (C–246/89) [1991] ECR I–4585.............................112 Commission v UK (Case C–222/94) [1996] ECR I–4025...................65, 72, 74, 75, 77, 78, 80 Confederación Española de Transporte de Mercancías (CETM) v Commission (Case T–55/99) [2000] ECR II-3207.................................................................................516 Consiglio Nazionale degli Ingegneri v Ministero della Giustizia, Marco Cavallera (CNDI) (Case C–311/06) [2009] ECR I–415............................................. xix, xx, xxiv, 401

Table of Cases  597 Cowan v Trésor Public (Case 186/87) [1989] ECR 195.........................................................16 CPM Meeusen v Hoofddirectie van de Informatie Beheer Groep (Case C–337/97) [1999] ECR I–3289............................................................................................................298 Cremer v Bundesanstalt für landwirtschaftliche Marktordnung (Case 125/76) [1977] ECR 1593..............................174, 203, 204, 208, 428, 526, 527, 529, 531, 533, 539, 542, 550 Crispoltoni v Fattoria Autonoma Tabacchi and Giuseppe Natale and Antonio Pontillo v Donatab Srl (Joined Cases C–133/93; C–300/93 and C–362/93) [1994] ECR I–4863..........................................................................................................................21 D and Sweden v Council (Joined Cases C–122/99P and C–125/99P) [2001] ECR I–4319..........................................................................................................................15 D P W Hendrix v Raad van Bestuur van het Uitvoeringsinstituut Werknemersverzekeringen (Case C–287/05) [2007] ECR I–6909........................................................................359 ex p Daily Mail and General Trust plc (Case 81/87) [1988] ECR 5483.............. 107, 135, 199, 229, 337, 522, 524, 533, 539 Defrenne v SABENA (Case 43/75) [1976] ECR 455......................................................13, 522 DEKA Getreideprodukte GmbH & Co KG, iL v European Economic Community (Case 250/78) [1983] ECR 421..................................................................................248, 249 Deniz Sahin v Bundesminister für Inneres (Case C–551/07) [2008] ECR I–10453......................................................................................................................321 Denkavit Internationaal BV, Vitic Amsterdam BV and Voormeer BV v Bundesamt für Finanzen (Joined Cases C–283/94, C–291/94 and C–292/94) [1996] ECR I–5063.......471 Denuit (Case C–14/96) [1997] ECR I–2785..................................65, 72, 74, 75, 76, 77, 78, 86 Départment du Loiret v Commission (Case T–369/00) [2003] ECR II-1789....................516 Deutsche Bahn v Commission (Case T–351/02) [2006] ECR II-1047................................501 Deutsche Paracelsus Schulen für Naturheilverfahren GmbH v Kurt Gräbner (Case C–294/00) [2002] ECR I–6515.................................................................................50 Deutsche Post AG v Gesellschaft für Zahlungssysteme mbH and Citicorp Kartenservice GmbH (Joined Cases C–147/97 and 148/97) [2000] ECR I–825......................................87 Deutscher Apothekerverband eV v 0800 DocMorris NV, Jacques Waterval (Case C–322/01) [2003] ECR I–14887.............................................................................205 D’Hoop v Office national de l’emploi (Case C–224/98) [2002] ECR I–6191........................................................................................ 318, 347, 348, 349, 359 Dionysios Diamantis v Elliniko Dimosio (Greek State) and Organismos Ikonomikis Anasygkrotisis Epicheiriseon AE (OAE) (Case C–373/97) [2000] ECR I–1705.............................................. 89, 108, 113, 114, 141, 171, 172, 173, 473, 522, 529, 530, 531, 532, 539, 546, 547, 554, 556, 560, 561 Direct Cosmetics Ltd and Laughtons Photographs Ltd v Commissioners of Customs & Excise (Joined Cases 138/86 and 139/86) [1988] ECR 3937................... 384, 385, 386, 387 Distribution Casino France SAS (formerly Nazairdis SAS) et al v Organic (Joined Cases C–266/04 to C–270/04, C–276/04 and C–321/04 to C–325/04) [2005] ECR I–9481........................................................................................................................501 Duphar BV et al v The Netherlands State (Case 238/82) [1984] ECR 523.........................209 Dynamic Medien Vertriebs GmbH v Avides Media AG (Case C–244/06) [2008] ECR I–505......................................................................................................................53, 54 Effer v Kantner (Case 38/81) [1982] ECR 825.....................................................................264 Egon Schempp v Finanzamt München V (Case C–403/03) [2005] ECR I–6421................................................................................................ 473, 474, 475, 477

598  Table of Cases Eichsfelder Schlachtbetrieb GmbH v Hauptzollamt Hamburg-Jonas (Case C–515/03) [2005] ECR I–7355........................212, 213, 214, 215, 530, 540, 542, 546 Elgafaji v Staatssecretaris van Justitie (Case C–465/07) [2009] ECR I–921................291, 314 Elliniki Radiophonia Tiléorassi AE (ERT) and Panellinia Omospondia Syllogon Prossopikou v Dimotiki Etairia Pliroforissis and Sotirios Kouvelas and Nicolaos Avdellas et al (Case C–260/89) [1991] ECR I–2925...........................................................10 Emsland-Stärke GmbH v Hauptzollamt Hamburg-Jonas (Case C–110/99) [2000] ECR I–11569.............................. xv, xvi, xvii, 4, 5, 19, 23, 27, 61, 87, 89, 102, 103, 113, 129, 171, 173, 203, 204, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 219, 220, 223, 224, 225, 227, 228, 230, 239, 240, 242, 244, 245, 247, 248, 249, 255, 258, 263, 264, 305, 322, 324, 326, 356, 360, 362, 368, 375, 381, 383, 384, 385, 389, 396, 397, 400, 401, 403, 406, 412, 417, 428, 441, 442, 443, 448, 449, 450, 452, 469, 470, 474, 476, 491, 520, 521, 522, 524, 526, 527, 528, 529, 530, 531, 532, 533, 534, 536, 537, 538, 539, 540, 541, 542, 545, 546, 551, 559 Erich Gasser Gmbh v MISAT srl (Case C–116/02) [2003] ECR I–14692................245, 246, 262, 264, 270, 272, 273, 274, 276, 280, 281, 287, 288, 557 Eurofood IFCS Ltd–Enrico Bondi v Bank of America et al (Case C–341/04) [2006] ECR I–3813................................................. 138, 139, 141, 151, 159, 166, 167, 179, 444, 445 European Commission v Italian Republic (Case C–387/05), judgment of 15 December 2009, nyr........................................................................................................11 European Commission v Italian Republic (Case C–239/06), judgment of 15 December 2009, nyr........................................................................................................11 European Commission v Republic of Finland (Case C–284/05), judgment of 15 December 2009, nyr........................................................................................................11 European Parliament v Council of the European Communities (Case C–70/88) [1990] ECR I–2041..............................................................................................................12 ex p Factortame Ltd et al (Case C–213/89) [1990] ECR I–2433......................................21, 58 ex p Factortame Ltd et al (Case C–221/89) [1991] ECR I–3905............................... 58, 70, 84, 112, 170, 228, 368 Familiapress v Bauer Verlag (Case C–368/95) [1997] ECR I–3689.................................13, 53 Faraj Hassan v Council of the European Union and European Commission (Case C–399/06P), judgment of 3 December 2009............................................................11 Fédération Charbonnière de Belgique v High Authority (Case 8/55) [1954–56] ECR 245..................................................................................................................................9 Fédération Nationale du Commerce Extérieur des Produits Alimentaires and Syndicat National des Négociants et Transformateurs de Saumon v French Republic (Case C–354/90) [1991] ECR I–5505........................................................498, 503 Félix Palacios de la Villa v Cortefiel Servicios SA (Case C–411/05) [2007] ECR I–8531..............................................................................................................14, 15, 16 Ferring SA v Agence centrale des organismes de sécurité sociale (ACOSS) (Case C–53/00) [2001] ECR I–9067.................................................................................511 Fidium Finanz AG v Bundesanstalt für Finanzdienstleistungsaufsicht (Case C–452/04) [2006] ECR I–9521...................................88, 89, 181, 184, 185, 186, 401, 447, 522, 523, 524, 526, 528, 531, 539, 541 Firma ING AUER - Die Bausoftware GmbH v Finanzamt Freistadt Rohrbach Urfahr (Case C–251/06) [2007] ECR I–9689...................................................................499

Table of Cases  599 Flemish Cable Network case (Case C–211/91) [1992] ECR I–6757......................... 64, 71, 72, 73, 82, 208, 470 Fleuren Compost BV v Commission (Case T–109/01) [2004] ECR II-127........................516 Foglia v Novello (Case 104/79) [1980] ECR 745............................................ 19, 171, 243, 526 Foglia v Novello (Case 244/80) [1981] ECR 3045................................................................171 Forcheri v Belgian State and asbl Institut Supérieur de Sciences Humaines Appliquées Ecole Ouvrière Supérieure (Case 152/82) [1983] ECR 2323.............................................16 Foto-Frost v Hauptzollamt Lübeck-Ost (Case 314/85) [1987] ECR 4199..............12, 13, 546 Franca Ninni-Orasche v Bundesminister für Wissenschaft, Verkehr und Kunst (Case C–413/01) [2003] ECR I–13187.................................22, 26, 298, 299, 302, 303, 304, 329, 346, 361, 474, 476, 491, 533, 535, 539, 550 France v Commission (Case C–241/94) [1996] ECR I–4551..............................................511 France v Commission (Case C–482/99) [2002] ECR I–4397..............................................501 Frank Weber (Case C–1/07) [2008] ECR I–8571.................................................................352 Franzén (Case C–189/95) [1997] ECR I–2471.......................................................................50 Freeport plc v Olle Arnoldsson (Case C–98/06) [2007] ECR I–8319 ............................................................................... 244, 256, 265, 284, 285, 286 Freistaat Thüringen v Commission (Case T–318/00) [2005] ECR II-4179........................517 Freskot v Greece (Case C–355/00) [2003] ECR I–5263.......................................................512 Futura Participations SA and Singer v Administration des contributions (Case C–250/95) [1997] ECR I–2471............................................... 460, 461, 462, 468, 486 Gabalfrisa and Others (Joined Cases C-110 to C-147/98) [2000] ECR I-1577..................452 Galeries Segoura v Société Rahim Bonakdarian (Case 25/76) [1976] ECR 1851..............235 Garage Molenheide BVBA (C-286/94), Peter Schepens (C-340/95), Bureau Rik Decan-Business Research & Development NV (BRD) (C-401/95) and Sanders BVBA (C-47/96) v Belgische Staat (Joined Cases C–286/94, C–340/95, C–401/95 and C–47/96) [1997] ECR I–7281....................................................................................386 Garcia Avello v Belgian State (Case C–148/02) [2003] ECR I–11613.........................331, 338 GB-INNO-BM v Confédération du commerce luxembourgeois (Case C–362/88) [1990] ECR I–667..............................................................................................................460 Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano (Case C–55/94) [1995] ECR I–4165................................................... 66, 160, 183, 318, 361 Gebroeders van Es Douane Agenten BV v Inspecteur der Invoerrechten en Accijnzen (Case C–143/93) [1996] ECR I–431.................................................................................355 Gebrüder Lück v Hauptzollamt Köln-Rheinau (Case 34/67) [1968] ECR 245..........451, 452 Gemeente Leusden v Staatssecretaris van Financiën and C–7/02 Holin Groep BV cs v Staatssecretaris van Financiën (Joined Cases C–487/01) [2005] ECR I–5337.......374, 470 General Milk Products GmbH v Hauptzollamt Hamburg-Jonas (Case C–8/92) [1993] ECR I–779..................27, 109, 174, 204, 208, 225, 226, 474, 526, 532, 537, 541, 558 Germany v Commission (Case C–156/98) [2000] ECR I–6857..........................................511 Germany v Commission (Case C–334/99) [2003] ECR I–1139..........................................516 Germany v Parliament and Council (Case C–376/98) [2000] ECR I–8419.......................330 Germany v Parliament and Council (Case C–380/03) [2006] ECR I–11573.....................330 Gestoras Pro Amnistía, Juan Mari Olano Olano and Julen Zelarain Errasti v Council of the European Union (Case C–354/04P) [2007] ECR I–1579................................ xxv, 11 GIL Insurance Ltd et al v Commissioners of Customs & Excise (Case C–308/01) [2004] ECR I–4777............................................................................ 502, 504, 505, 509, 512

600  Table of Cases GlaxoSmithKline Services v Commission (Case T–168/01) [2006] ECR II-2969..............225 GlaxoSmithKline Services Unlimited et al v Commission et al (Joined Cases C–501, 513, 515 and 519/06P)[2009] ECR I-9291........................................................................225 ex p Gloszczuk (Case C–63/99) [2001] ECR I–6369............................................................341 Government of Gibraltar and United Kingdom v Commission (Joined Cases T–211/04 and T–215/04) [2008] ECR II-3745................................. 498, 502, 503, 510, 512 Gravier v City of Liège (Case 293/83) [1985] ECR 593.........................................................16 Grzelczyk v Centre public d’aide sociale d’Ottignies-Louvain-la-Neuve (Case C–184/99) [2001] ECR I–6193........................299, 304, 311, 345, 346, 347, 348, 359 Groupement agricole d’exploitation en commun (GAEC) de la Ségaude v Council and Commission (Case 253/84) [1987] ECR 123............................................................513 Grant v South-West Trains (Case C–249/96) [1998] ECR I–621..........................................15 Günaydin v Freistaat Bayern (Case C–36/96) [1997] ECR I–5143............... 29, 324, 342, 558 Hadadi (Hadady) v Csilla Marta Mesko married name Hadadi (Hadady) (Case C–168/08) [2009] ECR I‒6871...................................................................... xxiv, 230 Halbritter v Freistaat Bayern (Case C–227/05) [2006] ECR I–49.......................................351 Halifax plc, Leeds Permanent Development Services Ltd, County Wide Property Investments Ltd v Commissioners of Customs & Excise (Case C–255/02) [2006] ECR I–1609...............................xv, xvi, xviii, xxii, xxiii, xxiv, 4, 5, 20, 23, 25, 27, 28, 31, 61, 88, 102, 129, 141, 171, 173, 174, 175, 181, 184, 185, 210, 219, 226, 227, 239, 240, 242, 244, 245, 247, 248, 249, 255, 258, 313, 324, 329, 366, 367, 368, 369, 371, 372, 374, 375, 376, 378, 379, 381, 383, 385, 386, 387, 388, 389, 390, 391, 392, 393, 395, 396, 397, 398, 399, 403, 404, 407, 412, 417, 418, 419, 420, 422, 423, 424, 427, 428, 429, 430, 431, 433, 441, 442, 443, 448, 449, 452, 453, 455, 465, 469, 470, 471, 472, 473, 477, 480, 487, 488, 489, 490, 491, 494, 497, 498, 499, 500, 502, 507, 508, 520, 522, 523, 524, 525, 527, 528, 529, 530, 531, 531, 532, 533, 534, 535, 536, 537, 538, 539, 540, 541, 542, 543, 544, 545, 546, 548, 555, 556, 557, 559, 562, 565, 568 Hans Just I/S v Danish Ministry for Fiscal Affairs (Case 68/79) [1980] ECR 501................21 Har Vaessen Douane Service BV v Staatssecretaris van Financiën (Case C‒7/08) [2009] ECR I‒5581...........................................................................................................xxiv Hartmann v Freistaat Bayern (Case C–212/05) [2007] ECR I–6303..................................298 Hauer v Land Rheinland-Pfalz (Case 44/79) [1979] ECR 3727............................................10 Heirs of H Barbier v Inspecteur van de Belastingdienst Particulieren/Ondernemingen buitenland te Heerlen (Case C–364/01) [2003] ECR I–15013................................176, 481 Heirs of MEA van Hilten- van der Heijden v Inspecteur van de Belastingdienst/ Particulieren/ Ondernemingen buitenland te Heerlen (Case C–513/03) [2006] ECR I–1957........................................................................................................474, 475, 478 Helmut Müller GmbH v Bundesanstalt für Immobilienaufgaben (Case C–451/08), ECJ Opinion of 17 November 2009, nyr..........................................................................227 HJ Banks & Co Ltd v The Coal Authority and Secretary of State for Trade and Industry (Case C–390/98) [2001] ECR I–6117........................................................494, 517 HMRC v Weald Leasing Limited (Case C–103/09), Opinion of 26 October 2010, nyr.............................................................................. xxiv, 371, 393, 522, 529, 530, 531, 533, 534, 535, 536, 538, 539, 540, 541, 544, 548 Hoechst v Commission (Joined Cases 46/87 and 227/88) [1989] ECR 2859.......................13

Table of Cases  601 Hoekstra (née Unger) v Bestuur der Bedrijfsvereniging voor Detailhandel en Ambachten (Case 75/63) [1964] ECR 177.......................................................................229 Holland Malt BV v Commission (Case T–369/06), [2009] ECR II-3313...........................501 Hoogovens v High Authority (Case 14/61) [1962] ECR 253...........................................9, 568 Hosse v Land Salzburg (Case C–286/03) [2006] ECR I–1771.............................................314 Hotel Cipriani SpA v Commission (Joined Cases T–254/00, T–270/00 and T–277/00) [2008] ECR II-3269...................................................................................................511, 516 Hubert Wachauf v Bundesamt für Ernährung und Forstwirtschaft (Case 5/88) [1989] ECR 2609................................................................................................................356 Hughes de Lasteyrie du Saillant v Ministère de l‘Économie, des Finances et de l‘Industrie (Case C–9/02) [2004] ECR I–2409.................369, 392, 397, 401, 437, 470, 471, 473, 474, 478, 482, 483, 491 Iannelli and Volpi SpA v Ditta Paolo Meroni (Case 74/76) [1977] ECR 557.............358, 451 Imperial Chemical Industries plc (ICI) v Kenneth Hall Colmer (Her Majesty’s Inspector of Taxes) (Case C–264/96) [1998] ECR I–4695.................. 4, 113, 368, 386, 387, 397, 431, 435, 436, 469, 474, 482, 551 International Chemical Corporation v Amministrazione delle finanze dello Stato (Case 66/80) [1981] ECR 1191..........................................................................................224 International Transport Workers’ Federation and Finnish Seamen’s Union v Viking Line ABP and OU Viking Line Eesti (Case C–438/05) [2007] ECR I–10779.................................................................................... 53, 54, 94, 104, 319, 327 Internationale Handelsgesellschaft v Einfuhr- und Vorratsstelle Getreide (Case 11/70) [1970] ECR 1125..................................................................................................................10 I/S Fini H v Skatteministeriet (Case C–32/03) [2005] ECR I–1599.............. 27, 452, 539, 558 Italy v Commission (Case 173/73) [1974] ECR 709........................................... 500, 501, 503, 504, 507, 511, 512 Italy v Commission (Case C–91/01) [2004] ECR I–4355............................................493, 506 Italy v Commission (Case C–66/02) [2005] ECR I–10901..........................................511, 516 Italy v Commission (Case T–211/05) [2009] ECR II-2777.................................................502 ITT Promedia NV v Commission (Case T–111/96) [1998] ECR II-2937..........................227 Jacqueline Förster v Hoofddirectie van de Informatie Beheer Groep (Case C–158/07) [2008] ECR I–8507................................................................................... 299, 302, 303, 304, 310, 311, 318, 347, 348, 349, 359 Janko Rottman v Freistaat Bayern (Case C–135/08), Opinion of 30 September 2009, judgment of 2 March 2010, nyr............................................... 229, 332, 333, 336, 352 Jany v Staatssecretaris van Justitie (Case C–268/99) [2001] ECR I–8615...........................297 Jenkins v Kingsgate (Case 96/80) [1981] ECR 911.................................................................97 Jessica Safir v Skattemyndigheten i Dalarnas Län, formerly Skattemyndigheten i Kopparbergs Län (Case C–118/96) [1998] ECR I–1897.................. 460, 473, 475, 476, 477 Jobra Vermögensverwaltungs-Gesellschaft mbH v Finanzamt Amstetten Melk Scheibbs (Case C–330/07) [2008] ECR I–9099................................................................440 Johann Gruber v Bay Wa AG (Case C–464/01) [2005] ECR 1–439............................248, 259 Johnston v Chief Constable of the Royal Ulster Constabulary (Case 222/84) [1986] ECR 1651....................................................................................................10, 13, 356 José Antonio da Silva Carvalho (Case C–408/02)................................................................351 Josefa Velasco Navarro v Fondo de Garantia Salarial (Fogasa) (Case C–246/06) [2008] ECR I–105................................................................................................................14

602  Table of Cases Kadi and Al Barakaat v Council (Joined Cases C–402/05P and C–415/05P) [2008] ECR I–6351....................................................................................................................10, 12 Kahla/Thüringen Porzellan GmbH v Commission (Case T–20/03) [2008] ECR II-2305...............................................................................................................511, 516 Kalfelis v Bankhaus Schröder, Münchmeyer, Hengst & Co et al (Case 189/87) [1988] ECR 5565........................................................................ 244, 265, 283, 284, 285, 286 Kamer van Koophandel en Fabrieken voor Amsterdam v Inspire Art Ltd (Case C–167/01) [2003] ECR I–10155................. 22, 65, 110, 111, 112, 113, 115, 120, 127, 129, 132, 137, 141, 145, 176, 178, 329, 409, 447, 448, 480, 482, 497, 525, 532, 535, 547, 551 Kapper (Case C–476/01) [2004] ECR I–5205......................................................................351 Keck and Daniel Mithouard (Joined Cases C–267/91 and C–268/91) [1993] ECR I–6097.................................................................................................. 52, 209, 328, 463 Kempf v Staatssecretaris van Justitie (Case 139/85) [1986] ECR 1741...............................298 Kerckhaert and Bernadette Morres v Belgian State (Case C–513/04) [2006] ECR 1-10967......................................................................................................................416 Klensch v Secrétaire d’Etat (Joined Cases 201/85 and 202/85) [1986] ECR 3477................15 Knoors v Staatssecretaris van Economische Zaken (Case 115/78) [1979] ECR 399.................................................................... 55, 56, 58, 59, 61, 67, 69, 109, 170, 176, 350, 351, 481, 522, 524, 545, 548, 552 Hans Markus Kofoed v Skatteministeriet (Case C–321/05) [2007] ECR I–5795.................................... xvi, xxiii, 20, 25, 28, 29, 30, 31, 171, 175, 176, 190, 369, 370, 371, 383, 388, 389, 399, 401, 407, 421, 422, 455, 456, 519, 521, 522, 532, 537, 539, 561, 562 Kolpinghuis Nijmegen (Case 80/86) [1987] ECR 3969..................................... 12, 21, 22, 456 Konstantinidis v Stadt Altensteig-Standesamt and Landratsamt Calw Ordnungsamt (Case C–168/91) [1993] ECR I–1191...............................................................................337 Konstantinos Adeneler et al v Ellinikos Organismos Galaktos (Case C–212/04) [2006] ECR I–6057..............................................................................................26, 249, 327 Konsumentombudsmannen (KO) v De Agostini (Svenska) Förlag AB (C-34/95) and TV-Shop i Sverige AB (Joined Cases C–34/95 to C–36/95) [1997] ECR I–3843.........................................................................65, 72, 78, 79, 80, 81, 82, 86, 548 Konsumentombudsmannen (KO) v Gourmet International Products AB (GIP) (Case C–405/98) [2001] ECR I–1795...............................................................................206 KP Scorpio Konzertproduktionen GmbH v Finanzamt Hamburg-Eimsbüttel (Case C–290/04) [2006] ECR I–9461...............................................................................454 Kraus v Land Baden-Württemberg (Case C–19/92) [1993] ECR I–1663...........109, 460, 461 Kremer (Case C–340/05) [2006] ECR I–98..........................................................................351 Kremzow v Austrian State (Case C–299/95) [1997] ECR I–2629.........................................10 Kücükdeveci v Swedex GmbH & Co KG (Case C–555/07), ECJ judgment of 19 January 2010, nyr..............................................................................................11, 16, 466 Kunqian Catherine Zhu, Man Lavette Chen v Secretary of State for the Home Department (Case C–200/02) [2004] ECR I–9925..............xvi, xix, 22, 177, 228, 229, 315, 334, 335, 338, 339, 359, 360, 361, 470, 491, 527, 530 Lair v Universität Hannover (Case 39/86) [1988] ECR 3161............... 69, 109, 296, 298, 301, 302, 303, 304, 310, 311, 328, 345, 346, 347, 349, 527, 533, 537, 539, 550, 558

Table of Cases  603 Lankhorst-Hohorst GmbH v Finanzamt Steinfurt (Case C–324/00) [2002] ECR I–11779............................................... 113, 176, 369, 397, 401, 431, 435, 436, 474, 482 Lasertec Gesellschaft für Stanzformen mbH v Finanzamt Emmendingen (Case C–492/04) [2007] ECR I–3775...............................................................................447 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet, Svenska Byggnadsarbetareförbundets avdelning 1, Byggettan and Svenska Elektrikerförbundet (Case C–341/05) [2007] ECR I–11767............................... 21, 22, 53, 54, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 319, 327, 525 Lawrie-Blum v Land Baden-Württemberg (Case 66/85) [1986] ECR 2121.......297, 298, 302 Lebon (Case 316/85) [1987] ECR 2811................................................................................348 Leclerc et al v SARL ‚Au Blé Vert‘ et al (Case 229/83) [1985] ECR 1........................ 19, 56, 59, 60, 61, 69, 109, 113, 170, 171, 205, 350, 522, 525, 533, 558 Leclerc et al v Syndicat des libraries de Loire-Océan (Case 299/83) [1985] ECR 2515............................................................................................................................205 Leur-Bloem v Inspecteur der Belastingdienst/Ondernemingen Amsterdam 2 (Case C–28/95) [1997] ECR I–4161................................................. 113, 386, 404, 471, 472 Levin v Staatssecretaris van Justitie (Case 53/81) [1982] ECR 1035..................................................... 231, 297, 298, 300, 301, 302, 306, 315, 340, 345 Lidl Belgium GmbH & Co KG v Finanzamt Heilbronn (Case C–414/06) [2008] ECR I–3601........................................................................................................................482 Lindorfer v Council of the European Union (Case C–227/04P) [2007] ECR I–6767..........................................................................................................................15 Lourenço Dias v Director da Alfandega do Porto (Case C‒343/90) [1992] ECR I‒4673........................................................................................................................526 LTU v Eurocontrol (Case 29/76) [1976] ECR 1541.............................................................523 Mainschiffahrts-Genossenschaft eG (MSG) v Les Gravières Rhénanes sarl (Case C–106/95) [1997] ECR I–911 (ECJ)......................................... 244, 245, 266, 270, 282, 283 ex p Man (Sugar) Ltd (Case 181/84) [1985] ECR 2889.......................................................355 Manfredi v Lloyd Adriatico Assicurazioni SpA (C-295/04), Antonio Cannito v Fondiaria Sai SpA (C-296/04) and Nicolò Tricarico (C-297/04) and Pasqualina Murgolo (C-298/04) v Assitalia SpA (Joined Cases C–295/04 to C–298/04) [2006] ECR I–6619........................................................................................................................239 Mangold v Rüdiger Helm (Case C–144/04) [2005] ECR I–9981.......... xxi, 10, 13, 14, 15, 16, 17, 18, 20, 30, 54, 175, 176, 306, 357, 358, 466, 549 Marks & Spencer plc v David Halsey (Her Majesty‘s Inspector of Taxes) (Case C–446/03) [2005] ECR I–10837............................................. 435, 437, 470, 474, 486 Marks & Spencer plc v Her Majesty‘s Commissioners of Customs and Excise (Case C–309/06) [2008] ECR I–2283...............................................................................452 Marleasing SA v La Comercial Internacional de Alimentacion SA (Case C–106/89) [1990] ECR I–4135........................................................................................ 5, 215, 369, 377 Marshall v Southampton and South-West Hampshire Area Health Authority (Case 152/84) [1986] ECR 723..........................................................................................221 Martínez Sala v Freistaat Bayern (Case C–85/96) [1998] ECR I–2691......... 16, 299, 311, 344 Maurice Donald Henn and John Frederick Ernest Darby (Case 34/79) [1979] ECR 3795............................................................................................................................206 Mellett v Court of Justice (Joined Cases T–66/96 and T–221/97) [1998] ECR SC II-1305................................................................................................................................516

604  Table of Cases Metallgesellschaft Ltd et al (C-397/98), Hoechst AG and Hoechst (UK) Ltd (C-410/98) v Commissioners of Inland Revenue and Her Majesty’s Attorney General (Joined Cases C–397/98 and C–410/98) [2001] ECR I–1727...........................397 Metin Bozkurt v Land Baden-Württemberg (Case C–303/08), judgment of 22 December 2010 nyr........................................................300, 522, 532, 533, 535, 539, 559 Metock et al v Minister for Justice, Equality and Law Reform (Case C–127/08) [2008] ECR I–6241................................................ 29, 31, 171, 291, 292, 293, 295, 296, 298, 304, 305, 306, 309, 311, 317, 321, 341, 360, 361 MG Eman and OB Sevinger v College van burgemeester en wethouders van Den Haag (Case C–300/04) [2006] ECR I–8055......................................................................333 Micheletti et al v Delegación del Gobierno en Cantabria (Case C–369/90) [1992] ECR I–4239........................................................................................ 331, 332, 334, 335, 336 ex p Milk Marque Ltd and National Farmers’ Union (Case C–137/00) [2003] ECR I–7975........................................................................................................................205 Minister voor Vreemdelingenzaken en Integratie v RNG Eind (Case C–291/05) [2007] ECR I–10719..................................................................................................309, 350 Ministero dell’Economia e delle Finanze v Cassa di Risparmio di Firenze SpA (Case C–222/04) [2006] ECR I–289.................................................................................511 Ministero dell’Economia e delle Finanze v Part Service Srl (Case C–425/06) [2008] ECR I–897..........................................xxiv, 227, 371, 375, 388, 392, 393, 397, 398, 407, 417, 420, 422, 423, 425, 427, 429, 443, 470, 471, 474, 476, 482, 520, 530, 534, 537, 538, 541, 542, 548 Ministero dell’Economia e delle Finance ; Agenzia delle Entrate v 3M Italia SpA (Case C‒417/10) [2010] OJ C288/23...............................................................................xxiv Ministre de l’Économie, des finances et de l’industrie v GEMO SA (Case C–126/01) [2003] ECR I–13769..........................................................................................................502 Ministero delle Finanze v IN.CO.GE. ‘90 Srl et al (Joined Cases C–10/97 to C–22/97) [1998] ECR I–6307............................................................................................................452 Ministero delle Finanze v Michele Franchetto (Case C–80/08)..........................................493 Mirja Juuri v Fazer Amica Oy (Case C‒396/07) [2008] ECR I‒8883................................xxiv Modehuis A Zwijnenburg BV (Case C–352/08), Opinion of 16 July 2009, nyr........................................................................................................ xxiii, 30, 227, 370, 379 Moskof AE v Ethnikos Organismos Kapnou (EOK) (Case C–244/95P) [1997] ECR I–6441........................................................................................................................171 Mouvement contre le racisme, l’antisémitisme et la xénophobie ASBL (MRAX) v Belgian State (Case C–459/99) [2002] ECR I–6591.........................................292, 309, 341 Mulder v Minister van Landbouw en Visserij (Case 120/86) [1988] ECR 2321......................................................................................................................13, 355 Müller-Fauré v Onderlinge Waarborgmaatschappij OZ Zorgverzekeringen UA, and between EEM van Riet and Onderlinge Waarborgmaatschappij ZAO Zorgverzekeringen (Case C–385/99) [2003] ECR I–4509.......................................221, 318 Mulox IBC (Case C–125/92) [1993] ECR I–4075................................................................264 N v Inspecteur van de Belastingdienst Oost/kantoor Almelo (Case C–470/04) [2006] ECR I–7409.............................................470, 471, 474, 475, 478, 479, 480, 482, 491 Nadin and Durré (Joined Cases C–151/04 and 152/04) [2005] ECR I–11203...................316 Nationale Raad van Dierenkwekers en Liefhebbers VZW, Andibel VZW v Belgische Staat (Case C–219/07) [2008] ECR I–4475........................................................................53

Table of Cases  605 NCC Construction Danmark A/S v Skatteministeriet (Case C–174/08), [2009] ECR I-10567.........................................................................................................................12 Netherlands v Commission (Case C–382/99) [2002] ECR I–5163.............................511, 515 Netherlands v Commission (Case C–159/01) [2004] ECR I–4461.............................511, 512 Nold, Kohlen- und Baustoffgrosshandlung v Commission (Case 4/73) [1974] ECR 491............................................................................................................................9, 10 Nuova Agricast Srl v Ministero delle Attività Produttive (Case C–390/06) [2008] ECR I–2577................................................................................................................506, 507 NV Lammers & Van Cleeff v Belgische Staat (Case C–105/07) [2008] ECR I–0173................................................................................................................438, 441 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn (Case C–36/02) [2004] ECR I–9609............... 18, 53, 176, 312, 447 Optigen Ltd, Fulcrum Electronics Ltd and Bond House Systems Ltd v Commissioners of Customs & Excise (Joined Cases C–354/03, C–355/03 and C–484/03) [2006] ECR I–483.............................................................................................. xxv, 388, 428 Orfanopoulos and Oliveri v Land Baden-Württemberg Joined Cases (C–482/01 and 493/01) [2004] ECR I–5257...............................................................................310, 356 Orkem v Commission (Case 374/87) [1989] ECR 3283........................................................13 Owusu v NB Jackson, trading as ‘Villa Holidays Bal-Inn Villas’ et al (Case C–281/02) [2005] ECR I–1383.............................................267, 268, 269, 270, 271, 272, 274, 275, 276 Oy AA (Case C–231/05) [2007] ECR I–6373.......................................................................470 P v S and Cornwall County Council (Case C–13/94) [1996] ECR I–2143...........................15 Panagis Pafitis et al v Trapeza Kentrikis Ellados AE et al (Case C–441/93) [1996] ECR I–1347..........................................................................21, 108, 113, 141, 172, 173, 217, 530, 532, 542, 546, 547, 554, 560, 568 Parodi v Banque H Albert de Bary (Case C–222/95) [1997] ECR I–3899..........................184 Parti écologiste ‘Les Verts’ v European Parliament (Case 294/83) [1986] ECR 1339...........12 Payir, Akyuz, Ozturk (Case C–294/06) [2008] ECR I–203..................................................342 Pedro IV Servicios SL v Total España SA (Case C‒260/07) [2009] ECR I‒2437...............xxiv Petri Manninem (Case C–319/02) [2004] ECR I–7477.......................................................486 Portugaia Construções Lda (Case C–164/99) [2002] ECR I–787.........................................99 Portugal v Commission (Case C–88/03) [2006] ECR I–7115.....................................503, 512 PPU Jasna Detiícek v Maurizio Sgueglia (Case C–403/09), judgment of 23 December 2009, nyr...............................................................................................................................11 Preston et al v Wolverhampton Healthcare NHS Trust et al (Case C–78/98) [2000] ECR I–3201........................................................................................................................356 Procureur du Roi v Dassonville (Case 8/74) [1974] ECR 837...............................................51 Pupino (Case C–105/03) [2005] ECR I–5285..............................................................222, 223 R v Her Majesty’s Treasury ex p British Telecommunications plc (Case C–392/93) [1996] ECR I–1631....................................................................................................212, 216 R v Immigration Appeal Tribunal and Surinder Singh ex p Secretary of State for the Home Department (Case C–370/90) [1992] ECR I–4265..................... 304, 305, 317, 338, 339, 340, 347, 350, 360, 522, 524, 548 R v Secretary of State for the Home Department ex p Kaur (Case C–192/99) [2001] ECR I–1237....................................................................................................229, 331 Raulin v Minister van Onderwijs en Wetenschapen (Case C–357/89) [1992] ECR I–1027................................................................................................ 298, 302, 303, 304

606  Table of Cases Regione Autonoma Friuli Venezia Giulia v Commission (Case T–288/97) [2001] ECR II-1169.......................................................................................................................516 Reisch v Kiesel (Case C–103/05) [2006] ECR I–6827..........................................................244 Reisebüro Broede v Gerd Sanker (Case C–3/95) [1996] ECR I–6511...................................50 Republik Österreich v Martin Huber (Case C–336/00) [2002] ECR I–7699......................250 Rewe-Zentrale AG v Bundesmonopolverwaltung für Branntwein (Case 120/78) [1979] ECR 649 (Cassis de Dijon).............................................51, 52, 53, 56, 328, 460, 461 Rewe Zentralfinanz eG v Finanzamt Köln-Mitte (Case C–347/04) [2007] ECR I–2647................................................................................................................470, 474 Reyners v Belgian State (Case 2/74) [1974] ECR 631....................................................56, 402 Rheinland-Pfalz v Alcan (Case C–24/95) [1997] ECR I–1591............................250, 515, 516 RHH Renneberg v Staatssecretaris van Financiën (Case C–527/06) [2008] ECR I–7735........................................................................................................................391 Rhiannon Morgan v Bezirksregierung Köln (C–11/06) and Iris Bucher v Landrat des Kreises Düren (C–12/06) (Joined Cases C–11/06 and C–12/06) [2007] ECR I–9161........................................................................................................................359 Rodríguez Caballero (Case C–442/00) [2002] ECR I–11915................................................14 Rolf Dieter Danner (Case C–136/00) [2002] ECR I–8147.......................... 462, 473, 475, 477 Roquette Frères v Commission (Case C–94/00) [2002] ECR I–9011...................................13 Rüffert v Land Niedersachsen (Case C–346/06) [2008] ECR I–1989............... 93, 96, 97, 100 Rutili v Minister for the Interior (Case 36/75) [1975] ECR 1219..........................................13 Säger v Dennemeyer & Co Ltd (Case C–76/90) [1991] ECR I–4221..................................460 Schmeink & Cofreth AG & Co KG v Finanzamt Borken and Manfred Strobel v Finanzamt Esslingen (Case C–454/98) [2000] ECR I–6973....................................175, 452 Schmidberger, Internationale Transporte und Planzüge v Republik Österreich (Case C–112/00) [2003] ECR I–5659............................................................. 18, 53, 54, 312 Schutzverband gegen unlauteren Wettbewerb v TK-Heimdienst Sass GmbH (Case C–254/98) [2000] ECR I–151.................................................................................356 Secretary of State for the Home Department v Hacene Akrich (Case C–109/01) [2003] ECR I–9607.....................22, 29, 71, 79, 171, 226, 291, 292, 302, 304, 305, 306, 307, 308, 309, 311, 315, 317, 340, 341, 343, 344, 350, 360, 361, 398, 401, 470, 474, 476, 491, 506, 522, 525, 532, 533, 535, 536, 540, 543, 548, 565 Segers v Bestuur van de Bedrijfsvereniging voor Bank- en Verzekeringswezen, Groothandel en Vrije Beroepen (Case 79/85) [1986] ECR 2375...................................................................................65, 68, 88, 119, 129, 176, 177 Segi, Araitz Zubimendi Izaga and Aritza Galarraga v Council of the European Union (Case C–355/04P) [2007] ECR I–1657...................................................................11 Serrantoni Srl, sorzio stabile edili Scrl v Comune di Milano (Case C–376/08), judgment of 23 December 2009, nyr..................................................................................11 Sevic Systems AG (Case C–411/03) [2005] ECR I–10805...................................................120 SIC - Sociedade Independente de Comunicaço SA v Commission (Case T–442/03) [2008] ECR II-1161...........................................................................................................501 Simmenthal v Ministero delle Finanze (Case 35/76) [1976] ECR 1871..............................210 Skandia, Ramstedt v Rikskatteverket (Case C–422/01) [2003] ECR I–6817......473, 475, 477 Skatteministeriet v Bent Vestergaard (Case C–55/98) [1999] ECR I–7641.................392, 474 Skatteverket v A (Case C–101/05) [2007] ECR I–11531......................................................477 Scott SA v Commission (Case T–366/00) [2007] ECR II-797.............................................516

Table of Cases  607 SNUPAT v High Authority (Joined Cases 42/59 and 49/59) [1961] ECR 53......................514 Société Comateb et al v Directeur Général des Douanes et Droits Indirects (Joined Cases C–192/95 to C–218/95) [1997] ECR I–165................................................21 Société thermale d’Eugénie-les-Bains v Ministère de l’Economie, des Finances et de l’Industrie (Case C–277/05) [2007] ECR I–6415........................................................257 Sonia Chacón Navas v Eurest Colectividades SA (Case C–13/05) [2006] ECR I–6467....................................................................................................................14, 15 SpA Alois Lageder et al v Amministrazione delle Finanze dello Stato (Joined Cases C–31/91 to C–44/91) [1993] ECR I–1761........................................................................356 Spagl (Case C–189/89) [1990] ECR I–4539...........................................................................13 Spain v Commission (Joined Cases C–278–280/92) [1994] ECR I–4103...........................516 Spain v Commission (Case C–169/95) [1997] ECR I–135..................................................516 Spain v Commission (Case C–351/98) [2002] ECR I–8031................................................512 Spain v Commission (Case C–480/98) [2000] ECR I–8717........................................502, 511 Spain v Commission (Case C–409/00) [2003] ECR I–1487........................................502, 512 Spain v Commission (Case C–276/02) [2004] ECR I–8091........................................498, 502 Spain v European Commission (Case C–240/97) [1999] ECR I–6571...............................257 Spain v UK (Case C–145/04) [2006] ECR I–7917...............................................................333 Spector Photo Group NV Chris Van Raemdonck v Commissie voor het Bank-, Financie- en Assurantiewezen (Case C–45/08), judgment of 23 December 2009, nyr.........................................................................................................................................11 Staatssecretaris van Financiën v J Heerma (Case C–23/98) [2000] ECR I–419..........................................................................................................................430 Staatssecretaris van Financiën v Orange European Smallcap Fund NV (Case C–194/06) [2008] ECR I–3747...............................................................................477 Standesamt Stadt Niebüll v Stefan Grunkin, Dorothee Regina Paul (Case C–96/04) [2006] ECR I–3561............................................................................................................338 Stefan Grunkin and Dorothee Regina Paul (Grunkin Paul) (Case C–353/06) [2008] ECR I–7639........................................................................................................................338 Stella Kunststofftechnik v OHIM (Case T–27/09), [2009] ECR II-4481..............................11 Stichting Collectieve Antennevoorziening Gouda v Commissariaat voor de Media (Case C–288/89) [1991] ECR I–4007...................................................................57, 66, 181 Stoke-on-Trent City Council and Norwich City Council v B & Q plc (Joined Cases 306/88; C–304/90 and C–169/91) [1992] ECR I–6457......................................52, 211, 462 Streekgeweest Westelijk Noord-Brabant v Staatsecretaris van Financiën (Case C–174/02) [2005] ECR I–85...................................................................................509 Stringer and Others v Her Majesty’s Revenue and Customs (Case C–520/06) [2009] ECR I-179.................................................................................................................20 Suat Kol v Land Berlin (Case C–285/95) [1997] ECR I–3069.............................. 29, 296, 300, 324, 342, 522, 558 Susanne Staubitz-Schreiber (Case C–1/04) [2006] ECR I–701................... 138, 141, 151, 160 Svensson and Gustavsson v Minstre du Logement et de l’Urbanisme (Case C–484/93) [1995] ECR I–3955............................................................................................................184 Swaddling v Adjudication Officer (Case C–90/97) [1999] ECR I–1075.............................348 Swedish Match (Case C–210/03) [2004] ECR I–11893.......................................................330 Syndesmos ton en Elladi Touristikon kai Taxidiotikon Grafeion v Ypourgos Ergasias (Case C–398/95) [1997] ECR I–3091...............................................................................460

608  Table of Cases Tadao Maruko v Versorgungsanstalt der deutschen Bühnen (Case C–267/06) [2008] ECR I–1757..........................................................................................................................15 Tas-Hagen and Tas v Raadskamer WUBO van de Pensioen- en Uitkeringsraad (Case C–192/05) [2006] ECR I–10451.............................................................................359 Technische Glaswerke Ilmenau GmbH v Commission (Case T–198/01) [2004] ECR II-2717.......................................................................................................................516 Territorio Histórico de Álava - Diputación Foral de Álava v Commission (Joined Cases T–127/99, T–129/99 & T–148/99) [2002] ECR II-1275........................................511 Territorio Historico de Alava - Diputación Foral de Alava et al v Commission of the European Communities (Joined Cases T–227/01 to T–229/01, T–265/01, T–266/01 and T–270/01), [2009] ECR I-3029..........................................................................501, 503 Territorio Histórico de Guipúzcoaù - Diputación Foral de Guipúzcoa v Commission (Joined Cases T–269/99, T–271/99 & T–272/99) [2002] ECR II-4217...........................511 Test Claimants in Class IV of the ACT Group Litigation (Case C–374/04) [2006] ECR I–11673..............................................................................................................431, 477 Test Claimants in the CFC and Dividend Group Litigation v Commissioners of Inland Revenue (Case C–201/05) [2008] ECR I–2875....................................................439 Test Claimants in the Thin Cap Group Litigation v Commissioners of Inland Revenue (Case C–524/04) [2007] ECR I–2107........................................................... 4, 5, 6, 29, 315, 397, 403, 404, 432, 433, 435, 438, 440, 441, 446, 447, 448, 457, 470, 472, 474, 477, 480, 482, 490, 491, 522, 524, 534, 537, 551 The Tatry (Case C–406/92) [1994] ECR I–5439.......................................... 266, 267, 270, 274 Theologos-Grigoris Khatzithanasis v Ipourgos Igeias kai Kinonikis Allilengiis (Case C–151/07) [2008] ECR I–9013...............................................................................351 Tiercé Ladbroke v Commission (Case C–353/95P) [1997] ECR I–7007............................511 Torfaen Borough Council v B & Q plc (Case C–145/88) [1989] ECR I–3851............................................................................................................13, 52, 462 Transalpine Ölleitung in Österreich GmbH et al v Finanzlandesdirektion für Tirol et al (Case C–368/04) [2006] ECR I–9957.......................................................493, 506, 517 Transocean Marine Paint v Commission (Case 17/74) [1974] ECR 1063..................236, 238 TransportService (Case C–395/02) [2004] ECR I–1991......................................................452 Trojani v Centre public de l’aide sociale de Bruxelles (CPAS) (Case C–456/02) [2004] ECR I–7573.................................................................................... 297, 301, 359, 361 TV10 SA v Commissariaat voor de Media (Case C–23/93) [1994] ECR I–4795....................................... 26, 57, 58, 59, 61, 64, 66, 68, 69, 70, 71, 73, 78, 79, 80, 81, 86, 87, 93, 109, 171, 177, 180, 181, 184, 185, 189, 208, 209, 226, 308, 375, 398, 470, 471, 524, 525, 542, 558 Turner v Grovit (Case C–159/02) [2004] ECR I–3565............................... 245, 246, 269, 270, 271, 272, 273, 274, 276, 287 Überseering BV v Nordic Construction Company Baumanagement GmbH (Case C–208/00) [2002] ECR I–9919................110, 115, 127, 137, 141, 145, 178, 329, 480 Unicredito Italiano SpA v Agenzia delle Entrate, Ufficio Genova 1 (Case C–148/04) [2005] ECR I–11137.................................................................. 502, 503, 505, 506, 511, 516 Union départementale des syndicats CGT de l’Aisne v SIDEF Conforama, Société Arts et Meubles and Société Jima (Case C–312/89) [1991] ECR I–997..................211, 462 Union Royale Belge des Sociétés de Football Association ASBL et al v Jean-Marc Bosman et al (Case C–415/93) [1995] ECR I–4921.........................................298, 318, 328

Table of Cases  609 United Pan-Europe Communications Belgium SA et al v Belgian State (Case C–250/06) [2007] ECR I–11135............................................................. 65, 72, 82, 83 University of Huddersfield Higher Education Corporation v Commissioners of Customs & Excise (Case C–223/03) [2006] ECR I–1751.....................xxiii, 26, 210, 397 Van Bennekom (Case 227/82) [1983] ECR 3883...................................................................50 Van Binsbergen v Bestuur van de Bedrijfsvereniging voor de Metaalnijverheid (Case 33/74) [1974] ECR 1299.........................xv, xxii, 19, 51, 58, 64, 66, 67, 68, 69, 70, 72, 73, 86, 90, 93, 109, 170, 180, 184, 248, 271, 296, 361, 384, 470, 499, 519, 522, 524, 527, 558 Van de Bijl v Staatssecretaris van Economische Zaken (Case 130/88) [1989] ECR 3039........................................................................................................................67, 69 Veli Tum and Mehmet Dari v Secretary of State for the Home Department (Case C–16/05) [2007] ECR I–7415...................................................xxiv, 29, 324, 341, 342 Verband Sozialer Wettbewerb eV v Clinique Laboratories SNC et Estée Lauder Cosmetics GmbH (Case C–315/92) [1994] ECR I–317..................................................358 Verein gegen Unwesen in Handel und Gewerbe Köln eV v Mars GmbH (Case C–470/93) [1995] ECR I–1923.................................................................................53 Vereniging Veronica Omroep Organisatie v Commissariaat voor de Media (Case C–148/91) [1993] ECR I–487........................................ 57, 64, 66, 67, 68, 69, 70, 71, 78, 79, 80, 81, 87, 93, 109, 171, 181, 208, 248, 308, 470, 524, 532 Vlaamse Televisie Maatschapij NV v Commission (Case T–266/97) [1999] ECR II-2329.........................................................................................................................79 Volker Graf v Filzmoser Maschinenbau GmbH (Case C‒190/98) [2000] ECR I‒493.......xix von Colson and Elisabeth Kamann v Land Nordrhein-Westfalen (Case 14/83) [1984] ECR 1891................................................................................................................215 von Lachmüller, Bernard Peuvrier, Roger Ehrhardt v Commission of the European Economic Community (Joined Cases 43/59, 45/59 and 48/59) [1960] ECR 463......................................................................................................................235, 250 Vonk Dairy Products BV v Productschap Zuivel (Case C–279/05) [2007] ECR I–239.................................................................................................. 209, 214, 215, 450 VT4 Ltd v Vlaamse Gemeenschap (Case C–56/96) [1997] ECR I–3143............................................................................... 65, 72, 76, 77, 78, 79, 83, 86\ Wagner Miret v Fondo de Garantía Salarial (Case C–334/92) [1993] ECR I–6911................................................................................................................222, 377 Walter Rau Lebensmittelwerke v De Smedt PVBA (Case 261/81) [1982] ECR 3961................................................................................................................13, 50, 356 Watts [Queen, The (on the application of Yvonne Watts) v Bedford Primary Care Trust and Secretary of State for Health] (Case C–372/04) [2006] ECR I–4325.....299, 318 Wienand Meilicke v ADV/ORGA F A Meyer AG (Case C–83/91) [1992] ECR I–4871..................................................................................................................19, 526 Wijsenbeek (Case C–378/97) [1999] ECR I–6207...............................................................348 Wolfgang Heiser v Finanzamt Innsbruck (Case C–172/03) [2005] ECR I–1627...............511 X and Y v Riksskatteverket (Case C–436/00) [2002] ECR I–10829........ 330, 369, 397, 431, 436 X v Commission (Case C–404/92P) [1994] ECR I–4737 .....................................................13 Yunying Jia v Migrationsverket (Case C–1/05) [2007] ECR I–0001...................293, 305, 340 Zelger v Salinitri (Case 129/83) [1984] ECR 2397...............................................................283

610  Table of Cases II International European Convention for the Protection of Human Rights and Fundamental Freedoms (1950) listed alphabetically Cable Music Europe Ltd v Netherlands (App no 18033/91).................................................64 Evans v UK, App no 6339/05, judgment of 10 April 2007...................................................320 Garaudy v France, App no 65831/01, Decision of 24 June 2003, ECHR 2003-IX..............319 Groppera Radio AG et al v Switzerland, App no 10890/84, (1990) Series A no 173............64 H, W, P and K v Austria, App no 12774/87, (1989) 62 DR 216...........................................325 Kühnen v FRG, App no 12194/86, (1988) 56 DR 205..........................................................325 Matthews v United Kingdom, App no 24833/94, (1999) 28 EHR 361................................333 Pavel Ivanov v Russia, App no 35222/04, 20 February 2007................................................313 Socialist Party of Turkey and others v Turkey, App no 21237/93, Reports 1998-III..........313 T v Belgium (App no 9777/82) (1983) DR 34......................................................................325 United Communist Party of Turkey and others v Turkey, App no 19392/92, Reports 1998-I...........................................................................................................313, 325 X SA v Netherlands, App no 21472/93, DR 76.......................................................................64

European Free Trade Area Joined Cases E-8/94 and E-9/94 Forbrukerombudet v Mattel Scandinavia and Lego Norge [1994–95] EFTA Court Reports 113..................................................................81, 82

International Court of Justice Nottebohm Case (Liechtenstein v Guatemala) (Second Phase), 6 April 1955, ICJ Reports 1955, 4............................................................................................................336

Permanent Court of International Justice Free Zones, PICJ, Series A, no 30, 1930, p 12........................................................................567 German Interests in Polish Upper Silesia, PCIJ, Series A, no 7, 1926, p 30.........................567

III National Australia CSR Ltd v Cigna Insurance Australia Ltd (1997) 189 CLR 345...........................................268 Régie National des Usines Renault SA v Zhang (2003) 210 CLR 491.................................268 Voth v Manildra Flour Mills Pty Ltd (1990) 171 CLR 538..........................................268, 269

Table of Cases  611 France Cour de cassation, Chambre civile, 3 August 1915 (Clément-Bayard), Recueil Dalloz 1917.I.79.....................................................................................................................220, 227 Conseil d’Ètat, 27 September 2006, No 260050 (Société Janfin).........................................378 Hong Kong Collector of Stamp Revenue v Arrowtown Assets Ltd [2003] HKCFA 46..................373, 375 Ireland, Republic of Cussens & Ors v Brosnan [2008] IEHC 169 (Irish High Court)................................. xxv, 508 Italy BL Macchine Automatische v Windmöller & Hölscher, Decision no 19550 of 19 December 2003, Corriere Giuridico 2/2004 162.........................................................272 United Kingdom Airtours et al v Commissioners of Customs & Excise, IPT 00007, 8 October 2001 (unreported)......................................................................................................................503 Astall v Revenue and Customs Commissioners [2009] EWCA Civ 1010...........................489 Atlantic Star, The (Owners) v Bona Spes [1974] AC 436 (HL)............................................268 Autologic Holdings plc v IRC [2005] STC 1357 (HL).........................................................467 Barclays Mercantile Business Finance Ltd v Mawson (Inspector of Taxes) [2005] 1 AC 684; [2004] UKHL 51............................................................... 372, 373, 376, 427, 489 Baylis v Gregory [1986] STC 22 (Ch)...................................................................................372 Betty’s Kitchen Coronation Street.........................................................................................242 Betws Anthracite Ltd v DSK Anthrazit Ibbenbüren GmbH [2003] EWHC 2403..............518 Bradford Corporation v Pickles [1895] AC 587 (HL)..........................................................170 Bulmer Ltd v Bollinger SA [1974] 2 CMLR 91 (CA)...........................................................221 Circuit Systems Ltd v Zuken-Redac (UK) Ltd [1997] 3 WLR 1177 (HL)...........................489 Commissioners of Her Majesty’s Revenue & Customs v Moorbury Limited, FTC/21/2009, 16 September 2010................................................................................... xxiii Commissioner of Inland Revenue v Duke of Westminster (1936) 1 AC 19 (HL)..............374 Cornhill Insurance plc v Improvement Services Ltd [1986] BCLC 26...............................152 Countess Fitzwilliam v IRC [1993] 1 WLR 1189 (HL)........................................................488 Craven (Inspector of Taxes) v White [1989] AC 398 (HL)..................................................488 Daisytek-ISA Ltd [2003] BCC 562........................................................................................153 DNick Holding plc [2007] (HC)...........................................................................................139 Donoghue v Stevenson [1932] AC 562 (HL)........................................................................237 Furniss (Inspector of Taxes) v Dawson [1984] AC 474 (HL)......................................488, 557 Ghaidan v Godin-Mendoza [2004] AC 557; [2004] UKHL 30...........................................377 GIL Insurance Ltd v Commissioners of Customs & Excise [2000] STC 204 (HC)............503 Hans Brochier Holdings Ltd v Exner [2007] BCLC 127; [2006] EWHC 2594 (Ch)....................................................................................................................142, 145, 159 Hatton v IRC [1992] STC 140...............................................................................................488 HMRC v IDT Card Services Ireland Ltd [2006] STC 1252; [2006] EWCA Civ 29.............377 Inland Revenue Commissioners v Burmah Oil Co Ltd [1982] STC 30......................427, 428 IRC v Brebner [1967] 2 AC 18...............................................................................................375

612  Table of Cases IRC v Scottish Provident Institution [2004] UKHL 52.......................................................489 Isaacs v Robinson [1985] AC 97 (PC)...................................................................................273 Krell v Henry [1903] 2 KB 740 (CA)....................................................................................530 Litster v Forth Dry Dock & Engineering Co Ltd (in receivership) [1990] 1 AC 546 (HL)....................................................................................................................377 Macniven (Her Majesty’s Inspector of Taxes) v Westmoreland Investments Ltd [2003] 1 AC 311; [2001] UKHL 6............................................................. 376, 427, 489, 558 MacShannon v Rockware Glass Ltd [1978] AC 795 (HL)...................................................268 Mayes v Revenue and Customs Commissioners [2009] EWHC 2443 (Ch).......................489 Mayor of Bradford v Pickles [1895] AC 587 (HL)...............................................................220 MG Rover Deutschland GmbH [2005] EWHC 874 (Ch)...................................................153 Michael Fleming (trading as Bodycraft) v HMRC and Condé Nast Publications Ltd v HMRC, Joined Cases [2008] 1 WLR 195 (HL)................................................454, 467, 468 Norglen Ltd v Reeds Rains Prudential Ltd [1997] 3 WLR 1177 (HL)................................489 Oakley v Animal [2005] EWHC 210 (Ch)............................................................................221 Page trading as Empowerment Enterprises Ltd v Commissioners of Customs & Excise [2004] UKVAT V18820...........................................................................................503 Pickstone v Freemans plc [1989] AC 66 (HL)......................................................................377 R v Attorney General ex p Imperial Chemical Industries Ltd [1985] CMLR 588 (Div Ct); [1987] 1 CMLR 72 (CA)....................................................................................503 R v Commissioners of Customs & Excise ex p Lunn Poly [1998] STC 649 (HC); [1999] STC 350 (CA); [1999] EuLR 653...........................................................501, 503, 512 R v IRC ex p Unilever plc [1996] STC 681............................................................................496 R ex p British Aggregates Association et al v Commissioners of Customs & Excise [2002] EuLR 394 (HC)......................................................................................................503 R ex p Professional Contractors’ Group et al v IRC [2001] STC 629 (HC); [2002] STC 165 (CA).................................................................................................460, 503 R (Aldogan) v Secretary of State for the Home Department [2007] EWHC 2586 (Admin)..............................................................................................................................342 R (Baiai) v Secretary of State for the Home Department [2009] 1 AC 287 (HL)..............344 St Pierre v South American Stores (Gath & Chaves) Ltd [1936] 1 KB 382 (CA)...............268 Schefenacker plc [2007] (HC)...............................................................................................139 Scott v Brown, Doering, McNab & Co [1892] 2 QB 724 (CA)............................................189 Snook v London & West Riding Investments Ltd [1967] 2 QB 786 (CA)..................365, 488 Street v Mountford [1985] AC 809 (HL)......................................................................266, 489 TC (Kenya) v Secretary of State for the Home Department [2008] EWCA Civ 543........................................................................................................................330, 344 Transbus International Ltd, Re [2004] EWHC 932 (Ch); [2004] 2 BCLC 550...................162 Turner v Grovit (CA).............................................................................................................561 University of Sussex v Commissioners of Customs & Excise [2001] STC 1495 (HC).......503 Vodafone 2 v HMRC (No 2) [2008] STC 2391 (HC); [2008] EWHC 1569 (Ch).................................................................................................... 377, 380, 454, 459, 463 Vodafone 2 v HMRC (No 2) [2009] STC 1480; [2009] EWCA Civ 446..............369, 466, 467 WT Ramsay Ltd v Inland Revenue Commissioners (IRC) [1982] AC 300 (HL).....................................................................................373, 376, 427, 487, 488, 489, 557

Table of Cases  613 United States of America Gregory v Helvering (1934) 69 F 2d 809..............................................................................374 Santa Fe Industries Inc v Green, 430 US 462 (1977)............................................................189 Schreiber v Burlington Northern, 472 US 1 (1985).............................................................189

Table of Legislation

I  European Union EC Treaty............................. 4, 5, 6, 9, 11, 12, 21, 22, 55, 57, 58, 59, 60, 99, 102, 109, 112, 128, 129, 130, 132, 148, 161, 170, 177, 178, 179, 191, 193, 210, 257, 296, 304, 307, 308, 309, 315, 316, 317, 318, 319, 320, 325, 328, 333, 347, 349, 350, 351, 358, 359, 360, 362, 369, 401, 412, 413, 415, 416, 421, 423, 431, 432, 433, 434, 437, 447, 460, 461, 463, 466, 481, 490, 497   Title VI..................................................................................................................................11    Art 2....................................................................................................................408, 463, 495    Art 3............................................................................................................................463, 495    Art 3(1)(c)..................................................................................................................408, 495    Art 3(1)(g)..........................................................................................................................495    Art 6(2).................................................................................................................................11    Art 10 (now 4(3) TEU)......................................................................................................456    Art 12 (now 18 TFEU)..............................................................16, 17, 18, 101, 337, 338, 348    Art 13 (now 19 TFEU)............................................................................... 13, 14, 15, 16, 357    Art 14(1)–(2)......................................................................................................................408    Art 17 (now 20 TFEU).......................................................................................................338    Art 17(1).............................................................................................................................332    Art 18 (now 21 TFEU)...............................................................................................338, 359    Art 18(1).............................................................................................................................362    Art 28 (now 34 TFEU).......................................................................................205, 350, 358    Art 30 (now 36 TFEU)...............................................................................................205, 206    Art 39(3) (now 45(3) TFEU).............................................................................................362    Art 43 (now 49 TFEU)......................................... 4, 5, 27, 100, 111, 129, 145, 180, 183, 337, 411, 439, 440, 447, 454, 459, 463, 464, 466, 467    Art 43(2).............................................................................................................................100    Art 46(1) (now 52(1) TFEU).....................................................................................179, 362    Art 47(2)...............................................................................................................................97    Art 48 (now 54 TFEU)................................ 110, 111, 129, 145, 337, 411, 439, 440, 447, 534    Art 49 (now 56 TFEU)..............................................................64, 73, 99, 100, 184, 185, 459    Art 50 (now 57 TFEU)...............................................................................................185, 318    Art 51(2) (now 58(2) TFEU).............................................................................................185    Art 52............................................................................................................ 58, 529, 531, 550    Art 54(3)(g)..........................................................................................................................59    Art 55 (now 62 TFEU).................................................................................................97, 362    Art 56 (now 63 TFEU)............................................................................... 184, 185, 459, 463    Art 58....................................................................................................................58, 529, 550

616  Table of Legislation    Art 87(1) (now 107(1) TFEU)...........................................................................................501    Art 88(2).............................................................................................................................514    Art 88(3) (now 108(3) TFEU)...........................................................................................503    Art 93..........................................................................................................................494, 509    Art 98..................................................................................................................................123    Art 139(2).............................................................................................................................16    Art 141 (now 157 TFEU).............................................................................................97, 356    Arts 189–190......................................................................................................................333    Art 220 (now 19(1) TEU)....................................................................................................11    Art 230 (now 263 TFEU)...........................................................................................243, 507    Art 234 (now 267 TFEU).....................................................................................12, 100, 110    Art 249 (now 288 TFEU).............................................................................................16, 456    Art 249(3)...........................................................................................................................456    Art 253................................................................................................................................507    Art 288 (now 340 TFEU)...................................................................................................235    Art 293................................................................................................................................416 EEC Treaty........................................................................................................................16, 318    Art 7 (later 12 EC, now 18 TFEU).......................................................................................16    Art 52 (now 49 TFEU).......................................................................................................461 ..................................................................................................................... Art 54(3)(g)…. 161    Art 95 (later 90 EC, now 110 TFEU).................................................................................451    Art 95(3).............................................................................................................................451    Art 215 (later 288 EC, now 340 TFEU).............................................................................248 EU Charter of Fundamental Rights......................................................................170, 313, 521    Art 21(1)...............................................................................................................................17    Art 28..................................................................................................................................104    Art 51(1).........................................................................................................................18, 22    Art 54...................................................................................170, 225, 313, 319, 325, 521, 556 European Coal and Steel Community (ECSC) Treaty.............................................................8    Art 31......................................................................................................................................8    Art 33......................................................................................................................................8    Art 40......................................................................................................................................8 Treaty of Amsterdam..................................................................................... 112, 132, 330, 357 Treaty on European Union (TEU)..................................................................................11, 333    Art 1......................................................................................................................................11    Art 2................................................................................................................................10, 12    Art 3............................................................................................................................463, 495    Art 3(3).......................................................................................................................408, 495    Art 4(2) (ex 10 EC)............................................................................................................499    Art 4(3) (ex 10 EC)............................................................................................................323    Art 6(1).................................................................................................................................17    Art 6(3) (ex 6(2)).................................................................................................................11    Art 14(1)–(3)......................................................................................................................333    Art 16..........................................................................................................................222, 223    Art 19(1) (ex 220)..................................................................................................................8    Art 48..................................................................................................................................222    Art F(2) (later 6(2)).............................................................................................................10

Table of Legislation  617    Art K.3................................................................................................................................559 Treaty on European Union (TEU, pre-Lisbon)............................................................132, 298    Art 35....................................................................................................................................11    Art 46....................................................................................................................................10    Final Act, Second Declaration...........................................................................................330 Treaty on the Functioning of the European Union (TFEU)..........5, 11, 18, 49, 50, 54, 66, 68, 87, 148, 156, 185, 221, 315, 318, 320, 435, 469, 494, 495, 510    Art 3............................................................................................................................463, 495    Art 3(1)(b)..................................................................................................................408, 495    Art 4............................................................................................................................463, 495    Art 4(2)(a)..................................................................................................................408, 495    Art 5 (ex 5 EC)...........................................................................................................211, 463    Art 5(2)(g)............................................................................................................................59    Art 6....................................................................................................................................463    Art 18 (ex 12 EC)................................................................................... 16, 95, 299, 472, 495    Art 19 (ex 13)................................................................................................... 13, 14, 17, 568    Art 19(1)...............................................................................................................................16    Art 20 (ex 17 EC)...............................................................................................................330    Art 21 (ex 18 EC)....................................................................................... 228, 359, 362, 472    Art 26(1)–(2)......................................................................................................................408    Art 34........................................................................................................................50, 52, 55    Art 36 (ex 30 EC)........................................................................................... 50, 51, 206, 296    Art 45 (ex 39 EC).......................................................................................................301, 359    Art 45(3).............................................................................................................................296    Art 49 (ex 43 EC)...................................................................4, 27, 58, 78, 79, 107, 108, 143, 144, 148, 180, 183, 461, 462, 464, 466, 529    Art 50(2)(g)..........................................................................................................................59    Art 52 (ex 46 EC).........................................................................................................50, 296    Art 53(1) (ex 47(2)EC)..................................................................................................94, 97    Art 54 (ex 48 EC)............................................................................... 108, 143, 144, 148, 529    Art 56 (ex 49 EC)..................................................... 50, 55, 64, 66, 67, 68, 70, 73, 87, 90, 93, 94, 95, 97, 99, 100, 103, 104, 221, 318    Art 56(1) (ex 49(1) EC).......................................................................................................75    Art 57 (ex 50 EC).................................................................................................................70    Art 62 (ex 55 EC)...................................................................................................50, 94, 296    Art 101 (ex 81 EC).............................................................................................................509    Art 102 (ex 82 EC)....................................................................... 19, 226, 227, 327, 496, 509    Art 106(1) (ex 86(1) EC)...............................................................................................78, 79    Art 107 (ex 87 EC)..................................................................................... 501, 502, 505, 513    Art 107(1)................................................................................... 500, 501, 502, 509, 510, 511    Art 107(2)........................................................................................... 494, 500, 510, 513, 514    Art 107(3)................................................................................................... 500, 510, 513, 514    Art 107(3)(b) (ex 87(3)(b)EC)..........................................................................................182    Art 108 (ex 88 EC).....................................................................................................498, 499    Art 108(2)........................................................................................... 506, 509, 512, 513, 514    Art 108(3)...........................................................................................................................517    Art 110 (ex 90 EC).............................................................................................................423

618  Table of Legislation    Art 114 (ex 100A EEC, later 95 EC)..................................................................................257    Art 115 (ex 100 EEC, later 94 EC).....................................................................................257    Art 116 (ex 96 EC).............................................................................................................509    Art 120 (ex 98 EC).............................................................................................................123    Art 155(2).............................................................................................................................16    Art 157..................................................................................................................................97    Art 258..................................................................................................................................21    Art 260..................................................................................................................................21    Art 263 (ex 230 EC)................................................................................... 8, 12, 21, 170, 323    Art 263(4)...........................................................................................................................507    Art 267 (ex 234 EC)..............................................12, 210, 212, 215, 227, 230, 526, 541, 546    Art 288 (ex 249 EC).....................................................................................................16, 221    Art 294................................................................................................................................222    Art 296 (ex 253 EC)...................................................................................................221, 507    Art 325................................................................................................................................558    Art 340 (ex 288 EC)...................................................................................................8, 12, 21    Art 340(2)...................................................................................................................494, 568 Treaty of Lisbon............................................................................................... 11, 330, 333, 416 Treaty of Maastricht 1992 see Treaty on European Union (TEU, pre-Lisbon) Directives (listed chronologically) 6th VAT Directive see Directive 77/388/EEC 10th Company Law Directive see Directive 2005/56/EC (Cross Border Directive) Citizenship Directive see Directive 2004/38/EC Merger Tax Directive see Directive 90/434/EC MiFID see Directive 2004/39/EC Parent-Subsidiary Directive see Directive 90/435/EEC Directive 64/221/EEC.................................................................... 292, 295, 317, 323, 345, 361 Directive 64/227/EEC, Art 3..................................................................................................350 Directive 67/226/EEC (Harmonisation of Turnover Tax Legislation)................................497   Recitals 1–4.........................................................................................................................497   Recital 12............................................................................................................................497 Directive 68/360/EEC.................................................................... 292, 295, 317, 323, 345, 361 Directive 69/335/EEC............................................................................................................397 Directive 72/194/EEC.................................................................... 292, 295, 317, 323, 345, 361 Directive 73/148/EEC.................................................................... 292, 295, 317, 323, 345, 361 Directive 75/34/EEC...................................................................... 292, 295, 317, 323, 345, 361 Directive 75/35/EEC...................................................................... 292, 295, 317, 323, 345, 361 Directive 75/117/EEC (Equal Pay)..........................................................................................97    Art 1......................................................................................................................................97 Directive 77/91/EEC (2nd Company Law)...........................................................................173    Art 25(1).....................................................................................................................172, 173 Directive 77/388/EEC (6th VAT)........................... 102, 173, 174, 240, 385, 395, 418, 420, 442, 443, 452, 453, 455, 458, 497, 522, 527, 529, 530    Art 5....................................................................................................................................471    Art 5(1)...............................................................................................................................471

Table of Legislation  619    Art 6(1)...............................................................................................................................471    Art 11A(1)(a).....................................................................................................................385    Art 17..................................................................................................................................418    Art 20..................................................................................................................................452    Art 22(8).............................................................................................................................452    Art 27............................................................................................................................31, 384    Art 28(2)(a)........................................................................................................................455 Directive 77/799/EEC............................................................................................................180 Directive 78/855/EEC, Art 13(2)...........................................................................................161 Directive 85/611/EEC....................................................................................................xvii, 193 Directive 86/653/EEC (Commercial Agents)................................................................250, 257    Art 3(1)...............................................................................................................................259    Art 4....................................................................................................................................250 Directive 88/361/EEC............................................................................................................185 Directive 89/48/EEC (Recognition of Diplomas)................................................................351 Directive 89/104/EEC............................................................................................................241 Directive 89/552/EEC (TV without Frontiers).............................63, 64, 65, 72, 73, 74, 75, 76, 77, 79, 80, 81, 82, 83, 84, 85, 87, 88, 90, 91, 93   Preamble...............................................................................................................................66     Recital 13..........................................................................................................................74     Recital 14........................................................................................................................520    Art 2.............................................................................................................. 64, 72, 74, 76, 83    Art 2(1).....................................................................................................................64, 75, 78    Art 2(2)............................................................................................................... 73, 83, 84, 85    Art 2(3).....................................................................................................................83, 84, 85    Art 2(3)(a)–(c).....................................................................................................................84    Art 2(4)...........................................................................................................................83, 85    Art 2(4)(a)–(c).....................................................................................................................84    Art 2(5).....................................................................................................................83, 84, 85    Art 3..............................................................................................................................85, 520    Art 3(1)–(2)..........................................................................................................................85    Art 3(3).................................................................................................................................86    Art 3(4)(a)–(b).....................................................................................................................86    Art 3(5).................................................................................................................................86    Art 3(6) (ex 3(2)).................................................................................................................64    Arts 4–5................................................................................................................................76    Art 23(a).........................................................................................................................83, 85 Directive 89/592/EEC............................................................................................................182 Directive 90/314/EEC............................................................................................................257 Directive 90/364/EEC.................................................................... 292, 295, 317, 323, 345, 361 Directive 90/365/EEC.................................................................... 292, 295, 317, 323, 345, 361 Directive 90/434/EEC (Merger Tax Directive).........................29, 30, 175, 369, 370, 371, 383, 389, 390, 421, 456, 457, 458, 475   Titles II–IV.................................................................................................................456, 520    Art 1....................................................................................................................................520    Art 11.......................................................................................................... 421, 455, 471, 472    Art 11(1).............................................................................................................................208

620  Table of Legislation    Art 11(1)(a)..............................30, 31, 175, 383, 390, 421, 422, 455, 456, 457, 475, 520, 521 Directive 90/435/EEC (Parent-Subsidiary)...........................................................379, 390, 475    Art 1.2.................................................................................................................................390    Art 2(1)(b)–(c)...................................................................................................................475    Art 3(2)...............................................................................................................................475 Directive 90/619/EEC............................................................................................................259 Directive 91/439/EEC (Driving Licences).....................................................................351, 352 Directive 91/533/EEC..............................................................................................................99 Directive 92/81/EEC..............................................................................................................501 Directive 93/6/EEC........................................................................................................xvii, 193 Directive 93/13/EEC (Unfair Contract Terms)............................................ 171, 247, 250, 257    Art 3(1).......................................................................................................................250, 259 Directive 93/22/EEC (Investment Services).........................................................xvii, 193, 197 Directive 93/96/EEC...................................................................... 292, 295, 317, 323, 345, 361 Directive 96/71/EC (Posted Workers)..............................93, 94, 96, 97, 99, 100, 101, 103, 104   Recital 5................................................................................................................................94   Recital 13..............................................................................................................................94   Recital 17..............................................................................................................................94    Art 3..............................................................................................................................96, 100    Art 3(1)..........................................................................................94, 95, 96, 97, 99, 103, 104    Art 3(1)(g)............................................................................................................................99    Art 3(7).....................................................................................................................94, 97, 98    Art 3(8)...........................................................................................................................95, 96    Art 3(10)...................................................................................................................94, 98, 99 Directive 97/7/EC (Distance Selling)............................................................................257, 259 Directive 97/36/EC............................................................................................... 63, 83, 84, 520   Preamble...............................................................................................................................83     Recital 14..........................................................................................................................83     Recital 19..........................................................................................................................83 Directive 97/81/EC (Part-time Workers)................................................................................99 Directive 98/27/EC.................................................................................................................259 Directive 98/44/EC.................................................................................................................242 Directive 99/44/EC.................................................................................................................247    Art 3(3)...............................................................................................................................259    Art 7....................................................................................................................................247 Directive 99/70/EC (Fixed Term Work)............................................ 16, 99, 249, 327, 357, 358    Art 5............................................................................................................................249, 250    Art 5(1)...............................................................................................................................249 Directive 2000/12/EC.....................................................................................................xvii, 193 Directive 2000/31/EC (E-Commerce)..................................................................................257 Directive 2000/35/EC.............................................................................................................257    Art 4....................................................................................................................................241 Directive 2000/78/EC........................................................................13, 14, 15, 16, 17, 357, 358    Art 18....................................................................................................................................14 Directive 2001/19/EC.............................................................................................................351 Directive 2002/22/EC (Universal Service)......................................................................82, 197    Art 31(1)...............................................................................................................................82

Table of Legislation  621 Directive 2002/65/EC.............................................................................................................259 Directive 2002/83/EC.............................................................................................................186    Art 6(3)...............................................................................................................................188 Directive 2003/6/EC (Market Abuse)............................................................ 182, 189, 190, 191   Recital 14............................................................................................................................182   Recital 17............................................................................................................................182    Art 1(2)(a)–(b)...................................................................................................................190    Art 2(1)(d)..........................................................................................................................182 Directive 2003/48/EC.............................................................................................................475 Directive 2003/86/EC (Family Reunification), Art 16(2)(b)...............................................343 Directive 2003/123/EC...................................................................................................390, 475 Directive 2003/124/EC...........................................................................................................190    Arts 4–5..............................................................................................................................190 Directive 2004/38/EC (Citizenship)...................... 171, 229, 291, 292, 293, 295, 296, 298, 317, 319, 323, 330, 345, 346, 347, 348, 349, 362, 556   Preamble.............................................................................................................................346   Recital 3...................................................................................................... 295, 299, 309, 345   Recital 4..............................................................................................................................348   Recital 6..............................................................................................................................311   Recital 8..............................................................................................................................311   Recital 18............................................................................................................................310   Recitals 23–24.....................................................................................................................310   Recital 28....................................................................................................................229, 343   Chapter VI..........................................................................................................................362    Art 3(1)...............................................................................................................................345    Art 7....................................................................................................................................345    Art 7(3)...............................................................................................................................299    Art 7(3)(d)..........................................................................................................................346    Art 14(4)(b)........................................................................................................................345    Art 24..........................................................................................................................328, 345    Art 24(2)..................................................................................................... 318, 345, 347, 348    Art 27....................................................................................................................98, 341, 344    Art 28..........................................................................................................................310, 344    Arts 30–31..................................................................................................................295, 344    Art 35........................................................................ 22, 29, 31, 171, 219, 223, 229, 293, 295, 300, 308, 313, 317, 323, 330, 341, 343, 344, 361, 520, 556 Directive 2004/39/EC (Market in Financial Instruments) (MiFID).......... xvii, 186, 187, 188, 189, 191, 193, 194, 195, 196, 197, 198, 549   Recital 22............................................................................................................................194   Recital 23............................................................................................................................195   Recital 32............................................................................................................................196    Art 5....................................................................................................................................186    Art 5(1).......................................................................................................................194, 195    Art 5(4)........................................................................188, 194, 195, 196, 197, 198, 199, 549    Art 6(3).......................................................................................................................195, 196    Art 7(1).......................................................................................................................194, 195    Art 8............................................................................................................................194, 195

622  Table of Legislation    Art 19..................................................................................................................................187    Arts 21–22..........................................................................................................................187    Art 25..................................................................................................................................187    Arts 27–28..........................................................................................................................187    Art 31..........................................................................................................................186, 196    Art 31(1)–(3)......................................................................................................................186    Art 32..........................................................................................................................186, 196    Art 32(7).............................................................................................................187, 188, 196    Art 32(8).............................................................................................................................188    Art 60..........................................................................................................................186, 196    Art 61..................................................................................................................................196    Art 61(1)–(2)......................................................................................................................187    Art 62..........................................................................................................................187, 196    Art 62(1).....................................................................................................................196, 197 Directive 2004/48/EC (Enforcement; Intellectual Property)...............................................242    Art 2(1)...............................................................................................................................242    Art 3....................................................................................................................................243    Art 3(1)...............................................................................................................................243 .................................................................................................................... Art 3(2).....243, 520 Directive 2004/72/EC.............................................................................................................190    Art 2....................................................................................................................................190 Directive 2005/19/EC.............................................................................................208, 390, 421 Directive 2005/56/EC (Cross-Border Mergers)................................... 115, 120, 122, 123, 135, 145, 154, 155, 160, 161    Art 4....................................................................................................................................161    Art 4(2)...............................................................................................................................145 Directive 2006/48/EC.............................................................................................................186    Art 11(2).............................................................................................................................188 Directive 2006/112/EC (Common VAT System)..........................................................489, 497   Recital 4..............................................................................................................................497    Art 395..........................................................................................................................31, 521 Directive 2006/123/EC...................................................................................................222, 223 Directive 2007/65/EC........................................................................................... 63, 83, 85, 520   Preamble     Recitals 31–33............................................................................................................85, 86     Recitals 40–42..................................................................................................................85    Art 1(a).................................................................................................................................85    Art 1(d).................................................................................................................................85    Art 1(3)...........................................................................................................................64, 85    Art 1(5).....................................................................................................................64, 85, 86 Directive 2008/95/EC (Trade Marks)............................................................................241, 242    Art 3(2)(d)..........................................................................................................................241 Directive 2008/122/EC (Timeshares)....................................................................................257 Directive 2009/136/EC.............................................................................................................82 Directive 2010/13/EU (Audiovisual Media Services)..................63, 79, 80, 84, 88, 91, 93, 104    Preamble, Recitals 40–42.....................................................................................................86    Art 2(2)...........................................................................................................................84, 85

Table of Legislation  623    Art 2(3).................................................................................................................................85    Art 2(3)(a)–(c).....................................................................................................................84    Art 2(4)–(5)....................................................................................................................84, 85    Art 4(1)–(2)..........................................................................................................................85    Art 4(3).................................................................................................................................86    Art 4(4)(a)–(b).....................................................................................................................86    Arts 16–17............................................................................................................................76    Art 29..............................................................................................................................83, 85 Regulations (listed chronologically) Brussels I Regulation see Regulation (EC) 44/2001 EIR see Regulation (EC) 1346/2000 Regulation (EEC) 1612/68............................................................. 292, 295, 317, 323, 345, 361   Recitals................................................................................................................................310    Art 7....................................................................................................................................299    Art 10.......................................................................................................... 292, 317, 339, 341    Art 10(1).............................................................................................................................292 Regulation (EEC) 1408/71.....................................................................................................318    Art 22..................................................................................................................................318 Regulation (EEC) 2730/79.............................................................................................206, 529 Regulation (EEC) 568/85.......................................................................................................206 Regulation (EC) 40/94...........................................................................................................241 Regulation (EC) 2988/95.......................................................................................................214   Recital 4..............................................................................................................................559   Recital 5..............................................................................................................................559    Art 1(1)...............................................................................................................................559    Art 1(2)...............................................................................................................................559    Art 4(1)...............................................................................................................................559    Art 4(3)................................................................................208, 213, 520, 521, 539, 540, 559 Regulation (EC) 659/99.........................................................................................505, 506, 509    Art 1(g)...............................................................................................................................506    Art 4(3)...............................................................................................................................506    Art 7(3)–(4)........................................................................................................................506    Art 9............................................................................................................................514, 515    Art 14..........................................................................................................................505, 515    Art 14(1)............................................................................................. 494, 505, 506, 515, 517    Art 14(3).............................................................................................................................515    Art 16..........................................................................................................................509, 517 Regulation (EC) 1346/2000 (Insolvency Proceedings)............... 137, 138, 139, 140, 143, 144, 145, 146, 148, 152, 153, 154, 155, 157, 158, 162, 168, 445   Preamble.............................................................................................................................162     Recital 2..........................................................................................................................146     Recital 4..........................................................................................................146, 160, 162     Recital 8..........................................................................................................................146     Recital 10........................................................................................................................152     Recital 13........................................................................................................153, 159, 169

624  Table of Legislation     Recital 16........................................................................................................................146     Recitals 19–20................................................................................................................146     Recital 22................................................................................................................158, 166    Art 1............................................................................................................................152, 153    Art 1(1)...............................................................................................................................152    Art 2(h).......................................................................................................................146, 165    Art 3(1)................................................ 138, 146, 147, 148, 153, 154, 158, 159, 166, 167, 445    Art 3(2).......................................................................................................................140, 165    Art 3(3)-(4)........................................................................................................................140    Art 4....................................................................................................................................139    Art 4(2)(m)........................................................................................................................158    Art 5....................................................................................................................................165    Art 10..........................................................................................................................140, 165    Art 16..................................................................................................................................158    Art 16(1).............................................................................................................................158    Art 17..................................................................................................................................158    Art 26..................................................................................................................................158    Art 38..................................................................................................................................151    Annex A..............................................................................................................152, 153, 166    Annex C..............................................................................................................................151 Regulation (EC) 44/2001 (Brussels I).................................................. 243, 244, 245, 261, 262, 264, 265, 269, 271, 275, 276, 284   Preamble.............................................................................................................................264   Recital 11............................................................................................................................256    Art 2............................................................................................................ 245, 256, 268, 269    Art 5....................................................................................................................................256    Art 5(1).......................................................................................................................244, 266    Art 5(1)(a)..................................................................................................................244, 245    Art 6............................................................................................................................256, 286    Art 6(1)................................................................................244, 256, 259, 266, 285, 286, 549    Art 6(2)....................................................................................... 244, 251, 255, 256, 286, 520    Arts 7–16............................................................................................................................256    Art 17..........................................................................................................................244, 256    Arts 18–21..........................................................................................................................256    Art 22..........................................................................................................................245, 256    Art 22(5).............................................................................................................................245    Art 23..................................................................................................................244, 245, 270    Art 27..................................................................................................................262, 263, 270    Art 71..................................................................................................................................270 Regulation (EC) 2157/2001 (European Company Statute)......................... 120, 122, 123, 160    Art 7............................................................................................................................115, 160    Art 8............................................................................................................................124, 160 Regulation (EC) 1/2003, Art 23(2)(a)...................................................................................226 Regulation (EC) 1882/2003...................................................................................................351 Regulation (EC) 883/2004.....................................................................................................317 Regulation (EC) 603/2005.....................................................................................................166 Regulation (EC) 694/2006.....................................................................................................166

Table of Legislation  625 Regulation (EC) 681/2007.....................................................................................................166 Regulation (EC) 864/2007 (Rome II), Recital 27.................................................................104 Regulation (EC) 593/2008 (Rome I).......................................................................................97 Regulation (EC) 788/2008.....................................................................................................166 Regulation (EC) 207/2009 (Community Trade Marks).......................................................241    Art 51(1)(b)........................................................................................................................241 Regulation (EC) 987/2009.....................................................................................................317

II  International Agreement between the EC and the Swiss Confederation in the audiovisual field,    Annex II, Art 1(2)(b)...................................................................................................86, 521 Association Agreement with Turkey.............................................................................341, 342    Additional Protocol 1970, Art 41(1).................................................................................341 Brussels Convention on Jurisdiction and the Enforcement of Judgments in Civil    and Commercial Matters (1968)....................... 243, 245, 261, 262, 264, 265, 267, 268, 269, 270, 271, 272, 274, 275, 276, 281, 282, 284, 285    Art 2............................................................................................................................265, 284    Art 5(1)...............................................................................................................270, 282, 283    Art 6(1)....................................................................................................... 244, 265, 283, 284    Art 6(2)....................................................................................................... 244, 251, 285, 520    Art 13..................................................................................................................................248    Arts 14–15..........................................................................................................................248    Art 16..................................................................................................................................245    Art 16(5).............................................................................................................................245    Art 17.......................................................................................... 235, 270, 280, 281, 282, 283    Art 21.................................................................................................. 262, 267, 270, 280, 281    Art 27..................................................................................................................................246    Art 57..................................................................................................................................270 European Convention for the Protection of Human Rights and Fundamental   Freedoms (1950).......................................................................................... 10, 170, 319, 325    Art 8....................................................................................................................................306    Art 10....................................................................................................................................64    Art 12..................................................................................................................................344    Art 17.....................................................................................22, 170, 225, 313, 319, 325, 521    4th Protocol, Art 3.............................................................................................................331 European Convention on Transfrontier Television (1989)....................................................80    Art 16....................................................................................................................................80   Protocol 1998.......................................................................................................................80 European Economic Area (EEA) Agreement (1994).............................................................81 International Covenant on Civil and Political Rights (1966)    Art 5....................................................................................................................................325    Optional Protocol, Art 3....................................................................................................313 International Covenant on Economic, Social and Cultural Rights (1966), Art 5...............325 Lugano Convention (1988)...................................................................................................261 Rome Convention (1980)........................................................................................................97

626  Table of Legislation    Art 6(2)(a)............................................................................................................................97 Statute of the International Court of Justice (1945)    Art 38..................................................................................................................................237    Art 38(1)(c)............................................................................................................................8 TRIPS Agreement (1994), Art 41(1).....................................................................................243 UN Universal Declaration of Human Rights (1948)    Art 18..................................................................................................................................319    Art 30..................................................................................................................................325    Art 54....................................................................................................................................22 Vienna Convention on the Law of Treaties (1969)    Art 26..................................................................................................................................312    Art 31..................................................................................................................................312

III  National Austria Income Tax Act, s 99 par 2 number 2....................................................................................454 Belgium Broadcasting Decree of the French Community, 27 February 2003, Art 2(7)......................83 Decree of the Council of the Flemish Community, 4 May 1994    Art 3......................................................................................................................................74    Art 5......................................................................................................................................74    Art 10(1) No 2......................................................................................................................77    Art 10(1) No 4......................................................................................................................77 Decree of the Council of the French Community, 17 July 1987............................................73    Art 26....................................................................................................................................73    Art 26(b)...............................................................................................................................73 Decree of the Council of the French Community, 19 July 1991............................................73 Decree of the Flemish Executive, 28 January 1987.................................................................77 Decree of the Flemish Executive, 12 June 1991......................................................................77 France Act of Fiscal Procedure (Livre des procedures fiscales).......................................................555    Art L 64...............................................................................................................................555 Civil Code.........................................................................................................................34, 220    Art 4........................................................................................................................................9    Art 1134..............................................................................................................................254    Arts 1382–1383..................................................................................................................254 Code de Commerce...............................................................................................................149    Art L 622–32.......................................................................................................................149    Art L 626-9.........................................................................................................................162    Art L 628–1.........................................................................................................................149    Art L 631–19.......................................................................................................................162    Art L 641–1.........................................................................................................................162 Code Général des Impôts (CGI), Art 167a...........................................................................437

Table of Legislation  627 Finance Law 2008...................................................................................................................379 Loi no 2008-776 du 4 août 2008 de modernisation de l’économie.....................................119 Germany Abgabenordnung, s 42(1)..............................................................................................373, 556   s 5........................................................................................................................................145 Aussensteuergesetz (foreign tax law)....................................................................................414 Civil Code (BGB).....................................................................................................................34   s 157....................................................................................................................................312   s 226................................................................................................................................34, 41   s 242..............................................................................................................................41, 312   s 475....................................................................................................................................247 Constitution (Grundgesetz), Art 18......................................................................................313 GmbH-Gesetz   s 4a......................................................................................................................................145   s 5(1)...................................................................................................................................119   s 5(a)...................................................................................................................................119 Missbrauchsbekämpfungs- und Steuerbereinigungsgesetz (law combating abuse and   harmonising taxation).......................................................................................................414 Greece Civil Code, Art 281......................................................................................... 169, 172, 173, 560 Ireland, Republic of Constitution...........................................................................................................................336 Italy Civil Code    Art 833................................................................................................................................536 Constitution...........................................................................................................................391    Art 42..................................................................................................................................374    Art 53..................................................................................................................................391 no 4 Legge Fallimentare, Art 9(2).........................................................................................147 Netherlands Media Act 2008 (Mediawet)    Art 57(1)...............................................................................................................................66    Art 57(4)...............................................................................................................................66    Art 66(1)...............................................................................................................................68 Wet op de Formeel Buitenlandse Vennootschappen (Law on foreign companies)............110 Spain Civil Code...............................................................................................................................331 Ley 2/1995 of 23 March.........................................................................................................118 Ley 7/2003 of 1 April..............................................................................................................118 no 1 Ley Concursal, Art 10....................................................................................................147

Index Introductory Note References such as ‘178–9’ indicate (not necessarily continuous) discussion of a topic across a range of pages. Wherever possible in the case of topics with many references, these have either been divided into sub-topics or only the most significant discussions of the topic are listed. Because the entire work is about ‘abuse of law’, the use of this term (and certain others which occur constantly throughout the book) as an entry point has been minimised. Information will be found under the corresponding detailed topics. abstract anti-abuse rules 114, 147, 379, 479–80 abus de droit 258 abuse see also Introductory Note    classic 4–5    concepts 3–4, 142–4, 154–5, 377–8, 407, 420, 422–4    definition of 387, 416    doctrine of 103–4, 169–71, 271, 281–6, 315–16, 319–21, 490–2    of freedom of establishment 138, 144, 160–1    intention of 69, 71, 87–8, 143, 156   of law see abuse    market 182, 189–91    objective element of 67, 69, 78, 180    potential 111, 138, 153, 182    of power 21, 258, 323, 496, 500–1    principle of 65–8, 70–4, 78–83, 87–8, 239–51, 472–3, 476–8    prohibition of 20–3, 64–5, 77–8, 83, 86, 88–9, 418–19    reasoning 482, 486–7, 490–1    of rights 53–5, 59–61, 169–99, 219–25, 258–9, 322–3, 328–9      in civil law tradition 33–46      definition of 20, 219      doctrine of 33–4, 36, 39–43, 46, 173, 220–4, 230–1      general doctrine of 170, 378      general test for 219–31    risk of 56, 75, 80, 85, 283    single principle of 469–83    subjective elements of 23, 69–71, 87, 102, 226–8, 239, 448–50    tax 377, 379, 419, 474–5, 477    tests 79, 87, 129, 207–8, 210–17, 393, 481–2 abusive practices 26–8, 174–5, 387–90, 419–21, 438–43, 452–3, 455–7    ground of prevention of 179, 402, 412, 438–9 abusive transactions 248, 369, 387, 393, 486 accrual of tax advantages 102, 174, 240, 367, 371, 419–20, 443 acquis communautaire 236–8, 240–1, 246, 249–51, 257, 259 actual establishment 112, 129, 132, 180, 412, 444, 463 actual pursuit 84, 112, 132, 228, 368, 402, 446–7 administrative law 9–10, 21, 36, 235, 355–6

advertising 63, 67, 69, 72, 76–7, 79–81, 86 aemulatio 33–4, 36, 38, 41, 43–4 age 13–17, 109, 167, 176, 197, 292, 357–8 agriculture 61, 87, 171, 203–17, 220, 257, 259 aid, State see State aid allocation of regulatory competence 49–50, 321, 349 anarchy 38–9 anti-abuse clauses 319, 477–82 anti-abuse concepts 408, 455, 473 anti-abuse doctrine 408, 421–2, 435, 492 anti-abuse legislation/measures/provisions/rules 113–14, 370–2, 378–9, 389–90, 401, 412–16, 477–82 anti-abuse principle 371, 391, 492, 494–6, 498–501, 505–8    general 494, 498, 500, 506–8 arbitrage, regulatory 60, 120, 127, 132–3, 135–6, 194–5, 197–8 argument, cohesion 473–4, 477–9 arm’s-length principle 403–4, 438, 446, 478, 480, 482 arrangement test, artificial 112–13, 130, 433, 448–9, 457, 475–7, 479 artificial arrangements see wholly artificial arrangements artificial constructions 402, 431–4 artificial establishments 131–2 artificial transactions 189, 191, 213, 369, 489, 491 artificiality 174, 326–7, 369, 374, 397–8, 431–2, 449 assertions of jurisdiction 265–6, 270 assets, debtors 138, 147, 149, 156–8, 162–3 assurances 515–16 attainment of tax advantages 102, 174, 240, 418–20, 430, 443 attribution    of income 474–5    of nationality 333–4, 336–7, 352    of powers 15    of rights 381 atypical situations 381, 389 Austria 262, 273, 280–1, 303, 346, 349, 440 authorities, tax 5, 31, 179, 418, 423–5, 444, 452 avoidance, tax 365–80, 384–90, 401, 421–3, 435–6, 447–8, 486–8

630 Index bad faith 23, 241–2, 249 bankruptcy 139–41, 145, 154–7, 165 see also insolvency banks 140, 182–3, 186, 517 behaviour    abusive 108, 110, 113–15, 123, 137, 166, 265–6    fraudulent 300, 317, 320–1    strategic 160, 166–7 Belgium 55–6, 64–5, 73–7, 82, 331–2, 414, 514–16    cable network 65, 72–4, 76–7, 80, 86 benefit tourism 311, 349 benefits, social 298–300, 302, 345, 348, 491–2 borders 55–6, 115, 124, 472 broadcasting services 57, 67, 73, 75, 81, 93, 180–1    free movement of 63–91 business purposes 373, 477, 487 cable networks 64–5, 68, 71–4, 76–7, 80, 82, 86 Cadbury Schweppes 4–6, 111–13, 128–33, 176–81, 368–9, 401–4, 407–68   and Centros 6, 112–13, 128, 130–1    test 130–3 CAP see common agricultural policy capital, paid-up 228, 438–9 capital requirements, minimum 117, 125, 134 cargo interests 266–7 carousel fraud 386, 388, 391, 428 case-by-case approach 27, 29, 111, 113–14, 133, 137, 145 centre of gravity of activities 70–1, 75, 91 centre of main interests see COMI Centros 58–9, 101, 107–9, 112–15, 127–31, 133–4, 176–8   and Cadbury Schweppes (CS) 6, 112–13, 128, 130–1 CESR see Committee of European Securities Regulators CFCs see controlled foreign companies children 53, 80–1, 228–9, 276, 302, 334–6, 338 circumvention 63–5, 68–9, 72–3, 76–80, 85–6, 127–32, 176–9    illicit 321, 326    legitimate 102, 246 citizenship 228, 289, 293, 295, 310–11, 321–62, 491 civil law 17, 33, 35, 171, 221, 230, 392    systems 169, 220–1    tradition, abuse of rights in 33–46 civil procedure 235, 240, 243–5, 250–1, 255–6    international 233, 240, 243, 253, 261–88 classic abuse 4–5 close links 333–4, 336 codification 35, 83, 85, 91, 257, 292–3, 295 cohesion argument 473–4, 477–9 collective action 53, 95–6, 101, 104, 319 collective agreements 95–8, 100–1, 249 collusion 102, 207, 210, 239, 442, 450 COMI (centre of main interests) 138–40, 142, 144–56, 159–68, 445    debtors 145, 147, 160–1, 167    new 145, 149–50, 165    shifts 144–53, 155–6, 164–7      abusive 161, 163–5, 167–8     isolated 145, 154

commercial justification 403–4, 433, 440, 446, 457 commercial operations, normal 20, 174–5, 371, 417, 421, 442, 455 commercial transactions 51, 204–5, 207, 225, 241, 257, 396 Committee of European Securities Regulators (CESR) 61, 182, 188–90, 197–8 common agricultural policy 61, 87, 171, 203–17, 220, 257, 259 common core concept 378, 388–90 common law 36, 189, 220–1, 223, 237, 272–5, 488–9    jurisdictions 220, 223, 276 Community exporters 102, 207, 210, 239, 442, 450 companies    brass plate 464–5    debtor 148, 167, 445    foreign 109–10, 116–17    front 111, 131–2, 179    groups of 146, 153, 393, 403, 438, 445    holding 148, 153, 155, 436    insurance 95, 476, 512    letter-box 113, 130–3, 136, 159, 179, 444–5, 464    non-letter-box 133, 136    non-resident 403, 438–9, 446, 461–2, 486    parent 159, 179–80, 390, 402, 408–9, 436–7, 445    private 127–8, 134, 136, 165    public 124, 172    resident 179–80, 403, 409–11, 424, 437–8, 440, 446    start-up 119–20, 123–4, 135 company law 58–60, 120–2, 139–41, 143–5, 154, 160–1, 171–3    regulatory competition 107–36 compensation 95, 239, 517 competences 49, 52, 57, 61, 75, 181, 195–6    Member State 195, 331, 498    regulatory 49–50, 53, 61, 318, 321, 349 competent authorities 30, 180, 186–8, 194–5, 197, 206, 461 competition 59, 103, 119, 121–3, 125, 505, 517–18    distortion of 497, 500–1, 507, 517–18    regulatory 59–60, 63, 79, 86, 90, 107–39, 154–5 competitiveness 132–3 compound feeding-stuffs 203–4 concessions 54, 485–6 constitutional status 12, 20, 23, 322 constitutional traditions 8, 10, 14 constitutionalisation 307, 328 construction 25–9, 65, 74–5, 80, 87, 282–3, 362    principles of 25–9, 282, 489    statutory 373–6, 427    strict 65, 72, 88–9, 103 constructions, artificial 402, 431–4 consumer contracts 171, 236, 246–7, 257, 259 consumer protection 51, 53, 86, 183, 246–8, 257, 259–60 continuity 304, 346, 404 contract law 246–57 contracts 37–8, 94, 244, 246–7, 249–51, 254–5, 257–60    consumer 171, 236, 246–7, 257, 259    fixed-term 17, 249

Index  631    freedom of 170, 247, 257, 259, 327    on-call 302–3 controlled foreign companies (CFCs) 27, 377, 411–14, 424, 439–40, 444, 464–7    legislation/rules 402–3, 410, 413, 454, 457 convenience, marriages of 29, 171, 295–6, 305–6, 320–1, 340–1, 343–4 corporate debtors 148, 159, 162–4, 166–8, 445 corporate emigration 161, 164, 337 corporate freedom of establishment 160–1 corporate law see company law country of residence 337–8, 351, 414 credit 164, 178, 185, 404, 414, 486 credit contracts, efficiency of 140, 146, 156 creditors 59, 138–40, 144–51, 153, 155–68, 177–8, 248–50    non-adjusting 140, 154, 156, 165    secured 162–3 cross-border insolvencies 137, 145–7, 158, 162 cross-border mergers 108, 115, 120, 123, 135, 145, 160–1 cross-border trade see inter-State trade cross-border transactions 56, 109, 380, 434 cultural policy 57, 66, 82, 86–7 customs formalities 173, 206, 213 damages 11–12, 21, 172, 220, 227, 235, 517–18 debtors 138, 146–52, 156–60, 162–7, 248–9    assets 138, 147, 149, 156–8, 162–3    COMI 145, 147, 160–1, 167    corporate 148, 159, 162–4, 166–8, 445 deception 143, 189, 296, 300, 352 deference 60, 487 delaying tactics 262, 264, 276, 281 Denmark 6, 30, 58–9, 108–9, 128, 177–8, 228    law 6, 30, 58–9, 101, 177–8, 228–9, 455–7 deregulation 116, 118, 133–4, 176 detriment 30, 87, 158, 172, 248–9, 264, 316 direct effect 7, 13–18, 21–2, 30, 221, 357–8, 383 direct taxation 180, 386–92, 400–1, 405–6, 427, 431, 501 directors 110–11, 133, 182, 438–9 discharge periods 150, 162 discharge provisions 149–50, 156 discretion 10, 12, 64–5, 70–2, 87, 204, 418–19 discretionary powers 35, 423–4 discrimination 13, 15, 17, 44, 176, 410, 415–16    prohibition of 15, 17, 176 disproportionality 259–60, 304 see also proportionality distortion of competition 497, 500–1, 507, 517–18 diversion of profits 180, 412 dividends 30, 438–9, 474–5    hidden 436 domestic broadcasters 67–8, 70–3, 76–7, 79, 85, 180–1 domestic law/rules 5, 377, 450–1, 453–4, 457, 473–4, 476–7 domicile 265, 284 dominant positions 226–7, 237, 327, 496 double taxation 414–16, 486 driving licences 351–2

economic activity 5, 55, 112–13, 228–9, 345, 418–19, 446–7    genuine 112, 131–3, 301, 305–6, 412–13, 444, 464 economic choices 445, 462–3 economic effects 140–1, 147, 150, 156, 372, 502, 504 economic freedoms 104, 225, 324–5, 327, 329, 344 economic justification 371, 375, 418 economic life 130–1, 402, 432, 446–7 economic motivation 113, 130–1 economic objectives 213, 420 economic operators 20, 90, 213, 428–9, 493, 498–500, 505–8    conduct of 493, 498–9 economic reality 178–9, 371–4, 402–3, 432–4, 438–40, 444, 502 economic substance 372–3, 428 educational qualifications 325, 349–50 effective administration 146–7 effective fiscal supervision 461–2 effectiveness 31, 71–2, 86–7, 210–12, 309, 356, 423 efficiency of credit contracts 140, 146, 156 emigration, corporate 161, 164, 337 employees 13, 15, 43–4, 68, 97–8, 162–3, 246 employers 16–17, 43–4, 95, 103–4, 246, 249, 309–10    private 356–7 employment 13–14, 17, 94, 97, 100–1, 298–9, 301–5 employment contracts, fixed-term 17, 249, 357 employment relationships 97, 302, 304, 346–7 Emsland-Stärke test 23, 87, 102, 203–17, 220, 228–9, 397 entities    corporate 161, 186, 315–16, 326 see also company law    legal 11, 124, 398, 436, 444–5, 448, 457    public 143, 152–4, 156, 358 entrepreneurs 115, 117, 119–20, 122–3, 133–4, 143–4, 150 entrepreneurship 150, 162 equal treatment 13, 100, 299, 321, 337, 347, 357–8    principle of 14–15, 17–19, 318, 333, 506 equality 3, 7, 17, 19, 54, 94, 469–70 see also equal treatment    principle of 11, 14–15, 18, 470, 473, 479 equipment 6, 129, 131, 133, 403, 444, 463–4 equity 34, 42, 170, 261, 275, 372 essential activities 78–9, 88, 91 essential aim 398, 429, 449 establishment    actual 112, 129, 132, 180, 412, 444, 463    artificial 131–2    fictitious 131, 444    fixed 84, 112–13, 130, 132, 228, 402, 446–7    free movement of 171, 402   freedom of see freedom of establishment    genuine 5–6, 424, 433    outbound 409–10    permanent 414, 461–2, 475    primary 131–2, 148, 462    secondary 59, 77, 88, 111, 131, 329, 337 European Court of Human Rights 54, 64, 319–20, 333 eve of bankruptcy 140, 155–6 evidence, objective 59, 109, 113, 128, 133, 226, 413

632 Index exceptio doli generalis 33, 35–6 exchange of shares 20, 175, 369, 405, 421–2, 455–6 exclusive jurisdiction 244–5, 275, 281, 283, 287 exempt transactions 395, 442 exit taxes 475, 478–9 expectations, legitimate 3, 13, 23, 424–5, 495, 505, 515–16 export refunds 61, 173, 204, 206–8, 212–14, 305, 441 exporters 102, 203–5, 207–8, 210, 212–14, 239, 450 expression, freedom of 57, 66, 82, 319, 325 fairness 51, 89, 94, 250, 495–6, 508 faith    bad 23, 241–2, 249    good 35, 235, 247–50, 254–5, 259–60, 312, 499–500 families 275, 310–11, 317–18, 335, 345, 360 family law 341, 343 family life 223, 306, 311, 321, 339, 341 family members 295, 298–300, 304–6, 308–9, 339, 341, 361 favourable tax treatment 398, 430, 432–3, 502 feeding-stuffs, compound 203–4 fictions, legal 478–82 fictitious establishments 131, 444 finality and function 36–7, 39–41 financial law 60, 169–99 financial markets 181–3, 196, 198 financial services 181, 183–4, 186, 259, 414 fiscal restrictions 4–5 fiscal State aid 500–4, 507 fiscal supervision, effective 461–2 fiscal unity 404–6 fiscal vacuum 473–4, 476–8 fixed establishments 84, 112–13, 130, 132, 228, 402, 446–7 fixed-term work 16–17, 249, 357 foreign companies 109–10, 116–17    controlled 27, 131, 377, 401, 408, 459, 464–6 foreign subsidiaries 179–80, 402, 409 foreseeability 139–40, 159, 174, 312, 396, 405 formal investigation procedure 513–15 formalism 39–42, 115 formalities, customs 173, 206, 213 forum shopping 63, 90, 118, 138–41, 146–7, 153–6, 189–90 four freedoms 4, 52, 65 France 36–7, 42–3, 51–3, 56, 118–19, 162–3, 169    insolvency law 149, 162–3 franchise taxes 123, 136 fraud 142, 295–6, 317, 323–4, 342–3, 350–3, 386–9    carousel 386, 388, 391, 428 fraude à la loi 193–4, 258, 283–5 fraus legis doctrine 376, 401 free choice 69, 137, 144–5 free movement 183–6, 299–302, 305–7, 311–13, 315–16, 358–62, 446–7    of broadcasting services 63–91    of goods 18, 46, 49–61, 87, 203, 209, 216–17    of persons 29, 31, 98, 307, 312–16, 319–20, 344    of services 47–104, 171, 184–5, 187, 473–5

   of workers 55, 171–2, 177, 231, 295–320, 345–6, 474 freedom of contract 170, 247, 257, 259, 327 freedom of establishment 108–15, 128–30, 143–5, 171, 368–9, 402, 411–12, 446–7, 464    abuse of 138, 144, 160–1    corporate 160–1    restrictions on 131, 179, 437–9, 490 freedom of expression 57, 66, 82, 319, 325 freedoms, fundamental 66–71, 423–4, 437–8, 459–60, 471–4, 480–2, 490–4 front companies 111, 131–2, 179 function, and finality 36–7, 39–41 fundamental freedoms 66–71, 423–4, 437–8, 459–60, 471–4, 480–2, 490–4 General Anti-Abuse Rule (GAAR) 470, 472–3, 478, 487 general principles of EU law 15, 90, 355–7, 377, 399–400, 494, 505–6    abuse of law as 3–4    definitions 7–32 general test for abuse of rights 219–31 generalisations, inductive 7, 236, 254–6 genuine economic activities 112, 131–3, 301, 305–6, 412–13, 444, 464 genuine establishment 5–6, 424, 433 genuine links 112, 229, 311, 334, 336, 347–9, 352 Germany 34–7, 41–5, 51, 115–17, 148–9, 184–5, 413–16    companies 121, 139–40, 145    courts 9, 14, 35, 41, 245, 309–10, 332–3 Gibraltar 332–3, 502–3 good faith 35, 235, 247–50, 254–5, 259–60, 312, 499–500 goods, free movement of see free movement, of goods gravity, centre of 70–1, 75, 91 Greece 89, 169, 172–3, 351, 397, 400–1, 515–16 groups of companies 146, 153, 393, 403, 438, 445 habitual residence 149–50, 155–6, 331, 338 Halifax model 368–9 Halifax test 245, 375 see also Emsland-Stärke test harassment 268–9 harming, for the sole purpose of 34, 37–8 head offices 82–4, 110, 164, 188–9, 194 health, public 51, 296 health care 150, 209, 318, 329 holding companies 148, 153, 155, 436 holding periods 404–6, 480 home State 49, 56, 59–61, 97–9, 186–9, 195–7, 349–51    control 49, 52, 56, 97, 99, 186, 337 horizontal effect 17, 28–9, 176, 319 host State 49–61, 77–8, 186–9, 195–7, 344–9, 409–12, 463–5    control 49–52, 54    investors 187, 196    rules 51, 55–7, 60–1, 100 human rights 10, 54, 64, 225, 312–13, 319–20, 325 illegal aid, recovery of 505–6, 515–16 illegal State aid 499, 501, 503, 505–6

Index  633 immigration laws 340, 342–3 immigration restrictions, national 360–1 immoral purpose 44–6, 297 import duties 206–7, 212–13 importers 56, 102, 205, 207, 210, 213, 450 imports 9, 51, 204, 206, 286, 350, 450 improper purpose 160–1, 189, 265, 323–4, 381 incentives 123–4, 136, 158, 160, 163, 167–8, 225 income 301, 414–16, 474–5, 486, 505, 510, 512    attribution of 474–5    passive 414, 416 incorporation 118–20, 122–3, 127–9, 134–6, 177–9, 337, 409 indefinite period 84, 112–13, 130, 132, 228, 402, 446–7 indirect taxation 170, 381–406, 433–4, 491, 501 individualised proportionality test 299, 311 inductive generalisation 7, 236, 254–6 industrial action 95, 104 injunctions 8, 242, 246, 274 input tax 418, 428, 452, 493 insider dealing 182, 190 insolvency 59, 138, 140–1, 152–3, 155–8, 160, 164–5    cross-border 137, 145–7, 158, 162    law 118, 137–68     France 149, 162–3    petitions 147–52, 155–6    proceedings 137–9, 146, 148–9, 151, 153–4, 157–60, 445      main 138–9, 142, 147, 149, 151, 153–6, 166 insurance 64, 95, 476, 512 Insurance Premium Tax (IPT) 504, 512 integration approach 348–9 intellectual property 223, 235–6, 240–3, 246, 251 intention 226–8, 239–40, 262–3, 373–5, 397–9, 412–13, 448–50    broadcasters 71, 77, 87    of delaying settlement 262–4, 276, 281    subjective 179, 210, 225–6, 396–9, 404–5, 430–2, 434 inter-State trade 49–53, 55–8, 60–1, 176, 434, 463 interest-bearing loan 438–9 interest payments 438–9, 475 interests    cargo 266–7    financial 67, 208, 213    legitimate 34, 55, 172, 311, 349–50   main see COMI    public 19, 51, 66–7, 70–1, 79, 83, 85–6 Internal Market 28–9, 52, 88–9, 326–8, 494–501, 507, 513–15 international civil procedure 233, 240, 243, 253, 261–88 international jurisdiction 153, 155–6, 445 international law, public 221, 312, 323, 329, 336, 383 interpretation    statutory 130, 226, 367, 369, 372, 375–7, 489    teleological 210, 309, 318, 398, 449, 478 interpretative concepts/functions/principles/tools 3, 78, 90–1, 142–3, 360–2, 472–3, 496 investigation procedure, formal 513–15 investment premiums 440–1

investment services/activities 183, 186, 189, 194–5 investments 167, 186–9, 194–6, 225, 511 IPT see Insurance Premium Tax Ireland 5–6, 179, 293, 305–6, 317, 334–6, 463–5 see also Northern Ireland    nationality 228–9, 334–6 irrebuttable presumptions 472, 477–8, 480–2 irreconcilable judgments, risk of 256, 266, 285 Italy 43, 52, 280–1, 378, 500–4, 506–7, 511–13    courts 262–4, 272–3, 280, 288, 378, 392–3, 443 judicial legislation 224, 237 judicial self-restraint 286, 454 jurisdiction 71–8, 83–5, 158–62, 165–7, 243–6, 262–76, 280–8    agreements 262, 280–1    assertion of 265–6, 270    clauses 77–8, 262–3, 276, 280–1    exclusive 244–5, 275, 281, 283, 287    international 153, 155–6, 445    low-tax 411, 432, 457, 459    special 244, 256, 265–6 jurisdictional rules 265, 268–9, 271, 285 justification 50–1, 59–61, 110–11, 436–7, 459–68, 481–2, 504    commercial 403–4, 433, 440, 446, 457    economic 371, 375, 418    level of 4–5, 447–8, 453, 481    and pre-emption 4–6 justificatory effect 68, 88–90 labour 45, 310, 326–7, 329 see also free movement, of workers    laws 94, 98–9, 448 Lamfalussy process 182, 197 Latvia 94–6, 98, 101–2 lawful residence, prior 292–3, 306 lawmaking power 259–60 legal basis 93, 97, 196, 357–8, 418, 425, 453–4 legal entities 11, 124, 398, 436, 444–5, 448, 457 legal fictions 478–82 legal form 115, 119–22, 134, 372    foreign 116, 119–20 legal order 7–8, 18, 89–90, 215–17, 223–4, 230, 322 legal personality 110, 229, 372 legal persons 57, 60, 70, 100, 138, 194, 209 legal uncertainty 83, 110, 147, 154, 376, 429, 472 legitimate circumvention 102, 246 legitimate expectations 3, 13, 23, 424–5, 495, 505, 515–16 legitimate interests 34, 55, 172, 311, 349–50 letter-box companies 113, 130–3, 136, 159, 179, 444–5, 464 level playing field, genuine 124, 135 liability    limited 115, 118–19, 134–5, 160    personal 110, 492    tax 174, 324, 379, 396, 418–19, 429–30, 442 limited companies 115, 118, 124, 134–5, 160    private 58, 119, 122, 422    public 57, 122

634 Index links    close 333–4, 336    genuine/real 112, 229, 311, 334, 336, 347–9, 352 lis pendens 246, 267, 280–1 loans 164, 299, 403, 438–9, 446 low-tax jurisdictions 411, 432, 457, 459 Luxembourg 57, 68–9, 82–4, 93, 98–9, 148, 461–2 main interests, centre of see COMI manipulation 182–3, 189–91, 204, 226, 264, 291, 474–6 market abuse 182, 189–91 market manipulation see manipulation marriages of convenience 29, 171, 295–6, 305–6, 320–1, 340–1, 343–4 measures, precautionary 187, 196 mergers 29, 160–1, 208, 390, 404–5, 421, 455–6    cross-border 108, 115, 120, 123, 135, 145, 160–1 migrant workers see free movement, of workers mind of management theory 152–4 minimum capital requirements 117, 125, 134 misuse    of aid 506–7, 516–18    of powers 21, 323 monetary compensatory amounts 204, 207, 442 motive test 180, 402, 412–14, 439, 454, 466 motives 69–70, 296, 300–9, 340, 375, 419, 432    non-tax 374, 419, 424    tax 419, 424, 430, 432, 437, 439–40, 444 must-carry status 72, 82–3 mutual recognition 195, 321, 324, 326, 349, 351, 353 mutual trust 166, 245–6, 272–3, 277, 287, 352 national courts, role of 203, 210–12, 215, 503 national law, circumvention of 109, 129–30, 132 national tax law 315, 408, 413, 415, 421, 423–4, 436–7 national territory 82, 96, 179, 187, 402–3, 438–40, 465 nationality 58, 229, 304, 326–7, 330–2, 334–8, 352    attribution of 333–4, 336–7, 352    designation 321, 331, 336–7    Irish 228–9, 334–6    law 228–9, 330–2, 334, 352 natural persons 58, 69–70, 138, 148–51, 155–6, 166, 209 naturalisation 332, 336 net assets 139–40, 146–50, 153, 156, 162–3 Netherlands 55, 57–8, 66–70, 84–5, 180–1, 351, 511–13    law 57, 66–8, 78, 110, 181 non-absolute host State control 50, 52 non-adjusting creditors 140, 154, 156, 165 non-discrimination 7, 11, 14–16, 18, 99, 299, 348 see also discrimination non-discriminatory basis 326, 344–7, 349, 353 non-letter-box companies 133, 136 non-resident companies 403, 438–9, 446, 461–2, 486 non-tax motives 374, 419, 424 normal commercial operations 20, 174–5, 371, 417, 421, 442, 455 normal family life 223, 311, 321 Northern Ireland 228, 334–5

objective circumstances 70, 87, 102, 210, 263, 396–8, 412 objective criteria 69, 185, 239, 304, 307–8, 475–6, 483 objective elements 67, 69–71, 78, 87, 169, 184–5, 228 objective evidence 59, 109, 113, 128, 133, 226, 413 objective factors 69–70, 102, 174, 240, 419, 439–40, 443–4 objective tests 239, 286, 374, 396–7, 399–400, 405, 512 observance, formal 87, 102, 180, 185, 191, 263, 499–500 occupational activity, very short period of 301, 303, 305, 346 opening hours, retail 462–3 ousting, sole object of 256, 266, 284, 286 outbound establishment 409–10 paid-up capital 228, 438–9 parent companies 159, 179–80, 390, 402, 408–9, 436–7, 445 passive income 414, 416 passport, single 183, 186, 188, 196–7 permanent establishments 414, 461–2, 475 personal liability 110, 492 personal taxation 469, 473–5, 478–9, 485, 487, 489–92 personality, legal 110, 229, 372 persons    free movement of 29, 31, 98, 307, 312–16, 319–20, 344    legal 57, 60, 70, 100, 138, 194, 209    natural 58, 69–70, 138, 148–51, 155–6, 166, 209    taxable 31, 174, 395, 412, 418, 442, 453 Poland 212–13 posted workers 93–4, 96, 98–104 powers    abuse of 21, 258, 323, 496, 500–1    discretionary 35, 423–4    supervisory 188–9 pre-emption 4, 52, 67–8, 87–90, 172 precautionary measures 187, 196 preliminary reference procedure 211, 215, 217, 227, 329–30 press diversity 53–4 presumptions, irrebuttable 472, 477–8, 480–2 prices 56, 167, 182, 189–90, 204–5 primacy 7, 21–2, 51, 300 primary establishment 131–2, 148, 462 prior authorisation 73–4 prior lawful residence 292–3, 306 private actors 143–4, 152, 156, 163, 327 private companies 127–8, 134, 136, 165 private law 7, 37, 40, 45–6, 169–70, 225, 235–60 private limited companies 58, 119, 122, 422 private parties 18, 20, 22, 90, 316, 319–20, 357–8 procedural law 170–1, 243, 246, 251, 269, 276 see also civil procedure procedural safeguards 293, 295, 323, 344, 362 profits 179–80, 402–3, 409–11, 414, 433, 438–40, 463–5    diversion of 180, 412 prohibiting age discrimination 16, 357 prohibition of abuse see Introductory Note and detailed entries

Index  635 property 33–4, 37–40, 44, 220, 225, 227, 241    intellectual 223, 235–6, 240–3, 246, 251    rights 39, 170, 220, 225, 241–2, 262 proportionality 175–7, 215–16, 385–7, 403–6, 446–7, 460–3, 465    tests 67, 71, 299, 311, 448 public companies 124, 172 public entities 143, 152–4, 156, 358 public health 51, 296 public interest 19, 51, 66–7, 70–1, 79, 83, 85–6 public international law 221, 312, 323, 329, 336, 383 public law 18, 54, 65, 170, 235, 322–3, 356–8 public limited companies 57, 122 public policy 94, 98–9, 158, 295–6, 362, 460 public security 296, 344, 362 purposive enquiry 324–5 pursuit, actual 84, 112, 132, 228, 368, 402, 446–7 qualifications, educational 325, 349–50 Ramsay doctrine/principle 373, 376, 427, 487–9 real links 112, 347–9 real seat theory 110, 118, 137, 144–5 reality, economic 178–9, 371–4, 402–3, 432–4, 438–40, 444, 502 reasonableness 114, 220, 250, 259–60, 392 recognition, mutual 195, 321, 324, 326, 349, 351, 353 recovery, of illegal aid 505–6, 515–16 refunds, export 61, 173, 204, 206–8, 212–14, 305, 441 registered office (RO) 137–9, 154–6, 159–61, 167–8, 188–9, 194–6, 445 regulatory arbitrage 60, 120, 127, 132–3, 135–6, 194–5, 197–8 regulatory competence 49–50, 53, 61, 318, 321, 349 regulatory competition 59–60, 63, 79, 86, 90, 107–39, 154–5    company law 107–36 regulatory protection 50–1 reincorporations 120, 122, 125, 136, 145, 154 relocations 111, 115, 120, 122–3, 135–6, 160, 164–5 remuneration 95–6, 297–8, 300, 309, 318, 440–1 residence 293, 304–5, 338–43, 345–6, 348, 474–5, 478–81    country of 337–8, 351, 414    habitual 149–50, 155–6, 331, 338    permits 300, 317, 342–3    rights 321, 339, 341–2, 344, 352 resident companies 179–80, 403, 407, 409–11, 424, 437–8, 440 resident subsidiaries 403, 436, 438 retail opening hours 462–3 retransmission 73–6, 80–1 right holders 37–8, 41, 322 rights    human 10, 54, 64, 225, 312–13, 319–20, 325    property 39, 170, 220, 225, 241–2, 262    residence 321, 339, 341–2, 344, 352    voting 43, 119, 333 Roman law 33–5, 38, 44, 169, 312 satellite services 63, 65, 68, 72, 74–8, 80–1, 84 seat 60, 75, 115, 436, 461

   real 110, 118, 137, 144–5    statutory 145 second best 453, 479, 481 secondary establishment 59, 77, 88, 111, 131, 329, 337 secured creditors 162–3 selectivity 502, 505, 511 self-employed persons 345–6, 350, 361, 402, 446 self-restraint, judicial 286, 454 services    broadcasting 57, 67, 73, 75, 81, 93, 180–1    financial 181, 183–4, 186, 259, 414    free movement of 47–104, 171, 184–5, 187, 473–5    investment 183, 186, 189, 194–5 set-off 245, 402 set-up speed 118–19, 121, 123 settlement, intention of delaying 262–4, 276, 281 sexual orientation 13, 15 shams 56–8, 60, 343, 365, 376, 445, 450 shareholders 17–19, 43, 58, 76, 120, 145, 172 shares 29, 115, 175, 189, 209, 402, 405–6    exchange of 20, 175, 369, 405, 421–2, 455–6 shield cases 485–7, 490–1 shifts, COMI 144–53, 155–6, 164–7 shipowners 266–7 single market 59, 93, 99, 101–3, 109, 177, 497 see also Internal Market single passport 183, 186, 188, 196–7 single principle of abuse 469–83 social benefits 298–300, 302, 345, 348, 491–2 sole object of ousting 256, 266, 284, 286 sole purpose 33–4, 37–8, 81, 173–4, 303–5, 342–3, 429    test 303, 375 Spain 118, 121, 147–8, 333, 502–3, 511–12, 516    courts 246, 269, 274 special jurisdiction 244, 256, 265–6 spouses 292, 306, 308, 320, 339–41 start-up companies 119–20, 123–4, 135 State aid 250, 493–518    definition 500–1, 505    fiscal 500–4, 507    illegal 499, 501, 503, 505–6    law/rules 182, 493–5, 500–5, 507, 509–11, 517 State liability 21, 212, 215 States, United 120, 122, 135, 141, 189, 274, 365–6 statutory construction 373–6, 427 statutory interpretation 130, 226, 367, 369, 372, 375–7, 489 strategic behaviour 160, 166–7 strict construction 65, 72, 88–9, 103 students 302–4, 328, 345–6, 348–9 subjective criteria 57, 185, 209–10, 213, 398, 448–50, 458 subjective elements of abuse 23, 69–71, 87, 102, 226–8, 239, 448–50 subjective intentions 179, 210, 225–6, 396–9, 404–5, 430–2, 434 subjective tests 174, 248, 375, 396–401, 405–6, 449 subsidiaries 4–6, 128–9, 173, 179–80, 186, 407–11, 463–5    foreign 179–80, 402, 409    resident 403, 436, 438 subsidies 170, 173, 248, 504, 510

636 Index substance    economic 372–3, 428    tests 464–5, 467 Sunday trading cases 52, 462 see also retail opening hours supervision 77, 186–8, 195, 198, 486, 493, 498 supervisory powers 188–9 suspect periods 147, 155–6 Sweden 15, 31, 80–1, 94–6, 98, 100–2, 265    trade unions 96–8 Switzerland 64, 86, 173, 184, 206–7, 226, 441 sword principle 487–98 tax   advantages      accrual of 102, 174, 240, 367, 371, 419–20, 443      attainment of 102, 174, 240, 418–20, 430, 443    authorities 5, 31, 179, 418, 423–5, 444, 452    avoidance 365–80, 384–90, 401, 421–3, 435–6, 447–8, 486–8     risk of 435–7    base 411, 434, 440, 485–7, 490, 492, 504    evasion 30–1, 369–70, 385–91, 421–3, 432, 436–7, 455–6    law 29, 107, 132, 177–8, 235, 239–40, 365–518      national 315, 408, 413, 415, 421, 423–4, 436–7    liability 174, 324, 379, 396, 418–19, 429–30, 442    rates 410–11, 463, 465, 504    treaties 383, 414, 477–8    treatment 133, 410–11, 414, 487, 514      favourable 398, 430, 432–3, 502 taxation    direct 180, 386–92, 400–1, 405–6, 427, 431, 501    double 414–16, 486    indirect 170, 381–406, 433–4, 491, 501    personal 469, 474, 478–9, 485, 487, 489–92    value-added 384–5, 387, 389, 391–3 taxes see also taxation    exit 475, 478–9    franchise 123, 136    input 418, 428, 452, 493 taxpayer purposes 373–4 TCNs (third country nationals) 298–300, 304, 306, 339–41, 343–4, 352, 359–62 teleological interpretation 210, 309, 318, 398, 449, 478 territory, national 82, 96, 179, 187, 402–3, 438–40, 465 tests   Emsland-Stärke 23, 87, 102, 203–17, 220, 228–9, 397    general 219–31   Halifax 245, 375    motive 180, 402, 412–14, 439, 454, 466    objective 239, 286, 374, 396–7, 399–400, 405, 512

   proportionality 67, 71, 299, 311, 448    sole purpose 303, 375    subjective 174, 248, 375, 396–401, 405–6, 449    substance 464–5, 467 third countries 84, 185, 212, 298, 333, 359–60, 362 third country nationals see TCNs third parties 139, 159, 244, 255–6, 403–4, 439–40, 444–6 trade, inter-State 49–53, 55–8, 60–1, 176, 434, 463 trade marks 241–2 trade unions 95–8, 100, 102–4, 274, 319 training 55–6, 346, 350–1 transmitting State principle 64–5, 72–4, 76, 80–2 transparency 95, 190–1, 414, 477 transport 21, 58, 70, 170, 207, 215, 228 trust, mutual 166, 245–6, 272–3, 277, 287, 352 TV broadcasters see broadcasting services u-turn transactions 470, 491 uncertainty, legal 83, 110, 147, 154, 376, 429, 472 Union citizenship see citizenship unions, trade 95–8, 100, 102–4, 274, 319 United Kingdom 6, 75–80, 133–4, 228–9, 333–6, 372–7, 409–12    company law 115, 118, 161, 409    controlled foreign companies provisions 408, 437, 459, 465–6    resident companies 179–80, 403, 407, 409–11, 424, 437–8, 440 United States 120, 122, 135, 141, 189, 274, 365–6 unlawful aid 505, 514–17 unreasonable burden 346–7 vacuum, fiscal 473–4, 476–8 value-added tax (VAT) 31, 382–93, 395–400, 404–7, 417–18, 427–8, 496–7 VAT see value-added tax very short period of occupational activity 301, 303, 305, 346 vexatious proceedings 268–9 voting rights 43, 119, 333 wages see remuneration warranty 244, 247, 285 wholly artificial arrangements 111–14, 128–33, 178–80, 412–13, 435–41, 443–53, 457–8    test 112–13, 130, 433, 448–9, 457, 474–7, 479 workers    free movement of 171–2, 177, 231, 295–320, 345–6, 474, 476    posted 93–4, 96, 98–104    status of 298, 300–4, 346 wrongful avoidance 57, 70