State Aid and the Energy Sector 9781509913701, 9781509913688, 9781509913695

This important new work offers a comprehensive and compelling account of State aid law and policy and its application to

184 15 15MB

English Pages [626] Year 2018

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

State Aid and the Energy Sector
 9781509913701, 9781509913688, 9781509913695

Citation preview

Editors’ Preface In 2014, the Council of the European Union launched an ambitious Climate and Energy Policy Framework, committing the EU to reduce greenhouse gas emissions by 40 per cent by 2030, to promote renewable energy, so that at least 27 per cent of all energy produced in Europe is sourced from renewable sources, and to push for greater energy efficiency (to achieve a 30 per cent improvement compared with 1990). The subsequent publication of the Commission’s plans for the ‘Clean Energy Transition’ and an extensive package of draft legislation on 30 November 2016 propel the urgency for major steps towards a low carbon economy so that, by 2050, Europe’s reliance on fossil fuels (many of which are imported) will be substantially reduced if not eradicated. Binding national targets for renewable energy production will disappear after 2020, however, and be replaced by a binding EU-wide target. The recently published recast of the 2009 Renewable Energy Directive 2009/28 envisages continued State support to RES production, in line with the European Commission ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG). Energy subsidies worldwide were projected at US$5.3 trillion in 2015, or 6.5 per cent of global GDP (that is equivalent to $10 million per minute). Within the European Union, public support (not including transport) is estimated at over €200 billion per annum. And it is not just the Member States that are providing support for energy projects: the 2018 EU budget prioritises energy projects through the European Fund for Strategic Investments (EFSI) programme (energy projects already topped the list of projects financed through the Connecting Europe Facility); EU agencies tasked with overseeing implementation of the EU energy policy will also benefit from increased funding (which is in line with their ever-growing list of competences). State aid to the energy sector is now the second largest category of aid in the Member States, according to the last available Commission State Aid Scoreboard. Much of the reported aid is targeted at the promotion of renewable energy (especially solar and wind). This type of aid is invariably deemed to be compatible aid. The State aid regime, as revised by the State aid modernisation initiative in 2013, will have an important role to play in ensuring that such support takes the form of what the Commission considers to be ‘good aid’. In particular, what role could and should the Treaty State aid regime play in this ambitious programme for transition? Can it lead to more efficient forms of support? Can it assist the transition away from coal and gas without jeopardising the security or reliability of supply? Can the State aid regime compensate for the failure of Member States to agree on what many would argue are more effective instruments—such as a carbon tax—to stimulate the phase out of fossil fuels? The impact of the State aid regime on the EU energy sector is not without legal and political controversy, and as State aid control is shaping up to be one of the major instruments available to the Commission, relegating the traditional anti-trust powers to second place, this controversy is only likely to grow.

vi

Editors’ Preface

The book fills a gap in the literature by systemically addressing the issues raised by the use of EU State aid control in the context of the energy transition. In the recent period, this area of European law has grown in importance for the energy sector and also in complexity. The increased use of the rules on State aid reflects the switch towards new policy objectives in energy—in particular decarbonisation—and the resulting reinvolvement of national public authorities. In this context, issues arising from State aid control in energy also made their way in the national courts. This book is the final result of a research programme led by the Energy Union Law Area of the Florence School of Regulation (EUI, RSCAS). Several of the chapters were discussed at a research seminar on ‘State aid in the energy sector’, which took place on 27 October 2016. The book should appeal to a wide audience composed of academics and professionals, including EU public officials, the staff of regulatory bodies, corporate economists and business lawyers. In terms of methodology, it has been our objective to pair authors with a variety of backgrounds and experience. As much as possible, we have therefore teamed lawyers with economists, practitioners with academics or private practitioners with public officials. We have also endeavoured to cover national developments, including actions brought before national courts for the suspension and/or recovery of illegal aid. Systematic research into the enforcement of the EU State aid regime to discipline energy subsidisation at national level has been largely neglected. Given the general policy shift towards the decentralisation of State aid control, the national perspective is of major importance. The book is structured in four parts. The first deals with the definition of State aid in the energy sector by providing a detailed discussion of the case law and Commission decisions on particular applications of Article 107(1) TFEU. The second part deals with compatibility issues. It explains the circumstances in which certain types of State aid are allowed. The third part focuses on topical procedural issues that are of particular interest to the energy sector. The final part covers State aid and the energy sector in selected national jurisdictions and, in particular, in jurisdictions where important cases have reached national courts. Part I is composed of four chapters on the definition of State aid. Chapter one, ‘What is State Aid?’ (Falk Schöning and Clemens Ziegler) sets the scene by exploring the definition of ‘State aid’ in the energy sector. It discusses the general issues in the TFEU and broadly considers all elements of the definition. It contains, in particular, a comprehensive analysis of welcome developments on the application of the market economy operator principle. Chapter two, ‘State Aid and Price Regulation in the Energy Sector’ (Guillaume Dezobry and Adrien de Hauteclocque) shows that the Union institutions have developed an important body of case law and decisional practice on price regulation and EU State aid law. Major conceptual difficulties nonetheless remain. According to the authors, regimes favouring a category of users (by exempting them from a component of the energy price) are likely to remain a common feature of the energy policy landscape. As renewable support schemes will progressively be rolled back while technologies are maturing, the key issue in the coming years will be the protection of energy-intensive users, in order to avoid carbon leakage.

Editors’ Preface vii Chapter three, ‘State Aid and Taxation: Special Focus on Energy and the Environment’ (Lena Sandberg) discusses the presence of State aid through tax exemptions and derogations from levies, or surcharges, as well as the compatibility of such tax measures with the internal market, in particular, pursuant to the EEAG. It includes an in-depth critique of developments on the application of the selectivity criterion. The complex application of this criterion is of major importance in the energy sector. Chapter four, ‘WTO Subsidy Rules: Implications for Energy’ (Anna-Alexandra Marhold) addresses the rationale of WTO subsidy rules and how they play out in the energy sector. It focuses on explaining the difference in nature between fossil fuels and clean energy subsidies and how this impacts their treatment under WTO subsidy disciplines. The WTO regime is attracting increased scrutiny from academics and practitioners, especially in the light of Brexit. Part II is composed of six chapters, discussing forms of compatible aid. After a critical review of the EEAG from a legal and economic perspective in chapter five, ‘Compatibility of RES: A Legal and Economic Approach’ (Francesco Maria Salerno, Sébastien Douguet and Vincent Rious), two chapters perform an in-depth analysis of Capacity Remuneration Mechanisms (CRMs): chapter six, ‘Capacity Mechanisms and Auctions’ (Leigh Hancher and Christoph Riechmann) provides a rich discussion of the multiple strands of EU law and policy informing CRMs, from State aid to sector specific regulation. The chapter also deals with the relations between the EEAG and the proposed rules on the new market design, finding that, through the State aid review of CRMs, the Commission has managed to spur the reform of shorter-term markets (eg, intraday and balancing markets) thus fast-forwarding the achievement of one of the goals of the Clean Energy package. Chapter seven on the cross-border aspects, ‘Cross-Border Participation in Capacity Mechanisms: Legal and Economic Issues’ (Fabien Roques, Charles Verhaeghe and Guillaume Dezobry) draws a distinction between implicit and explicit cross-border participation in these mechanisms, finding the latter more difficult to implement than the former. Chapter eight, ‘Aid to Nuclear and Coal’ (Leigh Hancher and Max Klasse) focuses on energy aid with a distinguished historical pedigree: aid to nuclear and coal, also providing insights into the relation between the EU State aid rules and the Euratom Treaty, which governs most aspects of the nuclear sector. Chapter nine on ‘Projects of Common Interest and Energy Infrastructure Projects’ (Julia Sack, Kristina Zych and Kai Uwe Pritzsche) illustrates the special rules for attaining PCI status and its impact on the development of energy infrastructure. The chapter also explains the (benign) treatment of energy PCIs for State aid purposes. Chapter ten, ‘Services of General Economic Interest and EU State Aid Control in Energy Markets’ (Francesco Maria Salerno, Adrien de Hauteclocque, and Simina Suciu) deals with Services of General Economic Interest, finding an emerging trend towards reducing Member States’ margin of discretion in areas subject to high EU harmonisation. Part III deals with procedures and, as this is a complex and rich area of State aid law in its own right, it cannot purport to be exhaustive. Instead, we have singled out a number of topical aspects which are of considerable significance for the application

viii Editors’ Preface of the State aid regime to the energy sector. This part includes a chapter on ‘State Aid Recovery and the Energy Sector’ (chapter eleven, Jacques Derenne); a chapter on ‘Arbitration’ (chapter twelve, Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz); and finally, a chapter on ‘Common EU Law Principals of Private Enforcement of State Aid’ (chapter thirteen, Georg Berrisch and Brian Byrne). Part IV comprises ten chapters on the role of State aid in the energy sector in selected national jurisdictions, the EEA, and Energy Community. Chapters fourteen to twenty-three cover relevant developments at national level in major EU jurisdictions, including Germany (chapter fourteen, Falk Schöning and Clemens Ziegler); France (chapter fifteen, Liliana Eskenazi); the Netherlands (chapter sixteen, Marinus Winters); Austria (chapter seventeen, Moritz Am Ende and Judith Grimm); Italy (chapter eighteen, Pierluigi Congedo); Greece (chapter nineteen, Antonis Metaxas); Belgium (chapter twenty, Wim Vandenberghe); Spain (chapter twenty-one, Jose Luis Buendía and Miguel Ángel Bolsa); the European Economic Area (chapter twenty-two, Jens Naas-Bibow and Svein Terje Tveit); and finally, the Energy Community (chapter twenty-three, Marco Botta and Rozeta Karova). We address our sincere thanks to each contributor for having participated in this endeavour and without whom it would have been impossible to deliver a project which covers so many facets of State aid law and practice. We would also like to thank the director of the Florence School of Regulation, Professor Jean-Michel Glachant, for his unwavering support for our activities. We have greatly benefited from invaluable know-how and editorial assistance led by Anne-Marie Kehoe, coordinator of the Energy Union Law Area of the Florence School of Regulation, and assisted by Jerzy Dudek and Ana Trías. Finally, we would like to express our gratitude to the team at Hart Publishing who did a great job in editing this book. We are very grateful to them for their considerable efforts throughout this process. Leigh Hancher Adrien de Hauteclocque Francesco Maria Salerno July 2017

List of Contributing Authors Georg Berrisch Dr Georg M Berrisch is a German Rechtsanwalt and partner in the Brussels office of Baker Botts LLP. He has over 25 years of experience in practising EU State aid, competition, and trade law in Brussels. Georg is widely recognised as an effective litigator before the Court of Justice and the General Court of the EU, where he has represented private and public sector clients, including the Council of the EU and the European Commission in over 150 cases. Georg has also litigated private State aid cases before the national courts in Germany. For example, he has represented an alleged beneficiary of illegal State aid in the cases before the German courts, including the German Supreme Court, and the Court of Justice, on the issue of the binding effect of Commission opening decisions in private State aid litigation before national courts. Miguel Ángel Bolsa Ferruz Miguel Ángel has worked in the antitrust department at the Brussels office of Garrigues since 2013. During his time at Garrigues he has worked in the areas of State aid and regulated sector matters, including dossiers in finance, telecommunications, energy and pharmaceuticals. Prior to joining Garrigues, he built up working experience at the European Investment Bank, and he is the author of a number of publications in specialised journals, including World Competition and European State Aid Law Quarterly, of which he currently holds the position of correspondent. He holds a Degree in Law and Business Administration from the University of Zaragoza (Spain), a Master’s degree in Economic Analysis of Law from the College of Europe (Bruges, Belgium) and a Master’s degree in Law and Finance at the University of Oxford (UK). Marco Botta Marco Botta has worked at the Max Planck Institute for Innovation and Competition (MPI), Munich (Germany) since October 2016. Before joining the MPI, he worked for five years as Assistant Professor at the Institute for European Integration Research of the University of Vienna (Austria), where he taught a number of courses on general EU law, competition law and State aid law. Since 2011 he has worked as a Research Fellow at the Robert Schuman Centre for Advanced Studies (RSCAS) of the European University Institute (EUI), Florence (Italy). Marco Botta holds a Bachelor Degree in International Relations from the University of Turin (Italy), LL.M. in European Business Law from Leiden University (the Netherlands) and PhD from the EUI Law Department.

xii List of Contributing Authors José Luis Buendía Sierra José Luis Buendía is in charge of Garrigues’ Brussels office and is a Partner in the Firm’s EU and Competition Law Department. Previously, he spent fourteen years as a European Commission civil servant, specifically in the Legal Service, the DGs of Competition and the Internal Market, and the Cabinet of Commissioner Mr Oreja. José Luis is a renowned expert in EU and antitrust law, with a particular emphasis on State aid and other public measures impacting on competition, as well as handling appeals before the EU courts. Since 2009, he has been a visiting lecturer on State aid at King’s College London and coordinates the section on aid in the courses on competition law at IEB Madrid. He is currently co-editor of European State Aid Law Quarterly (EStAL). He holds a PhD and a Degree in Law from Universidad de Zaragoza and a Master’s Degree in European Legal Studies from the College of Europe, Bruges. Brian Byrne Brian R Byrne is an Irish solicitor and Senior Associate in the Brussels office of Baker Botts LLP. Brian is an experienced litigator with a diverse practice encompassing EU State aid law, EU competition law, and international trade law. He has represented clients, including the Council of the EU, in several high profile cases before the General Court and Court of Justice of the EU. Brian regularly advises and represents clients in State aid matters before the European Commission. He is currently representing an alleged State aid beneficiary in multiple cases before the General Court, challenging negative State aid decisions adopted by the Commission. Pierluigi Congedo Pierluigi Congedo is a module lecturer of EU Competition Law at LUISS Guido Carli, Rome. He holds a PhD from King’s College London and an LL.M. from the Institut d’Etudes Européennes, Brussels. He is a Fellow of the Centre of European Law at King’s College London. He is also counsel to Macchi di Cellere Gangemi LLP, London, with double qualification as Avvocato in Italy and Solicitor of the Law society of England and Wales. Jacques Derenne Jacques Derenne is the head of the EU Competition & Regulatory practice at Sheppard Mullin’s Brussels office. Jacques has over 25 years of competition law experience in all areas (mergers, cartels, abuses of dominance and State aid), in EU regulatory and related competition law issues in a variety of sectors (in particular in regulated industries and services such as energy, the postal sector, aviation, railways, telecoms, satellites, the audiovisual sector and tobacco products). Jacques regularly appears at competition hearings before the European Commission, and pleads cases before the General Court and the Court of Justice of the EU, national competition authorities, the Belgian and French courts and various regulatory bodies. Jacques’ State aid experience spans more than two decades, during which time he has acted

List of Contributing Authors xiii for beneficiaries, competitors and Member States before the European Commission, EU courts and national courts. Jacques co-directed and co-authored studies for the European Commission on the enforcement of State aid rules at national level (2006 and 2009), which contributed to the Commission’s Recovery and Enforcement Notices in 2007 and 2009 respectively. He co-edited a book on the Enforcement of EU State aid law at national level—2010—Reports from the 27 Member States (Lexxion, October 2010), and has written quarterly comments on State aid case law and the Commission’s decisional practice in the journal Concurrences since 2004 (together with EU officials). Jacques also publishes widely on various other EU constitutional, competition and regulatory issues. He is a founding member of the Global Competition Law Centre (College of Europe, Scientific Council and Executive Committee). He graduated from the University of Liège (Belgium, 1987) and from the College of Europe (Bruges, 1988), and teaches competition law (State aid aspects) at the University of Liège and at the Brussels School of Competition. Guillaume Dezobry Guillaume Dezobry is of Counsel to FIDAL and a senior lecturer in Energy Law (University Paris-Dauphine, Strasbourg, Amiens and Sciences-Po Paris). He offers clients sharp insight into public business law and energy law. His hands-on experience spans all aspects of the energy sector: transmission, distribution, generation and supply of electricity and gas as well as new aspects of the market design in the electricity sector (capacity mechanisms, participation of demand response etc). He is also a researcher and a member of the European Electricity Market Chaire (University Paris-Dauphine) and focuses his research on the legal issues related to the liberalisation of the electricity and gas markets. Sébastien Douguet Sébastien Douguet is a Manager in the Energy & Regulation team of Deloitte Economic Consulting. He graduated in 2012 from Mines ParisTech with a Master degree in engineering. Sébastien has worked for more than four years in the practice on various studies focused on energy and electricity economics. His experience in these sectors covers market and regulation designs as well as asset valuation and financial assessments. Sébastien has developed a thorough knowledge and proficiency on issues involving electricity and gas network regulation, the economic relationships between network investments and efficiency and the compatibility of support mechanisms with State aid rules. Moritz Am Ende Moritz Am Ende is a German attorney practising law at Starlinger Mayer attorneysat-law in Vienna. He advises clients on all aspects of competition and antitrust law, with a special focus on EU competition and State aid law. Before joining the bar, he worked several years in Luxembourg as chef de cabinet for the President of the EFTA Court.

xiv List of Contributing Authors Alejandro Escobar Alejandro Escobar is a public international law and arbitration partner at Baker Botts, London. He advises businesses and states in disputes arising out of investment protection treaties and in arbitrations arising out of project agreements. He has handled numerous claims of expropriation and abusive regulation in various industries, including power, oil and gas, telecommunications, and public concessions and procurement. He has extensive experience advising corporate and government clients on investment treaty disputes, and on other areas of public and private international law, including international trade agreements and human rights instruments. He regularly advises on complex matters involving multiple dispute-settlement procedures and overlapping obligations under treaties, general international law, domestic law and contracts. His disputes work has involved proceedings under all major arbitration rules. Alejandro advises on other international law issues, including boundary disputes, relevant to international business. Mr Escobar practises bilingually in English and Spanish, and has a working knowledge of French and Portuguese. He was previously Senior Counsel at the International Centre for Settlement of Investment Disputes (ICSID). Mr Escobar holds a PhD from the University of Cambridge, a Bachelor in Social and Juridical Sciences from the University of Chile and a Certificate of Merit from The Hague Academy of International Law. Liliana Eskenazi Liliana Eskenazi holds a Doctorate in Law from the University of Paris—PanthéonSorbonne and is a qualified lawyer registered with the Paris Bar. She works as Counsel to French boutique law firm Fréget—Tasso de Panafieu, specialising in EU, competition and regulatory law. Previously, Liliana worked for eleven years in the Brussels and Paris offices of Allen & Overy LLP, where she advised on various EU and antitrust law matters. A versatile professional, Liliana has developed an expertise in State aid law and has focused, in recent years, on regulated sectors (energy, telecom, pharma), as well as on complex issues at the intersection of IP and antitrust law. Currently, Liliana teaches competition law at École Centrale-Supelec (Paris). Ernesto Féliz Dr Ernesto Féliz De Jesús is an associate at Baker Botts LLP in London. He has provided advice on investment treaty arbitration law, on the international legal protection available for transnational investments, on the domestic law applicable to international investments, on the enforcement of international arbitral awards, on construction law and on numerous areas of public international law, such as maritime delimitation issues, the law of the sea, the law of state responsibility, the law of treaties and international humanitarian law. Prior to joining Baker Botts in 2014, Dr Féliz De Jesús was an associate at an international law and international arbitration boutique law firm in London. He holds a doctorate in public international law from the University of Oxford. He was also a university trainee (2010–2011) at the International Court of Justice (ICJ) in The Hague, where he clerked for Judges Abdulqawi Ahmed Yusuf and Sir Christopher Greenwood.

List of Contributing Authors xv Laurie Frey Laurie Frey is an associate in the International Arbitration and Dispute Resolution group at Baker Botts (UK) LLP in London. Prior to joining Baker Botts, she practised with a major law firm in New York and served as a law clerk to the Honorable Allyne R Ross of the United States District Court for the Eastern District of New York. She attended Princeton University and received a JD from Columbia Law School. She is qualified to practice in New York and England and Wales. Judith Grimm Dr Judith Grimm is an associate lawyer (Rechtsanwaltsanwärterin) in Vienna. She specialises in the areas of competition and antitrust law. Leigh Hancher Leigh Hancher is a part-time Professor and the Director of the Energy Union Law Area at the Florence School of Regulation. She is also Professor of European Law at Tilburg University, the Netherlands, and Of Counsel to Allen & Overy in Amsterdam. She has written and lectured widely on both EU State aid law and European energy law and has acted for clients in numerous State aid matters. Adrien de Hauteclocque Dr Adrien de Hauteclocque is a Law Clerk (Référendaire) with Vice-President M. van der Woude at the Court of Justice of the European Union (General Court), Luxembourg, and an Adviser of the Florence School of Regulation. His research interests include EU competition and State aid law, competition policy in network industries and the law and economics of energy regulation. Before moving to the EU institutions, he pursued his academic career at the University of Manchester and at the European University Institute where he co-founded the ‘EU Energy Law & Policy’ Area (now the Energy Union Law Area) of the Florence School of Regulation. He is a regular speaker at international conferences and has published numerous books, academic articles and book chapters in the last ten years. He also holds regular teaching commitments for the Legal Task Force of the Council of European Energy Regulators (Belgium), École Nationale d’Administration (France) and HEC Paris (France). Dr de Hauteclocque obtained his PhD in Law from the University of Manchester (UK) and holds a MSc in Management from EM Lyon (France) and a MSc in Economic Policy from Strathclyde University (UK). Rozeta Karova Rozeta Karova is a Senior Energy Lawyer at the Energy Community Secretariat and she is responsible for coordination of the dispute settlement procedure and enforcement of Energy Community law. She also monitors the implementation and reviews the compliance of measures taken by the Contracting Parties for the implementation of energy acquis. She holds an LL.M in European Business Law from the Faculty of Law at University of Leiden (The Netherlands) and a PhD in law from the

xvi List of Contributing Authors European University Institute in Florence. Her thesis focused on the liberalisation of the electricity market in South East Europe, and was published as a monograph by Kluwer in 2012. She has published numerous articles on all aspects of Energy Community law. Max Klasse Max Klasse is a partner in the antitrust group of Blomstein. He focuses on all aspects of German and European antitrust and competition, EU State aid and merger control law. In the area of antitrust, he advises companies in relation to internal investigations, in cartel proceedings, in compliance and behavioural matters. In State aid cases, he represents both companies and state authorities before the European Commission and EU courts. He recently provided counsel to the German Federal Government before the European Court of Justice. Before founding Blomstein, Max was a lawyer in the antitrust and competition group of an international law firm from 2005 until 2016. Max regularly publishes on aspects of German and European antitrust and EU State aid law. He is co-author of a legal textbook on antitrust law in the pharmaceutical sector (2nd edition, 2015) and author in Heidenhain, European State Aid Law. Johannes Koepp Dr Johannes Koepp is a partner in the Baker Botts International Arbitration & Dispute Resolution Group. His practice focuses on international arbitration, both public and private. In addition to serving as an arbitrator, Dr Koepp has represented clients before most major arbitral institutions, as well as in ad hoc proceedings. He is experienced in multijurisdictional disputes arising out of energy, joint ventures, mergers and acquisitions, telecommunications and financial services. Dr Koepp worked as a research fellow at the University of Geneva for two years. During this time, he completed his doctoral thesis, which analysed and compared various thirdparty procedures in the field of international dispute settlement. He regularly publishes on arbitration-related topics and has taught law at universities in Switzerland, Germany and Spain. Anna Marhold Anna Marhold is Assistant Professor and Senior Researcher at the Tilburg Law and Economics Center (TILEC), a centre of excellence at Tilburg University in the Netherlands, where she researches and teaches on various topics of international and European law. Her main interests lie at the intersection of international economic law and energy regulation. Anna is also a fellow at the Cambridge University-based C-EENRG Platform on Global Energy Governance. In addition to her academic activities, Anna regularly provides policy advice for international and European think tanks. Anna obtained her PhD in Law at the European University Institute (EUI) under the supervision of Professor Petros Mavroidis. During her PhD, she was an EU-US Fulbright Schuman Grantee and Visiting Scholar at NYU School of Law. Additionally, she was a Marie Curie Early Research Fellow at the Graduate Institute

List of Contributing Authors xvii in Geneva. Anna holds parallel degrees in Law (LLB, LLM) and Russian (BA, MA) from the University of Amsterdam. Antonis Metaxas Professor Dr A Metaxas is Faculty Member at the Department of Political Science and Public Administration of the National and Kapodistrian University of Athens and Visiting Professor of EU and EU energy law at various universities in Greece and abroad. He is also Managing Partner of Law Firm Metaxas & Associates (M&A). Professor Metaxas is Chairman of the Hellenic State Aid Institute (HSAI) and of the Hellenic Energy Regulation Institute (HERI). Jens Naas-Bibow Jens heads the Thommessen Law Firm’s renewable energy practice. He deals with matters concerning energy law, environmental law and litigation. He has particular experience with energy licences, European and Norwegian regulation of the energy sector, economic regulation impacting energy distribution and transmission, district heating, rights of way for infrastructure, expropriation, appraisal cases and energy law related litigation. Kai Uwe Pritzsche Dr Kai Uwe Pritzsche, LLM (Berkley), is a partner of Linklaters LLP and a corporate lawyer resident in Berlin. He has special expertise in M&A transactions, privatisations and joint ventures, in particular in the energy industry. Kai Pritzsche practiced law in Cologne, New York and Berlin. Since the 1990s he has advised clients in numerous transactions and disputes in the energy industry. Kai Pritzsche is part of the Linklaters global energy sector leadership, he is the president of the board of the Humboldt European Law School Foundation and a member of the DIS (German Institution for Arbitration). In 2017 Kai Pritzsche published a book on German energy law, he is co-editor of the German energy law journal EnWZ and a member of the advisory board of the energy law journal RdE. He has publicly given presentations and has published extensively in the fields of corporate law, M&A and energy law. Christoph Riechmann Christoph Riechmann is a Director in the Energy Practice of Frontier Economics. He provides market design, regulation, competition, valuation, strategy and dispute support advice to clients in Europe. Christoph has over 20 years of consulting experience in the energy sector and has recently led several studies on aspects of energy market design at the EU level and in different EU Member States. He regularly advises clients on renewable support and climate policy. Christoph was previously a researcher at the Institute of Energy Economics (EWI) at the University of Cologne. He has a holds a Doctorate in Economics (Dr rer pol) from Cologne University, an MSc (Econ) from Glasgow University/Scottish Postdoctoral Programme and Dipl-Ök from Justus-Liebig-University, Gießen (Germany). Christoph has frequently

xviii List of Contributing Authors published on market and regulatory designs in academic and industry journals as well as in several handbooks. Vincent Rious Vincent is now economist in the economic department of power system in the French TSO RTE. He was previously director of Energy & Regulation in Microeconomix— Deloitte Economic Consulting and an associate professor at Supelec, a French Grande École specialising in electricity. He is also advisor in engineering at the Florence School of Regulation in the framework of the Loyola de Palacio research programme concerning European energy policy. With more than ten years of experience, Vincent has extensive expertise related to the assessment of market designs, the financial situation of TSOs, the impact of regulatory decisions, proposals of innovative market and regulatory designs and modelling of their microeconomic properties. He has producednumerous expert reports for industry and published more than ten papers in peer-reviewed journals and multi-authored volumes. Since 2008, he has taught economic analysis of the liberalisation of network industries at Supelec and University Paris-Sud 11. Fabien Roques Fabien Roques is an Executive Vice President with the economics consultancy Compass Lexecon and an Associate Professor in Economics at University Paris Dauphine. His expertise spans European power, gas and carbon dioxide (CO2) markets, with a specific focus on competition and State aid issues, market design and regulation, and asset valuation. Fabien is a regular contributor to academic and practitioners’ publications and conferences. Over the past decade, Fabien has also worked on numerous assignments for clients, including major European utilities, financial investors, and European regulators. Previous roles included leading the IHS CERA European power and carbon research and consulting team, and contributing to the World Energy Outlook as a Senior Economist with the International Energy Agency. Fabien holds an MSc in engineering from the École Centrale Lyon and a PhD in Economics from the University of Cambridge. Julia Sack Julia Sack, LLM, is a Managing Associate of the law firm Linklaters LLP in Berlin. She advises international clients on M&A transactions and joint ventures in the energy sector as well as on various regulatory matters of European and German energy law. Before she joined Linklaters in 2013, Julia studied law at the HumboldtUniversity of Berlin carried out her legal traineeship at the higher regional court of Berlin. In addition, she is a participant of the North Sea Energy Law Programme and holds a Master’s degree from the universities of Groningen, Copenhagen, Aberdeen and Oslo. Julia regularly publishes and gives presentations in the field of energy law. Francesco Maria Salerno Francesco Maria Salerno has been a Managing Partner of the Brussels office of Gianni, Origoni, Grippo, Cappelli & Partners (GOP) since April 2017. Francesco

List of Contributing Authors xix Maria has extensive experience in EU law, having acted in several matters before the EU Commission and the courts in Luxembourg. Besides representing companies in antitrust matters, Francesco Maria assists on cases arising from state measures such as aids, regulation in sectors covered by EU harmonisation rules (such as digital, energy, transports, postal services, banking and finance), as well as other state measures impacting the freedoms guaranteed by the internal European market. Before joining GOP, Francesco Maria worked for 18 years with leading international law firm Cleary Gottlieb where he began, in 1999, working in its Rome offices. In 2005 he transferred to the Brussels office of the same firm, and was appointed a Senior Attorney in 2010. After earning both Masters and PhD degrees from the London School of Economics, Francesco Maria continued his academic and research interests and several of his articles and essays on State aids, independent authorities and antitrust issues have been published by prestigious, European and American legal publishing houses. He speaks Italian, Spanish, French and English. He is a Member of the bar in Italy (Catania) and Spain (Madrid). Lena Sandberg Lena Sandberg is a Danish qualified Of Counsel and a member of the Brussels Bar, based in the Brussels office of Gibson, Dunn & Crutcher. She is a member of the firm’s Antitrust and Competition Practice Group. She has held positions both in the private sector (law firms and companies) and in the public sector (the European Commission and the EFTA Surveillance Authority). Ms Sandberg’s work experience covers all aspects of competition law, including State aid, abuse of dominant position cartels, and merger filings. She also has strong experience in Public Procurement and Trade Defence Investigations. In addition, she has extensive sectoral regulatory experience in the energy/environmental sectors (including renewables, gas and electricity production and transmission). Her sectoral experience also includes electronic communications sectors (covering telecommunications, Internet and IT). Her work experience includes negotiations with international institutions such as the European Commission and national competition/regulatory bodies as well as litigation before various European Courts. Ms Sandberg also has solid experience in State aid and corporate tax benefits such as carry forward of losses, capital tax, depreciation, property tax and parafiscal duties in a variety of sectors. In particular, she has handled tax benefits for the aluminium sector in Iceland and non-tax entities in the tax haven of Liechtenstein. Ms Sandberg is recommended in the Belgium: Competition section of Legal 500 EMEA 2016. Prior to joining Gibson Dunn, Ms Sandberg served as Senior Officer in the Competition and State Aid Directorate at the EFTA Surveillance Authority (ESA) from 2004 to 2011. Through, in particular, her time with ESA she gained experience as a State aid and competition regulator as well as in the field of public procurement. She has covered complex questions in the energy and environmental area particularly in the field of the EU Emissions Trading Scheme, renewable energy, energy taxes, electricity supply, carbon capture, and a string of related issues including research and development and innovation, and risk capital. Her work involved drafting decisions and policy legislation and negotiating with relevant counter parts in the European Commission both on cases and policy guidelines. Ms Sandberg is a regular contributor and member of the Editorial Board

xx List of Contributing Authors of the European Energy Journal, and a frequent speaker on energy, State aid, Public Procurement and energy issues. She is fluent in English, Danish, Swedish, German and French. Falk Schöning Falk Schöning is a partner in Brussels and heads the EU State aid practice at Hogan Lovells. In State aid cases, he structures projects and transactions and counsels in the proceedings of the European Commission, national authorities or courts. Falk has advised numerous clients on their State aid issues including Phase 1 and 2 Commission investigations. Areas of focus include State aid and Tax, TMT projects, R&D and regional aid, and security and defence. Falk is a regular speaker on current State aid developments and publishes frequently on State aid law in law journals and other publications. Falk is also part of Hogan Lovells’ Brexit strategy group and advices clients on the implication of Brexit on State aid law. Simina Suciu While contributing to this book, Ms Simina Suciu was an associate in the Antitrust and Competition Practice of Baker Botts’ Brussels office. Her practice concentrated on EU competition law matters, including antitrust, merger control and State aid. She has represented international companies both before the European Commission and the Romanian Competition Council. In addition, Ms Suciu has counselled clients on EU energy law related matters, in particular on the implementation of the EU Legislative Energy Package in the Natural Gas sector, REMIT, EU trade law and EU sanctions involving Ukraine and Russia. Ms Suciu has a LLB from the University of Bucharest and an LLM in Competition Law from Queen Mary University of London. Ms Suciu currently holds the position of Regulatory Advisor at Trans Adriatic Pipeline. Svein Terje Tveit Svein works in Thommessen Law Firm’s EU/EEA practice. He has around ten years of experience in advising clients on matters relating to State aid, public procurement, competition and general EU/EEA law. Svein particularly provides advisory services in relation to regulated markets (energy, telecom, transport and postal services). He has previously worked as a deputy judge in Stavanger City Court and as a seconded associate with Freshfields Bruckhaus Deringer LLP (London). Wim Vandenberghe Wim Vandenberghe is a partner in the Brussels office of Sheppard Mullin Richter & Hampton LLP, specialising in EU regulatory, competition and energy law. He represents gas and electricity grid operators, investors, power producers and large energy consumers. He advises on regulatory law matters, including TPA, unbundling, tariffs, investment structuring, network rules and licences and he also represents clients in fossil fuels and renewable energy contracts and projects, EU energy liberalisation packages, and climate change and emissions trading matters. He has been recognised as a leading energy lawyer by independent legal directory Legal 500 EMEA, which

List of Contributing Authors xxi noted his ‘wealth of experience in the regulation of fossil fuels and renewable energy’ and ‘his expertise in EU energy liberalisation packages and issues relating to climate change’. He is a guest lecturer for the Competition Law and Regulated Network Industries LLM at King’s College (UK), and a former research fellow at KU Leuven Law Faculty (Belgium). Charles Verhaeghe Charles Verhaeghe is a Vice President of a consulting firm, Compass Lexecon, in Paris. Over the past four years, Charles has led or brought expertise to more than fifty consulting projects for energy companies, regulators and financial institutions on regulatory and market design issues. Previously, Charles worked for six years at the French energy regulator. As head of department, he managed a team of six economists in charge of European integration and power market design. He led the implementation of a regulatory framework for demand response in France. Charles was also very active at European level: he chaired several projects for ACER, interacting frequently with regulators, the European Commission, system operators and other stakeholders. Charles started his carrier in the renewable energy sector at EDF Energies Nouvelles more than ten years ago. Charles graduated as an engineer from École des Ponts ParisTech and holds a Master Degree in Economics. Marinus Winters Marinus Winters is of Counsel to Allen & Overy LLP, Amsterdam office. He has in-depth experience of advising on energy law, competition law, and the regulatory aspects of the electricity and gas markets. Marinus has acted for several energy and petrochemical companies in civil and administrative procedures relating to almost every aspect of the energy sector, such as tariff regulation for grid operators, access to the grid, unbundling issues, stranded costs, and CO2 emission allowances. In addition, he often advises on the regulatory aspects of the construction of power plants, gas storages, and (offshore) wind farms. Clemens Ziegler Clemens Ziegler is a senior associate in Brussels and practises EU and German competition law as well as EU State aid law at Hogan Lovells. In State aid cases, he counsels in proceedings of the European Commission, the EU Courts and national authorities. Clemens has advised numerous clients on their State aid issues including Phase 1 and 2 Commission investigations. Areas of focus include State aid in the energy and natural resources sectors, as well as in the transport sector. Clemens is a regular speaker on current State aid developments, including on the implications of Brexit in the State aid area. Clemens is licensed to practise law in Germany and in the State of New York and is fluent in German, English, French and Italian. Kristina Zych Kristina Zych, LLM (Brussels), is a counsel of Linklaters LLP, resident in Berlin. She has special expertise in German public (commercial) law (Öffentliches Wirtschaftsrecht) and European law, including State aid law, public procurement

xxii List of Contributing Authors law and law of regulated industries (energy/health care), privatisations and public– private partnerships. Kristina Zych has ten years of experience in a wide variety of projects advising the Government and state-owned entities, private companies and financial institutions in complex infrastructure projects, privatisations as well as huge and complex award procedures, notably in the energy sector, relating to the realisation of energy infrastructure projects in Germany and Central & Eastern Europe. Another focus of Kristina Zych’s work lies in advising the public banking sector (public sector banks, development banks as well as their public owners) particularly with regard to State aid and subsidy law, institutional as well as public procurement law.

List of Abbreviations ACER AG BAA BEE BITs BNetzA CACM CBCA CEER CEF CF CJEU DG DSR DSOs EBRD ECT EEA EEC EEG EEPR EFSI EFTA EIB EIIs ENTSO-E ENTSO-G ERDF ESIF EU EU ETS FSR GATS GATT GBER GDP IIAs INEA IPCEI

Agency for the Cooperation of Energy Regulators Advocate-General British Aggregate Association German Association for Renewable Energies Bilateral Investment Treaties Federal Network Agency for Electricity, Gas, Telecommunications, Post and Railway (Bundesnetzagentur) Capacity Allocation and Congestion Management Cross-Border Cost Allocation Council of European Energy Regulators Connecting Europe Facility Cohesion Fund Court of Justice of the European Union Directorate-General Demand-Side Response Distribution System Operators European Bank for Reconstruction and Development Energy Charter Treaty European Economic Area European Economic Community Renewable Energies Act (Erneuerbare-Energien-Gesetz) European Energy Programme for Recovery European Fund for Strategic Investment European Free Trade Association European Investment Bank Energy-Intensive Industries European Network of Transmission System Operators for Electricity European Network of Transmission System Operators for Gas European Regional Development Fund European Structural Investment Funds European Union European Emissions Trading System Florence School of Regulation General Agreement on Trade in Services General Agreement on Tariffs and Trade General Block Exemption Regulation Gross Domestic Product International Investment Agreements Innovation and Networks Executive Agency Important Project of Common Economic Interest

xxiv List of Abbreviations NOME NRA PCIs PPAs R&D R&D&I REMIT RES SGEI SGI SME TEN-E TEN-T TEU TFEU TSOs TTIP TYNDP VAT WTO

Nouvelle Organisation du Marché de l’Electricité National Regulatory Authority Projects of Common Interest Power Purchase Agreements Research and Development Research, Development and Innovation Regulation on wholesale energy market integrity and transparency Renewable Generation Sources Service of General Economic Interest Service of General Interest Small or Medium-sized Enterprise Trans-European Energy Networks Trans-European Transport Networks Treaty on the European Union Treaty on the Functioning of the European Union Transmission System Operators Transatlantic Trade and Investment Partnership Ten-Year Network Development Plan Value Added Tax World Trade Organization

Table of Cases European Union Court of Justice Case 30/59 De Gezamenlijke Steenkolenmijnen in Limburg v High Authority, EU:C:1961:2................................................................................................................. 472 Case C-30/59 Steenkolenmijnen v High Authority, EU:C:1961:2 ......................................... 6 Case 6/64 Costa v ENEL, EU:C:1964:66 ......................................................... 357, 459, 524 Case C-56/65 Société Technique Minière v Maschinenbau Ulm, EU:C:1966:38 ............... 118 Joined Cases 6/69 and 11/69 Commission v France, EU:C:1969:68 ................................. 473 Case 70/72 Commission v Germany, EU:C:1973:87......................................................... 304 Case 120/73 Gebrüder Lorenz GmbH v Federal Republic of Germany et Land de Rhénanie-Palatinat (Lorenz), EU:C:1973:118 .......................................... 357, 459, 476 Case C-173/73 Italy v Commission, EU:C:1974:71 ........................................ 10, 18, 25, 520 Case C-74/74 CNTA v Commission, EU:C:1975:59 ........................................................ 329 Case C-33/76 Rewe, EU:C:1976:188 ..........................................................................363–64 Joined Cases C-64/76, 113/76, 167/78, 239/78, 27/79, 28/79 and 45/79 Dumortier Freres SA and others v EEC (Dumortier), EU:C:1982:185 .......................... 461 Case C-78/76 Steinike and Weinlig, EU:C:1977:52 ...............................................70–71, 361 Ruling 1/78 Physical Protection of Nuclear Materials, EU:C:1978:202 ............................ 205 Case C-61/79 Amministrazione delle finanze dello Stato v Denkavit italiana [1980] ECR 1205 ......................................................................................................... 345 Case C-172/80 Gerhard Züchner v Bayerische Vereinsbank AG, EU:C:1981:178 ............ 277 Joined Cases C-188-190/80 France, Italy and the UK v Commission (Transparency Directive), EU:C:1982:257 ........................................................................................... 204 Case C-283/81 CILFIT, EU:C:1982:335 ................................................................... 393, 445 Case C-7/82 GVL v Commission, EU:C:1983:52 ............................................................. 277 Case 72/83 Campus Oil Limited and Others v Minister for Industry and Energy and others, EU:C:1984:256 .......................................................................................... 277 Case C-72/83 Campus Oil, EU:C:1984:256 ..................................................................... 198 Case C-290/83 Commission v France (Poor Farmers), EU:C:1985:37 ................................ 23 Case C-169/84 COFAZ v Commission, EU:C:1990:301 ...............................................34–35 Case C-284/84 Germany v Commission, EU:C:1987:437 .................................................. 18 Joined Cases C-67/85, C-68/85 and C-70/85 Van der Kooy v Commission, EU:C:1988:38................................................................................................16, 22–23, 37 Case C-213/85 Commission v Netherlands, EU:C:1988:39 .............................................. 384 Case C-223/85 Rijn-Schelde-Verolme (RSV) Machinefabrieken en Scheepswerven NV v Commission, EU:C:1987:502 .............................................................................. 331 Case C-249/85 Albako v BALM, EU:C:1987:245 ............................................................ 383 Case C-259/85 France v Commission, EU:C:1987:478 ...................................................... 71 Case C-265/85 Van den Bergh en Jurgens BV v Commission, EU:C:1987:121 ................. 331 Case C-310/85 Deufil v Commission, EU:C:1987:96 ....................................................... 481 Case 328/85 Deutsche Babcock v Hauptzollamt Lübeck-Ost, EU:C:1987:548 ................. 223 Case C-57/86 Greece v Commission, EU:C:1988:284 ........................................................ 71 Case C-111/86 Delauche v Commission, EU:C:1987:562 ................................................ 372

xxvi Table of Cases Case C-30/87 Bodson v Pompes funèbres des régions libérées, EU:C:1988:225.................... 9 Joined Cases C-62/87 and C-72/87 Exécutif regional wallon v Commission, EU:C:1988:132............................................................................................................. 263 Case C-102/87 France v Commission, EU:C:1988:391 .................................................... 315 Joined Cases C-106-120/87 Astéris AE et al v Greece and European Economic Community, EU:C:1988 ........................................................................................... 9, 345 Case 142/87 Belgium v Commission, EU:C:1990:125 ........... 362, 460, 476, 481–82, 525–26 Case 301/87 France v Commission, EU:C:1990:67 .................................. 256, 302, 362, 476 Case C-2/88-Imm, Zwartveld and Others, EU:C:1990:440 ...................................... 360, 366 Joined Cases C-143/88 and C-92/89 Zuckerfabrik, EU:C:1991:65 .......................... 324, 384 Case C-152/88 Sofrimport SARL v Commission EU:C:1990:259..................................... 329 Case C-303/88 Italy v Commission (ENI-Lanerossi), EU:C:1991:136 ...........................12–13 Case C-5/89 Commission v Germany, EU:C:1990:320 ............................................. 332, 372 Case C-96/89 Commission v Netherlands [1991] ECR I-2461 ......................................... 330 Joined Cases C-104/89 and C-37/90 Mulder and Others v Council and Commission, EU:C:2000:38 .................................................................................. 329 Case C-234/89 Delmitis, EU:C:1991:91 ........................................................................... 366 Case C-305/89 Italy v Commission, EU:C:1991:142 ........................................................ 117 Joined Cases C-6/90 and C-9/90 Francovich and Bonifaci v Italy, EU:C:1991:428............................................................................................. 307, 374, 460 Case C-33/90 Commission v Italy, EU:C:1991:476 .......................................................... 366 Case C-313/90 CIRFS and Others v Commission, EU:C:1993:111 .................................. 254 Case C-354/90 Fédération nationale du commerce extérieur des produits alimentaires (FNCE) and Others v France, EU:C:1991:440 ............357, 359–62, 372, 383, 459–60, 476, 524 Case C-17/91 Lornoy et al v Belgian State, EU:C:1992:514 ..................................... 360, 524 Case C-47/91 Italy v Commission EU:C:1994:358 ........................................................... 378 Joined Cases C-72/91 and C-73/91 Sloman-Neptun, EU:C:1993:97 .................................. 71 Joined Cases C-144/91 and C-145/91 Demoor and Others v Belgian State, EU:C:1992:518............................................................................................................. 362 Case C-183/91 Commission v Greece, EU:C:1993:233 .................................................... 117 Case C-189/91 Kirsammer Hack, EU:C:1993:907 ..................................................... 71, 361 Case C-38/92 Italian State to Alumix ................................................................................. 40 Case C-188/92 TWD Textilwerke Deggendorf GmbH v Germany [1994] ECR I-833 ...... 384 Case C-275/92 Schindler, EU:C:1994:119 ........................................................................ 185 Case C-393/92 Gemeente Almelo, EU:C:1994:171 .......................................................... 184 Joined Cases C-46/93 and C-48/93 Brasserie du pêcheur and Factortame, EU:C:1996:79....................................................................................... 307, 364, 374, 460 Case C-56/93 Belgium v Commission, EU:C:1996:64 .............................................16, 35–36 Case C-348/93 Commission v Italian Republic, EU:C:1995:95 ................................ 304, 366 Case C-350/93 Commission v Italian Republic, EU:C:1995:96 ................................ 304, 308 Case C-465/93 Atlanta, EU:C:1995:369 .................................................................. 324, 384 Case C-39/94 SFEI and Others, EU:C:1996:285 ...................... 9–10, 304, 358–62, 366, 369, 372–73, 375, 377, 383, 476 Case C-159/94 Commission v France, EU:C:1997:501 .................................................... 121 Joined Cases C-178/94, C-179/94, C-188/94, C-189/94 and C-190/94 Dillenkofer and Others v Germany, EU:C:1996:375 ..................................................... 374 Case C-24/95 Alcan Deutschland, EU:C:1997:163................................................... 372, 384 Case C-169/95 Spain v Commission EU:C:1997:10 ......................................................... 481 Case C-278/95 P Siemens v Commission, EU:C:1997:240 ............................................... 305

Table of Cases xxvii Case C-288/96 Germany v Commission, EU:C:2000:537 .......................................... 14, 254 Case C-342/96 Spain v Commission, EU:C:1999:210 .......................................................... 9 Case C-415/96 Spain v Commission, EU:C:1998:533 ........................................................ 26 Case C-6/97 Italy v Commission, EU:C:1999:251 ............................................................ 473 Case C-75/97 Belgium v Commission (Maribel bis/ter), EU:C:1999:311 .............................. 5 Case C-104/97 P Atlanta v European Community, EU:C:1999:498 ................................. 329 Case C-126/97 Eco Swiss China Time Ltd v Benetton International NV, EU:C:1999:269..................................................................................................... 345, 350 Case C-174/97 P FFSA and Others v Commission, EU:C:1998:130 ................................. 283 Case C-200/97 Ecotrade v Altiforni e Ferriere di Servola, EU:C:1998:579 ..................... 6, 26 Case C-251/97 France v Commission, EU:C:1998:572 ...................................................... 10 Case C-256/97 DM Transport, EU:C:1999:332 ......................................................... 12, 472 Case C-295/97 Piaggio, EU:C:1999:313............................................................................... 6 Case C-372/97 Italy v Commission, EU:C:2004:234 .................................................. 41, 473 Case C-404/97 Commission v Portugal EU:C:2000:345................................................... 481 Case C-39/98 Aide de la part d’EDF à certaines firmes de l’industrie papetière .................. 38 Case C-82/98 P Kögler v Court of Justice, EU:C:2000:282 .............................................. 372 Case C-83/98 P France v Ladbroke Racing and Commission, EU:C:2000:248 ................... 22 Case C-107/98, Teckal Srl v Comune di Viano and Azienda Gas-Acqua Consorziale (AGAC) di Reggio Emilia, EU:C:1999:562 ............................................... 295 Case C-156/98 Germany v Commission, EU:C:2000:467 ................................................ 522 Joined Cases C-180/98 to C-184/98 Pavlov et al, EU:C:2000:428........................................ 7 Case C-351/98 Spain v Commission, EU:C:2002:530 ........................................................ 26 Case C-379/98 PreussenElektra, EU:C:2001:160 ............... 6, 17–21, 57–64, 70, 72–78, 119, 170, 231, 388–90, 435, 445, 456, 486, 543 Case C-390/98 Banks, EU:C:2001:456................................................................312–13, 525 Case C-480/98 Spain v Commission (Magefesa), EU:C:2000:559 ........................................ 6 Case C-29/99 Commission v Council, EU:C:2002:734..................................................... 203 Case C-143/99 Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke, EU:C:2001:598 ............................................................... 24, 358, 442, 444 Case C-310/99 Italy v Commission, EU:C:2002:143 ................................................ 254, 315 Case C-312/90 Spain v Commission, EU:C:1992:282 ...................................................... 378 Joined Cases C-328/99 and C-399/00 Italy v Commission, EU:C:2003:252 ................312–13 Case C-334/99 Germany v Commission, EU:C:2003:55 .................................................. 366 Case C-382/99 Netherlands v Commission, EU:C:2002:368 ............................................ 363 Case C-390/99 Canal Satélite Digital [2002] ECR I-607 .................................................. 380 Case C-400/99 Italy v Commission, EU:C:2005:275 ........................................................ 378 Case C-403/99 Italian Republic v Commission, EU:C:2001:507 ...................................... 329 Case C-475/99 Ambulanz Glöckner, EU:C:2001:577 ........................................................... 9 Case C-482/99 France v Commission (Stardust Marine), EU:C:2002:294 ...............13, 22–23 Case C-503/99 Commission v Belgium, EU:C:2002:328 .................................................. 277 Case C-53/00 Ferring, EU:C:2001:627 ............................................................................. 283 Case C-129/00 Commission v Italy, EU:C:2003:656 ........................................................ 321 Case C-209/00 Commission v Germany, EU:C:2002:74 ................................................... 317 Case C-277/00 Germany v Commission, EU:C:2004:238 ................................................ 313 Case C-280/00 Altmark Trans and Regierungspräsidium Magdeburg, EU:C:2003:415................................... 17, 27–28, 54, 63, 169–70, 211, 273, 280, 283–84, 286, 288–89, 293, 295, 297–98, 422, 570 Case C-298/00 P Italy v Commission, EU:C:2004:240 ....................................................... 27 Case C-399/00 SIM 2 Multimedia v Commission, EU:C:2003:252 .................................... 18

xxviii Table of Cases Case E-04/01 Karl K Karlsson hf v The Icelandic State [2002] EFTA Ct Rep 240 ............ 518 Joined Cases C-34/01 to C-38/01 Enirisorse v Ministero delle Finanze, EU:C:2003:640................................................................................................................. 9 Case C-126/01 GEMO SA, EU:C:2003:622 ............................................................16–17, 71 Case C-167/01 Inspire Art [2003] ECR I-10155 .............................................................. 380 Case C-224/01 Köbler, EU:C:2003:513 .................................................................... 307, 363 Joined Cases C-261/01 and C-262/01 van Calster and Cleeren, EU:C:2003:571................................................................................358–60, 362, 370, 472 Joined Cases C-264/01, C-306/01, C-354/01 and C-355/01 AOK-Bundesverband et al, EU:C:2003:304 ...................................................................... 8 Case C-297/01 Sicilcassa and Others, EU:C:2003:416 ..................................................... 361 Case C-314/01 Siemens and ARGE Telekom [2004] ECR I-2549..................................... 380 Joined Cases C-65/02 P and C-73/02 P ThyssenKrupp v Commission [2005] ECR I-6773 ....................................................................................................... 330 Case C-174/02 Streekgewest, EU:C:2005:10 .......................................................364, 368–69 Joined Cases C-183/02 P and C-187/02 P Demesa and Territorio Histórico de Álava v Commission, EU:C:2004:701 .............................................. 358, 372 Case C-276/02 Spain v Commission, EU:C:2004:521 ...................................................... 472 Case C-345/02 Pearle et al, EU:C:2004:448 ......................................................... 22, 71, 360 Case C-17/03 Vereniging voor Energie, Milieu en Water v Directeur van de Dienst uitvoering en toezicht energie, EU:C:2005:362 .................................. 219, 329, 428 Case C-88/03 Portugal v Commission (Azores), EU:C:2006:511........................................ 18 Joined Cases C-128/03 and C-129/03 AEM SpA and AEM Torino SpA v Autorita’ per l’energia elettrica e per il gas and others, EU:C:2005:224 ........457, 471–72 Case C-152/03 Ritter-Coulais, EU:C:2006:123 ................................................................ 380 Case C-172/03 Heiser, EU:C:2005:130 ................................................................ 5, 8, 20, 28 Case C-173/03 Traghetti del Mediterraneo, EU:C:2006:391 .................................... 307, 374 Case C-182/03 Belgium v Commission, EU:C:2006:416 .................................................... 27 Joined Cases C-182/03 and C-217/03 Forum 187 v Commission, EU:C:2006:416 ........... 22, 329–32, 358, 372 Case C-276/03 P Scott v Commission, EU:C:2005:590 .................................................... 305 Joined Cases C-346/03 and C-529/03 Atzeni and Others [2006] ECR I-1875 .................. 384 Case C-415/03 Commission v Greece, EU:C:2005:287 ...........................6, 306, 312–13, 366 Joined Cases C-442/03 P and C-471/03 P P&O European Ferries (Vizcaya) v Commission, EU:C:2006:356 ............................................................................ 369, 375 Case C-71/04 Xunta de Galicia, EU:C:2005:493.............................................. 362, 372, 383 Case C-148/04 Unicredito Italiano, EU:C:2005:774................................................. 358, 372 Case C-222/04 Cassa di Risparmio di Firenze et al, EU:C:2006:8 .................................. 8, 25 Case C-237/04 Enirisorse v Sotocarbo, EU:C:2006:197 ............................................... 7, 361 Joined Cases C-295/04 to C-298/04 Manfredi, EU:C:2006:461 ....................................... 364 Case C-368/04 Transalpine Ölleitung in Österreich, EU:C:2006:644 ........... 357–64, 370–71, 375, 383, 443–44, 460 Joined Cases C-392/04 and C-422/04 i-21 Germany, EU:C:2006:586 .................363–64, 371 Joined Cases C-393/04 and C-41/05 Air Liquide Industries Belgium, EU:C:2006:403....................................................................................... 70, 357, 360, 362 Case C-404/04 P Technische Glaswerke Ilmenau v Commission, EU:C:2007:6 .................. 16 Case C-526/04 Laboratoires Boiron, EU:C:2006:528..........................305, 363–64, 394, 440 Case C-119/05 Lucchini, EU:C:2007:434................................................................. 307, 323 Case C-232/05 Commission v France (Scott), EU:C:2006:651 ....305, 307, 323–24, 364, 384

Table of Cases xxix Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission, EU:C:2008:461 ............................................. 349 Case C-415/05 Commission v Greece, EU:T:2010:386 ..................................................... 312 Case C-426/05 Tele2 Telecommunication, EU:C:2008:103 .............................................. 365 Case C-176/06 P Stadtwerke Schwäbisch Hall ao v Commission, EU:C:2007:730 ........... 212 Case C-187/06 Commission v Belgium, EU:C:2008:527 .................................................. 321 Case C-199/06 CELF and ministre de la Culture and de la Communication (CELF I), EU:C:2008:79 ............................................... 360, 362, 368, 372, 375, 383, 524 Case C-206/06 Essent Netwerk Noord et al, EU:C:2008:413.....................23, 57, 59, 61–62, 70, 72–73, 78, 119, 389, 429, 486 Case C-341/06 P Chronopost v Ufex, EU:C:2008:375 ....................................................... 16 Case C-419/06 Commission v Greece, EU:C:2008:89 ...................................................... 314 Case C-441/06 Commission v France, EU:C:2007:616 .................................................305–6 Case C-487/06 P British Aggregates v Commission, EU:C:2008:757........................ 5, 25, 68 Case C-28/07 P Ter Lembeek International EU:C:2007:764 ............................................. 321 Case C-49/07 MOTOE, EU:C:2008:376 ...................................................................... 8, 190 Case C-113/07 P SELEX Sistemi Integrati v Commission, EU:C:2009:191 .......................... 9 Case C-214/07 Commission v France, EU:C:2008:619 ............................................ 318, 366 Case C-290/07 P Commission v Scott, EU:C:2010:480 ............................................ 305, 324 Case C-333/07 Société Régie Networks, EU:C:2008:764 ................................................. 413 Case C-334/07 P Commission v Freistaat Sachsen, EU:C:2008:709 ............................. 5, 375 Case C-384/07 Wienstrom, EU:C:2008:747 ........................................375, 383, 443, 445–46 Case C-369/07 Commission v Hellenic Republic, EU:C:2009:428 ................................... 317 Case C-441/07 P Commission v Alrosa, EU:C:2010:377.................................................... 55 Case C-2/08 Fallimento Olimpiclub, EU:C:2009:506 ....................................................... 323 Case C-78/08 Ministero dell’Economia e delle Finanze v Paint Graphos, EU:C:2011:550............................................................................................................... 66 Joined Cases C-78/08 to C-80/08 Paint Graphos and Others, EU:C:2011:550 ................. 357 Case C-169/08 Presidente del Consiglio dei Ministri v Regione autonoma della Sardegna, EU:C:2009:420 ................................................................................................ 5 Case C-265/08 Federutility and Others v Autorità per l’energia elettrica e il gas, EU:C:2010:205............................................................................................................. 282 Case C-279/08 P Commission v Netherlands, EU:C:2011:551 ............................6, 20–21, 24 Case C-507/08 Commission v Slovakia, EU:C:2010:802.......................................... 320, 322 Case C-1/09 CELF and ministre de la Culture and de la Communication (CELF II), EU:C:2010:136 .................................................................... 360, 364, 372, 377 Joined Cases C-71/09 P, C-73/09 P and C-76/09 P Comitato ‘Venezia vuole vivere’ et al v Commission, EU:C:2011:368.............................................................. 25, 27 Case C-106/09 P Commission and Spain v Government of Gibraltar and United Kingdom, EU:C:2011:732 ........................................................................ 5, 67 Case C-210/09 Scott SA et Kimberly Clark SAS v Ville d’Orléans, EU:C:2010:294.......... 325 Case C-305/09 Commission v Italy, EU:C:2011:274 ................................................ 320, 323 Case C-331/09 Commission v Poland, EU:C:2011:250 .................................................... 320 Case C-369/09 P ISD Polska and Others v Comission, EU:C:2011:175 ........................... 372 Joined Cases C-465/09 P and C-470/09 P Territorio Histórico de Vizcaya— Diputación Foral de Vizcaya, Territorio Histórico de Álava et al v Commission, EU:C:2011:372............................................................................................................. 321 Case C-496/09 Commission v Italy, EU:C:2011:740 ........................................................ 325 Case C-549/09 Commission v France, EU:C:2011:672 .................................................... 320

xxx Table of Cases Case C-81/10 P France Télécom v Commission, EU:C:2011:811 ..................... 310, 317, 321 Case C-124/10 P Commission v EDF, EU:C:2012:318 ..................................................11–12 Case C-242/10 Enel Produzione SpA v Autorita per l’energia elettrica e il gas, EU:C:2011:861............................................................................................................. 282 Case C-243/10 Commission v Italy, EU:C:2012:182 ........................................................ 482 Case C-275/10 Residex Capital IV CV v Gemeente Rotterdam, EU:C:2011:814........................................................................308, 311, 359–60, 362, 372 Case C-399/10 P Bouygues and Bouygues Télécom v Commisison et al, EU:C:2013:175..........................................................................................................21–22 Case C-403/10 P Mediaset SpA v Commission, EU:C:2011:533 ................ 24, 308, 315, 321 Case C-448/10 P Thyssen Krupp Acciai Speciali Terni SpA v European Commission ....... 468 Joined Cases C-448/10 P to C-450/10 P ThyssenKrupp Acciai Speciali Terni SpA et al .... 469 Case C-452/10 P BNP Paribas v Commission, EU:C:2012:366 ............................................ 5 Case C-459/10 P Freistaat Sachsen v Commission, EU:C:2011:515 ..................................... 5 Case C-490/10 Parliament v Council, EU:C:2012:525 ..................................................... 204 Case C-584/10 P European Commission and Others v Yassin Abdullah Kadi, EU:C:2013:518............................................................................................................. 349 Case C-610/10 Commission v Spain, EU:C:2012:781 ...................................................... 325 Case C-138/11 Compass-Datenbank, EU:C:2012:449 ......................................................... 9 Case C-184/11 Commission v Spain, EU:C:2014:33 ........................................................ 325 Case C-288/11 P Mitteldeutsche Flughafen and Flughafen Leipzig-Halle v Commission, EU:C:2012:821 ........................................................................................ 7 Case C-677/11 Doux Élevage and Coopérative agricole UKL-ARREE, EU:C:2013:348....................................................................................................18, 70–71 Case C-6/12 P Oy, EU:C:2013:525 .............................................................................359–61 Joined Cases C-204/12 to C-208/12 Essent Belgium EU:C:2013:294 ............................... 185 Case C-262/12 Vent de Colère et al, EU:C:2013:851 ....... 20, 22, 57–59, 61–63, 74, 78, 119, 403, 412–14, 486–87, 504, 543 Case C-284/12 Deutsche Lufthansa, EU:C:2013:755 .........350, 357, 359–62, 366, 370, 377, 379–80, 383, 390, 524 Joined Cases C-501/12 to C-506/12, C-540/12 and C-541/12 Specht and Others, EU:C:2014:2005 ....................................................................................... 374 Case C-573/12 Ålands Vindkraft, EU:C:2014:2037 ................................................. 142, 291 Case C-27/13 Flughafen Lübeck, EU:C:2014:240 ............................................................ 379 Case C-69/13 Mediaset, EU:C:2014:71 .................................................................... 315, 383 Case C-219/13 Ferraci v Commission, EU:C:2014:2207 .................................................. 318 Case C-275/13 Elcogás, EU:C:2014:2314 .......................................................... 63, 486, 504 Case C-518/13 Eventech v The Parking Adjudicator, EU:C:2015:9 ............................ 28, 553 Case C-672/13 OTP Bank, EU:C:2015:185.........................................357, 359–60, 372, 383 Case C-5/14 Kernkraftwerke Lippe-Ems, EU:C:2015:354 ........................ 205, 208, 212, 394 Case C-37/14 Commission v France, EU:C:2015:90 ........................................................ 318 Case C-63/14 Commission v France, EU:C:2015:458 ...................................................... 384 Case C-99/14 P Carbunión v Council, EU:C:2014:2446 .................................................. 227 Joined Cases C-352/14 and C-353/14 Iglesias Gutiérrez and Elisabet Rion Bea v Bankia SA and Others, EU:C:2015:691 .................................................358–60 Case C-367/14 Commission v Italy, EU:C:2015:611 ........................................................ 325 Case C-441/14 DI, EU:C:2016:278 .................................................................................. 374 Case C-493/14 Dilly’s Wellnesshotel, EU:C:2016:577 .........................................446, 449–51 Case C-505/14 Klausner Holz Niedersachsen, EU:C:2015:742 ...........323, 363–64, 371, 377 Case C-524/14 P Commission v Hansestadt Lübeck, EU:C:2016:971 .............................. 379

Table of Cases xxxi Case C-574/14 PGE, EU:C:2016:686 ..........................................................359–60, 366, 370 Case C-590/14 P DEI and Commission v Alouminion tis Ellados, EU:C:2016:797 ........... 46, 359–60, 366, 371, 383, 479 Case C-604/14 P Alcoa Trasformazioni v Commission, EU:C:2016:54 .............................. 42 Case C-121/15 Association nationale des opérateurs détaillants en énergie (ANODE) v Premier ministre and Others, EU:C:2016:637 .......................................282, 410–11, 511 Case C-268/15 Ullens de Schooten, EU:C:2016:874......................................................... 374 Case C-379/15 Association France Nature Environnement, EU:C:2016:603 .................... 412 Case C-234/16 Kapferer, EU:C:2006:178 ......................................................................... 323 Case C-405/16 P Germany v Commission, pending ....................................... 19, 78, 87, 232 Case C-640/16 P Greenpeace Energy v Commission ........................................................ 274 Case C-585/17 Dilly’s Wellnesshotel................................................................................. 451 General Court/Court of First Instance Joined Cases T-244/93 and T-486/93 TWD Deggendorf v Commission, EU:T:1995:160 ............................................................................................................. 306 Case T-459/93 Siemens v Commission, EU:T:1995:100 ........................................... 305, 522 Case T-239/94 Association des Aciéries Européennes Indépendantes (EISA) v Commission, EU:T:1997:158 ............................................................................. 364, 374 Case T-353/94 Postbank v Commission, EU:T:1996:119 ................................................. 366 Case T-358/94 Air France v Commission, EU:T:1996:194 ........................................... 18, 22 Case T-380/94 AIUFASS and AKT v Commission, EU:T:1996:195 .................................. 254 Case T-106/95 FFSA and Others v Commission, EU:T:1997:23 ....................................... 283 Joined Cases T-129/95, T-2/96 and T-97/96 Neue Maxhütte Stahlwerke and Lech-Stahlwerke v Commission, EU:T:1999:7 ......................................................... 10 Case T-214/95 Vlaamse Gewest v Commission, EU:T:1998:77 .................................. 28, 254 Joined Cases T-126/96 and T-127/96 BFM and EFIM v Commission, EU:T:1998:207 ............................................................................................................... 12 Case T-46/97 SIC v Commission [2000], EU:T:2000:123 ................................................. 283 Case T-296/97 Alitalia v Commission (Alitalia I), EU:T:2000:289...................................... 12 Case T-36/99 Lenzing v Commission, EU:T:2004:312 .......................................................... 6 Joined Cases T-227/99 and T-134/00 Kvaerner Warnow Werft GmbH v Commission, EU:T:2002:54 ............................................................................................................... 330 Joined Cases T-228/99 and T-233/99 Westdeutsche Landesbank v Commission, EU:T:2003:57 ..................................................................................................9–10, 12, 18 Joined Cases T-269/99, T-271/99 and T-272/99 Diputación Foral de Guipúzcoa and Others v Commission, EU:T:2002:258 ................................................. 379 Joined Cases T-346/99 to T-348/99 Diputación Foral de Álava et al v Commission, EU:T:2002:259 ................................................................................................................. 5 Case T-98/00 Linde v Commission, EU:T:2002:248 ........................................................... 11 Joined Cases T-254/00, T-270/00 and T-277/00 Hotel Cipriani v Commission, EU:T:2008:537 ............................................................................................................. 309 Case T-308/00 Salzgitter v Commission, EU:T:2013:30 ........................................... 369, 375 Case T-324/00 CDA Datenträger Albrechts v Commission, EU:T:2005:364..................... 117 Case T-366/00 Scott v Commission, EU:T:2003:113 ................................................ 305, 324 Case T-109/01 Fleuren Compost v Commission, EU:T:2004:4 ..................................... 5, 331 Joined Cases T-116/01 and T-118/01 P & O European Ferries (Vizcaya) v Commission, EU:T:2003:217 ........................................................................329, 331–32 Joined Cases T-195/01 and T-207/01 Government of Gibraltar v Commission, EU:T:2002:111 ............................................................................................................. 310

xxxii Table of Cases Case T-198/01 Technische Glaswerke Ilmenau v Commission, EU:T:2004:222 .................. 16 Case T-217/01 Forum des migrants v Commission [2003] ECR II-1563 .......................... 330 Case T-92/02 Stadtwerke Schwäbisch Hall ao v Commission, EU:T:2006:26 ................... 212 Case T-93/02 Confédération nationale du Crédit Mutuel (Livret bleu), EU:T:2005:11 ....... 28 Case T-217/02 Ter Lembeek International v Commission, EU:T:2006:361....................... 321 Case T-351/02 Deutsche Bahn AG v Commission, EU:T:2006:104 .................................. 345 Case T-20/03 Kahla Thüringen Porzellan v Commission, EU:T:2008:395 .......................... 11 Case T-68/03 Olympiaki Aeroporia Ypiresies v Commission, EU:T:2007:253 .................. 313 Case T-271/03 Deutsche Telekom v Commission, EU:T:2008:101 ..................................... 56 Case T-289/03 BUPA v Commission, EU:T:2008:29 ................................................. 277, 290 Case T-156/04 EDF v Commission, EU:T:2009:505 ........................................................... 12 Case T-196/04 Ryanair v Commission, EU:T:2008:585...................................................... 11 Case T-211/04 Government of Gibraltar v Commission, EU:T:2008:595 ........................... 67 Joined Cases T-304/04 and T-316/04 Italian Republic and Wam SpA v Commission, EU:T:2006:239 ............................................................................................................. 474 Joined Cases T-427/04 and T-17/05 France v Commission, EU:T:2009:474 ..................... 310 Case T-163/05 DEP-Bundesverband deutscher Banken v Commission, EU:T:2010:59........ 12 Case T-211/05 Italy v Commission, EU:T:2009:304 ........................................................... 26 Joined Cases T-415/05, T-416/05 and T-423/05 Hellenic Republic, Olympiakes Aerogrammes AE, and Olympiaki Aeroporia Ypiresies AE v Commission, EU:T:2010:386 ............................................................................................................. 314 Case T-415/05 Greece v Commission, EU:T:2010:386 ..................................................... 312 Joined Cases T-80/06 and T-182/09 Budapesti Erömü Zrt v Commission, EU:T:2012:65 ........................................................................................................5, 11–12 Case T-80/06 Budapesti Erőmű v Commission ................................................................. 286 Joined Cases T-298/07, T-312/97, T-313/97, T-315/97, T-600/97 to T-607/97, T-1/98, T-3/98 to T-6/98 and T-23/98 Alzetta et al, EU:T:2000:151................................ 26 Case T-62/08 ThyssenKrupp Acciai Speciali Terni v Commission, EU:T:2010:268 ............................................................................................... 45, 329, 468 Case T-268/08 Land Burgenland and Austria v Commission, EU:T:2012:90 ........................ 5 Case T-384/08 Elliniki Nafpigokataskevastiki et al v Commission, EU:T:2011:650 ........... 22 Joined Cases T-394/08, T-408/08, T-453/08 and T-454/08 Regione autonoma della Sardegna and others v Commission, EU:T:2011:493 ............................................ 308 Case T-394/08 Regione autonoma della Sardegna v Commission, EU:T:2011:493 ........... 330 Case T-396/08 Freistaat Sachsen und Land Sachsen-Anhalt v Commission, EU:T:2010:297 ............................................................................................................. 522 Joined Cases T-443/08 and T-455/08 Mitteldeutsche Flughafen and Flughafen Leipzig/Halle v Commission, EU:T:2011:117 ....................................................................................... 14 Case T-443/08 Freistaat Sachsen, EU:T:2011:117 ............................................................ 316 Case T-452/08 DHL Aviation SA/NV and DHL Hub Leipzig GmbH v Commission, EU:T ............................................................................................................................. 311 Case T-468/08 Tisza Erömü kft v Commission, EU:T:2014:235 ............... 22, 24, 28, 54, 384 Case T-565/08 Corsica Ferries France v Commission, EU:T:2012:415 ..........................12–13 Joined Cases T-115/09 and T-116/09 Electrolux and Whirlpool v Commission, EU:T:2012:76 ............................................................................................................... 383 Case T-123/09 Ryanair v Commission, EU:T:2012:164...................................................... 12 Case T-139/09 France v Commission, EU:T:2012:496 ............................................... 18, 486 Case T-179/09 Dunamenti Erömü Zrt v Commission, EU:T:2014:236 ..........................11–12 Case T-182/09 EU:T:2012:65 ........................................................................................... 286 Case T-79/10 France v Commission, EU:T:2013:463 ....................................................... 273

Table of Cases xxxiii Case T-137/10 CBI v Commission, EU:T:2012:584 .......................................................... 121 Case T-177/10 Alcoa Trasformazioni v Commission, EU:T:2014:897 ...........................41–42 Case T-499/10 MOL v Commission, EU:C:2010:55............................................................. 6 Case T-57/11 Castelnou Energía v Commission (T-57/11), EU:T:2014:102 ..................................................................... 121, 225, 232, 278, 287, 291 Case T-176/11 Carbunión v Council, EU:T:2013:686 ...................................................... 227 Joined Cases T-233/11 and T-262/11 Greece v European Commission and Ellinikos Chrysos v Commission, ECLI:EU:T:2015:948 .............................................................. 321 Case T-251/11 Austria v Commission, EU:T:2014:1060 .....................5–6, 20, 22, 26, 70, 74 Case T-308/11 Eurallumina v Commission, EU:T:2014:894 ............................................... 43 Case T-319/11 ABN Amro Group v Commission, EU:T:2014:186 ................................... 254 Case T-542/11 Alouminion v Commission, EU:T:2014:859 ............................................... 46 Case T-674/11 TV2/Danmark v Commission, EU:T:2015:684 ......................................... 383 Case T-242/12 Société nationale des chemins de fer v Commission, EU:T:2015:1003 ...... 313 Case T-295/12 Germany v Commission, EU:T:2014:675 ........................................... 27, 281 Case T-461/12 Hansestadt Lübeck v Commission, EU:T:2014:758 .................................. 379 Case T-473/12 Aer Lingus Ltd v Commission, EU:T:2015:78 .......................................... 305 Case T-517/12 Alro v Commission, EU:T:2014:890 ............................................378, 382–83 Case T-129/13 Alpic v Commission, EU:T:2014:895 ........................................................ 382 Case T-454/13 Société nationale maritime Corse Méditerranée (SNCM) v Commission, EU:T:2017:134 ............................................................................................................. 298 Joined Cases T-515/13 and T-719/13 Spain v Commission, EU:T:2015:1004 ..................... 66 Case T-172/14 Stahlwerke Bous v Commission, EU:T:2015:402 ...................................... 381 Case T-298/14 Erbslöh Aluminium v Commission, EU:T:2015:349 ................................. 381 Case T-300/14 Fricopan Back v Commission, EU:T:2015:336 ......................................... 381 Case T-301/14 Michelin Reifenwerke v Commission, EU:T:2014:747 .............................. 381 Case T-694/14 EREF v Commission, EU:T:2015:915 ....................................................... 113 Case T-788/14 MPF Holdings v Commission ........................................................... 167, 396 Case T-793/14 Tempus Energy and Tempus Energy Technology v Commission ....... 167, 396 Case T-812/14 R BPC LUX 2 and Others v Commission, EU:T:2015:119 ....................... 384 Case T-47/15 Germany v Commission, EU:T:2016:281 .......................18–19, 51, 63, 77, 87, 119, 232, 389 Case T-57/15 Trajektna luka Split v Commission, EU:T:2016:470 ..................................... 57 Case T-356/15 Austria v Commission, pending ........................................................ 217, 275 Case T-382/15 Greenpeace Energy and Others v Commission, EU:T:2016:589 ........ 217, 274 Case T-624/15 European Food and Others v Commission ............................................... 510 Case T-704/15 Micula ea v Commission [2016] OJ C68/30 ..................................... 348, 510 Case T-620/17 TVO v Commission .................................................................................. 210 Other International EFTA Court Case E-4/97 of 3 March 1999 .......................................................................................... 522 Case E-06/98 The Kingdom of Norway v EFTA Surveillance Authority ................... 520, 522 Case E-1/100 State Debt Management Agency v. Islandsbanki-FBA ................................. 522 Case E-02/02 of 19 June 2003.....................................................................................522–23 Joined Cases E-5/04, 6/04 and 7/04 Fesil et al v ESA [2005] EFTA Ct Rep 117 ......................................................................................................520, 522–23 Case E-09/04 of 7 April 2006........................................................................................... 522 Case E-2/05 of 24 November 2005 .................................................................................. 522

xxxiv Table of Cases Case E-02/06 EFTA Surveillance Authority v The Kingdom of Norway (Waterfall case) [2007] EFTA Ct Rep 164............................................................. 518, 528 Case E-5/07 of 21 February 2008..................................................................................... 522 Case E-06/09 Magasin-og Ukepresseforeningen v EFTA Surveillance Authority ............... 522 Joined Cases E-04/10, E-6/10 and E-07/10 of 10 May 2011 ............................................ 522 Case E-14/10 of 22 August 2011...................................................................................... 522 Joined Cases E-17/10 and E-06/11 of 30 March 2012...................................................... 522 Joined Cases E-10/11 and E-11/11 of 8 October 2012 ..................................................... 522 Case E-12/11 of 17 August 2012...................................................................................... 522 Case E-16/11 EFTA Surveillance Authority v Iceland ....................................................... 522 Case E-1/12 of 11 December 2012 ................................................................................... 522 Case E-1/13 of 27 January 2014 ...................................................................................... 522 Case E-08/13 of 29 August 2014...................................................................................... 522 International Arbitration AES Summit Generation Limited and AES-Tisza Erőmű Kft v Hungary, ICSID Case No ARB 07/22..............................................................................336, 339–40 Charanne BV and Construction Investments SÀRL v Kingdom of Spain, SCC Arb No 062/2012 ................................................................................................. 335 Crystallex International Corporation v Bolivarian Republic of Venezuela, ICSID Case No ARB(AF)/11/2...................................................................................... 334 Duke Energy Electroquil Partners and Electroquil SA v Republic of Ecuador, ICSID Case No ARB/04/19......................................................................................335–36 ECE Projektmanagement International GmbH and Kommanditgesellschaft Panta Achtundsechzigste GmbH & Co v Czech Republic, PCA Case No 2010-5 (UNCITRAL) ..........................................................................................................334–35 EDF International v Hungary, UNCITRAL Award (3 December 2014)............................ 336 Electrabel SA (Belgium) v Republic of Hungary, ICSID Case No ARB/07/19 ..............337–39 Eureko BV v Slovak Republic, UNCITRAL arbitration, PCA Case No 2008-13 .............. 347 Franck Charles Arif v Republic of Moldova, ICSID Case No ARB/11/23 ........................ 335 Frontier Petroleum Ltd v Czech Republic, PCA (UNCITRAL) ......................................... 336 International Thunderbird Gaming Corp v United Mexican States, UNCITRAL (NAFTA) Award ........................................................................................................... 333 Ioan Micula, Viorel Micula, SC European Food SA, SC Starmill SRL and SC Multipack SRL v Romania, ICSID Case No ARB/05/20 ......................... 340, 353, 509 LG&E Energy Corp, LG&E Capital Corp and LG&E International Inc v Argentine Republic, ICSID Case No ARB/02/1 .....................................................334–35 Parkerings Compagniet AS, ICSID Case No ARB/05/8 ............................................ 333, 335 Saluka Investments BV, PCA (UNCITRAL) .................................................................334–35 Ulysseas, Inc v Republic of Ecuador, PCA Case No 2009-19 (UNCITRAL) ..................... 335 WTO Appellate Body Reports Canada—Certain Measures Affecting the Renewable Energy Generation Sector / Canada—Measures Relating to the Feed-in Tariff Program, adopted 24 May 2013, DSR 2013:I, 7 .........................................................97, 104, 106–7 United States—Final Countervailing Duty Determination with Respect to Certain Softwood Lumber from Canada, WT/DS257/AB/R, adopted 17 February 2004, DSR 2004:II, 571 ................................................................................................... 95, 101

Table of Cases xxxv Panel Reports Canada—Certain Measures Affecting the Renewable Energy Generation Sector / Canada—Measures Relating to the Feed-in Tariff Program, adopted 24 May 2013, as modified by Appellate Body Reports WT/DS412/ AB/R/WT/DS426/AB/R, DSR 2013:I, 237..........................................................................................97, 104, 106–7 Canada—Measures Affecting the Export of Civilian Aircraft, WT/DS70/R, adopted 20 August 1999, upheld by Appellate Body Report WT/DS70/AB/R, DSR 1999:IV, 1443 ........................................................................................................ 96 India—Certain Measures Relating to Solar Cells and Solar Modules, WT/DS456/R, circulated to WTO Members 24 February 2016 (appeal pending) .................................. 97 United States—Measures Treating Exports Restraints as Subsidies, WT/DS194/R and Corr 2, adopted 23 August 2001, DSR 2001:XI, 5767 .......................................... 100 United States—Subsidies on Upland Cotton, WT/DS267/R, Add.1 to Add 3 and Corr 1, adopted 21 March 2005, as modified by Appellate Body Report WT/DS267/AB/R, DSR 2005:II, 299 ........................................................................................................... 96 National Austria Case 2003/17/0001 [2004] Supreme Administrative Court, referral decision ................... 444 Case 2004/05/0274 [2007] Supreme Administrative Court .........................................444–45 Case 2004/17/0078 [2007] Supreme Administrative Court .............................................. 444 Case 2006/17/157 [2006] Supreme Administrative Court ................................................ 444 Case 2008/17/0138 [2009] Supreme Administrative Court .............................................. 446 Case 2012/17/0175 [2012] Supreme Administrative Court .............................................. 448 Case 2012/17/0469 [2013] Supreme Administrative Court .............................................. 448 Case 2013/15/0053 [2013] Supreme Administrative Court .............................................. 449 Case A 10/08 [2009] Constitutional Court....................................................................... 446 Case B 321/12 [2012] Constitutional Court ..................................................................... 448 Case B 1348/02 [2002] Constitutional Court ................................................................... 443 Case B 2251/97 [2001] Constitutional Court ........................................................... 442, 450 Case RV/0188-I/12 [2012] UFS, Außenstelle Innsbruck ................................................... 447 Case RV/5100360/2013 [2016] BFG, Außenstelle Linz .................................................... 450 Cases B2251/97 and B2594/97 [1999] Constitutional Court, VfSlg 15450/2001 ............. 441 Supreme Court of Austria, Spa Gardens ......................................................................370–71 Belgium Brussels, Commercial Court of Brussels Ryanair v Brussels Airlines [29 April 2015] AR 2014/52 .................................................................................................................. 370 Commercial Court of Brussels, Ryanair v Brussels Airlines [29 April 2015] A 2014/52.655 ............................................................................................................. 368 Commercial Court of Kortrijk, Gimvindus and Flemish Region v Idealspun, De Clerck and Others [20 September 1994] Case No 1310/90 ..................................... 372 Constitutional Court decision 17 July 2014, Case 106/2014 .......................................................................... 489 decision 17 September 2015, Case 114/2015 ................................................................ 489 President of the Commercial Court of Brussels, Breda Fucine Meridionali v Manoir Industries (13 February 1995) .............................................................................. 368, 370 Raad van State decision of 30 June 2015, Case 231.827 T-Power v Belgian State ................................. 490

xxxvi Table of Cases NV Wattplus/NV Essent Belgium v Flemish Region, Case 221.374, judgment of 13 November 2012 ............................................................................................... 486 Trib Com Bruges, of 12 February 2009, Zeebrugse Visveiling v Exploitatie Vismijn Oostende NV/Pakhuizen Oostende NV, 00886/08........................................................ 309 Tribunal de première instance francophone de Bruxelles, S Civile, Jugement, 25 January 2016, 15/7241/A and 15/7242/A ................................................................ 349 France Administrative Tribunal of Bastia (TA Bastia), Corsica Ferries [23 February 2017], Case No 1500375......................................................................................................... 375 CE 1 July 2010, No 321595............................................................................................. 404 CE 9 October 2015, No 369417 ...................................................................................... 418 CE 10 July 2012, No 353356........................................................................................... 411 CE 11 April 2014, No 365219 ......................................................................................... 404 CE 15 December 2014, No 370321 ................................................................................. 411 CE 15 May 2012 Association Vent de Colère! Fédération Nationale et autres No 324852 ..............................................................................................................412–13 CE 19 July 2017, No 370321......................................................................................411–12 CE 20 November 2012, No 363572 ................................................................................ 411 Conseil Constitutionnel, Decision No 2006-543 DC of 30 November 2006—Act pertaining to the energy sector .................................................................... 409 Germany BGH Case V ZR 314/02 02 of 4 April 2003.......................................................................... 476 Case VIII ZR 156/16 .................................................................................................... 390 Case XI ZR 53/03 (2004) ............................................................................................. 476 judgment of 9 February 2017—I ZR 91/15 Flughafen Lübeck ............................. 381, 387 judgment of 10 February 2011—I ZR 136/09, EuZW 2011, 440 ................................. 387 Press Release regarding judgment of 9 February 2017—I ZR 91/15 ............................. 391 BVerfG decision of 6 September 2016—1 BvR 1305/13—, juris ............................................... 393 decision of 11 October 1994—2 BvR 633/86 ............................................................... 389 BVerwG judgment of 10 October 2012—7 C 11/10—, juris ..................................................392–93 judgment of 26 October 2016—10 C 3/15 ................................................................... 391 Judgment of 6 May 2015—VIII ZR 56/14—BGHZ 155 .................................................. 388 Judgment of 14 May 2013—I-19 U/12, juris .................................................................... 389 Judgment of 25 June 2014—VIII ZR 169/13—BGHZ 201 .............................................. 389 OLG Düsseldorf decision of 15 July 2015—VI-3 Kart 83/14 (V)—, juris ...........................................396–97 decision of 28 April 2015—VI-3 Kart 332/12 (V)—, juris .......................................395–96 OLG Hamburg, judgment of 5 July 2016—9 U 156/15—, juris ....................................... 390 OLG München, judgment of 18 October 2000—20 U 2503/00—, juris........................... 395 Greece Council of State, Judgment No 2406/2014....................................................................... 477 Decision 3910/1988 of the 4th Chamber of the Hellenic Council of State ........................ 368

Table of Cases xxxvii Italy Case No RG 34082/2015, Milan Tribunal ....................................................................... 463 Cassation Court, joint chambers, 13 December 2016, No 25516..................................... 463 Consiglio di Stato, Sez 3, 13 May 2015, No 2401............................................................ 463 Consiglio di Stato No 18/2002 (Reg Ric 237/2002) of 14 January 2003, with respect to an appeal for annulment of a judgment of the Lombardy Administrative Tribunal, Sez II, No 6265 of 3 October 2001 ....................................................... 457, 470 Court of Appeal of Cagliari Exol SpA v Nuova Cartiera di Arbatax SpA [21 July 1999] .........................................................................................................368–69 Court of First Instance of Genova Grandi traghetti di navigazione SpA v Viamare di navigazione Spa and Finmare SpA [26 April 1993] .................................. 369 Judgment RG No 481/208 of 10 April 2008 of Modena Commissione Tributaria Provinciale, Sezione VI................................................................................. 469 Lombardy Administrative Tribunal, Sez II, no 6265 of 3 October 2001 ................... 457, 470 Regional Administrative Court of Lazio 11 June 1990, No 1071 ..................................... 472 Netherlands Administrative Jurisdiction Division of the Council of State Vogelaarwijken [6 February 2013] NL:RVS:2013:BZ0794 .................................................................... 369 CBb, 9 December 2009 AWB 08/433 and 08/434 .............................................................................................. 427 AWB 08/532, 08/533 and 08/534 ................................................................................. 427 Court Arnhem, 22 June 2009, E.ON Benelux N.V./Ministry of Economic Affairs, NL:RBARN: 2009:BJ1171 ........................................................................................... 431 Court Groningen, 22 June 2005, NJF 2005, 296 ............................................................. 429 Court of Appeal The Hague, 16 October 2015, NL:GHDA:2015:3415 ........................... 440 Court of The Hague, 1 August 2008, Dutch State/NEA, NL:RBSGR:2007:BB1424 ........ 430 High Administrative Court (Afdeling bestuursrechtspraak van de Raad van State) 8 February 2012, NL:RVS:2012:BV3215 ..................................................................... 438 22 October 2008, NL:RVS:2008:BG1152 .................................................................... 438 President of the District Court of Groningen Essent Kabelcom BV v Gemeente Appingedam [3 September 2004 NL:RBGRO:2004:AQ8920 ....................................... 371 Q10 Offshore Wind BV (Case SA.34742) ........................................................................ 436 Norway Borgarting Court of Appeal Case LB-1998-1805 ............................................................. 524 Gulating Court of Appeal Case LG-2008-16104 .............................................................. 524 Gulating Court of Appeal Case LG-2015-150132 ............................................................ 524 Oslo Court of First Instance, Case TOSLO-2010-45497 .................................................. 525 Spain Constitutional Court 19/2016 of 4 February 2016 ........................................................... 499 Constitutional Court 29/2016 of 18 February 2016 ......................................................... 499 Constitutional Court 61/2016 of 17 March 2016 ............................................................ 499 Constitutional Court 270/2015 of 17 December 2015 ..................................................... 499 Constitutional Court of 19 May 2016.............................................................................. 513 Supreme Court of 1 July 2016 Case 1259/2016 INCAM ................................................. 501

xxxviii

Table of Cases

Supreme Court of 3 October 2016 Case TRADELDA ..................................................... 503 Supreme Court of 7 February 2012 (STS 419/2010) ........................................................ 511 Supreme Court of 20 July 2016 Case Compañia Europea General Textil SL .................... 503 Supreme Court of 24 October 2016, STS 4526/2016 (ES:TS:2016:4526) and STS 4527/2016 (ES:TS:2016:4527)........................................................................ 510 Supreme Court of 27 February 2015 Case Elcogás........................................................... 504 Switzerland Republic A v B International, 4A 34/2015, Federal Supreme Court of Switzerland, 6 October 2015 ............................................................................................................ 352 United Kingdom Air Canada v Emerald Supplies [2015] EWCA Civ 1024 ................................................. 349 Micula v Romania [2017] EWHC 31 (Comm) ............................................................349–50 United States Micula v Gov’t of Romania Case No 15 Misc 107 (LGS), 2015 WL 5257013 (SDNY 9 September 2015) .............. 351 Case No 15 Misc 107 (Part I), 2015 WL 4643180 (SDNY 5 August 2015).............351–52

Table of Legislation – International and European Aarhus Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters ........................................................... 247 Agreement between the EEA EFTA States on the Establishment of a Surveillance Authority and a Court of Justice (SCA) Art 31 ......................................................................................................................521–22 Art 32 ........................................................................................................................... 522 Art 34 ........................................................................................................................... 522 Art 36 ........................................................................................................................... 522 Protocol 3 ........................................................................................521, 524, 534, 547–48 Agreement between the Government of the United Kingdom of Great Britain and Northern Ireland and the Government of the Republic of Mozambique for the Promotion and Protection of Investments, Art 2(2) ........................................... 333 ASEAN Comprehensive Investment Agreement, Art 11(1) ............................................... 334 Association Agreement between the European Union and its Member States, on the one part, and Ukraine, on the other part............................................................ 557 Association Agreement between the European Union and the European Atomic Energy Community and their Member States, on the one part, and the Republic of Moldova, on the other part........................................................................................................... 557 Directive 96/92/EC concerning common rules for the internal market in electricity ................................................................... 13, 28, 219–20, 426, 489, 492–93 Art 24 ........................................................................................................................... 219 Directive 1998/30 concerning common rules for the internal market in natural gas ......... 220 Directive 2000/60/EC establishing a framework for Community action in the field of water policy .................................................................................... 399, 530 Directive 2003/54/EC ............................................... 182, 274, 287, 291, 403, 492, 511, 517 Art 2(4) ........................................................................................................................ 189 Art 3(4) ........................................................................................................................ 561 Art 9 ............................................................................................................................. 189 Art 11(4) .............................................................................................................. 233, 278 Directive 2003/55/EC ....................................................................................................... 409 Directive 2003/87/EC establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC ............ 21, 392 Directive 2003/96/EC restructuring the Community framework for the taxation of energy products and electricity ..................................................26, 69, 79–82, 208, 439 Art 2(1) ........................................................................................................................ 208 Art 2(4) ........................................................................................................................ 523 Art 4 ............................................................................................................................... 80 Art 14 ................................................................................................................... 208, 439 Art 14(1)(a) .................................................................................................................. 208 Art 17 ............................................................................................................................. 81 Art 17(2) ........................................................................................................................ 81

xl

Table of Legislation – International and European

Directive 2004/18/EC ....................................................................................................... 295 Directive 2005/89/EC concerning measures to safeguard security of electricity supply and infrastructure investment ............................................................ 156, 182, 516 Art 4 ............................................................................................................................. 193 Directive 2006/111/EC on the transparency of financial relations between Member States and public undertakings as well as on financial transparency within certain undertakings ................................................................................................................. 293 Art 1(2) ........................................................................................................................ 293 Art 2(d) ........................................................................................................................ 293 Directive 2006/67/EC imposing an obligation on Member States to maintain minimum stocks of crude oil and/or petroleum products .............................................. 516 Directive 2009/28/EC on the promotion of the use of energy from renewable sources, amending and subsequently repealing Directives 2011/77/EC and 2003/30/EC ............................................................................................. 70, 124, 503 Directive 2009/29/EC amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community (ETS Directive) ............................................................................................21, 87–90, 393 Art 4 ............................................................................................................................... 87 Art 9 ............................................................................................................................... 88 Art 10 ............................................................................................................................. 88 Art 10(2) ........................................................................................................................ 88 Art 10(2)(a) .................................................................................................................... 88 Art 10(2)(c) .................................................................................................................... 88 Art 10(c) ......................................................................................................................... 21 Art 10a ..................................................................................................................... 88, 90 Art 12(1) ........................................................................................................................ 88 Art 16(3) ........................................................................................................................ 88 Directive 2009/72/EC on the internal electricity market ........... 182, 189, 274, 492, 511, 529 Art 3(1) ........................................................................................................................ 282 Art 3(15) .............................................................................................................. 169, 418 Art 3(2) ........................................................................................................................ 423 Art 42 ........................................................................................................................... 193 Directive 2009/73/EC concerning common rules for the internal market in natural gas ................................................................................ 189, 282, 409, 517, 571 Directive 2011/70/Euratom establishing a Community framework for the responsible and safe management of spent fuel and radioactive waste ............................................ 203 Directive 2012/27/EU on energy efficiency ..........................................................398–99, 453 Directive 2014/104/EU on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union................................................................................. 474 Directive 2014/25/EU on procurement by entities operating in the water, energy, transport and postal services sectors ............................................................................. 294 Art 24(2) and (3) .......................................................................................................... 294 Art 44(2) ...................................................................................................................... 294 Art 50 ........................................................................................................................... 295 EEA Agreement ...................................................................515–21, 527, 536, 544, 547, 551 Recital 4 ....................................................................................................................... 515 Art 2(a)......................................................................................................................... 518 Art 7 ............................................................................................................................. 518 Art 14 ........................................................................................................................... 520

Table of Legislation – International and European xli Art 15 ........................................................................................................................... 520 Art 24 ........................................................................................................................... 515 Art 31 ........................................................................................................................... 525 Art 40 ........................................................................................................................... 527 Art 54 ........................................................................................................................... 527 Art 59(1) ...................................................................................................................... 527 Art 61 ........................................................................................................... 521, 525, 534 Art 61(1) .........................................................................................519–20, 523, 545, 551 Art 61(2) .............................................................................................................. 519, 524 Art 61(3) ........................................................................... 519, 522–24, 536, 539, 545–46 Art 61(3)(c) ............................................................... 519, 535–36, 541–42, 545, 551, 553 Art 62 ................................................................................................................... 521, 534 Art 62(1) ...................................................................................................................... 520 Art 102 ......................................................................................................................... 516 Art 102(1) .................................................................................................................... 518 Protocol 26 ................................................................................................................... 521 Ann IV.......................................................................................................... 515, 517, 527 EFTA Convention ............................................................................................................ 518 Energy Charter Treaty (ECT) .......................................... 328, 334, 337–39, 352, 500–1, 506 Art 10 ........................................................................................................................... 500 Art 10(1) ...................................................................................................................... 334 Art 13 ........................................................................................................................... 500 Energy Community Treaty (EnCT)................................................ 227, 555, 557–59, 568–71 Art 18 ........................................................................................................................... 568 Art 18(1)(c) .................................................................................................................. 566 Art 18(c) ....................................................................................................................... 557 Art 19 ........................................................................................................................... 568 Art 67(b) ...................................................................................................................... 558 Art 67(d) ...................................................................................................................... 558 Art 70 ........................................................................................................................... 558 Art 90(1) ...................................................................................................................... 567 Arts 90–93.................................................................................................................... 566 Procedural Act on the Rules of Procedure for Dispute Settlement under the Treaty (DSR) ....................................................................................................... 566 Art 2 ................................................................................................................. 569, 571 Art 2(1) ..................................................................................................................... 567 Art 2(2) ..................................................................................................................... 567 Art 4 ......................................................................................................................... 567 Art 11 ....................................................................................................................... 567 Art 19(2) ................................................................................................................... 567 Espoo Convention on Environmental Impact Assessment in a Transboundary Context ........................................................................................................................ 247 EU–Canada Comprehensive Economic and Trade Agreement Art 8(10)(2) .................................................................................................................. 336 Art 8(10)(4) .......................................................................................................... 334, 336 Euratom Treaty ........................................................................... 7, 201–5, 212, 214–17, 219 Recital 3 ....................................................................................................................... 203 Art 1 ............................................................................................................. 202, 205, 219 Art 2 ........................................................................................................202, 204, 218–19 Art 2(b) ........................................................................................................................ 202

xlii Table of Legislation – International and European Art 2(c) ......................................................................................................................... 217 Art 2(d) ........................................................................................................................ 205 Art 4 ............................................................................................................................. 202 Art 5 ............................................................................................................................. 202 Art 6 ............................................................................................................................. 202 Art 6(d) ........................................................................................................................ 202 Art 30 ........................................................................................................................... 203 Art 34 ........................................................................................................................... 202 Art 37 ........................................................................................................................... 202 Arts 40–44............................................................................................................ 202, 204 Art 52 ........................................................................................................................... 219 Art 67 ........................................................................................................................... 203 Art 106(a)(3) .................................................................................................................... 7 Art 106a ....................................................................................................................... 204 Art 106a(3)................................................................................................................... 204 Art 187 ......................................................................................................................... 204 Art 192 ................................................................................................................. 203, 205 Art 197 ......................................................................................................................... 202 Ann 1 ........................................................................................................................... 202 European Convention on Human Rights.......................................................................... 503 Free Trade Agreement between the European Union and the Socialist Republic of Vietnam, Art 14........................................................................................................ 336 ICSID Arbitration Rules, Art 37(2) .................................................................................. 337 ICSID Convention ......................................................................... 339, 343, 348–49, 351–52 Art 52 ........................................................................................................................... 348 Art 53 ........................................................................................................... 345, 348, 351 Art 53(1) ...................................................................................................................... 343 Art 54 ......................................................................................................345, 348, 351–52 Art 54(1) ...................................................................................................................... 343 Implementing Rules to Decision No 4/2000 of the EU-Romanian Association Council, Art 1............................................................................................................... 342 Kyoto Protocol ........................................................................................................... 88, 392 New York Convention ........................................................................................343–46, 350 Art III ........................................................................................................................... 343 Art V ............................................................................................................................ 343 Art V(1)(a).................................................................................................................... 343 Art V(1)(b) ................................................................................................................... 343 Art V(1)(c) .................................................................................................................... 343 Art V(1)(d) ................................................................................................................... 343 Art V(1)(e) .................................................................................................................... 343 Art V(2) ........................................................................................................................ 343 Art V(2)(b) ......................................................................................................345–46, 350 North American Free Trade Agreement, Art 1105(1) ....................................................... 334 Partnership and Cooperation Agreement between the European Communities and their Member States and the Republic of Moldova ................................................ 556 Partnership and Cooperation Agreement between the European Communities and their Member States and Ukraine........................................................................... 556 Regulation (EEC) No 17/62 First Regulation implementing Articles 85 and 86 of the Treaty ................................................................................................................. 116

Table of Legislation – International and European xliii Regulation (EEC) No 1983/83 on the application of Article 85(3) of the Treaty to categories of exclusive distribution agreements......................................................... 118 Regulation (EC) No 2894/94 concerning arrangements for implementing the Agreement on the European Economic Area................................................................. 518 Regulation (EC) No 659/1999 ...................................................301, 304, 318, 321, 323–25, 391, 447, 450, 469 Regulation (EC) No 1407/2002 on State aid to the coal industry ..........28, 223–28, 254, 488 Art 4(e) ......................................................................................................................... 225 Art 6 ............................................................................................................................. 225 Regulation (EU) No 1/2003 ............................................................................................. 118 Regulation (EC) No 713/2009 Establishing an Agency for the Cooperation of Energy Regulation ............................................................................................................ 517, 527 Regulation (EC) No 714/2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003 ......... 182, 190, 517 Art 7 ............................................................................................................................. 527 Art 16 ................................................................................................................... 193, 195 Art 16(6) ...................................................................................................................... 190 Art 17 ........................................................................................................................... 190 Regulation (EC) No 715/2009 on Conditions for Access to the Natural Gas Transmission Networks ................................................................................................ 517 Regulation (EU, Euratom) No 617/2010 concerning the notification to the Commission of investment projects in energy infrastructure within the European Union .................. 204 Regulation (EU) No 360/2012 of 25 April 2012 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid granted to undertakings providing services of general economic interest................... 29 Regulation (EU) No 360/2012 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid granted to undertakings providing services of general economic interest............................................................. 272 Regulation (EU) No 648/2012 on OTC derivatives, central counterparties and trade repositories ................................................................................................................... 532 Regulation (EU, Euratom) No 966/2012 on the financial rules applicable to the general budget of the Union ...................................................................................................... 253 Regulation (EU) No 347/2013 on guidelines for trans-European energy infrastructure .................................................................... 235–38, 240–43, 246, 248, 250, 252, 255, 268–69 Recital 7 ....................................................................................................................... 236 Recital 30 ..................................................................................................................... 246 Recital 41 ..................................................................................................................... 248 Art 3(1) ........................................................................................................................ 242 Art 3(3) ........................................................................................................................ 242 Art 3(4) ........................................................................................................................ 242 Art 4 ............................................................................................................................. 241 Art 4(1) ...................................................................................................................237–38 Art 4(2) ...................................................................................................................237–38 Art 4(4) ........................................................................................................ 237, 239, 242 Art 5(1) ........................................................................................................................ 242 Art 5(4) ........................................................................................................................ 243 Art 5(5) ........................................................................................................................ 243 Art 5(9) ........................................................................................................................ 243 Art 7 ............................................................................................................................. 245

xliv Table of Legislation – International and European Art 7 et seq ................................................................................................................... 245 Art 7(2) and (3) ............................................................................................................ 245 Art 8 ............................................................................................................................. 246 Art 8(2) ........................................................................................................................ 246 Art 8(3) ........................................................................................................................ 246 Art 8(4) ........................................................................................................................ 246 Art 8(5) ........................................................................................................................ 246 Art 9(3) ........................................................................................................................ 247 Art 9(4) ........................................................................................................................ 247 Art 9(5) ........................................................................................................................ 247 Art 9(7) ........................................................................................................................ 247 Art 10 ........................................................................................................................... 246 Art 10(2) ...................................................................................................................... 246 Art 10(6) ...................................................................................................................... 246 Art 11 ........................................................................................................................... 247 Art 11 et seq ................................................................................................................. 245 Art 13 ........................................................................................................................... 247 Art 14 ........................................................................................................................... 249 Art 14(1) ...................................................................................................................... 249 Art 14(2) ...................................................................................................................... 250 Art 14(4) ...................................................................................................................... 250 Art 15(1) ...................................................................................................................... 249 Ann I ............................................................................................................................ 237 Pt III.......................................................................................................................... 249 Ann II ..............................................................................................................237–38, 261 Ann II(1)(a)–(d) and (2) ................................................................................................ 250 Ann III s 1(1)......................................................................................................................... 240 s 1(4)......................................................................................................................... 240 s 1(5)......................................................................................................................... 240 s 1(7)......................................................................................................................... 240 s 2(1)................................................................................................................. 241, 247 s 2(11)....................................................................................................................... 241 s 2(12)....................................................................................................................... 242 s 2(13)....................................................................................................................... 242 s 2(14)....................................................................................................................... 242 s 2(3 .......................................................................................................................... 241 s 2(4)......................................................................................................................... 241 s 2(6)......................................................................................................................... 241 s 2(7)......................................................................................................................... 241 s 2(8)......................................................................................................................... 241 s 2(9)......................................................................................................................... 241 Ann IV(1) ..................................................................................................................... 238 Ann IV(2)–(5) ............................................................................................................... 237 Ann IV(2)(a) ................................................................................................................. 238 Ann IV(2)(c) ................................................................................................................. 238 Ann IV(3)(a) ................................................................................................................. 239 Ann IV(3)(b) ................................................................................................................. 239 Ann IV(3)(c) ................................................................................................................. 239

Table of Legislation – International and European xlv Ann IV(3)(d) ................................................................................................................. 239 Ann IV(4) ..................................................................................................................... 238 Ann IV(5)(a) ................................................................................................................. 239 Ann IV(5)(b) ................................................................................................................. 239 Ann IV(5)(c) ................................................................................................................. 239 Ann VI.......................................................................................................................... 247 Regulation (EU) No 1300/2013 on the Cohesion Fund .................................................... 250 Regulation (EU) No 1301/2013 on the European Regional Development Fund and on specific provisions concerning the investment for growth and jobs goal ............ 250 Regulation (EU) No 1303/2013 on the Cohesion Fund ............................................ 251, 254 Regulation (EU) No 1316/2013 establishing the Connecting Europe Facility ................... 248 Art 5 ............................................................................................................................. 248 Art 7 ............................................................................................................................. 249 Art 10(3) ...................................................................................................................... 249 Art 14(2) ...................................................................................................................... 249 Art 15(4) ...................................................................................................................... 248 Ann I ............................................................................................................................ 250 Regulation (EU) No 1391/2013 ....................................................................................... 242 Regulation (EU) No 1407/2013 of 18 December 2013 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid ......................................................................................... 28, 254 Art 3(1) .......................................................................................................................... 29 Regulation (EU) No 283/2014 on guidelines for trans-European networks in the area of telecommunications infrastructure and repealing Decision 1336/97/EC .................................................................................................... 248 Regulation (EU) No 651/2014 declaring certain categories of aid compatible with the internal market in application of Art 107 and 108 of the Treaty ....................................................................... 4, 78–82, 119, 123, 164, 253–54, 361, 398–400, 446–50, 521 Art 3(1) ........................................................................................................................ 449 Art 4(1)(s)................................................................................................................398–99 Art 4(1)(v) .................................................................................................................... 400 Art 4(1)(w) ................................................................................................................... 401 Art 4(1)(x) .................................................................................................................... 401 Art 9 ............................................................................................................................. 448 Art 9(1) ........................................................................................................................ 449 Art 9(2) ........................................................................................................................ 449 Arts 9–11........................................................................................................................ 80 Art 17(1) ...................................................................................................................... 449 Art 17(1)(a) .................................................................................................................... 81 Art 25 ......................................................................................................................447–49 Art 25(3) ...................................................................................................................... 449 Art 38 ........................................................................................................................... 398 Art 41 ........................................................................................................................... 399 Art 41(6)(a) and (b) ...................................................................................................... 399 Art 41(6)(c) .................................................................................................................. 400 Art 42 ........................................................................................................................... 400 Art 44 ............................................................................................................................. 80 Art 44(3) ........................................................................................................................ 80

xlvi Table of Legislation – International and European Art 46 ........................................................................................................................... 401 Art 48 ........................................................................................................................401–2 Art 58(4) ........................................................................................................................ 80 Art 59 ............................................................................................................................. 80 Ann III .......................................................................................................................... 447 Regulation (EU) No 2015/1017 on the European Fund for Strategic Investments, the European Investment Advisory Hub and the European Investment Project Portal ................................................................................................................ 251 Regulation (EU) No 2015/1222 establishing a guideline on capacity allocation and congestion management ................................................................................. 175, 193 Regulation (EU) No 2015/1588 on the application of Art 107 and 108 of the Treaty on the Functioning of the European Union to certain categories of horizontal State aid (codification) ................................................................................................ 4, 28 Regulation (EU) No 2015/1589, laying down detailed rules for the application of Article 108 of the TFEU .........................................4, 161, 301, 304, 309–11, 321, 359, 361–62, 366–67, 379, 384, 391, 481–82 Art 1(f) ......................................................................................................................... 359 Art 3 ................................................................................................................................. 4 Art 4 ............................................................................................................................. 359 Art 4(3) ........................................................................................................................ 391 Art 4(4) ........................................................................................................................ 391 Art 9 ............................................................................................................................. 359 Art 9(2) ........................................................................................................................ 382 Art 9(3) ........................................................................................................................ 382 Art 9(4) ........................................................................................................................ 382 Art 9(5) ........................................................................................................................ 382 Art 12 ........................................................................................................................... 359 Art 13(2) .............................................................................................................. 362, 379 Art 15 ........................................................................................................................... 359 Art 16 ................................................................................................................... 311, 329 Art 16(1) ...................................................................................................................... 304 Art 16(2) .............................................................................................................. 305, 383 Art 16(3) .................................................................................................372, 383, 482–83 Art 17 ......................................................................................................................309–10 Art 25(1) ...................................................................................................................... 161 Art 29 ................................................................................................................... 359, 480 Art 29(1) .............................................................................................................. 360, 366 Art 29(2) .............................................................................................................. 360, 367 Regulation (EU) No 2015/2282 amending Regulation (EC) No 794/2004 as regards the notification forms and information sheets .............................................. 305 Arts 9–11...................................................................................................................... 305 Regulation (EU) No 2016/89 ........................................................................................... 242 SAA between the European Communities and their Member States and the Former Yugoslav Republic of Macedonia ................................................................................. 556 SAA between the European Communities and their Member States and the Republic of Albania..................................................................................................................... 556 SAA between the European Communities and their Member States and the Republic of Bosnia and Herzegovina ........................................................................................... 556 SAA between the European Communities and their Member States and the Republic of Montenegro.............................................................................................................. 556

Table of Legislation – International and European xlvii SAA between the European Communities and their Member States and the Republic of Serbia ....................................................................................................................... 556 Treaty on European Union (TEU) .................................................................................... 204 Art 4(3) ...............................................................................................6, 365–66, 391, 480 Treaty on the Functioning of the European Union (TFEU) Art 4(3) ........................................................................................................................ 349 Art 30 ................................................................................................................. 6, 86, 260 Art 34 ....................................................................................................................... 6, 198 Arts 34–36.................................................................................................................... 184 Art 35 ....................................................................................................................... 6, 198 Art 36 ......................................................................................................................198–99 Arts 56–57.................................................................................................................... 184 Art 101 ......................................................................................................56, 113–23, 203 Art 101(1) ...............................................................................................................117–18 Art 101(3) ...............................................................................................................117–19 Art 102 ................................................................................................................... 56, 203 Art 106 ................................................................................................................. 282, 409 Art 106(2) ......... 55, 63, 113, 118, 121, 211, 220–21, 272–73, 275, 280, 282–83, 287–91 Art 107 ....................................................... 6, 28–29, 31–32, 57, 78, 80, 113–14, 116–17, 119–21, 143, 184, 204–21, 223, 225, 253–54, 272, 298, 358, 361–62, 398–99, 431, 435, 442, 463, 469–72, 480, 504–5, 558 Art 107(1) ............. 3–5, 7–9, 13, 20–24, 27–28, 33, 35, 37–38, 45–47, 55, 57, 59–60, 63, 71, 78, 118–20, 164, 169, 183, 205–6, 211–12, 214, 231–32, 252–54, 261, 283–85, 288, 316, 328, 361, 363, 365, 367, 371, 377, 379–80, 383, 390, 395, 429, 475–80, 483, 485, 519, 562, 564–65 Art 107(2) ...................................................................................................... 46, 254, 519 Art 107(3) .............46, 55, 68, 113, 118, 120–22, 213, 218, 236, 254, 275, 280, 519, 538 Art 107(3)(a) and (c) ...................................................................................................... 37 Art 107(3)(b) ...................................................................................................261–66, 269 Art 107(3)(c) .................................. 55, 79, 205, 213–15, 218, 232, 254–56, 261–62, 269, 279, 398–401, 407, 437, 487, 489, 566 Art 107(3)(d) ................................................................................................................ 519 Art 108 ......................................31, 204, 223, 329, 350, 367, 443, 458, 469, 475–81, 504 Art 108(2) .....................................33, 36, 38, 167, 288, 319, 327, 377–79, 388, 476, 482 Art 108(2)(a) ................................................................................................................ 481 Art 108(3) ............................ 4, 116, 183, 253, 272, 303, 307, 311, 331–32, 357, 360–62, 364–72, 374–79, 383, 392, 418–19, 441, 450, 459–60, 463, 476, 478, 480, 483 Art 108(4) .............................................................................................................. 78, 521 Art 110 ................................................................................................................... 86, 260 Art 170(2) ............................................................................................................ 252, 256 Art 194 ......................................................................................................................... 181 Art 194(1)(d) ................................................................................................................ 256 Art 260 ................................................................................................................. 318, 325 Art 263 ......................................................................................................................... 384 Art 267 ......................................................................................... 361, 380, 390, 447, 478 Art 278 ................................................................................................................. 324, 384 Art 290 ......................................................................................................................... 254 Art 311 ......................................................................................................................... 253

xlviii Table of Legislation – International and European United Nations Framework Convention on Climate Change (UNFCCC) ............. 4, 103, 110 Vienna Convention on the Law of Treaties (VCLT) .......................................................... 340 Art 30 ........................................................................................................................... 340 Art 30(2) ...................................................................................................................... 348 Art 30(3) ...................................................................................................................... 340 Art 30(4) ...................................................................................................................... 348 Art 31 ........................................................................................................................... 341 Art 59 ........................................................................................................................... 340 WTO Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994..................................................................................... 94, 97 Agreement on Subsidies and Countervailing Measures (ASCM) ........... 91–95, 100, 102–3, 107–10 Art 1 ................................................................................................................... 95, 106 Art 1.1 .............................................................................................................. 101, 103 Art 1.1(a)(1)................................................................................................................ 95 Art 1.1(a)(1)(iii) ........................................................................................................ 100 Art 1.1(a)(1)(iv) ........................................................................................................ 100 Art 1.1(b) .................................................................................................................. 106 Art 1.2 ...................................................................................................................... 106 Art 2 ..............................................................................................94–96, 101, 103, 110 Art 2.1 ........................................................................................................ 96, 101, 105 Art 2.1(a) .................................................................................................................... 96 Art 2.1(b) .................................................................................................................... 96 Art 2.1(c) .................................................................................................................. 102 Art 2.2 ........................................................................................................................ 96 Art 2.3 ........................................................................................................................ 96 Art 3 ........................................................................................................................... 94 Art 3.1(a) .................................................................................................................... 94 Art 3.1(b) .........................................................................................................94, 106–7 Art 3.2 ...................................................................................................................... 107 Art 5 ................................................................................................................... 94, 106 Art 8 ..............................................................................................................94–95, 108 Art 8.2(b) .................................................................................................................. 108 Art 8.2(c) .................................................................................................................. 108 Art 12.9 ...................................................................................................................... 93 Art 14 ................................................................................................................... 93, 96 Art 14(d) ................................................................................................................... 100 Art 15.1 ...................................................................................................................... 93 Art 19 ......................................................................................................................... 93 Art 31 ....................................................................................................................... 109 Agreement on Trade-Related Investment Measures (TRIMS) ....................................106–7 Art 2.1 ...................................................................................................................106–7 General Agreement on Tariffs and Trade (GATT) ..............................94–95, 107, 109, 556 Art III(4) ................................................................................................................... 107 Art XVI....................................................................................................................... 95 Art XX.................................................................................................................109–10 Marrakesh Agreement Establishing the World Trade Organization .................. 91, 94, 106

Table of Legislation – National Albania Law No 9374 ............................................................................................................... 558 Art 16 ....................................................................................................................... 559 Law No 9470/2006 ...................................................................................................... 560 Austria Act Against Unfair Competition, s 1 ............................................................................. 371 Green Electricity Act........................................................................6, 73–74, 78, 445, 540 Law accompanying the budget (Budgetbegleitgesetz), Law No 111/2010 ..................... 446 Law No 71/2003 .......................................................................................................... 443 Law No 158/2002 ........................................................................................................ 443 Law on the rebate of energy taxes (EAVG) ................................................ 441–44, 446–51 Belgium Law dated 28 June 2015 amending the Nuclear Phase-out Act..................................... 488 Nuclear Phase-out Act of 31 January 2003................................................................... 487 Programme Law of 26 December 2013 ........................................................................ 486 Bosnia and Herzegovina State aid law ................................................................................................................. 558 Croatia State aid Act ................................................................................................................. 558 France Amending Finance Law of 2015 Art 5...................................................................................................................... 416 Art 14.................................................................................................................... 416 Capacity Decree............................................................................................................ 169 Decree No 2011-466 of 28 April 2011 ......................................................................408–9 Decree No 2012-1405 of 14 December 2012 ............................................................... 416 Decree No 2015-1823 of 30 December 2015 ............................................................... 408 Energy Code ................................................................................................................. 408 Art L 123-1 et seq ..................................................................................................... 417 Art L 335-1 et seq ..................................................................................................... 417 Arts L.335-1 to L.335-6 ............................................................................................ 169 Arts L 336-1 to L 336-10 .......................................................................................... 408 Arts L 337-13 to L 337-16 ........................................................................................ 408 Art L L 271-1............................................................................................................ 417 Law No 2000-108 .......................................................................................................... 60 Law No 2000-321, Art 4 .............................................................................................. 324

l

Table of Legislation – National Law No 2008-776, Art 166 .......................................................................................... 405 Law No 2010-1488 of 7 December 2010 (Loi NOME) ...............55, 169, 405, 407–8, 416 Law No 2013-312 of 15 April 2013 ............................................................................. 417 Ministerial Order (Arrêté) of 22 January 2015 ............................................................. 416

Georgia Law on competition, No 6148-IS ................................................................................. 558 Germany ABLAV Ordinance ...................................................................................................172–73 Coal Financing Act ....................................................................................................... 227 Code of Civil Procedure, ss 66–71 ................................................................................ 370 Erstes Gesetz zur Änderung des Erneuerbare-Energien-Gesetzes vom 7. November 2006, BGBl 2006 I .............................................................................................................. 388 Law revising the legal framework for the promotion of electricity production from renewable energy (EEG) of 28 July 2011 .............................. 18–19, 51, 63–64, 74, 76–77, 85–87, 105, 255, 262, 388–90, 402 Renewable Energy Act ...................................................................................... 19, 85, 381 Zuteilungsgesetz 2012 vom 7. August 2007 ................................................................. 392 Gibraltar Corporate Tax Act .......................................................................................................... 67 Greece Civil Procedure Code .................................................................................................... 484 Law No 2362/1995 ...................................................................................................... 483 Law No 3468/2006 ...................................................................................................... 477 Law No 4002/2011 ...................................................................................................... 483 Law No 4093/2012 ...................................................................................................... 477 Law No 4254/2014 ...................................................................................................... 477 Italy Civil Code .................................................................................................................... 463 Art 2043 ................................................................................................................... 463 Art 2598 ................................................................................................................... 463 Constitution, Art 103 ................................................................................................... 464 Law 80/2005 .......................................................................................................... 43, 466 Art 11(11) ................................................................................................................. 467 Law of 11 May 2012 no 56 .......................................................................................... 453 Legislative Decree No 3/2017 ....................................................................................... 474 Ordinary Law No 234 of 2012..................................................................................... 463 Kosovo Law 2011/04-L-024, Law on State aid ................................................................. 558, 568 Arts 7–8 .................................................................................................................... 559 Macedonia Law on State aid Control.........................................................................................558–59

Table of Legislation – National li Moldova Law on State aid ........................................................................................................... 558 Art 8 ......................................................................................................................... 559 Montenegro Law of Control of State aid Support and Aid ............................................................... 558 Art 11 ....................................................................................................................... 559 Netherlands Besluit stimulering duurzame energieproductie, Staatsblad 2007, 410 (SDE-Decree) .......................................................................................................431–33 Electricity Act 1998 .................................................................................................426–27 Environmental Taxes Act .............................................................................................. 439 General Administrative Act .......................................................................................... 435 Regulation of the Ministry of Economic Affairs, 20 January 2005, no WJZ 4081042 ...................................................................................................... 431 Transition Act for the Electricity Generation Sector .................................................426–30 Norway Constitution, Art 93 ..................................................................................................... 521 Electricity Certificate Act ...................................................................................... 529, 537 Electricity Certificate Regulation .................................................................................. 537 Energy Act .......................................................................................................526–27, 529 Greenhouse Gas Emission Trading Act ......................................................................... 525 Industrial Concession Act (ICA) ................................................................................... 530 Industrial Licensing Act ........................................................................................ 530, 552 Petroleum Tax Act ...................................................................................................541–42 Waterfall Resources Act (WRA)............................................................................ 530, 552 Oman Model Exploration and Production Sharing Agreement of 2002, Art 26.2 ................... 333 Poland Energy Act ...................................................................................................................... 52 Serbia Law on State aid Control.............................................................................................. 558 Art 6 ......................................................................................................................... 558 Spain Constitution .............................................................................................. 499–500, 502–3 Art 9(3) ..................................................................................................................... 500 Industry Ministry Order IET/1045/2014 of 16 June on the remuneration methodology for certain energy production installations based on renewable sources, cogeneration and waste ....................................................................... 498, 501 Law 3/2013 of 4 June 2013 creating the National Commission for Markets and Competition ....................................................................................................... 493

lii

Table of Legislation – National Law 15/2012 of 27 December on tax measures to ensure the financial sustainability of the energy sector ................................................................................................... 496 Law 24/2013 of 26 December 2013 on the Electricity Sector ..........................492, 497–98 Law 54/1997 of 27 October 1997 on the Electricity Sector .......................................... 492 Ministerial Order ITC/3353/2010 of 28 December setting electricity system tariffs and primes for the year 2011 .......................................................................... 504 Regional Law 11 of 2015 on renewable energies .......................................................... 513 Royal Decree 134/2010 ...........................................................................................287–88 Royal Decree 413/2014 of 6 June on the regulation of energy production based on renewable sources, cogeneration and waste.................................................. 497, 501 Royal Decree 661/2007 of 25 May on the regulation of electric energy under a special regime ......................................................................................................... 495 Royal Decree 900/2015 of 9 October 2015 on the administrative, technical and economic regulation of production and self-consumption of electrical energy .... 512 Royal Decree 968/2014 of 21 November on the financing mechanisms for the social bond .................................................................................................... 510 Royal Decree 1565/2010 of 19 November on the regulation and modification of certain aspects relating to electricity production under the special regime ............. 496 Royal Decree-Law 7/2016 of 23 December on the regulation of the financing mechanism of the social bond and other consumer protection measures in the electricity market............................................................................................. 511 Royal Decree-Law 9/2013 of 12 July on urgent measures to ensure the financial stability of the electrical system ........................................................ 497–99, 501–3, 506 Royal Decree-Law 14/2010 of 23 October on the establishment of urgent measures to reduce the tariff deficit in the electricity sector ....................................... 496

Ukraine Law of State aid to Business Entities ........................................................................558–59 United Kingdom Arbitration Act 1966 ...............................................................................................349–50

1 What is State Aid? FALK SCHÖNING AND CLEMENS ZIEGLER

I. INTRODUCTION

I

T CAN MAKE an important difference if a State measure providing support to an undertaking in whatever form does or does not fall under EU State aid rules. As a general rule, all State aid measures are illegal unless they have been notified to and authorised by the European Commission. Even after decades of State aid practice in the EU, what is and what is not aid still appears to be one of the most controversially debated issues in the field.1 The debate around the notion of State aid is of particular importance in the energy sector. From 2009 to 2014 alone, the European Commission (the Commission) registered approved and block-exempted aid measures aimed at environmental protection including energy saving in a total EU-wide amount of close to €117 billion.2 This corresponds to nearly a quarter of the total EU-wide amount of all State aid published on the website of the Commission’s DG COMP.3 However, a sizeable increase in the environmental protection-related numbers for 2014 indicates that in previous years, the actual amount of State aid for environmental protection including energy saving might have been higher.4 Therefore, the overall significance of

1 Over two years after the Commission had published the corresponding draft notice, it published its ‘Notice on the notion of State aid as referred to in Article 107(1) of the Treaty on the Functioning of the European Union’ ([2016] OJ C262/1) (Commission Notice on the notion of State aid). In September 2015, the Commission also published a document called ‘Analytical Grids of the application of State aid rules to the financing of infrastructure projects’ (Analytical Grids), available at: ec.europa.eu/competition/ state_aid/studies_reports/state_aid_grids_2015_en.pdf. The Analytical Grids go beyond a discussion of the notion of State aid in the context of the financing of infrastructure projects and also set out more detailed guidance, where aid is involved, whether a notification is or is not required. 2 Commission, Directorate General for Competition (DG COMP), State aid Scoreboard 2015, available at: ec.europa.eu/competition/state_aid/scoreboard/index_en.html. 3 This excludes aid regarding railways and services of general economic interest and crisis aid to the financial sector. 4 DG COMP writes: ‘Member States (EU28) spent 101.2 billion EUR ie 0.72% of GDP on State aid [in 2014 (excluding aid for railways)]. In nominal terms, this represents an increase of about 50% compared to 2013 expenditures (+33.4 billion EUR). However, the increase in expenditure is largely (85%) due to the inclusion of more renewable energy support schemes (RES) in the reporting. That stems, among others, from the increased awareness by MS of the State aid nature of subsidies to RES following the adoption of the 2014 Energy and Environmental Aid Guidelines. Indeed, 2014 shows an increase of the reported State aid on environmental protection and energy savings of about +28.5 billion EUR at EU level’ (DG COMP’s State aid Scoreboard 2015 (n 2)). In 2014, State aid for environmental protection and

4

Falk Schöning and Clemens Ziegler

State aid in the energy sector is likely much greater than €117 billion over a period of six years, not least because any State aid falling under the de minimis rules would also have to be taken into account, as well as State aid measures the Commission declared incompatible with the common market. The importance of State aid in the energy sector is not surprising. First, the sector is characterised by huge infrastructure and industrial projects (pipelines, grids, power and/or heat plants, storage facilities etc) requiring considerable financial investments. Second, State aid measures have been and will continue to be an important instrument of political steering. The role of and need for political steering can even be expected to grow over the coming few decades, in particular in the energy sector, not least due to the COP 21 climate agreement adopted in Paris in December 20155 to fight climate change: the conversion of the European energy sector from a carbon-intensive to a low- or zero-carbon industry is hard to imagine without considerable involvement of the State. The Commission’s EU Energy Union project6 is also likely to require substantial additional investments to overcome the fragmentation of EU energy markets. Classifying a measure as State aid has significant procedural consequences: in principle, all State aid measures are illegal unless they have been authorised by the Commission. For already notified measures, Article 108(3) TFEU and the Procedural Regulation7 provide for a standstill obligation until ‘the Commission has taken, or is deemed to have taken, a decision authorising such aid’.8 Thus, the delineation of the scope of State aid determines if a project can go ahead without any administrative steps being required or if the standstill and notification obligation trigger a proceeding leading to months and possibly years of delay. Exceptions apply (i) under the GBER9 or a block exemption regulation in the meaning of the Enabling Regulation;10 (ii) under the de minimis exemption;11 (iii) in the case of existing aid in the sense of Article 1(b) of the Procedural Regulation; or (iv) where a measure falls under a State aid scheme that has already been approved by the Commission. This first chapter discusses the concept of State aid as defined in Article 107(1) TFEU. It will, however, not focus on State aid and (i) tariffs in the energy sector and

energy saving (and reported by the Commission) represented not less than 42% of the total amount of State aid (excluding aid for railways and services of general economic interest, as well as crisis aid to the financial sector), compared with a share in the order of 18%–22% from 2009 to 2013 (calculated based on ‘Country fiches, European Union’, State aid Scoreboard 2015 (n 2)). 5 Paris agreement FCCC/CP/2015/L.9, published on 12 December 2015, available at: unfccc.int/ resource/docs/2015/cop21/eng/l09.pdf. 6 Commission, ‘A Framework Strategy for a Resilient Energy Union with a Forward-Looking Climate Change Policy, Energy Union Package’ (Communication) COM(2015) 82 final. 7 Council Regulation (EU) 2015/1589 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union (codification) [2015] OJ L248/9 (‘Procedural Regulation’). 8 Art 3 of the Procedural Regulation. 9 Commission Regulation (EU) 651/2014 declaring certain categories of aid compatible with the internal market in application of Art 107 and 108 of the Treaty [2014] OJ L187/1 (‘GBER’). 10 Council Regulation (EU) 2015/1588 on the application of Art 107 and 108 of the Treaty on the Functioning of the European Union to certain categories of horizontal State aid (codification) [2015] OJ L248/1 (‘Enabling Regulation’). 11 See under section II.C below.

What is State Aid? 5 issues regarding price formation, (ii) direct taxation, (iii) energy/environmental taxes, as well as (iv) energy infrastructure.12 These issues will be discussed in chapters two to nine.

II. WHAT IS STATE AID?

A. General Considerations on the Notion of State Aid The notion of State aid is defined in Article 107(1) TFEU. It reads as follows: 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, in so far as it affects trade between Member States, be incompatible with the internal market.

State aid under EU law is an objective, legal concept.13 Whether a certain measure falls under the notion of State aid or not has to be assessed with regard to the specific facts of the case, leaving only a limited margin of discretion.14 Even if a State measure does not primarily aim at providing support to one or several undertakings and is presented as pursuing goals of public interest (be it described as a measure to achieve certain fiscal, economic or social aims15 or in the interests of the environment or regional policy),16 it may be State aid (and the environmental or other policy objectives may usefully be taken into account in the compatibility assessment regarding the State aid measure in question).17 Only where the State acts ‘by exercising public power’ or where public entities act ‘in their capacity as public authorities’, no State aid can be involved.18 State aid being an objective concept, the Commission cannot be bound by an earlier decision in which it may have assessed a measure differently.19 The Commission has to assess each measure based on the specific features of the case as it finds it on the date on which it decides.20

12

On energy infrastructure, see 13 f below. Case C-452/10 P BNP Paribas v Commission, EU:C:2012:366, para 100; Case C-487/06 P British Aggregates v Commission, EU:C:2008:757, para 111; Case T-268/08 Land Burgenland and Austria v Commission, EU:T:2012:90, para 76; Joined Cases T-80/06 and T-182/09 Budapesti Erömü Zrt v Commission, EU:T:2012:65, para 83. 14 British Aggregates (n 13) paras 111–12; Commission Notice on the notion of State aid (n 1) para 4. 15 Case C-75/97 Belgium v Commission (Maribel bis/ter), EU:C:1999:311, para 25; Case C-172/03 Heiser, EU:C:2005:130, paras 44–46; Joined Cases T-346/99 to T-348/99 Diputación Foral de Álava et al v Commission, EU:T:2002:259, paras 54 and 63; Case T-251/11 Austria v Commission, EU:T:2014:1060, para 115. 16 Case C-169/08 Presidente del Consiglio dei Ministri v Regione autonoma della Sardegna, EU:C:2009:420, Opinion AG Kokott, para 140; Case T-109/01 Fleuren Compost v Commission, EU:T:2004:4, para 54. 17 British Aggregates (n 13) para 92. 18 Commission Notice on the notion of State aid (n 1) para 17. 19 Case C-459/10 P Freistaat Sachsen v Commission, EU:C:2011:515, para 50; Case C-106/09 P Commission and Spain v Government of Gibraltar and United Kingdom, EU:C:2011:732, para 136; Land Burgenland and Austria (n 13) paras 76–77. 20 Case C-334/07 P Commission v Freistaat Sachsen, EU:C:2008:709, para 50. 13

6

Falk Schöning and Clemens Ziegler

The notion of State aid is a fairly broad concept: it does not only apply to subsidies, ie, direct financial support, but to any form of aid. The European Court of Justice (ECJ) established this relatively early on by stating: A subsidy is normally defined as a payment in cash or in kind made in support of an undertaking other than the payment by the purchaser or consumer for the goods or services which it produces. An aid is a very similar concept, which, however, places emphasis on its purpose and seems especially devised for a particular objective which cannot normally be achieved without outside help. The concept of aid is nevertheless wider than that of a subsidy because it embraces not only positive benefits, such as subsidies themselves, but also interventions 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.21

This broad concept of State aid means that not only direct grants can be State aid. In fact, indirect financial measures, for example, waivers of public debts and exemptions from obligations to pay fines22 or the non-enforcement of obligations (eg, to pay taxes and other charges)23 account for a large portion of State aid measures. Examples include certain measures/schemes in the environmental area, for example, the free award of emissions allowances under emissions trading schemes24 or ‘green electricity’ purchase obligations.25 As regards the timing for the assessment if aid is present or not, the point in time of the measure under consideration is relevant, not subsequent changes: [W]here a Member State concludes with an economic operator an agreement which does not involve any State aid element for the purposes of Article 107 TFEU, the fact that, subsequently, conditions external to such an agreement change in such a way that the operator in question is in an advantageous position vis-à-vis other operators that have not concluded a similar agreement is not a sufficient basis on which to conclude that, together, the agreement and the subsequent modification of the conditions external to that agreement can be regarded as constituting State aid.26

Unlike the prohibition of customs duties and quantitative restrictions pursuant to Articles 30, 34 and 35 TFEU in the area of the free movement of goods, EU State aid law does not know any parallel concept of ‘measures of equivalent effect’ as State aid or a teleological extension of the notion of State aid by applying Article 4(3) TEU,27 21 Case C-30/59 Steenkolenmijnen v High Authority, EU:C:1961:2, 19, regarding Art 4(c) of the European Coal and Steel Community (ECSC) Treaty. 22 Case C-200/97 Ecotrade v Altiforni e Ferriere di Servola, EU:C:1998:579, para 43; Case C-295/97 Piaggio, EU:C:1999:313, para 42. 23 Case C-480/98 Spain v Commission (Magefesa), EU:C:2000:559, para 21; Case C-415/03 Commission v Greece, EU:C:2005:287, paras 37–38; Case T-36/99 Lenzing v Commission, EU:T:2004:312, paras 136–37. 24 Case C-279/08 P Commission v Netherlands, EU:C:2011:551, paras 75–76, 86–96 and 106–12. This is discussed further below, see 20 f. 25 Cases NN 162/A/2003 and N 317/A/2006 Austria Support of electricity production from renewable sources under the Austrian Green Electricity Act (feed-in tariffs) [2006] OJ C221/8, paras 48–59; Case 2007/580/EC Slovenian green electricity scheme [2007] OJ L219/9, paras 66–74; Austria v Commission (n 15) paras 83 and 88. 26 Case T-499/10 MOL v Commission, EU:C:2010:55, para 64. 27 Case C-379/98 PreussenElektra, EU:C:2001:160, paras 63–65, as well as Opinion AG Jacobs, EU:C:2000:585, paras 180–85.

What is State Aid? 7 stating the Member States’ obligation to ‘facilitate the achievement of the Union’s tasks and [to] refrain from any measure which could jeopardise the attainment of the Union’s objectives’. The notion of State aid in Article 107(1) TFEU applies to the entire energy sector, including nuclear energy.28

B. The Elements of the Notion of Aid in Article 107(1) TFEU According to the definition of State aid in Article 107(1) TFEU, the following elements are needed to classify a measure as State aid: (i) one or several undertakings; (ii) they need to receive a certain economic advantage; (iii) the advantage must be granted directly or indirectly through State resources and must be imputable to the State; (iv) the advantage must favour certain undertakings or the production of certain goods (the ‘selectivity’ criterion); and (v) the measure must distort or threaten to distort competition and affect trade between Member States. These elements will be discussed in the following sections. The Commission bears the burden of proof to show that a measure in question falls under the notion of State aid, ie, that all of its elements are present.29 i. ‘Undertaking’ as the Recipient A measure that does not have an ‘undertaking’ as beneficiary cannot be State aid. a. General Considerations According to the jurisprudence of the European courts, an ‘undertaking’ is an entity engaged in an economic activity, regardless of whether it is governed by private or public law and how it is financed.30 The concept of undertaking under EU State aid law is the same as under general EU competition law.31 The concept of undertaking being one of EU law, the status of an entity under national law (private or public) is irrelevant.32 b. Economic Activities According to the ECJ, any activity consisting in offering goods and/or services on a market is an economic activity.33 Also non-profit entities can be undertakings 28 Pursuant to Art 106(a)(3) of the Treaty Establishing the European Atomic Energy Community ([2016] OJ C203/1), the ‘provisions of the Treaty on European Union and of the Treaty on the Functioning of the European Union shall not derogate from the provisions of … [the Euratom Treaty]’. Therefore, the TFEU applies as long as the TFEU does not imply derogation from the Euratom Treaty. Commission Decision N 506/2010 on Partial financing of decommissioning of two already shut down nuclear plants (A1 and V1) [2013] OJ C162/3, recitals 65–66. 29 K Bacon, European Union Law of State Aid, 2nd edn, 2013, para 2.37. 30 Joined Cases C-180/98 to C-184/98 Pavlov et al, EU:C:2000:428, para 74. 31 Case C-237/04 Enirisorse v Sotocarbo, EU:C:2006:197, paras 28–29; Case C-288/11 P Mitteldeutsche Flughafen and Flughafen Leipzig-Halle v Commission, EU:C:2012:821, para 50. 32 Commission Notice on the notion of State aid (n 1) para 8. 33 Pavlov et al (n 30) para 75.

8

Falk Schöning and Clemens Ziegler

under EU State aid law, as they too can offer goods and services on a market.34 Selfemployed professionals engaged in offering goods and/or services on a market, for example, dentists,35 architects, lawyers, fall within the notion of an undertaking. It has been suggested that a basic test to assess if an activity is economic in nature or not would be to see whether it could, at least in principle, be carried on by a private undertaking in order to make profits. If there were no possibility of a private undertaking carrying on a given activity, there would be no purpose in applying the competition rules to it.36

However, since non-profit entities may also fall under the notion of undertaking,37 such activities should be tested by asking whether they could possibly be seen as competing with the activities of entities that carry out the same activities, but with the aim of making profits. Where an entity is engaged in both economic and non-economic activities, it has to be considered an undertaking only regarding its economic activities.38 On the other hand, where only the activities of several entities together can be considered an economic activity, those entities can be considered an undertaking only if taken together.39 The mere holding of a controlling stake in an undertaking, without being directly or indirectly involved in the management of that undertaking cannot be considered an economic activity.40 In turn, the activities of the controlled undertaking would have to be assessed separately. Recently discussed examples of economic activities in the energy sector include the activities of a private company established by a Member State government to pursue a public good mission of mitigating climate change through accelerating the move to a low-carbon economy41 and the engagement in investment projects that will lead to the production of bio-energy.42 Research activities by energy companies may also be considered economic activities.43 A foundation whose activities are limited

34

Case C-49/07 MOTOE, EU:C:2008:376, paras 27–28. Case C-172/03 Heiser (n 15) para 26. 36 Joined Cases C-264/01, C-306/01, C-354/01 and C-355/01 AOK-Bundesverband et al, EU:C:2003:304, Opinion AG Jacobs, para 27. 37 eg, the hiring of unemployed people, to reintegrate them into the employment market, to collect recyclable waste in order has been considered as economic and the entity in question an undertaking. Commission Decision of 16 September 1997 on State aid for Gemeinnützige Abfallverwertung GmbH (published under document no C (1997) 2903) [1998] OJ L159/58. 38 Commission Notice on the notion of State aid (n 1) para 10. 39 ibid, para 11. 40 Case C-222/04 Cassa di Risparmio di Firenze et al, EU:C:2006:8, paras 107–18. 41 Case SA.24895 of 31.7.2013 on Public investment in wind power development projects [2013] OJ C279/1, recitals 15, 94-95 (among other activities, the entity was providing expert advice to the industrial commercial sector; at least part of the entity’s services were also provided by private actors, as for-profit activities). 42 Commission Decision SA.35458 Aid for investments to produce bio-energy [2013] OJ C144/1, recital 40. 43 Only research projects by universities and certain research organisations may under certain conditions qualify as non-economic activities. Such projects need to concern the primary activities of the relevant organisation, such as (i) education for more and better skilled human resources; (ii) the conduct of independent research and development for more knowledge and better understanding, including collaborative research and development; and (iii) the dissemination of research results. Technology transfer 35

What is State Aid? 9 to managing the eternal liabilities stemming from coal mining does not engage in an economic activity.44 c. No State Functions/State Action (Including Discussion of Public Undertakings) An entity that the relevant Member State has entitled to exercise public powers does not, in principle, qualify as an undertaking.45 It may, however, have to be considered an undertaking with regard to possible economic activities that can be distinguished from its exercise of public powers.46 Should the entity engage in economic activities that cannot be separated from its exercise of public powers, that entity’s activities as a whole cannot be qualified as those of an undertaking.47 In turn, just because a Member State has entrusted a certain entity with some tasks in the sphere of public interest does not exempt that entity from the application of State aid law to all its activities.48 As long as it has at least one economic activity, the fact alone that an entity is wholly or partially State-owned does not exempt it from State aid rules.49 That said, the particular features of a transaction relating to a publicly owned beneficiary can affect the assessment of whether the transaction would or would not have taken place under normal market conditions.50 ii. Aid in the Form of an Economic Advantage a. Economic Advantage Under Article 107(1) TFEU, any economic benefit that an undertaking would not have obtained under normal market conditions, ie, without State intervention, qualifies as economic advantage in the meaning of State aid law.51 Therefore, the State measure in question must improve the economic or financial position of the (alleged) beneficiary undertaking.52 Improvement in this context means as compared with the ‘normal’ condition that should prevail for the undertaking in question,

activities (eg licensing, the creation of spin-offs, etc) would be considered non-economic provided they are internal in nature and all income is reinvested into the primary activities of the relevant research organisation. Commission Notice on the notion of State aid (n 1) paras 31–32, with reference to point 19 of the ‘Framework for State aid for research and development and innovation’ [2014] OJ C198/1. 44 Commission Decision SA.33766 Finanzierung der Beendigung des subventionierten Steinkohlenbergbaus bis Ende 2018 (Steinkohlebeihilfen für 2011) [2011] recital 86. 45 Case C-30/87 Bodson v Pompes funèbres des régions libérées, EU:C:1988:225, para 18. 46 Case C-138/11 Compass-Datenbank, EU:C:2012:449, para 38; Case C-113/07 P SELEX Sistemi Integrati v Commission, EU:C:2009:191, paras 72–79. 47 SELEX Sistemi (n 46) paras 72–80; Compass-Datenbank (n 46) para 38. 48 Case C-475/99 Ambulanz Glöckner, EU:C:2001:577, para 21; Enirisorse (n 31) para 34. 49 Joined Cases C-34/01 to C-38/01 Enirisorse v Ministero delle Finanze, EU:C:2003:640. 50 Joined Cases T-228/99 and T-233/99 Westdeutsche Landesbank v Commission, EU:T:2003:57, para 270. 51 Case C-39/94 SFEI et al, EU:C:1996:285, para 60; Case C-342/96 Spain v Commission, EU:C:1999:210, para 41. 52 Compensation for damages incurred as a result of State action does not confer an advantage to the undertaking receiving compensation. Joined Cases C-106-120/87 Astéris AE et al v Greece and European Economic Community, EU:C:1988, paras 23 and 24; Commission Decision SA.42536 of 23 May 2016, Closure of German lignite plants [2016] recitals 34 and 49–51.

10 Falk Schöning and Clemens Ziegler ie, compared with other undertakings in the same or a corresponding situation, not to the alleged beneficiary’s prior situation. For example, where the State has in the past granted an advantage only to one company on the market and later partially reduces that advantage without entirely discontinuing it, this (revised) measure still sets the alleged beneficiary apart from other undertakings in corresponding situations and thus is to be considered as an ‘improvement’ of the alleged beneficiary undertaking’s situation. For the assessment of the economic advantage, one does not need to consider the State’s underlying motivation regarding the measure—only that the measure’s effect is relevant.53 It therefore does not matter whether the undertaking would be able to avoid or refuse the economic advantage.54 As long as an undertaking does not receive more from the State than it could equally have received from any other commercial entity that acts to earn a reasonable profit, the alleged advantage does not qualify as State aid. This applies both where the State acts as an investor and as a creditor. That said, since the considerations underlying both tests are not the same, the two tests/principles will be discussed separately below.55 b. Market Economy Operator Principle (MEOP) General Considerations While Member States are in principle free to carry out economic transactions, in doing so, they might trigger the applicability of EU State aid rules. This does not apply where the State acts as if it were a private actor investing under normal commercial conditions.56 The Commission has to use the MEOP (including its more specific subcategories ‘market economy investor principle’ (MEIP) (also called ‘market economy investor test’ or ‘private investor test’, ‘private creditor test’ and ‘private vendor test’)57 in all cases where it might prima facie apply. This is the case if it appears that, under comparable conditions, a private entity of comparable size operating under normal market conditions could reasonably have been convinced to make the same investment58 (or, in turn, to grant a loan or to sell under the same terms). It is not for the Member State to request the application of the MEOP, but

53

Case C-173/73 Italy v Commission, EU:C:1974:71, para 13. Commission Decision 2004/339/EC on the measures implemented by Italy for RAI SpA [2004] OJ L119/1, recital 69; Case C-251/97 France v Commission, EU:C:1998:572, Opinion AG Fennelly, para 26. 55 The European Courts have also developed a ‘private vendor test’. Here, the Commission tests whether a sale by a public entity involves State aid, assessing if a private seller could have obtained the same or an even better price under normal market conditions. Under this test, non-economic considerations cannot be accepted as reasons for accepting a lower price. Commission Decision SA.34721 On the measures in favour of HoKaWe Eberswalde GmbH [2014] OJ L109/30, recitals 38–69 (delivery of wood from a State-owned forest to a company operating a wood-fired biomass power plant: see recitals 9–11 and 38–69). 56 SFEI et al (n 51) paras 60–61. 57 Commission Notice on the notion of State aid (n 1) paras 74–75. 58 Joined Cases T-129/95, T-2/96 and T-97/96 Neue Maxhütte Stahlwerke and Lech-Stahlwerke v Commission, EU:T:1999:7, paras 104–05; Westdeutsche Landesbank (n 50) paras 176–77 and 181. 54

What is State Aid? 11 for the Commission to request all required information from the Member State to assess whether the test applies.59 The Commission can only take into account information that was available and developments that were foreseeable at the time of the investment decision.60 Applying the MEIP involves a complex economic appraisal. The Commission therefore enjoys a wide discretion and judicial review is limited to verifying whether the Commission complied with the relevant rules governing procedure and the statement of reasons, whether there was any error of law, whether the facts … have been accurately stated and whether there has been any manifest error of assessment of those facts or any misuse of powers … [T]he Court is not entitled to substitute its own economic assessment for that of the author of the decision.61

Under the MEIP, the Commission only takes into account the benefits and obligations that the State enjoys/has as an economic actor, not those arising from its role as a public entity. The relevant comparator cannot be what might be rational behaviour by a public entity that aims at achieving certain public policy objectives, because a market economy investor would normally merely act in its own interests. Therefore, any considerations of social, regional, environmental or other sectorial public policies have to be disregarded.62 Also, it is irrelevant whether a private actor can rely on the same specific means the State employs to engage in the investment, for example, fiscal means.63 The General Court’s (GC) decisions in Budapesti Erömü Zrt v Commission64 and Dunamenti Erömü Zrt v Commission65 and the underlying Commission decision66 show how the MEIP can apply in the energy sector. They discuss long-term power purchase agreements (PPAs) between a Hungarian State-owned public undertaking and Hungarian electricity generators. The PPAs foresaw the purchase of a fixed quantity of electricity at a fixed price, to ensure security of supply. The Commission’s reference point was a market operator with the same obligations and opportunities as the Hungarian State-owned public company, faced with the same legal and economic conditions as those existing in Hungary during the period under examination. The Commission ‘compared the … PPAs with the main features of standard forward and spot contracts, “drawing rights” contracts, long-term contracts concluded by large end-consumers, and contracts concluded between generators and [Transmission System Operators; “TSOs”] for the provision of balancing services’.67 59

Case C-124/10 P Commission v EDF, EU:C:2012:318, paras 103–04. ibid, para 105. 61 Budapesti Erömü Zrt v Commission (n 13) para 65; Case T-179/09 Dunamenti Erömü Zrt v Commission, EU:T:2014:236, para 81; see also Case T-196/04 Ryanair v Commission, EU:T:2008:585, para 41. 62 Commission v EDF (n 59) paras 79–81; Case T-20/03 Kahla Thüringen Porzellan v Commission, EU:T:2008:395, para 242; Case T-98/00 Linde v Commission, EU:T:2002:248, paras 46–49. 63 Commission v EDF (n 59) para 88. 64 Budapesti Erömü Zrt v Commission (n 13). 65 Dunamenti Erömü Zrt v Commission (n 61). The appeal against this decision was unsuccessful. Case C-357/14 P Electrabel SA, Dunamenti Erömü Zrt v Commission EU:C:2015:642. 66 Commission Decision SA.17365 on the State aid C 41/05 awarded by Hungary through Power Purchase Agreements [2009] OJ L225/53. 67 Budapesti Erömü Zrt v Commission (n 13) para 70; Dunamenti Erömü Zrt v Commission (n 61) para 206. 60

12 Falk Schöning and Clemens Ziegler The purpose was to see if the PPAs ‘provided the generators concerned with guarantees that a buyer would not accept if it were acting on purely commercial grounds’.68 According to the GC, ‘the Commission correctly … [concluded] that, structurally, PPAs provide power generators with a better guarantee than that provided under standard commercial contracts’.69 The MEIP applies also in the area of State aid and taxes. In 1997, France waived some corporation tax in the course of restructuring a State-owned energy company. The GC disagreed with the Commission’s finding that this was a capital injection in the same amount and found the MEOP to be inapplicable.70 The ECJ insisted that the MEOP applies even where advantages are provided by fiscal measures, provided that although the means of granting the advantage were instruments of State power, the Member State conferred that advantage as a shareholder of an undertaking under its control.71 The Commission’s assessment has to consider the advantage to the company in which the investment is made. It is irrelevant whether the State could have made a more profitable investment, whether through a different measure regarding the same entity or by investing somewhere else.72 The assessment has to be made for the relevant time period during which the measure is actually implemented—foreseeable developments are to be taken into account, but not unforeseeable developments after and due to which the investment no longer fulfils the conditions of the MEIP.73 The average private investor wants to achieve the highest possible return on investment.74 Nevertheless, the behaviour of the State does not have to be compared with the behaviour of a private investor seeking to realise a profit in the short term. The State as shareholder should be compared to a private holding company or private group of undertakings pursuing a structural policy, seeking profitability in the longer term.75 Also, a State-owned parent or holding company may, for a certain period, absorb losses to enable a subsidiary to close under the most favourable conditions possible, just as a private investor might be willing to inject additional capital to secure the survival of an undertaking to earn profits in the future.76 An

68 Budapesti Erömü Zrt v Commission (n 13) para 68; see also Dunamenti Erömü Zrt v Commission (n 61) para 85. 69 Budapesti Erömü Zrt v Commission (n 13) para 79; see also Dunamenti Erömü Zrt v Commission (n 61) para 92. 70 Case T-156/04 EDF v Commission, EU:T:2009:505, paras 221–84. 71 ibid, paras 91–92. 72 Case T-163/05 DEP-Bundesverband deutscher Banken v Commission, EU:T:2010:59, paras 58, 175; Case T-123/09 Ryanair v Commisson, EU:T:2012:164, paras 119–21. 73 Commission v EDF (n 59) paras 83–85, 105. The Commission has in the meantime ordered France to recover the unlawful State aid from EDF. Commission Decision (EU) 2016/154 of 22 July 2015 on State aid SA.13869 (ex NN 80/2002)—reclassification as capital of the tax-exempt accounting provisions for the renewal of the high-voltage transmission network (RAG) implemented by France in favour of EDF [2016] OJ L34/152. 74 Westdeutsche Landesbank (n 50) para 314. 75 Case C-256/97 DM Transport, EU:C:1999:332, para 24; Case T-296/97 Alitalia v Commission (Alitalia I), EU:T:2000:289, paras 84 and 96; Case T-565/08 Corsica Ferries France v Commission, EU:T:2012:415, para 80. 76 Case C-303/88 Italy v Commission (ENI-Lanerossi), EU:C:1991:136, para 21; Joined Cases T-126/96 and T-127/96 BFM and EFIM v Commission, EU:T:1998:207, para 80.

What is State Aid? 13 injection of public funds, however, requires at least a certain prospect of long-term profitability. Otherwise, it is State aid.77 In general, if a public capital contribution was made at the same time as a significant private capital contribution in comparable circumstances (the ‘concomitance’ test), it complies with the MEIP.78 To assess this, the Commission asks the following questions: (i) Is the private investor in question a market investor and do its investments have real economic significance? (ii) Do both parties concerned invest at the same time (‘concomitance’)? (iii) Are the terms and conditions of the investment identical for all shareholders? (iv) Does the private investor in question have other relationships with the State outside the relevant investment which would need to be taken into account while assessing if the equivalence in the mere investment terms suffices.79

Also without a concomitant private investment, the MEIP may still be met, provided that … the market structure and future prospects of the beneficiary … [allowed] a normal return acceptable to a hypothetical private investor … [to be] expected in a reasonable period of time, as compared, if possible, to similar … [private investments].80

In such circumstances, the Commission takes into account, inter alia, the following additional elements: (i) market conditions at the time the decision was made;81 (ii) existence of a reliable (ie, objective and possibly independent) ex-ante evaluation of the project; and (iii) a realistic business plan without any erroneous assumptions.82 Investments in Energy Infrastructure Traditionally, the public funding of infrastructure was considered to fall outside the scope of the State aid rules:83 in 1995, ie, still before the adoption of the first liberalisation directives for electricity and gas,84 the Commission wrote: Governments have always used financial intervention as an essential tool in their policy of infrastructure development … [A]s long as access and usage remain public and general, such intervention will not constitute aid within the meaning of Article 92(1) [(now Article 107(1) TFEU)] but will be normally regarded as being in the public interest.85

77

ENI-Lanerossi (n 76) para 22; Corsica Ferries France v Commission (n 75) para 84. Commission Decision SA.24895 of 31 July 2013, Public investment in wind power development projects [2013] OJ C279/1, recital 112. 79 ibid para 116. 80 ibid, para 113. 81 Case C-482/99 France v Commission (Stardust Marine), EU:C:2002:294, para 71; Public investment in wind power development projects (n 78) para 113. 82 Public investment in wind power development projects (n 78) para 113. 83 Commission Notice on the notion of State aid (n 1) para 201. 84 European Parliament and Council Directive 96/92/EC of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L27/20; and 98/30/EC of 22 June 1998 concerning common rules for the internal market in natural gas [1998] OJ L204/1. 85 European Commission, XXVth Report on Competition Policy 1995, available at: ec.europa.eu/ competition/publications/annual_report/1995/en.pdf, point 175. 78

14 Falk Schöning and Clemens Ziegler Energy infrastructure as defined in the Commission’s ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG)86 allows energy supply against payment and thereby an economic activity. In principle, the future use of an infrastructure determines whether its construction is an economic activity and accordingly whether its public funding constitutes State aid or not.87 Different from the historic situation considered above, energy infrastructure is nowadays, in most cases, built by commercial actors and usually financed through tariffs ultimately levied from the final consumers. Public funds involved in energy infrastructure projects favour an economic activity, are likely to affect trade between Member States and therefore, in general, are subject to State aid rules.88 c. Private Creditor Test and State Guarantees A Member State is a creditor where it intervenes on the debt of a company, by granting a credit, a guarantee or where it waives or reschedules the repayment of debts. There can only be an advantage if the State behaves differently from a private creditor. Under a State guarantee, the State assumes the risk to be taken up for payment. According to the Commission’s Guarantee Notice,89 a State guarantee is not aid under the following conditions: (i) the borrower is not in financial difficulty; (ii) its extent can be properly measured when it is granted; (iii) it does not cover more than 80 per cent of the outstanding loan or other financial obligation; and (iv) a market-oriented price is paid for it.90 Where a market-based remuneration cannot be determined, the aid would consist in the difference between the market interest rate the recipient would have borne without the guarantee and the interest rate obtained by means of the State guarantee after any premium paid has been taken into account,91 or, in other words, in the difference between the guarantee premium demanded by the State and the premium requested for an equivalent guarantee by a commercial bank.92 The only exception applies when, due to the dire financial situation of the beneficiary, no financial institution would be ready to issue a loan to it without a State guarantee. In such cases, the Commission has considered the entire guaranteed loan amount as a State aid.93 However, the assumption that the complete amount (eg, loan) that would not have been given by a private creditor is State aid warrants criticism, because also under bankruptcy proceedings, at least a part of the obligations can regularly be met (and there have been rescue operations with even 100 per cent recovery). Therefore, from an economic point of view, the aid amount 86 Commission, ‘Guidelines on State aid for environmental protection and energy 2014–2020’ [2014] OJ C200/1, recital 31. 87 Joined Cases T-443/08 and T-455/08 Mitteldeutsche Flughafen and Flughafen Leipzig/Halle v Commission, EU:T:2011:117, paras 92–100. 88 Commission Notice on the notion of State aid (n 1) para 217. 89 Commission Notice on the application of Articles 87 and 88 of the EC Treaty to State aid in the form of guarantees [2008] OJ C155/10 (Guarantee Notice). 90 Guarantee Notice (n 89) point 3.2. 91 ibid, point 4.2. 92 Commission Decision of 26 January 2011 on State aid Case N 630/2009 Project T3 Est Parisien ‘Réalisation d’une canalisation structurante et développement du réseau (Compagnie Parisienne de Chauffage Urbain)’ [2013] OJ C1/1, recital 40. 93 ibid, citing to C-288/96 Germany v Commission, EU:C:2000:537, paras 30–31.

What is State Aid? 15 should be somewhere between 50 per cent and 100 per cent of such a State loan/ guarantee. This discussion impacts the aid amount significantly, because instead of being limited to the low amount of the guarantee premium (usually a low, one-digit percentage figure), the aid in question would correspond to a much higher proportion of the loan/guarantee amount. Overall, the context of a measure will also have to be taken into account: the State’s motives to give a particular loan or guarantee may have a bearing on the answer to this question as well. In a recent decision regarding a Lithuanian LNG terminal, the Commission discusses how to analyse whether a guarantee price is market conform or not. If it is not, the aid element therein is established using a benchmarking method to determine the market-conform guarantee remuneration with Credit Default Swap (CDS) spreads or bond spreads, or with cost-based approaches such as Risk Adjusted Return on Capital (RAROC).94 Both the benchmarking method on the basis of CDS spreads or bond spreads and the costing approach may … [lead to an appropriate] estimate for a market oriented price in the case of guarantors with a very strong credit standing. [They may, however, overstate] … the value of the guarantee … when the state guarantor has a significantly lower credit standing. In those cases, an indirect approach based on a comparison of interest rates with and without state guarantee may be envisaged (rate differential approach) … [T]he market conform guarantee premium for a state guarantee is defined as the difference in financing cost with and without state guarantee.95

A State guarantee that is limited to ensuring ‘payment of liabilities of an undertaking only after the undertaking is bankrupt and wound up does not provide a benefit to the undertaking in question’.96 In the case in question, the Commission saw the State’s guarantee for a company operating coal mines as a State intervention ‘on grounds of public security, as a matter of public order … to ensure the continued financing of eternal liabilities in the public interest, notably to prevent environmental damage’.97 In principle, a State guarantee should not cover more than 80 per cent of the outstanding financial obligation.98 This is to increase the lender’s incentive to ‘properly assess, secure and minimise the risk arising from the lending operation, and in particular to properly assess the borrower’s creditworthiness … [and to contain] the amount of higher-risk guarantees in the State’s portfolio’.99 However, more than 80 per cent of the loan can be covered under the following conditions: (1) the beneficiary is entrusted with … [a service of general economic interest (SGEI)], (2) the guarantee is provided by the same public authority that entrusted the SGEI and (3) the

94 Commission Decision SA.36740 of 20 November 2013, Klaipėdos nafta LNG terminal [2016] OJ C161/1, recitals 101-129. Another recent energy-related example for an assessment of a State guarantee is contained in Commission Decision SA.36323 of 31 July 2014, Liquidity support in the energy sector: PPC [2014] OJ C348/1, recitals 60–65 (here both publicly funded loans and State guarantees). 95 Klaipėdos nafta LNG terminal (n 94) recital 103. 96 Commission Decision SA.33766 of 7 December 2011, Finanzierung der Beendigung des subventionierten Steinkohlenbergbaus bis Ende 2018 (Steinkohlebeihilfen für 2011) (Coal mine closure plan 2008–2018, here coal mine aid for 2011) [2012] OJ C284/6, recital 85. 97 ibid, recital 87. 98 Guarantee Notice (n 89) s 3.2(c). 99 ibid.

16 Falk Schöning and Clemens Ziegler beneficiary has no other economic activities. The Guarantee Notice explicitly requests a Member State who wishes to provide a guarantee above the 80% threshold … to duly substantiate … [why this does not constitute aid] and notify it to the Commission so that the guarantee can be properly assessed.100

The SGEI must comply with the Commission’s SGEI package, consisting of decision,101 communication102 and framework.103 d. Sale or Supply of Assets or Services Where the State sells or otherwise provides public assets under normal market conditions, for example, not supply or sale below value or otherwise under preferential terms, State aid rules do not apply.104 For utilities (eg, gas- or electricity-related), under certain conditions, terms that a priori look preferential might be necessary to meet competition on the market, so that a private operator might also have supplied gas under such terms.105 Also, there is no State aid where the difference between a ‘market level’ and an apparently preferential tariff is justified due to other commercial reasons, for example, overcapacity.106 State aid rules apply to State-owned undertakings providing services, if they are not compensated for them where under normal market conditions a supplier would107 or where they charge lower prices than would be charged under normal market conditions or offer otherwise preferential terms.108 e. Purchase of Goods and Services (Including Activities under Public Service Obligations) by the State The State can purchase goods and services from commercial operators without triggering the applicability of State aid law. However, such purchases have to be based on objective and transparent criteria. No State aid rules apply if, first, the transaction is based on normal commercial terms and, second, the State genuinely needs the goods or services (thus does not confer an advantage on the undertaking selling the goods or services). This applies in particular to payments for the conduct of public

100 Commission Decision SA.31261 of 10 May 2011, Municipal Guarantee for Loan for Geothermal Heat Distribution Network [2011] OJ C262/1, recital 25. In this case, the 80% limitation did not apply; ibid, recital 36. 101 Commission Decision of 20 December 2011 on the application of Art 106(2) of the Treaty on the Functioning of the European Union to State aid in the form of public service compensation granted to certain undertakings entrusted with the operation of services of general economic interest [2012] OJ L7/3. 102 Commission, ‘On the application of the European Union State aid rules to compensation granted for the provision of services of general economic interest’ (Communication) [2012] OJ C8/4. 103 Commission, ‘European Union framework for State aid in the form of public service compensation’ (Communication) [2012] OJ C8/15. 104 Cases C-341/06 P Chronopost v Ufex, EU:C:2008:375, para 123; T-198/01 Technische Glaswerke Ilmenau v Commission, EU:T:2004:222, paras 98–99; Case C-404/04 P Technische Glaswerke Ilmenau v Commission, EU:C:2007:6, para 95. 105 Joined Cases C-67/85, C-68/85 and C-70/85 Van der Kooy v Commission, EU:C:1988:38, para 30; Case C-56/93 Belgium v Commission, EU:C:1996:64, para 10. 106 Commission Decision of 11 April 2000 on the measure implemented by EDF for certain firms in the paper industry [2001] OJ L95/18, recitals 68–75. 107 Case C-126/01 GEMO SA, EU:C:2003:622, paras 18–19 and 30–33. 108 Chronopost (n 104) para 123.

What is State Aid? 17 service obligations. Since this area will be discussed in more detail in chapter nine below, its discussion here is limited to the following general remarks. The ECJ’s landmark Altmark Trans ruling shows how State aid can be excluded in the area of public service obligations.109 For a State measure to be limited to compensation for services under a public service obligation: (i) there needs to be a clearly defined public service obligation, imposed over and above the duties of an undertaking due to sectorial regulations or collective agreements;110 (ii) the way to calculate the compensation must be established in advance in an objective and transparent manner;111 (iii) overcompensation is excluded;112 and (iv) the undertaking providing the service under the public service obligation must either be chosen in a public procurement procedure that allows for the selection of the bidder capable of providing those services at the least cost to the community, or the compensation must correspond to the costs which a typical and well-run undertaking would have incurred in performing the service under the obligation, based on actual costs and a reasonable profit.113 iii. Granted Directly or Indirectly through State Resources/Imputable to the State a. PreussenElektra: Only Advantages Granted through State Resources The element of ‘granted by a Member State or through State resources’ has, after decades of debate, been clarified by an ECJ decision in the energy sector, the landmark ruling of PreussenElektra.114 A State aid measure must be both financed directly or indirectly through public resources and be imputable to the State. Those conditions are cumulative, not alternative.115 According to the German Stromeinspeisungsgesetz, private electricity distributers had to purchase certain amounts of electricity from renewable energy sources at State-fixed minimum prices. Private conventional electricity suppliers had to partially pay the difference between those prices and the market electricity prices. According to the ECJ, the purchase obligation, while constituting an advantage, did not involve a direct or indirect transfer of State resources, ie, no State aid was involved. The granting of an advantage by a Member State was not sufficient. The advantage had also to be financed directly or indirectly through public resources.116 The ECJ held that the purchase obligation imposed by the Stromeinspeisungsgesetz did not involve any direct or indirect transfer of State resources to undertakings producing energy from renewable sources.117 Therefore, the Court was of the opinion that ‘the allocation of the financial burden arising from that obligation for those

109 110 111 112 113 114 115 116 117

Case C-280/00 Altmark Trans and Regierungspräsidium Magdeburg, EU:C:2003:415, para 87. ibid, para 89. ibid, para 90. ibid, para 92. ibid, para 93; Commission Decision Project T3 Est Parisien (n 92) recital 37. PreussenElektra (n 27). GEMO (n 107) para 24. PreussenElektra (n 27) para 58. ibid, para 59.

18 Falk Schöning and Clemens Ziegler private electricity supply undertakings as between them and other private undertakings cannot constitute a direct or indirect transfer of State resources either’.118 b. Directly or Indirectly Granted through State Resources ‘State’ has to be understood in a broad sense, including not only central, federal or federal State authorities, but also regional and local bodies, regardless of their legal status and name.119 Any public body or institution thus qualifies as a possible grantor of State aid.120 State resources include all resources under State control and thus available to the public authorities,121 such as government funds and assets,122 assets in the hands of a public or even private undertaking or a public– private partnership,123 funds derived from parafiscal charges or other compulsory contributions124 and funds of voluntary private deposits, even if they can be withdrawn by the private savers.125 In other words, the private nature of resources alone does not prevent their qualification as State resources.126 In cases involving compulsory payments by third parties, for example, regarding support for renewable energy, it has never been easy to draw the line between mechanisms involving and not involving State resources. No State aid was involved in the PreussenElektra case, as well as in the successor scheme in Germany to the Stromeinspeisungsgesetz, as well as several Belgian schemes imposing a purchase obligation for ‘green’ electricity on electricity supply companies.127 However, a later German successor scheme, the law on renewable energy of 2012 (EEG 2012),128 was found to involve State aid. It required network operators (mostly distribution system operators (DSOs) in the first instance and transmission systems operators (TSOs) in the second) to purchase electricity from renewables at a statutory price exceeding market levels.129 The TSOs sold the electricity produced by renewables on the spot 118

ibid, para 60. Case C-284/84 Germany v Commission, EU:C:1987:437, para 17; Case C-88/03 Portugal v Commission (Azores), EU:C:2006:511, para 55. 120 Case T-358/94 Air France v Commission, EU:T:1996:194, para 56. 121 Case C-173/73 Italy v Commission (n 53) para 16; Case C-399/00 SIM 2 Multimedia v Commission, EU:C:2003:252, para 37. 122 Westdeutsche Landesbank (n 50) paras 181–82. 123 Commission Decision of 23 January 2013 on State aid SA.24123 (2012/C) implemented by the Netherlands Alleged sale of land below market price by the Municipality of Leidschendam-Voorburg (notified under document C(2013) 87) [2013] OJ L148/52, recital 65. 124 Case C-173/73 Italy v Commission (n 53) para 16; Slovenian green electricity scheme (n 25) paras 73–74; Case C-677/11 Doux Élevage and Coopérative agricole UKL-ARREE, EU:C:2013:348, para 26; more on this below, in ch 3: ‘State aid and Taxation: Special Focus on Energy and the Environment’. 125 Air France v Commission (n 120) paras 63–68. 126 ibid; Case T-139/09 France v Commission, EU:T:2012:496, paras 63–64. 127 PreussenElektra (n 27) paras 56–66; Commission Decision of 22 May 2002 NN 27/2000 Law on promotion of electricity generation from renewable energies [2002] OJ C164/3, 3–4 of version available at: ec.europa.eu/competition/state_aid/cases/138917/138917_410173_26_2.pdf; Commission Decision of 25 July 2001 N 550/2000 Green electricity certificates [2001] OJ C330/3, 6–10 of version available at: ec.europa.eu/competition/state_aid/cases/134040/134040_1153314_21_2.pdf; Commission Decision of 24 October 2006 N 254/2006 Belgique Panneaux Photovoltaiques C (2006) 4954 final, recitals 9–15. 128 Gesetz zur Neuregelung des Rechtsrahmens für die Förderung der Stromerzeugung aus Erneuerbaren Energien (Law revising the legal framework for the promotion of electricity production from renewable energy) of 28 July 2011 BGBl. 2011 I (Federal Official Gazette 2011, part I) 1634. 129 Case T-47/15 Germany v Commission, EU:T:2016:281, paras 4–5. 119

What is State Aid? 19 market of the electricity exchange to either recover their higher costs automatically where spot prices were sufficiently high or by passing on a so-called ‘EEG Surcharge’ to customers.130 However, some undertakings, for example, such with electricityintensive manufacturing activities (EIUs), obtained a cap on that EEG Surcharge to protect their international competitiveness. Germany applied for annulment of a Commission decision that qualified the entire scheme as State aid, including the cap on the EEG Surcharge for EIUs. The GC found that (i) the funds generated by the EEG Surcharge remained under the dominant influence of the public authorities (the TSOs administering the funds collectively could be assimilated to an entity executing a State concession, and could not administer them freely, on their own behalf); and (ii) the resources in question were generated by charges that the EEG 2012 ultimately imposed on private persons.131 The GC rejected Germany’s arguments based on the ECJ’s PreussenElektra decision,132 because the mechanism in PreussenElektra neither required the additional costs to be passed on to the final consumers nor the intervention of intermediaries such as the TSOs in this case in charge of administering the relevant funds collectively.133 The advantage in PreussenElektra was limited to guaranteeing that beneficiary undertakings were able to resell all renewable energy at a price exceeding the market price; a scheme financing that price supplement was entirely missing.134 Further, the private undertakings in PreussenElektra were not appointed by law to manage a State resource, but only had to purchase by means of their own financial resources.135 The GC also confirmed that the reduction in the EEG surcharge for EIUs qualified as an advantage under EU State aid law, because it released them from a charge that they normally would have had to bear (ie, normal operating aid).136 Germany has brought an appeal against the judgment of the GC.137 If one looks at the two RES-schemes that were the object of the discussion in PreussenElektra and the most recent EEG 2012-related decisions, doubts may be warranted as to whether both schemes in fact were that different. Ultimately, in both cases, the idea was to reimburse producers of electricity from renewable sources for

130

ibid, para 7. ibid, paras 92–96. 132 See 17 f above. 133 Case T-47/15 Germany v Commission (n 129) para 99. See also the similar reasoning in Commission Decision SA.43735 of 24 October 2016, Germany ABLAV Interruptibility Scheme, recitals 38 ff. 134 Case T-47/15 Germany v Commission (n 129) para 100. 135 ibid, para 101. See also Commission Decision SA.36511 of 27 March 2014 Mécanisme de soutien aux énergies renouvelables et plafonnement de la CSPE, recital 81. 136 Case T-47/15 Germany v Commission (n 129) para 126; see also European Commission, Decision SA.42536 Closure of German lignite plants, recitals 28–32. 137 Appeal brought on 19 July 2016 by the Federal Republic of Germany against the judgment of the General Court (Third Chamber) of 10 May 2016 in Case T-47/15 Federal Republic of Germany v Commission (Case C-405/16 P, [2016] OJ C326/18): according to the German government, its authorities do not exercise control over the financial means of TSOs and energy supply undertakings. In the German government’s view, while the German Renewable Energy Law creates private law contractual relations between single undertakings, it establishes no State control over their financial means. The German government further does not deem the Renewable Energy Law to provide an advantage amounting to aid to energy-intensive undertakings as final customers, contrary to the case law relating to structural disadvantages and selectivity of aid. 131

20 Falk Schöning and Clemens Ziegler their additional costs compared with conventional electricity production. In both cases, it was done with a system that ultimately led to the passing on of the additional costs to final electricity customers. There may be a tendency that the Commission tries to increase its influence on State measures that may have an impact on the internal market by interpreting the notion of State aid more broadly. Since the corrective to a slightly broader understanding of the notion of State aid may be a slightly less strict compatibility assessment under Article 107(3) TFEU, this development may not necessarily result in stricter State aid enforcement or fewer measures of State support being approved. Where the State finances support measures by imposing parafiscal charges or contributions that are managed and apportioned according to legislation, State aid rules may apply, even if the resources are managed by entities separate from the public authorities.138 A mechanism that fully compensates undertakings selling wind-generated electricity at a price above the market price, financed by the final consumers of electricity throughout that Member State qualifies as being ‘granted through State resources’.139 Conversely, in PreussenElektra, the funds were not under public control and the private operators under the purchase obligation did not benefit from a mechanism that set off any additional costs arising from the obligation, to ensure that they would be fully compensated for the additional costs,140 nor were they appointed by law to manage a State resource.141 In contrast to Member States’ fundings, EU resources are not State resources for the purposes of Article 107(1) TFEU; however they are, once EU resources are controlled by a Member State (eg, European Structural Fund finance).142 The funds must be awarded ‘directly’ or ‘indirectly’ through State resources.143 This is easy to establish for a subsidy or other transfer of State resources, such as capital investments, loans and payments for goods and services. At the same time, no actual transfer is required,144 ie, the State may also just renounce on revenue that it would otherwise have received.145 An example of this would be the ECJ’s judgment in the Dutch NOx trading scheme, which according to the ECJ involved State resources, because the Netherlands allocated emissions allowances free of charge instead of selling them in an auction.146

138 Case C-262/12 Vent de Colère et al, EU:C:2013:851, para 25. The entity administering the purchase obligation in this case was a public law corporation; ibid, para 29. See also Austria v Commission (n 15) para 82. 139 Case C-262/12 Vent de Colère et al (n 138) para 37. 140 ibid, para 36. 141 ibid, para 35. See also above, before n 134. In addition, this will be discussed further in chs 2 and 3 of this volume. 142 Commission Decisions SA.34056 of 27 June 2012 Cable Car for London [2012] OJ C220/6, recital 40; SA.34938 of 16 April 2013 Aid to increase the capacity of PGNiG’s gas storage facility in Husow [2013] OJ C173/4, recital 25; SA.37068 of 18 December 2013 Underground electricity transmission line between Aghios Stefanos and Nea Chalkedona [2014] OJ C460/5, recital 17; and SA.39050 of 17 July 2015 Aid to gas infrastructure in Poland [2015] OJ C352/4, recital 34. 143 Cases C-72/91 and C-73/91 Sloman Neptun, EU:C:1993:97, para 19; PreussenElektra (n 27) para 58. 144 Austria v Commission (n 15) para 55. 145 Case C-172/03 Heiser, EU:C:2004:678, Opinion AG Tizzano, para 52. 146 Commission v Netherlands (n 24) paras 106–12.

What is State Aid? 21 The ECJ also found the scheme to constitute an additional burden for the public authorities, because the undertakings in question could avoid the payment of fines by instead buying the emissions allowances.147 The ECJ finally also distinguished the NOx case from PreussenElektra with regard to the State’s loss of tax revenue as an ‘inherent feature’ of the scheme at issue:148 in the NOx case, the foregoing of resources … [is not] ‘inherent’ in any instrument designed to regulate emissions of atmospheric pollutants … Furthermore … there is a sufficiently direct connection between the measure in question and the loss of revenue, a link which did not exist between the imposition of the obligation to purchase and the possible diminution in tax receipts at issue in the case which led to the judgment in PreussenElektra.149

This reasoning has also been applied in a number of other cases in the electricity sector. For instance, the allocation, by eligible Member States, of optional transitional free allowances in the context of implementing the ETS Directive150 may involve State aid within the meaning of Article 107(1), ‘because Member States forego revenues by granting free allowances and give a selective advantage to power generators … State aid is also involved at the level of investments that recipients of free allowances will undertake at a reduced cost’.151 Another example is the Commission’s Decision opening a formal investigation procedure into the French electricity capacity mechanism. Here, the French authorities issued capacity certificates to electricity producers for free. At the same time, they created a market for these certificates by imposing a quota obligation on the electricity suppliers. The quotas were linked with the demand peeks of their customers. They thereby created a demand for the certificates and attributed a corresponding value to them. Furthermore, instead of selling the certificates to producers or to auction them, the State issued them for free, and thereby forwent public resources.152 A number of other Commission decisions discuss such cases.153 The advantage to the beneficiary of the measure does not have to be equivalent to the effect on State resources.154

147

ibid, para 106. See last two sentences of quote from PreussenElektra before n 118 above. 149 Commission v Netherlands (n 24) para 112. 150 Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC [2003] OJ L275/32. 151 ‘Guidelines on certain State aid measures in the context of the greenhouse emission allowance trading scheme post-2012’ [2012] OJ C158/4, s 1.3 (para 17). 152 Commission Decision SA.39621 of 13 November 2015 French country-wide capacity mechanism [2016] OJ C46/35, recital 115. 153 Commission Decisions SA.34457 of 10 July 2014 National Investment plan for transitional free allocation of emission allowances under Article 10C(5) of Directive 2003/87/EC [2016] OJ C241/1, recital 21; SA.34674 of 22 January 2014 Free allowances to power generators under Article 10c of the ETS Directive [2015] OJ C24/1, recital 21; SA.34086 of 19 December 2012 Investments aiming at the modernisation of the Hungarian energy sector under Article 10(c) EU ETS Directive [2013] OJ C43/13, recital 28; and SA.37345 of 2 August 2016 Polish certificates of origin system to support renewables and reduction of burdens arising from the renewables certificate obligation for energy-intensive users, recitals 140 ff. 154 Case C-399/10 P Bouygues and Bouygues Télécom v Commisison et al, EU:C:2013:175, paras 109–10. 148

22 Falk Schöning and Clemens Ziegler Article 107(1) TFEU includes all financial means by which authorities may actually support undertakings, be they permanent State assets or not.155 State guarantees are grants through State resources: although the State may never have to actually pay anything under the guarantee, it represents a contingent liability and would turn into an actual burden for the State should it be called upon.156 Tax exemptions and other waivers of tax revenue will be discussed in chapter three below. c. Imputable to the State Only advantages that are imputable to the State qualify as State aid.157 Acts by private entities are not State aid even if they may have the same or similar economic effects as aid measures imputable to the State. Often, the nature of a measure allows deducing State imputability. An advantage resulting from a (non-EU) legislative measure will usually be imputable to the State.158 In general, it is necessary to examine whether the public authorities have been involved in the adoption of the measure in question.159 Therefore, State imputability cannot be assumed only because an advantage is granted through the resources of a public or publicly owned undertaking.160 At least day-to-day business decisions of a publicly owned undertaking in which the public authorities do not interfere should not trigger the applicability of the State aid rules.161 To determine whether an action by a public undertaking is imputable to the State, the circumstances and context of a decision have to be considered,162 for example, (i) how an undertaking was established;163 (ii) its legal status (public or corporate law?);164 (iii) if it is integrated into the structures of the public administration;165 (iv) to what extent it is supervised and managed by the public authorities;166 (v) the type of its activities on the market and

155 Austria v Commission (n 15) para 56; Case C-482/99 France v Commission (Stardust Marine), EU:C:2002:294, para 37; Case C-83/98 P France v Ladbroke Racing and Commission, EU:C:2000:248, para 50. 156 Case T-384/08 Elliniki Nafpigokataskevastiki et al v Commission, EU:T:2011:650, paras 89 and 92. See also Bouygues (n 154) paras 106–09. 157 Air France v Commission (n 120) para 55; Case C-482/99 Stardust Marine (n 81), para 24; Joined Cases C-182/03 and C-217/03 Forum 187 v Commission, EU:C:2006:416, para 127. 158 Air France v Commission (n 120), para 59. 159 Vent de Colère et al (n 138) para 17 and case cited; Austria v Commission (n 15) para 86. 160 Case C-482/99 Stardust Marine (n 81) para 52; Commission Decision of 5 June 2013 SA.32184 Alleged State aid to an electricity supplier [2013] OJ C230/23, recital 118. 161 Case C-482/99 France v Commission (Stardust Marine), EU:C:2001:685, Opinion AG Jacobs, para 55. 162 Commission Decision SA.32184 (n 160) in recital 121, offers a non-exhaustive list of indicators that the Commission would take into account when assessing whether the actions of a TSO as public company are imputable to the State. See also Case C-345/02 Pearle et al, EU:C:2004:448, paras 36–37. 163 Commission Decision of 9 November 2005 on the State aid which the Federal Republic of Germany has implemented for the introduction of terrestrial television (DVB-T) in Berlin-Brandenburg [2006] OJ L200/14, recital 53. 164 Case C-482/99 Stardust Marine (n 81) para 56. 165 ibid; Case T-468/08 Tisza Erömü kft v Commission, EU:T:2014:235, para 170. Formal independence or autonomy is insufficient, as the State aid rules cannot be circumvented by the creation of autonomous institutions that allocate aid; Air France v Commission (n 120) para 62. 166 Van der Kooy v Commission (n 105) para 36; Tisza Erömü kft v Commission (n 165) ibid.

What is State Aid? 23 whether those compete with the activities of other private operators;167 (vi) whether the decision in question had to be approved by, or was subject to requirements set by the public authorities;168 and (vii) how the public authorities presented the measure.169 Imputability does not require the public authorities to incite the public undertaking to engage in the relevant aid measures.170 However, the State as originator of a measure indicates imputability to the State even if it is possibly carried out by a private undertaking.171 On the other hand, measures originating from and being organised by private bodies are prima facie not imputable to the State, hence not aid. For example, in the Commission’s British Energy decision, British Nuclear Fuel Limited (BNFL) renegotiated its supply contracts with British Energy (BE), but there was no evidence that the renegotiation had been instigated by the State, so (the lower rates BNFL got from BE) were not considered to be State aid.172 The transposition of EU legislation into a Member State’s national law providing advantages to an undertaking would be considered to be imputable to the EU rather than the State. Therefore, Article 107(1) TFEU would not apply, provided that the Member State does not introduce additional direct or indirect advantages in excess of those granted through EU legislation. iv. Selective (Favouring Certain Undertakings or the Production of Certain Goods) The criterion of selectivity often plays a significant role for the State aid assessment. The relevant question is to identify the correct comparison group in order to establish whether the alleged beneficiary is indeed in a comparable situation to other companies that do not receive State support. a. Aid to Individual Undertakings A measure that provides benefits exclusively to one individual undertaking is presumptively selective and therefore likely to be State aid. However, sometimes, no other comparable undertakings are present in the Member State in question, or some apparently comparable undertakings are in fact not comparable. The principles underlying the assessment of comparability in this regard do not differ from the assessment of aid under the schemes below.173

167

Case C-482/99 Stardust Marine (n 81) para 56. ibid, para 55. 169 Case C-290/83 Commission v France (Poor Farmers), EU:C:1985:37, para 15. 170 Case C-482/99 Stardust Marine (n 81) paras 53–54. 171 Case C-206/06 Essent Netwerk Noord et al, EU:C:2008:413, para 73; Joined Cases C-67/85, C-68/85 and C-70/85 Van der Kooy v Commission, EU:C:1987:177, Opinion AG Slynn, 250. 172 Commission Decision of 22 September 2004 on the State aid which the United Kingdom is planning to implement for British Energy plc (notified under document no C(2004)3474), [2005] OJ L142/26, recital 294. 173 See under ‘Undertakings in a Comparable Situation’ below. 168

24 Falk Schöning and Clemens Ziegler b. Aid Schemes Exclusion of General Measures Purely general measures without advantages for certain undertakings or a certain sector do fall out of the scope of Article 107(1) TFEU. However, even measures that prima facie appear to be general in nature may be selective to a certain extent and fall under Article 107(1) TFEU.174 Even measures applying to a large number of undertakings (possibly all undertakings in a sector) or measures in diverse and/ or particularly large sectors may still not qualify as general measures of economic policy, if not all economic sectors can benefit from it.175 That an aid does not aim at one or more specific, predefined recipients, but that it may be granted based on a series of objective criteria, within the framework of an overall budget allocation, to an indefinite number of beneficiaries who are not initially individually identified, is insufficient to call into question the selective nature of the measure. Undertakings in a Comparable Situation There can only be selectivity where undertakings are in a comparable situation in light of the objective pursued by the measure.176 The Commission distinguishes between a so-called material and regional selectivity.177 Materiality and regionality can help to identify the relevant comparison framework. Regarding material selectivity, the Commission distinguishes between de jure and de facto selectivity. De jure selectivity is due to legal criteria that formally apply a measure only to certain undertakings (of a certain size and/or legal form, active in certain sectors, incorporated or newly listed on a regulated market; belonging to a group with certain characteristics or entrusted with certain functions within a group, ailing companies, or export undertakings or undertakings performing export-related activities).178 De facto selectivity exists where, although the criteria for the application of the measure have been neutrally phrased in general and objective terms, the measure is structured to significantly favour a particular group of undertakings. Conditions or restrictions that Member States impose that prevent a particular group or particular groups of undertakings from benefiting from the measure may lead to de facto selectivity. An important aspect of establishing selectivity is the identification of the relevant reference system. The Commission calls the reference system the ‘benchmark against which the selectivity of a measure is assessed’.179 It is a set of rules of general application, based on objective criteria, to all undertakings within its scope as defined

174

Case C-403/10 P Mediaset SpA v Commission, EU:C:2011:533, para 81. Case C-143/99 Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke, EU:C:2001:598, para 48. 176 ibid para 41; Commission v Netherlands (n 24) para 52; Tisza Erömü kft v Commission (n 165) para 160. 177 See Commission Notice on the notion of State aid (n 1) para 120. 178 For references see Commission Notice on the notion of State aid (n 1) para 121. 179 ibid, para 132. 175

What is State Aid? 25 by its objective. The reference system does not only define the scope of the system, but also the conditions under which the system applies, the rights and obligations of undertakings under it and the details regarding the functioning of the system.180 The reference geographic area cannot be wider than national. Selectivity does not mean that undertakings in Member State B in a comparable situation as undertakings in Member State A do not enjoy the same benefits as the undertakings in Member State A. It is established case law that a measure can be a selective advantage although it aims at compensating for disadvantages affecting undertakings in one Member State compared with their competitors in other Member States.181 Selective Measures Within the reference system, the assessment asks whether the measure in question derogates from the system by differentiating between undertakings. Does a measure favour certain undertakings or groups of undertakings over other undertakings in a similar factual and legal situation? It is irrelevant whether such differentiation follows an objective of the reference system or not. Even external policy objectives, for example, regional, environmental or industrial policy objectives, cannot justify a different treatment of undertakings in this regard and hence cannot exclude a finding of selectivity.182 A measure that favours a certain undertaking or group of undertakings in a comparable legal and factual situation as other undertakings or groups of undertakings is prima facie selective.183 In certain cases, the reference geographic area may even be regional.184 Measures Justified by the Nature of the System Some measures a priori look selective, but nevertheless do not qualify as State aid, because advantages arising from them for one or a number of undertakings are justified ‘by the nature or general scheme’ of the system under review,185 also referred to as the ‘nature or structure’186 or ‘logic’187 of the system. Examples for such justification were (i) a sectoral tax differentiation in Germany, under which energy products sold for heating and fuel purposes were taxed, but energy products used for certain energy-intensive processes (such as metal production, mineralogical processes) were not;188 (ii) a tax exemption for green electricity;189 and (iii) for energy used for non-fuel purposes.190 Other examples can be found in 180

ibid, para 133. Joined Cases C-71/09 P, C-73/09 P and C-76/09 P Comitato ‘Venezia vuole vivere’ et al v Commission, EU:C:2011:368, para 96. 182 Commission Notice on the notion of State aid (n 1) para 135. 183 ibid, para 137. 184 This relates mostly to taxation cases and is therefore dealt with in ch 3 of this volume. 185 British Aggregates (n 13) para 83. 186 Case C-173/73 Italy v Commission (n 53) para 15. 187 Cassa di Risparmio di Firenze et al (n 40) para 137. 188 Case N 820/2006 Tax exemptions for certain energy intensive processes (7 February 2007). 189 Cases NN 30b/2000 and N 678/2001 Zero tariff for green electricity (28 November 2001). 190 Commission Decision of 3 April 2002 on the dual-use exemption which the United Kingdom is planning to implement under the Climate Change Levy and the extended exemption for certain competing processes [2002] OJ L229/15. 181

26 Falk Schöning and Clemens Ziegler the context of the regulation of small- and medium-sized enterprises (SMEs). Special, usually lighter administrative and regulatory requirements imposed on SMEs may possibly not qualify as State aid where they are justified by the smaller size and fewer resources of the SMEs in question.191 On the other hand, the GC refused to recognise a justification ‘by the nature of the system’ where certain undertakings engaging in energy-intensive manufacturing processes in Austria were allowed to limit their green electricity charges due to the purchase of green electricity to 0.5 per cent of the net value of their production:192 in this case, the GC saw no ‘unity of object and purpose between the general system provided for in the [law in question] as a whole and the special provision [catering for the exemption for certain undertakings with energy-intensive production]’.193 Similarly, the Commission refused to recognise this justification for an excise duty exemption specifically for the production of certain biofuels: The excise duty exemption is destined to alleviate costs incumbent on a specific sector, namely that of the production of certain biofuels. Consequently, the exemption foreseen … [is outside] the logic and scheme of the general tax system applicable to fuels in Italy.194

v. Distorting Competition or Threatening to Distort Competition and Affecting Trade While the two elements of distorting or threatening to distort competition and affecting trade between Member States are distinct, in practice, they are often discussed together.195 Nevertheless, since only one of these two conditions may be met for a particular measure, they will be discussed separately. Even for already implemented measures (ie, actual effects could be evaluated), actual effects on competition and trade do not have to be shown. Rather, the Commission has to show that a measure may potentially affect competition and trade.196 This assessment has to be based on an ex ante view, even where the measure has already been implemented. Those will be cases in which the measure has not been notified. In order not to reward Member States for a failure to notify, the Commission will not take into account that as a matter of fact, different from what could have been expected on an ex ante basis, there are no meaningful effects on competition.197 At the same time, the Commission is allowed to use evidence of actual effects to strengthen/confirm its ex ante assessment.198

191 Case C-200/97 Ecotrade v Altiforni e Ferriere di Servola, EU:C:1998:378, Opinion AG Fennelly, para 26. 192 Austria v Commission (n 15) para 122. 193 ibid, para 106. 194 Commission Decision of 12 September 2011 on State aid N 529-2008, Application of Directive 2003/96/EC on restructuring the Community framework for the taxation of energy products and electricity [2011] OJ C303/4, recital 24. 195 Joined Cases T-298/07, T-312/97, T-313/97, T-315/97, T-600/97 to T-607/97, T-1/98, T-3/98 to T-6/98 and T-23/98 Alzetta et al, EU:T:2000:151, para 81. 196 Case T-211/05 Italy v Commission, EU:T:2009:304, para 152. 197 Case C-351/98 Spain v Commission, EU:C:2002:530, paras 66–67. 198 Case C-415/96 Spain v Commission, EU:C:1998:533, para 41.

What is State Aid? 27 a. Distortion of Competition or Threatening to Distort Competition A measure distorts or threatens to distort competition when it improves the competitive position of the alleged beneficiary compared with its competitors.199 Meeting this condition is not particularly difficult: there is no need for a detailed definition of the relevant product and geographic markets. Generally, there is a distortion of competition pursuant to Article 107(1) TFEU where the State grants a financial advantage to an undertaking in a liberalised sector where there is (or at least could be) competition.200 The assignment of a public service to a State-owned entity (where the service could also have been assigned to a private third party) may constitute a distortion of competition. This would not be the case where a service not competing with other services is subject to a legal monopoly (in compliance with EU law) that excludes competition not only on the market but also for the market, ie, any possible competition to become the exclusive provider of the service in question.201 Where such a service provider is active in another (geographical or product) market open to competition, State aid rules prohibit any cross-subsidisation from the funding received for the service under the legal monopoly into the activities subject to competition.202 It is not necessary that an advantage granted enables the recipient undertaking to outright strengthen its competitive position. It is sufficient that the aid allows the undertaking to maintain a stronger competitive position than it would have had without the aid having been provided. In this regard, any aid that relieves an undertaking of the expenses it would otherwise have had to bear in its day-to-day business operations distorts or threatens to distort competition.203 The Commission has, in a case regarding a State grant to fund energy infrastructure, also found a possible distortion in the more generally understood systems competition between energy sources: ‘[f]avouring the development of gas infrastructure has the potential of influencing the patterns of competition between energy sources, which are partly substitutable with gas’.204 The competitive conditions in other Member States are irrelevant to assess a possible distortion of competition. Only the competitive situation in the Member State granting the advantage is to be considered. Even a measure that a Member State introduced to offset a competitive disadvantage suffered by undertakings compared with undertakings in other Member States still can distort or threaten to distort competition.205 b. The Effect on Trade between Member States The element of ‘effect on trade between Member States’ aims at drawing a line between the jurisdiction of the Commission and the Member States: without effect

199

Cases C-182/03 Belgium v Commission, EU:C:2006:416, para 131. Altmark Trans (n 109) paras 77–79. 201 Case T-295/12 Germany v Commission, EU:T:2014:675, para 158; Commission Decision of 7 July 2002 on State aid N 356/2002, United Kingdom—Network Rail [2002] OJ C232/2, recitals 75–77. 202 Commission Notice on the notion of State aid (n 1) para 188. 203 Comitato ‘Venezia vuole vivere’ et al v Commission (n 180) para 136. 204 Commission Decision SA.39050 (n 141) recital 37. 205 Case C-298/00 P Italy v Commission, EU:C:2004:240, paras 61–62. 200

28 Falk Schöning and Clemens Ziegler on trade between Member States, an aid measure would not threaten to negatively affect the internal market.206 The requirement is satisfied where an aid measure strengthens the position of an undertaking that competes in intra-Union trade.207 This is always met where the alleged beneficiary undertaking is active in cross-border trade or where there is a sizeable level of trade in the sector in question. In the EU energy markets, measures favouring undertakings in one Member State should always qualify for having an effect on trade between Member States.208 The Commission has stated that, once the electricity market was opened up to competition and, in particular, once Directive 96/92 entered into force, measures favouring undertakings in the energy sector in one Member State were capable of impeding the ability of undertakings from other Member States to export electricity to the first Member State, or of favouring the export of electricity from that State to other Member States.209

The condition can still be met where the alleged beneficiary does not export its products or services or operates only on a local or regional level. In such cases, trade will be affected where aid increases domestic production, reducing the chances of companies in other Member States to export their products.210 Neither a relatively small amount of aid nor a relatively small-sized recipient undertaking as such can preclude the possibility that trade between Member States might be affected.211 Overall, cases without the effect of aid on trade are relatively rare. Examples include activities with low levels of competition in intra-Union trade such as car repairs, taxi services or sectors with very high transport costs (some raw materials, construction materials etc), relatively small aid amounts for small undertakings which are predominantly active locally.212

C. De minimis Aid Article 107(1) TFEU does not generally exclude de minimis aid measures from the application of EU State aid rules: where an undertaking only reaps limited benefits from a State aid measure, the level of distortion of competition may be lower, but there still will be distortion of competition.213 There is no significance threshold below which competition and trade may not be affected under Article 107(1) TFEU. That said, since the early 2000s, the Commission used to adopt de minimis regulations based on the Enabling Regulation. The current De Minimis Regulation214 206

Case T-93/02 Confédération nationale du Crédit Mutuel (Livret bleu), EU:T:2005:11, para 82. Case C-518/13 Eventech v The Parking Adjudicator, EU:C:2015:9, para 66. 208 Tisza Erömü kft v Commission (n 165) para 186 (for the electricity sector); Commission Decision SA.39050 (n 141) recital 38 (for the gas sector). 209 Tisza Erömü kft v Commission (n 165) para 186. 210 Altmark Trans (n 109) paras 77–78. Commission Decision SA.38769 of 11 August 2015 on Green Deal for Electric Vehicle Charging Infrastructure [2017] OJ C20/6, recital 34. 211 Eventech v The Parking Adjudicator (n 207) para 68. 212 Heiser (n 145) Opinion AG Tizzano, para 58. 213 Case T-214/95 Vlaamse Gewest v Commission, EU:T:1998:77, para 46. 214 Commission Regulation (EU) No 1407/2013 of 18 December 2013 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid [2013] OJ L352/1. 207

What is State Aid? 29 is applicable to public support if (i) it is provided to a single undertaking; (ii) it does not exceed €200,000 over three years; and (iii) the Regulation’s other requirements are met. Such support will, pursuant to Article 3(1) of the De Minimis Regulation, not be considered State aid and thus does not have to be notified. In addition, in 2012, the Commission adopted a special de minimis regulation for services of general economic interest.215 The Regulation applies to public funding allocated for the provision of services of general economic interest of no more than €500,000 over three years, insofar as the other conditions of the Regulation are also satisfied. A measure within the scope of the Regulation is not to be considered as State aid.

III. CURRENT AND FUTURE FOCUS AREAS REGARDING THE NOTION OF STATE AID IN THE ENERGY SECTOR

EU State aid law has played a major role in shaping European energy policy in the past. Recent industry trends also involve significant investments that will often be supported by State resources: building new, highly efficient power plants, the decommissioning of old nuclear power plants, rolling out additional infrastructure for the transmission of renewable energy, capacity mechanisms and other measures to ensure security of supply, regarding both electricity and gas, or e-mobility projects. It can be expected that the definition of State aid and its elements continue to be of significant importance for any major energy projects, and will in fact contribute to the further development of EU and Member States’ practice in the area of State aid law.

215 Commission Regulation (EU) No 360/2012 of 25 April 2012 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid granted to undertakings providing services of general economic interest [2012] OJ L114/8.

2 State Aid and Price Regulation in the Energy Sector GUILLAUME DEZOBRY AND ADRIEN DE HAUTECLOCQUE

I. INTRODUCTION

T

HE ISSUE OF energy prices is a sensitive one because it determines the competitiveness of national economies and the social acceptance of energy and climate policies in the long term. It is little wonder, then, that energy prices have been subjected to massive public interventions, both direct and indirect, on a regular basis. There is no denying the fact that EU law impacts the components of the price paid by the end users of energy. Wholesale energy prices are for instance affected by the carbon price as defined in the context of the European emissions trading system, or by the so-called capacity mechanisms that are being implemented at national level but supervised by the EU Commission under the rules on State aid. EU law also regulates, first and foremost, the network component of energy prices, in particular through rules about ownership unbundling or third-party access. As far as retail prices are concerned, the influence of EU law is felt, for instance, in the field of consumer protection or in the performance of public service obligations. Also, the legal rules of the EU that regulate duties, taxes and other assistance funds, such as those in support of renewable energies, must be mentioned because they constitute a large and ever-increasing share of the final energy price. Of course, the enforcement of rules on competition (anti-cartels, abuse of a dominant position and merger control), freedom of movement and even the energy diplomacy of the EU also have repercussions on prices. The figure below (Figure 1) formalises, without claiming to be exhaustive, the relationship between energy prices, intervention and EU law. The relationship between EU State aid control, under Articles 107 and 108 of the TFEU and the regulation of energy prices is a very broad subject. Some important aspects will be treated elsewhere in this book. In particular, the public interventions aimed at controlling the regulated prices granted to producers of renewable energies are addressed by Francesco Maria Salerno, Sébastien Douguet and Vincent Rious in chapter five. The issue of the interface between State aid control and public service rules, which is for instance at the heart of the debate surrounding capacity

32 Guillaume Dezobry and Adrien de Hauteclocque

Figure 1: Retail Price Regulation and EU Law: Mapping the Issues Source: A de Hauteclocque, ‘Retail Price Regulation and EU Law: Mapping the Issues’, speech delivered at the CREG Annual Conference (2015).

mechanisms, will be discussed by Leigh Hancher and Christoph Riechmann in chapter six and by Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu in chapter ten. The first section of the present chapter will address the issue of preferential tariffs granted to certain so-called energy-intensive industries. The second section will tackle the specific issue of preferential sales tariffs, and the third will consider aid regimes for residential customers. The fourth section will focus on the relationship between State aid control and actions undertaken by the national regulator to control prices. The last part will review in greater detail the jurisprudence of the Court of Justice of the EU, as for public intervention to be qualified as State aid under Article 107 of the Treaty, it must be granted by the State, or through State resources, which is key when the question of public intervention on energy prices comes up.

II. STATE AID AND PREFERENTIAL TARIFFS GRANTED TO ENERGY-INTENSIVE INDUSTRIES

As globalisation gained momentum in the second half of the twentieth century, access to energy at competitive prices became a prerequisite for wooing and retaining large electricity consuming industries or so-called energy-intensive industries. Competing as they are against one another, States do not hesitate to adopt systems that guarantee a high-quality energy supply at reasonable prices to industrial undertakings.

State Aid and Price Regulation 33 Throughout the European Union, such regimes are forced to conform to the requirements of State aid law. The examination of whether the systems comply with the law on State aid is complex, bordering between industrial, energy and competition policies. First, our analysis will focus on the systems put in place at a time when the energy sector was still vertically integrated and was mainly dependent upon a State undertaking enjoying a monopoly. Second, it will concentrate on the systems adopted following the implementation of liberalisation directives and the segmenting of the value chain.

A. State Aid and So-Called Integrated Preferential Tariffs At the end of the 1950s, preferential rates for supplying energy to gas- and electricityintensive industries were introduced in many Member States. At the time and until liberalisation directives were promulgated, both gas and electricity sectors had a monopoly. As a result, energy prices were integrated prices, without any distinction being made between the production, the transmission, the distribution or the supply of energy. The following section aims to present the most emblematic cases and to show the diversity and complexity of issues that have arisen. Moreover, the examination of various court rulings will result in our attempting to systematise the lines of reasoning of the Commission or the EU judges in order to qualify these tariffs as State aid and to assess whether they are compatible. We will conclude that it is not easy to apply the lessons of that period to the current situation due to broad changes in market structures. i. The Dutch Cases a. The Case of Preferential Tariffs as Applied to Ammonia Producers (F Tariff) The so-called Tariff F case may be presented through an analysis of the following four decisions. Commission Decision of 17 April 1984 Financial conditions under which Dutch ammonia producers were able to buy gas gave rise to several decisions, which it is appropriate to revisit. On 25 October 1983, the Commission decided to trigger the procedure provided for by Article 93 paragraph 2 of the EEC Treaty—now Article 108 paragraph 2 of the TFEU—against the tariff system applied in the Netherlands for the supply of gas destined for the production of ammonia. Through a reasoned opinion delivered on 13 March 1984, the Commission observed that the tariff structure in question was tantamount to State aid within the meaning of Article 92 paragraph 1 of the ECC Treaty (now Article 107 paragraph 1 TFEU) and that it was not entitled to any of the exceptions provided for in paragraph 3 of that Article.

34 Guillaume Dezobry and Adrien de Hauteclocque In response, the Dutch government informed the Commission, on 14 April 1984, that Gasunie had abolished the disputed tariff and that the tariff structure for gas had been augmented with a so-called Tariff F intended for those very large industrial users that fulfilled the following conditions: —

Consume an amount of gas higher than, or equal to, 600 million cubic metres per annum. — Present a load factor of 90 per cent or more. — Accept total or partial cessation of deliveries, at Gasunie’s discretion. — Accept the supply of gas with different energy values. This new Tariff F was being billed 5 cents/CM less than Tariff E being charged to large industrial customers. Through a decision on 17 April 1984, the Commission ruled that ‘the tariffs now in force did not contain any State aid components’. To reach this conclusion, the Commission picked out three pieces of information. First, the Commission underscored the ‘substantial savings in the supply of gas’ made by Gasunie owing to the consumption profile of those large industrial users. Second, the preferential tariffs were justified both economically and commercially, especially ‘when compared to prices charged to other large consumers’. Third, from the point of view of the Commission, the price differential of 5 cents/CM between Tariff E and Tariff F could be explained by the difference in the cost of delivering services. Judgment of 12 July 1990, COFAZ v Commission, C-169/84, EU:C:1990:301 Two French producers, CdF Chimie azote et fertilisants (Nitrogen and Fertilisers) and the chemical undertaking Grande Paroisse, brought an action against the decision of the Commission of 17 April 1984, being of the opinion that Tariff F constituted State aid, which provided an undue advantage to Dutch ammonia producers. The question that arose was to know whether, with regard to the consumption profile of large industrial undertakings that benefited from the preferential tariff, the contracts signed by Gasunie at Tariff F were likely to generate savings for the supplier. From the Commission’s standpoint, the 5-cent/CM difference between Tariff E and Tariff F was justified owing to the savings made by Gasunie on the basis of the three consumption characteristics of the industrial undertakings paying Tariff F: — The huge volumes consumed and the regularity of off-takes. — Interruptibility of deliveries. — Possibility to vary gas quality. To assess those various elements and, especially, to measure the savings that Gasunie might make due to the consumption profile of industrial undertakings that paid Tariff F, the Court decided to call upon experts independent from the parties. In its ruling of 12 July 1990, the Court found that: In the experts’ opinion, it follows that the value of the savings in Gasunie’s costs resulting from the conditions applied to Tariff F customers, by comparison with those applied to tariff E customers, is much lower than the price difference between those two tariffs.

State Aid and Price Regulation 35 The experts infer that the rebates granted to Tariff F customers must be attributable to other considerations (emphasis added).1

The Court ruled that, ‘by stating that the price difference of 5 cents/CM between tariffs E and F corresponded to the difference in the cost of the services provided, the Commission committed a manifest error of appraisal’, and therefore quashed the decision. Commission Notice Pursuant to Article 93(2) of the EEC Treaty to Other Member States and Other Parties Concerned Regarding a Preferential Tariff System Applied to the Netherlands for Supplies of Natural Gas to Dutch Nitrate Fertiliser Producers2 Acknowledging that the savings made from deliveries by Gasunie to large industrial undertakings, on the basis of their consumption profile, were insufficient in themselves to account for the 5 cent/CM-discount, the Commission finally found that Tariff F was also justified for commercial reasons. First, the Commission underscored the pressures brought to bear by exporters of nitrogen fertilisers from Eastern Europe. Second, it noticed that Dutch nitrogen fertiliser producers consumed annually some 30 per cent of the gas delivered by Gasunie to the industry. Under these conditions, Gasunie had a vested interest in keeping this customer base, since the volumes of gas sold made it possible to ‘achieve faster depreciation of investments’. Third, according to the Commission, Tariff F enabled Gasunie to cover both its variable and fixed costs. It pointed out in this regard that, such costs are well below Tariff F, and the undertaking was thus able to increase its net revenue through sales at the Tariff F price, while at the same time ensuring the maintenance of a significant group of customers which it was in danger of losing if it did not grant Tariff F.3

In the final analysis, the Commission considered that Tariff F was justified and did not fall within the scope of Article 92 paragraph 1 (now Article 107, paragraph 1 TFEU). Judgment of 29 February 1996, Belgium v Commission, C-56/93, EU:C:1996:64 Following an action brought by Belgium seeking annulment, the Court again examined Tariff F and whether this complied with the rules on State aid. The Court recalled that any preferential tariff must be viewed as aid, save if this measure is based on economic grounds. It stated that: According to established case-law, the charging by a Member State, or an entity on which it exerts influence, of a tariff fixed at a level lower than that which would normally be chosen may be regarded as aid for the purposes of Article 92(1) of the Treaty. In such

1

Case C-169/84 COFAZ v Commission, EU:C:1990:301, para 50. Commission Notice pursuant to Article 93(2) of the EEC Treaty to other Member States and other parties concerned regarding a preferential tariff system applied to the Netherlands for supplies of natural gas to Dutch nitrate fertiliser producers [1992] OJ C344/4. 3 ibid, 6. 2

36 Guillaume Dezobry and Adrien de Hauteclocque circumstances, the State, or the entity on which it exerts influence, does not apply the preferential tariff as an ordinary economic agent but uses it to confer a financial advantage on certain undertakings by forgoing the profit which it would normally realise. On the other hand, a preferential tariff does not constitute aid if, in the context of the market in question, it is objectively justified by economic reasons such as the need to withstand competition on the same market.4

In the case at hand, ‘the decisive question is whether the conditions of competition on the market were such as to justify Gasunie in granting Dutch ammonia producers a lower price than that charged to other industrial sectors’.5 The Court found that, if the tariff in question had not been consented, ammonia producers might have relocated, thereby depriving Gasunie of a substantial market outlet and resulting in a significant drop in volume, which it could not offset by a corresponding increase in exports.6 Moreover, in response to Belgium’s rejoinder, according to which ‘Tariff F was not justified on commercial grounds but politically motivated, its purpose being to place Dutch ammonia producers at an advantage’,7 the Court recalled that: According to established case-law, Article 92 of the Treaty does not make a distinction according to the causes or aims of the measures of State intervention concerned but defines them according to their effects (cases 173/73 and 310/85). Where a practice is objectively justified on commercial grounds, the fact that it also furthers a political aim does not mean to say that it constitutes State aid for the purposes of Article 92 of the Treaty.8

Finally, the appeal was dismissed and Tariff F not regarded as a State aid. b. The Preferential Tariff Case as Applied to Dutch Glasshouse Growers In the Netherlands, horticulture enjoys special tariff conditions for the supply of natural gas. This tariff structure was put in place in the mid-1970s to encourage glasshouse growers to stop using heavy fuel and replace it with less polluting gas. Back in 1981, the Commission had found that horticultural tariffs constituted incompatible aid (Decision 82/73).9 In the wake of that decision, a new tariff was negotiated whose terms were set in an agreement signed between Gasunie and the Landbouwschap (the body representing glasshouse growers). This new tariff was notified to the European Commission which decided to trigger the procedure provided for in Article 93 paragraph 2 (Article 108, paragraph 2 TFEU) of the Treaty. The final decision of the Commission was adopted on 13 February 1985, and challenged before the Court of Justice of the EU which gave its ruling on 2 February 1988.

4

Case C-56/93 Belgium v Commission, EU:C:1996:64, para 10. ibid, para 30. 6 ibid, para 34. 7 ibid, para 15. 8 ibid, para 79. 9 Commission Decision 82/73/EEC on the preferential tariff charged to glasshouse growers for natural gas in the Netherlands [1982] OJ L37/29. 5

State Aid and Price Regulation 37 Commission Decision of 13 February 1985 on the Preferential Tariff Charged to Glasshouse Growers for Natural Gas in the Netherlands 85/215 ([1985] OJ L97/49) In its decision, the Commission recalled that the three conditions under which a preferential rate is caught by the prohibition of Article 92 paragraph 1 EC (Article 107, paragraph 1 TFEU) are: —

— —

It favours certain undertakings or certain products that are competing with undertakings or products in other Member States, the products concerned being subject to intra-community trade. It was imposed by public authorities. It results in compensation paid by the State to the distributing company, or in lower State revenue.

Following this line of argument, the Commission concluded that the aid in question met those conditions and that the exemption provided for in paragraphs 2 and 3 of this Article, could not apply. As regards Article 107 paragraphs 3(a) and (c), the Commission specified that: Aid must be regarded as operating aid granted to the undertakings concerned: such type of aid was, in principle, opposed by the Commission because granting it is unrelated to the conditions which would enable them to qualify for one of the exemptions under paragraph 3 points a) and c) of Article [107].10

Finally, the Commission decided that aid must be recovered. Judgment of 2 February 1988, Kwekerij van der Kooy ea v Commission, C-67/85, C-68/85 and C-70/85, EU:C:1988:38 By its ruling of 2 February 1988, the Court approved the line of reasoning of the Commission and dismissed the action. The Court began by recalling that: It should be observed at the outset that this case raises the question whether the fixing of the tariff charged for a source of energy at a level lower than that which would normally have been chosen may be regarded as ‘aid’ when it is attributable to action by the Member State concerned and the other conditions laid down by Article 92 are satisfied. In such circumstances the State, or the entity on which it exerts influence, does not apply the tariff as an ordinary economic agent but uses it to confer a pecuniary advantage on energy consumers, in the same way as it grants aid to certain undertakings, forgoing the profit which it could normally realise (emphasis added).11

Thereupon, it added that: That would not be the case, however, if it were demonstrated that the preferential tariff was, in the context of the market in question, objectively justified by economic reasons such as the need to resist competition on the same market from other sources of energy the

10 Commission Decision 85/215/EEC on preferential tariff charged to glasshouse growers for natural gas in the Netherlands [1985] OJ L97/49. 11 Cases C-67/85, C-68/85 and C-70/85 Kwekerij van der Kooy ea v Commission, EU:C:1988:38, para 28.

38 Guillaume Dezobry and Adrien de Hauteclocque price of which was competitive. In determining whether such competition is a real prospect account should be taken not only of the different price levels but also of the costs involved in conversion to a new source of energy, such as replacement and depreciation costs for heating equipment (emphasis added).12

It must be pointed out that these judgments are consistent with those made in connection with the cases relating to Tariff F, since they make aid conditional upon a close investigation of the behaviour of the energy supplier itself. If the behaviour of the latter is driven by some degree of economic rationality, then there is no point in prohibiting the measure in question. ii. The Case of the Preferential Tariff Granted by EDF to Papermakers Commission Decision of 11 April 2000 Concerning the Measure Put in Motion by EDF in Favour of Certain Papermaking Mills ([2001] OJ L95/18) In its ruling of 11 April 2000, the Commission had to evaluate whether the preferential tariff accorded by EDF to papermakers complied with State aid rules. Facing overcapacity at the end of the 1980s, due to a surplus of energy from nuclear sources, EDF, a vertically integrated State-owned undertaking, took to offering certain customer rebates on the price of the energy consumed, provided they invested in industrial processes likely to increase their power consumption. For instance, the French incumbent was offering to lower power rates for papermaking undertakings that were willing to invest in infrared drying equipment. Suspecting that the French State was paying aid, the Commission decided to trigger the procedure provided for in Article 88 paragraph 2 (Article 108, paragraph 2 TFEU) of the Treaty. The key issue was to determine whether or not the preferential rate granted by EDF to paper mills constituted an undue advantage under the provisions of Article 107 paragraph 1 of the Treaty. In light of its analysis, the Commission found that the preferential tariff did not constitute an undue advantage conferred on papermakers. Its line of reasoning hinged on the three following arguments. First, the Commission acknowledged the fact that EDF had, for several years, been grappling with overcapacity from nuclear sources. Under those conditions, and as emphasised by French authorities, ‘EDF had a vested interest in promoting the development of technologies and equipment likely to boost power demand and create new markets for its products’.13 Second, the Commission checked that the tariff levels made it possible for the historic French operator to cover its variable costs and a part of its fixed costs. More to the point, the additional electricity generated by setting up infrared dryers was billed at a price that ‘covered variable costs and at least 35% and on average 57% of fixed costs’.14

12

ibid, para 30. Case C 39/1998 Aide de la part d’EDF à certaines firmes de l’industrie papetière [2001] OJ L95/18, especially point 40. 14 ibid, especially point 74. 13

State Aid and Price Regulation 39 Third, the Commission noted that ‘there is nothing in the dossier to indicate that EDF could have obtained a higher price elsewhere’.15 Finally, the Commission ruled that the measure in question had merit for business reasons and did not constitute State aid within the meaning of Article 87 paragraph 1. iii. The Italian Cases a. The Case of the Preferential Tariff Granted to Alcoa This case, which gave rise to many decisions, revolved around the price of electricity paid by two of Italy’s largest aluminium production centres: the Portovesme plant in Sardinia and the Fusina plant in Venetia. Commission Notice Pursuant to Article 93(2) of the EC Treaty to Other Member States and Interested Parties Concerning Italian State Aid to Alumix ([1996] OJ C288/4) Alumix SpA was a subsidiary of the EFIM Group, an Italian holding company that owned 99.9 per cent of its share capital, and had in turn several subsidiaries operating in the aluminium sector. In July 1992, the Italian government decided to wind up EFIM. Several aid and support measures were adopted through Legislative Decrees No 362 of 14 August 1992 and No 414 of 20 October 1992. One of the measures adopted consisted of granting Alumix a preferential power supply rate. On 23 December 1992, the Commission decided to trigger the procedure provided for in Article 93 paragraph 2 (Article 108, paragraph 2 TFEU) of the Treaty in connection with those support measures. The procedure targeted the preferential electricity rates charged to the aluminium smelters at Portovesme (Sardinia) and Fusina (Venetia). In the 1990s, these two aluminium smelters were the first and the third largest electricity users in Italy in terms of demand and energy consumption hours. To appraise whether those measures were compatible with State aid rules, the Commission began analysing the local environment. As for the Alumix smelter at Portovesme in Sardinia, the Commission noted some specifics about the Sardinian power grid and, in particular, the following three points: — —

Sardinia had its own power generation sources. There were already power generation overcapacities on the island. In 1994, excess capacity stood at 70 per cent, at least, according to estimates by the Commission. Such huge overcapacity was due to the fact that Sardinia had failed to industrialise as fast as anticipated. — Interconnections between Sardinia and continental Italy are poor (the capacity of the undersea cable does not exceed 257 MW) so that the island’s overcapacity cannot be used to supply continental Italy. Against this backdrop, the Portovesme smelter represented an important market outlet for the Italian incumbent operator ENEL and closure of the site would simply 15

ibid.

40 Guillaume Dezobry and Adrien de Hauteclocque increase surplus capacity and worsen the imbalance between supply and demand for electricity in Sardinia. The Commission concluded that Alumix had a rather strong bargaining power vis-a-vis ENEL, reinforced by the fact that the aluminium smelter requires a constant 24-hour supply instead of peak deliveries.16 Therefore, [i]t is in ENEL’s economic interest to continue to supply the Portovesme smelter, provided the latter buys electricity at a higher price than the average marginal cost of production based on the actual mix of fuels used by power plants in Sardinia … The tariff covers marginal costs and makes a small contribution towards fixed costs. By so doing, ENEL prevents an even bigger overcapacity in the region and receives a positive cash flow because of the contribution to the fixed costs that would be absent in the case of a closure of the aluminium smelter.17

The Commission estimated, under those circumstances, that ‘ENEL is behaving as a normal commercial operator and that the tariff does not constitute State aid within the meaning of Article 92(1) of the EC Treaty’. The situation of the Fusina plant was very similar to that of the Portovesme plant. In fact, Venetia, at the time, was in the throes of deindustrialisation, thus accentuating the power generation overcapacity throughout the region. Faced with the same problem as in Sardinia, it would have been logical for ENEL to apply the same solution as that for the Portovesme smelter and to ‘charge a price that covered the average marginal production costs plus a contribution to fixed costs’.18 Finally, the Commission decided that [o]n the basis of the above facts, ENEL’s behaviour reflects the behaviour of a private operator in the electricity market by making sure that it delivers electricity at a price that covers its marginal costs and contributes to its fixed costs, that under-utilisation of the system is not aggravated and that it keeps its biggest industrial consumer as a client.19

Commission Decision of 19 November 2009, Relating to State Aid C 38/A/04 and C 36/B/06 enforced by Italy in favour of Alcoa Trasformazioni Initially planned to remain in force for 10 years, the preferential tariff granted to Alumix, an undertaking bought out by Alcoa in 1995, was prolonged and substantially modified on the basis of several pieces of legislation enacted between 1999 and 2005. This legislation resulted in the extension of the application of the preferential tariff until 2010 and the modification of its regime. Since decisions N 204/1999 of 29 December 1999 and N 148/2004 of 9 August 2004 were enacted by the Italian national regulatory authority, Alcoa was charged the standard prices by ENEL and was receiving compensation from a public entity (Cassa Conguaglio) up to the difference between the standard tariff charged by ENEL and the tariff set in the 1995 legislative decree. This regime was financed by a parafiscal levy on all electricity consumers. 16 17 18 19

Case C-38/92 Italian State to Alumix [1996] OJ C288/4, 16. ibid. ibid, 17. ibid.

State Aid and Price Regulation 41 At issue was whether this new regime would avoid being regarded as aid within the meaning of the Treaty provisions. First, the Commission considered that the preferential tariff constituted an undue advantage to Alcoa. It underlined the fact that Alcoa would not have been in a position to obtain similar prices from a supplier on the market without intervention by the State. Moreover, it dismissed Alcoa’s argument to the effect that the preferential tariff equalled the price it could have obtained on an arm’s length basis. The Commission noted, in this regard, that well-established jurisprudence stated that the fact that a Member State seeks, by unilateral measures, to approximate conditions of competition in a particular sector of the economy to those prevailing in other Member States cannot deprive the measures in question of their character as State aid.20 The Commission also underlined the fact that, in order to ascertain the price that an electro-intensive industrial undertaking would pay in normal conditions on an arm’s length basis, one should not use the most competitive marginal cost of production, but a ‘weighted average that factors in both the high and low rates charged during peak and off-peak periods, respectively’.21 Finally, the Commission found that the ‘tariff reduces the costs arising from the contract signed with ENEL, which normally should burden the undertaking’s budget, so that, in accordance with settled case law, the measure confers an economic advantage to Alcoa’ (paragraph 158). In such conditions, the measure was actually tantamount to aid within the meaning of Article 87 paragraph 1 of the EC Treaty.22 After reviewing the different exemptions provided for in paragraphs 2 and 3, as well as the establishment of a Virtual Power Plant (VPP) to increase competition on the Sardinian electricity market,23 the Commission finally found that aid was incompatible. Judgment of 16 October 2014, Alcoa Trasformazioni v The Commission, T-177/10, EU: T:2014:897 After an action was brought before the Court for annulment of the above-mentioned decision, the Court approved the line of reasoning of the Commission and dismissed the action. It also confirmed that the preferential tariff for the supply of energy was, in fact, operating aid. The EU judge pointed out that: It is settled case-law that operating aid is aid which is intended to release an undertaking from costs which it would normally have had to bear in its day-to-day management or normal activities. Consequently, the aid in question, which permitted the applicant to reduce the costs connected with its electricity consumption, which is by definition part of day-to-day management, was indeed operating aid, a fortiori because, the aluminium

20

See also Case C-372/97 Italy v Commission, EU:C:2004:234, para 67. Case C 38a/2004 Preferential electricity tariff in favour of Alcoa [2010] OJ L227, para 154. 22 ibid, para 190. It must be emphasised that the Commission did not examine the measure from ENEL’s standpoint; namely, it did not consider the question of whether this contract was advantageous for that operator. 23 ibid, paras 249–59. 21

42 Guillaume Dezobry and Adrien de Hauteclocque smelting process being particularly energy-intensive, the purchase of electricity is of key importance to the applicant’s operations.24

Last, by a court order of 21 January 2016, the Court of Justice dismissed the appeal.25 b. The Case of the Preferential Tariff Applied to the Portovesme and Eurallumina Corporations Decision of 23 February 2011 Relating to State Aid C 38/B/04 and C 13/06 Enforcement Thereof by Italy in Favour of Portovesme Srl, ILA SpA, Eurallumina SpA and Syndial SpA The decree enacted by the Council of Ministers of 6 February 2004 extending the preferential tariff available to Alcoa since 1995 also had the effect of introducing preferential electricity rates for Portovesme Srl, a zinc, silver and lead producer, and Eurallumina SpA, an alumina producer. Building on the procedure initiated in the context of the preferential rates accorded to Alcoa, the Commission focused on the tariffs granted to these two industrial undertakings. In the observations sent to the Commission, both industrial undertakings stressed the importance of maintaining those rates in order to address the failures of the electricity market in Italy (and especially in Sardinia), and of re-establishing the market conditions that would have prevailed had the market worked properly. Hence, according to Portovesme, ‘the preferential tariff intends to address a market failure, namely the failure of the recently liberalised Sardinian electricity market to deliver competitive prices owing to the market power of the incumbent operators’.26 The company further noted that ‘Portovesme contended that the preferential price enjoyed corresponded to the price the company would have obtained under normal market conditions, ie if the electricity market functioned correctly’.27 In other words, ‘the measure merely offset the competitive disadvantage suffered by the company as a result of Sardinia’s structural problems (which included the absence of natural gas)’.28 For its part, Eurallumina recalled that ‘alumina production costs are higher in Europe than in other regions of the world, because of higher fuel and electricity prices’.29 Moreover, both industrial undertakings claimed that, in the absence of preferential tariffs, they would be forced to shutter their facilities in the short or longer term. Finally, Italy recalled that ‘similar measures were implemented in other Member States, where energy-intensive users benefitted as a result of regulated electricity tariffs’.30 24

Case T-177/10 Alcoa Trasformazioni v Commission, EU:T:2014:897, para 92. Case C-604/14 P Alcoa Trasformazioni v Commission, EU:C:2016:54. 26 Preferential electricity tariff in favour of Portovesme, ILA and Euroallumina (Case SA.15390) Commission Decision 2011/746 [2011] OJ L309/1, para 56. 27 ibid, para 57. 28 ibid, para 58. 29 ibid, para 63. 30 ibid, para 72. 25

State Aid and Price Regulation 43 In reply to these observations, the Commission stressed the following points. First, it recalled that ‘neither the Union law courts nor the Commission have ever taken into account the conditions that would prevail in a hypothetical market that functioned better’.31 Next, as to the existence of similar measures in the other Members States, the Commission stressed that ‘the existence of unlawful aid in some Member States cannot justify the adoption of similar measures by another Member State’.32 Finally, concerning the necessity to preserve the competitiveness of undertakings domiciled in the European Union and the risks of relocation, the Commission noted that: There is no precedent in the Commission’s decisions or in the judgments of the Union’s law courts for accepting this argument as a justification for the granting of State aid. Decisions to relocate production, in whole or in part, are part of a company’s business strategy, and the Commission has never held that they constitute a market failure.33

The General Court confirmed the Commission’s analysis.34 c. The Case of the Terni Tariff Commission Decision of 20 November 2007 on the State Aid C 36/A/06 Implemented by Italy in Favour of ThyssenKrupp, Cementir and Nuova Terni Industrie Chimiche By a law dated 6 December 1962, Italy had nationalised its electricity sector. This law called for the transfer of electrical power plants to ENEL, a newly established State-owned undertaking that was to be granted a monopoly on the production, distribution and supply of electricity. Terni—a State-owned undertaking operating in the steel, chemical and cement industry—owned hydropower stations whose production was mainly destined to supply power to its industrial sites. In accordance with the provisions of the nationalisation law, these Terni hydropower plants were transferred to ENEL. By a decree from the President of the Italian Republic, the government compensated Terni for the transfer of its electrical assets by granting it a preferential electricity tariff over the period from 1963 to 1992, from which Terni’s three successor undertakings, ThyssenKrupp, Cementir and Terni Industries, benefited. After a first extension for the period from 1992 to 2007, justified by the renewal of hydroelectric concessions, the preferential tariff was prolonged a second time—that is, until 2010, by Law No 80/2005. The Commission, to which this second prolongation had not been notified, decided to examine if the scheme was complying with State aid rules. As part of the procedure initiated by the Commission, the observations contributed by Italian authorities revolved around two strands of thought. On the one hand, prolongation of the tariff followed on from the prolongation of the original

31 32 33 34

ibid, para 108. ibid, para 143. ibid, para 190. See Case T-308/11 Eurallumina v Commission, EU:T:2014:894.

44 Guillaume Dezobry and Adrien de Hauteclocque scheme and was justified by the concern to ensure that the expropriated undertaking would be adequately compensated. On the other, the preferential tariff was indispensable to guarantee the industrial undertakings concerned a price similar to that paid by their competitors. In this regard, Italian authorities pointed out that the tariff is necessary to establish a level-playing field between these energy-intensive companies active in Italy and their competitors in the EU which also benefit from reduced energy prices (tariff or contract-based). If the tariff was abolished, the companies in question would delocalise their operations outside the EU. This would inevitably lead to an industrial crisis and sever job losses in the affected regions.35

It was easy for the Commission to conclude that aid had been given. After having underscored that such a tariff conferred ‘a clear economic advantage for the beneficiaries, who see their production costs reduced and their competitive position strengthened’,36 it verified that the aid had been paid selectively and that it had had an impact on trade and competition.37 In assessing whether aid was compatible, the Commission underscored the harmfulness of that measure as well as its detrimental effects on competition. It stressed that, operating aid granted in the form of a preferential electricity price to an energy-intensive undertaking, that is, an undertaking having electricity as one of its major cost factors, is a particularly distortive form of support since the aid has a substantial and direct impact on that undertaking’s production cost and resulting competitive position.38

In addition, with respect to the industrial policy concerns raised by Italy, the Commission recalled that such elements could not justify grants of aid. It commented that: Italy has in fact extensively explained the industrial policy reasons for the second prolongation of the Terni tariff. The main thrust of Italy’s argument in favour of the tariff is that energy intensive companies in other Member States can also benefit from reduced energy prices and the tariff is required as a transitional measure to avoid delocalisation outside the EU pending the full liberalisation of the energy market and the improvement of infrastructure.39

But the Commission also added that the Court has ruled that ‘the fact that a Member State seeks to approximate, by unilateral measures, conditions of competition in a particular sector of the economy to those prevailing in other Member States cannot deprive the measures in question of their character as State aid’. Further, the Italian argument that such a State aid would be justified by the existence of other (equally distortive) State aids in the EU is to be dismissed altogether. Such an approach would lead to subsidy races and would run counter to the very objective of State aid control. As regards the alleged risk of delocalisation outside the EU, the Commission

35 State aid in favour of ThyssenKrupp, Cementir and Nuova Terni Industrie Chimiche (Case C 36a/2006) Commission Decision 2008/408/EC [2008] OJ L144/37, para 61. 36 ibid, para 99. 37 ibid, para 117. 38 ibid, para 139. 39 ibid, para 144.

State Aid and Price Regulation 45 notes that there is no precedent in its decisional practice or in the jurisprudence of the Community courts, where such an argument has been accepted as a justification for the grand of State aids.40

This decision was confirmed by the Court judgment of 1 July 2010, ThyssenKrupp Acciai Speciali Terni v Commission.41 iv. The Case of the Preferential Tariff Granted to Aluminium of Greece SA42 Commission Decision of 13 July 2011 on the State aid SA.26117—C 2/10 implemented by Greece in favour of Aluminium of Greece SA A preferential supply tariff had been granted to a large aluminium producer located in Greece—Aluminium of Greece—beginning in 1960. By virtue of an agreement signed between that producer and the Greek State, this preferential tariff was to be valid until 2006. In a decision of 23 January 1992, the European Commission ruled that the preferential tariff granted to Aluminium of Greece constituted a State aid compatible with the internal market. In the wake of a dispute between Aluminium of Greece and DEI, Greece’s Stateowned power undertaking, about the end of the tariff, two decisions were rendered by Greek courts to extend the preferential tariff for the period running from 5 January 2007 to 6 March 2008, thus allowing the aluminium producer to save some €17 million. In July 2008, the Commission received two complaints alleging that Greece had granted aid to Aluminium of Greece and thereupon, on 27 January 2010, it decided to trigger the procedure provided for in Article 108 paragraph 2. In order to demonstrate that the preferential tariff, which was available to Aluminium of Greece between January 2007 and March 2008, did indeed amount to State aid within the meaning of Article 107 paragraph 1 of the Treaty, the Commission made the following points. In its decision of 13 July 2011, the Commission ruled, first that the preferential tariff accorded to Aluminium of Greece for the supply of electricity constituted an advantage, which was not due to the operation of market forces. According to the Commission, two items of proof could be adduced. On the one hand, the Stateowned electricity undertaking had demonstrated repeatedly its opposition to continuing with this rate. On the other, the Greek government had to grant subsidies to the State-owned undertaking in the early 2000s to compensate it for the stranded costs related to the application of the preferential tariff that had been available to Aluminium of Greece since 1960, implying that the preferential tariff did not make it possible for the operator to cover all its costs.

40 41 42

ibid, para 145. Case T-62/08 ThyssenKrupp Acciai Speciali Terni v Commission, EU:T:2010:268. See also ch 19 in this volume on Greece.

46 Guillaume Dezobry and Adrien de Hauteclocque As to the use of State resources, the Commission stressed that ‘State resources had indeed been used’, because a preferential tariff had been granted by the State-owned power undertaking, 51 per cent of which was held by the Greek State.43 Finally, after ruling that the exemptions provided for in paragraphs 2 and 3 of Article 107 TFEU did not apply, the Commission deemed the aid to be incompatible.44 After an action was brought before the General Court for annulment of the abovementioned decision, the EU Court quashed the aforementioned decision on the grounds that the aid in question was existing aid and could not be considered new aid.45 The General Court decision was, it must be noted, abrogated by the Court of Justice by a judgment of 26 October 2016 on account of errors of law about the qualification as new aid and existing aid.46 Four Lessons First, the crux of the matter is in knowing whether or not the measure constitutes an advantage within the meaning of Article 107 paragraph 1 TFEU. Should this be the case, qualification as aid generally applies, to the extent that all other criteria in Article 107 paragraph 1 are fulfilled and that exemptions in paragraphs 2 and 3 are seldom applicable. Next, the Commission seems to have changed its decision-making processes: previous analysis of its decisions shows that it was willing to recognise that no special advantage had been granted to an industrial undertaking, if the following conditions are met: —

An overcapacity for the energy undertaking (which is often caused by poor interconnections in the electric or gas network in the area where the industrial site is located). — A price that covers variable/marginal costs and a part of the fixed costs. — A risk of relocation of the industrial site if the preferential tariff is not applied, which would make the overcapacity situation even worse. Yet, owing to the increased integration of energy markets—and, in particular to better interconnections between the States—the chances of the success of arguments based on power grid overcapacity seem slim, if not nil. That also goes for the risks of relocation, which are no longer considered by the Commission. In addition, when the application of a preferential tariff is justified by the savings made by the power company owing to the consumption profile of the industrial undertaking, the Commission will still review the proportionality of the measure (cf Gasunie, Tariff F). Last, although reference to the application of the private investor test in a market economy has not been expressly specified by the Commission, use of that criterion has informed several of the abovementioned decisions (see the Gasunie and Alumix cases above). 43 Aluminium of Greece (Case SA.26117) Commission Decision 2012/339/EU [2012] OJ L166/83, para 28. 44 ibid, para 56. 45 Case T-542/11 Alouminion v Commission, EU:T:2014:859. 46 Case C-590/14 P DEI and Commission v Alouminion tis Ellados, EU:C:2016:797.

Tariff for Alcoa (prolonged)

Tariff for Alcoa (Alumix)

Tariff for glass house growers

iii)

ii)

i)

iii)

ii)

Overcapacity of the power grid in Sardinia and low-power interconnection to the continent. Lack of other market outlets for the energy undertaking. Tariff covers marginal costs and part of the fixed costs.

The specifics of consumption profiles of industrial undertakings (especially in terms of regularity and interruptibility) do not make it possible to justify the high rebates. International competitors and risks of relocation have been factored. Price would cover variable costs and a part of the fixed costs.

i)

Dutch case

Tariff F

The measure does not constitute an advantage within the meaning of Article 107 paragraph 1

Cases

Change in the nature of the scheme. Industrial undertaking unable to obtain such a price on the market. Operating aid. No exemption may apply.

iii)

i) ii)

Tariff is an advantage and the other criteria are met.

The measure creates an advantage within the meaning of Article 107 paragraph 1 and the other aid criteria are fulfilled

(continued)

Aid is incompatible

Aid is incompatible That is operating aid that is eligible to none of the exemptions provided for by the Treaty.

Aid is compatible Aid is incompatible

Table 1: Preferential Electricity Tariffs and the Existence of an ‘Advantage’ within the Meaning of Article 107 TFEU

State Aid and Price Regulation 47

Tariff for Aluminium of Greece

iii)

EDF had excess capacity. Tariff made it possible to cover variable costs and a part of fixed costs. There was no indication that electricity could have been sold at a higher price to other consumers.

The measure does not constitute an advantage within the meaning of Article 107 paragraph 1

French Tariff for i) Case papermakers ii)

Greek Case

Tariff for Terni

Italian Tariff for Case Portovesme and Eurallumina

Cases

Table 1: (Continued)

ii)

i)

Tariff was not negotiated with the operator who wanted to end it. Operator was partly compensated for stranded costs generated by tariff application.

Tariff confers a net advantage to its beneficiaries.

ii)

Aid is incompatible

Aid is incompatible i) That is operating aid which distorts free competition. ii) Existence of aid in other States is no justification. iii) Risk of relocation is no justification for granting aid.

Aid is incompatible Risks of an industry’s relocating does not justify grants of aid by a State.

i)

Tariff is an advantage, even when set at the level that would prevail in a competitive market. Enactment of similar measures in other Member States does not justify the tariff.

Aid is compatible Aid is incompatible

The measure creates an advantage within the meaning of Article 107 paragraph 1 and the other aid criteria are fulfilled

48 Guillaume Dezobry and Adrien de Hauteclocque

State Aid and Price Regulation 49 B. State Aid and Preferential Tariffs for Electro-Intensive Customers in the Context of an Unbundled Energy Sector The gas or electricity bill for electro-intensive customers usually includes three components: — The ‘energy’ component — The ‘transmission’ component — The ‘tax’ component The integration of the internal energy market and the elimination of regulated sales tariffs for non-residential customers, especially, electro-intensive customers, force Member States that wish to introduce a special scheme for electro-intensive customers to set up a system that targets one or several components of the price of consumed energy. Therefore, the various options available to Member States and the limitations that derive from the application of State aid rules must be reviewed. The two components (transmission and tax) are mainly used by Member States in order to lower energy prices paid by electro-intensive customers. i. Preferential Tariffs and Reduction of or Exemption from the ‘Transmission’ Component A significant part of the costs of energy supply for electro-intensive undertakings is made up of expenses related to the transmission of power, ie, charges related to the use of transmission (and, the case may be, distribution) networks. Several Member States have introduced exemptions and/or cap systems, with respect to these charges, for electro-intensive customers in order to maintain their competiveness. To date, only one decision resulted in an examination of these schemes through the prism of EU State aid rules: that is, the decision, taken on 6 March 2013 by the Commission, to initiate a detailed investigation of the German exemption system relating to the transmission costs for electro-intensive customers.47 In this opening decision, the Commission concluded that this measure constituted operating aid. It noted that: The Commission comes to the preliminary conclusion that the exemption mechanism constitutes operating aid, because it relieves the beneficiaries from network charges, which they would normally have had to bear in their day-to-day operation. According to caselaw, operating aid is in principle not compatible with the internal market because it has

47 Exemption from network charges for large electricity consumers (para 19 StromNEV) in Germany (Case SA.34045) [2013] OJ C128/43.

50 Guillaume Dezobry and Adrien de Hauteclocque the effect in principle of distorting competition in the sectors in which it is granted, whilst nevertheless being incapable, by its very nature, of achieving any of the objectives of the aforesaid exceptions.48

Three stages in the reasoning of the Commission must be underscored. First, the Commission is of the opinion that the measure constitutes an advantage for electro-intensive customers to the extent that exemption will relieve them from a charge they should normally assume. Therefore, the Commission stresses that: Electricity consumers normally have to pay a charge for using the electricity network. The exemption constitutes therefore a deviation from the principles applied in Germany for the establishment of network charges, as described under section 2.1 of this Decision. By exempting large electricity consumers from the network charges, the State has thus relieved them from a charge they normally have to bear as network users. The Commission therefore considers at this stage that the exemption thus confers an advantage to LEC.49

It noted that this exemption would not be granted rationally by network managers.50 In addition, it will be noted that the German government attempted to justify the exemption as being in return for services that electro-intensive customers, having regard to their consumption profile, render to the system. The Commission simply noted in this respect that: ‘Germany has merely stated that the exemption from network charges is justified on the basis of the stabilising effects that large electricity users have on the network without, however, substantiating this stabilising effect’.51 Second, the Commission observed that such a measure was selective: ‘The Commission considers that the advantage is selective. It is granted only to undertakings having an energy consumption of 7,000 hours of use and 10 gigawatt hour of energy’.52 Third, the Commission pointed out that the measure was attributable to the State and was financed by State resources.53 Finally, the reduction of charges, or exemption therefrom, related to the transmission of energy seemed to be an overriding issue for ensuring that Europe’s most energy-intensive industries remained competitive. However, compatibility of these schemes with applicable rules in the matter of State aid remains unclear. It should be noted that the existence of an advantage is not clear-cut. As a matter of fact, the consumption profile of electro-intensive undertakings makes it possible to provide services to network managers, who thereby can make savings (in terms of load balancing). Those savings may prove insufficient to justify a reduction (even exemption) of network access charges.

48

ibid, para 98. ibid, paras 27–28. 50 In paragraph 34 of its Decision, it notes that: ‘it does therefore not seem to be in the logic of the network charge system to fully exempt undertakings having a large energy consumption’. 51 ibid, para 97. 52 ibid, para 31. 53 ibid, paras 36 and 41. 49

State Aid and Price Regulation 51 ii. Preferential Tariffs and the Contribution of Electro-Intensive Undertakings to the Financing of Support Mechanisms for the Production of Renewable Energies The objective here is not to examine whether support mechanisms for renewable energies conform to State aid rules. The present study, conducted in chapter five, showed that such schemes could be considered as State aid, which might be compatible, provided certain conditions are met. The issue here is different. It has to do with the allocation of the expenses that finance such support mechanisms. In other words, who should contribute to the financing of these support mechanisms for renewables. More to the point, should the exemption or cap on the contribution of electro-intensive undertakings be regarded as State aid? And, if so, under which condition(s) can such aid be regarded as compatible? The EEG case deals with the conformity to State aid rules of the German law of 28 July 2011 relating to the new regulation of the legal framework for promoting electricity generated from renewables (the so-called ‘EEG’). The EEG calls for an increase in the share of renewables in the electricity supply to at least 35 per cent by 2020, and thereafter, incrementally up to a minimum of 80 per cent by 2050. In order to bolster the development of renewable energies, the legal framework calls for a support mechanism, which relies, in practice, on the end users of electricity. However, in order to protect electro-intensive undertakings, a cap has been provided for, depending on the level of consumption. In its decision of 25 November 2014,54 the European Commission ruled that this measure constituted State aid and that its compatibility was only recognised for certain categories of company. After an action was brought before the Court of Justice of the EU for annulment, the General Court delivered its judgment on 10 May 2016.55 In this judgment, which is discussed in greater detail in chapter fourteen on Germany, two points need highlighting for our purpose. The first relates to the rejection by the Commission and the Court of the arguments brought forward by the German government as to the absence of advantages conferred on electro-intensive undertakings. From the point of view of the German government, capping the contribution of electro-intensive undertakings does not constitute an advantage. Indeed, the mechanism put in place only aims at preventing a decline in the competitiveness of German companies, which have a regime imposed on them that either does not exist in the other States or, where it exists, includes specific terms and conditions for electro-intensive undertakings. In other words, for the German government, this cap makes it possible to offset a disadvantage that might be detrimental to industries operating in sectors highly exposed to international competition.

54 Support of renewable electricity and reduced EEG surcharge for energy-intensive users (Case SA.33995) Commission Decision 2105/1585 [2015] OJ L250/122. 55 Case T-47/15 Germany v Commission, EU:T:2016:281.

52 Guillaume Dezobry and Adrien de Hauteclocque The EU judge recalled in this regard that the fact that a Member State seeks, by unilateral measures, to approximate conditions of competition in a particular sector of the economy to those prevailing in other Member States cannot deprive the measures in question of their character as State aid.56

The Court finally ruled that the measure in question did indeed constitute an advantage for electro-intensive undertakings. The second point concerns the statement on the part of the Commission that the cap provided for electro-intensive undertakings would relieve those of a ‘burden they should normally bear’. This statement would imply that, for the Commission, the costs of financing the development of renewable energy must ‘normally’ be borne, at least partly, by electro-intensive undertakings.57 Yet, no link can be established a priori between the activities of electro-intensive undertakings and the financing of renewable energies. The decision to ask these to contribute to the development of renewable energies is purely a political choice. In other words, the fact of considering, with respect to certain customers, the cap on or exemption from the contribution to the financing of renewables as ‘an advantage’ is questionable, within the meaning of State aid rules. It could then be argued that there exists a difference in nature between the measures granting exemption from or reductions of the network access costs paid by the electro-intensive undertakings and the exemption from or cap on the expenses related to the development of renewables. Whereas the former is intrinsically linked to the activity of electro-intensive undertakings, which would result in regarding any reduction or exemption as an advantage granted to them within the meaning of the State aid rules, the latter are not. If one considers that the cap on the contribution of electro-intensive industries to the financing of renewable energies should be viewed as State aid, then the question of its compatibility arises. The Commission, in its 2014 EEAG guidelines, addresses this issue by pointing to the necessity for Member States to preserve the competitiveness of electro-intensive undertakings by limiting their contribution to the financing of renewables. The Commission underscored that: The aid should be limited to sectors that are exposed to a risk to their competitive position due to the costs resulting from the funding of support to energy from renewable sources as a function of their electro-intensity and their exposure to international trade. Accordingly, the aid can only be granted if the undertaking belongs to the sectors listed in Annex 3. This list is intended to be used only for eligibility for this particular form of compensation.58

56

ibid, para 56. In the State aid SA.37345 Decision related to Polish certificates of origin system to support renewables and the reduction of burdens arising from the renewables certificate obligation for energy-intensive users, the Commission noted that the Polish Energy Act ‘relieves [EIUs] from a burden related to the financing of the CO system that they would normally have to bear in the same way as all other electricity consumers’ (para 169). 58 Commission, ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG) [2014] OJ C200/1, para 185. 57

State Aid and Price Regulation 53 III. STATE AID, LONG-TERM CONTRACTS AND POWER-PURCHASE AGREEMENTS

Among the cases of State aid granted in the form of long-term purchasing contracts, two cases stand out. They involve Hungarian agreements (A) and Hinkley Point C (B).

A. The Hungarian Long-Term Contracts In the mid-1990s, the Hungarian energy sector had to be modernised. The most urgent objective was security of supply at the lowest possible cost. Therefore, the Hungarian government decided to enter into long-term power purchase agreements (PPAs) with investors who would agree to invest in the construction and modernisation of power plants in Hungary. In this context, the public company MVM signed contracts with power generators. As described by the Commission in its decision, [t]he PPAs require the power generators to duly maintain and operate their generation facilities. They reserve all or the bulk of the power plants’ generation capacities (MW) for MVM. This capacity allocation is independent of the actual use of the power plant. Beyond the reserved capacities, the PPA requires MVM to purchase a specific minimum quantity of electricity (MWh) from each power plant.59

While carrying out its control over such contracts and assessing their compatibility with State aid rules, the most interesting issue dealt with by the Commission was about the existence of an economic advantage. In order to find out whether or not such advantage existed, the Commission made use of the ‘private operator in a market economy test’. The Commission took as a reference a market operator subject to the same obligations as MVM and having the same opportunities. Noting that PPAs structurally provide more guarantees to generators than standard commercial contracts. The generators are thus in a more advantageous situation that the one they would face on the free market without their PPA. In order to complete the assessment of the existence of an advantage, it is necessary to assess the positive and negative effects that the public authorities could expect from the PPAs60

the Commission finally concludes that the contracts entail an advantage to the power generators beyond a normal commercial advantage. The same conclusion was reached by the General Court in its judgment: ‘the combination of ‘long-term capacity reservation, a minimum guaranteed off-take and

59 Decision on State aid C 41/05 awarded by Hungary through Power Purchase Agreements [2009] OJ L225/53, para 37. 60 ibid, para 217.

54 Guillaume Dezobry and Adrien de Hauteclocque price-setting mechanisms covering variable, fixed and capital costs’, as provided for by the PPAs, do not correspond to usual contracts on European wholesale markets’.61 The Commission finally decided that the PPAs were incompatible State aids. The General Court upheld this decision.

B. The Hinkley Point C Case In this case, the future operator of the Hinkley Point power plant had entered into a ‘contract for difference’ intended to ensure it would receive an income during the operational phase of the plant’s Unit C. Under this contract, the beneficiary is to receive a level of income to be determined on the basis of the wholesale market prices at which it sold electricity and on a compensatory payment corresponding to the difference between the predetermined exercise price (‘Strike Price’) and the Reference Price as recorded during the previous reference period. This exercise price was set at about €130 per MWh. In its Decision of 8 October 2014, the Commission found that State aid had been paid in accordance with the meaning of the Treaty provisions and that this measure was compatible with the internal market.62 As to this being qualified as aid, the Commission, after having duly noted that the conditions in the Altmark case law had not been fulfilled in the case at hand, quickly found that State aid had been given within the meaning of Article 107 paragraph 1 TFEU.63 As part of the analysis conducted on aid compatibility, three points must be made: first, the support mechanism put in place constitutes, in the opinion of the EU Commission, investment aid and not operating aid; second, the Commission acknowledges that investing in a power plant such as Hinkley Point C would not be feasible without British government intervention due to several market failures; and third, the Commission ruled that the mechanism was proportionate as far as the exercise price level was concerned.

IV. AID SCHEMES FOR RESIDENTIAL CUSTOMERS

Very few decisions of the Commission relate to actual aid schemes, aside from those support schemes earmarked for renewable energies. As far as schemes which introduce regulated retail tariffs are concerned, this chapter will address in more detail the French decision of June 2012 on standard tariffs (yellow and green) and return tariffs (see also chapter fifteen by Liliana Eskenazi). As to the selective nature of the advantage, the Commission considered that regulated tariffs favoured electricity users as compared with consumers using other energy sources. It also based its decision on the fact that there existed several types of tariff: standard tariffs and return tariffs as well as several categories (green and yellow). 61 62 63

Case T-468/08, Tisza Eromu kft v Commission, EU:T:2014:235, para 99. Hinkley Point (Case SA.34947) Commission Decision 2015/658 [2015] OJ L109/44. ibid, para 342.

State Aid and Price Regulation 55 Under such conditions, it was impossible, according to the Commission, to not conclude that regulated tariffs were selective. As to the concept of advantage, the Commission compared the tariffs with those prevailing in the market. After making adjustments for network access rates, the Commission noted that, each year, standard rates were systematically below market prices. As for the yellow and green return tariffs, this was on average the case over the period under consideration. As preferential tariffs were below market price and customer switching rates remained limited, this was a sign that an advantage had been granted. As for the issue of State resources and their imputability, discussion revolved around the role of the Caisses des Dépôts et Consignations (CDC—Deposit and Consignments Funds). To the extent that EDF is a State-owned enterprise and that the French State is heavily involved in this sector, that issue, as well as that of the impact on trade, did not raise any difficulties. The Commission therefore ruled that there was aid within the meaning of Article 107 paragraph 1 TFEU. The far more interesting question was whether such aid was compatible with the internal market. The Commission, first, quickly dealt with the question of how Article 106 paragraph 2 TFEU applied, and then moved on to discuss the basis of Article 107 paragraph 3, under c) TFEU. It found that regulated electricity rates in France may be regarded as compatible with the internal market under two conditions: the phase-out of regulated tariffs and implementation of the NOME law on the new organisation of electricity markets. Requiring implementation of the NOME law by the French government may be surprising, in light of the criteria laid down in Article 107 paragraph 3 under c) TFEU. The French decision of June 2012 has often been perceived as a decision of the Commission on the NOME law, which it is not. Implementation of the NOME law is only part of the commitments made by the French government, along with that of phasing out regulated tariffs, in the context of the aid procedure. The question raised, in this case, still remains how to apply the criteria provided for in Article 107 paragraph 3 under c) TFEU, in particular the one pertaining to proportionality. In fact, it is difficult to see to what extent the combination of regulated rates and implementation of the NOME law may be viewed as proportionate to a market deficiency, which has not been clearly defined. In such conditions, there also arises the issue of how much leeway is given to the Commission to accept commitments that have such a structural impact on the market in connection with aid procedures and a compatibility analysis under Article 107 paragraph 3 TFEU. It is obvious that the proportionality criterion cannot, under such conditions, be assessed in a strict manner. Such a debate brings to mind the discussions that took place in connection with the many twists and turns of the Alrosa antitrust case.64 It is useful to recall that the Commission has for a long time been engaging in market design by granting aid (eg, for shipyards), as was partly confirmed by its recent rulings of June 2012 in the financial sector in the wake of the 2009 crisis. Finally, it is also advisable to remember that, as regards the French decision of June 2012, there was necessarily an interaction between this aid

64

Case C-441/07 P Commission v Alrosa, EU:C:2010:377.

56 Guillaume Dezobry and Adrien de Hauteclocque procedure and the infringement proceedings, the former overshadowing, as it were, the latter. In the final analysis, the issue of the existence of an advantage never raised major problems in the case of aid schemes—in which it is quite obvious that this criterion was met. As a matter of fact, the private investor test does not seem to apply to compensatory mechanisms funded by a parafiscal tax. Selectivity is also being interpreted in a broad manner as far as aid schemes are concerned, as evidenced by the French decision of June 2012.

V. STATE AID AND NATIONAL PRICE REGULATION: A FEW REMARKS

The respective areas of competence of the domestic regulator and of the Commission, where the latter makes use of its antitrust powers (Articles 101 and 102 TFEU), are governed by established case law, which does not admit any ambiguity. Whatever the actions of the national regulator, the Commission always has jurisdiction to enforce competition law throughout the European Union. According to EU law, in order to fulfil the role assigned to it by the Treaty, the Commission cannot be bound by a decision rendered by a State authority and is entitled to take individual decisions at any time in order to apply Articles 101 TFEU and 102 TFEU, even though a domestic court has already ruled on an agreement or practice, and the decision considered by the Commission contravenes this ruling. The EU Court also confirmed that the existence of sectoral rules was of no importance when it came to assessing the proportionality of a Commission’s ruling on competition issues.65 In practice, the domestic energy regulator has often been involved in negotiating and monitoring commitments hammered out between enterprises and the Commission as part of antitrust proceedings. The implementation of ‘Virtual Power Plants’ or ‘Gas Release Programmes’ immediately springs to mind.66 However, the question of the optimal interface between sectoral regulation and competition policy remains open.67 The relationship between State aid control as practised by the Commission and price regulation as monitored by the regulator is an issue that is less well delineated. However, there is no denying the fact that decisions taken by the regulator in matters of tariffs may sometimes constitute an advantage for some operators. Since State aid control is based on the effects of a measure and not on its form, it is obvious that decisions taken by regulators may favour an enterprise over another. Moreover, according to established case law, infra-State bodies may hand out State subsidies. Admittedly, in the banking, or even in the energy sector, the Commission at times based its analysis on that of domestic regulators, for instance, with regard to the future shortfall in electricity generation capacity. Nevertheless, the prospect

65

See Case T-271/03 Deutsche Telekom v Commission, EU:T:2008:101, paras 120 and 199. See Commission, Report on Competition Policy 2002 (Luxembourg, Office for Official Publications of the European Communities, 2003) 39; Commission, ‘Commission clears Irish Synergen venture between ESB and Statoil following strict commitments’ (Press release of 31 May 2002) IP/02/792. 67 G Dezobry and F Fontaine, ‘Régulation et droit de la concurrence—Regards croisés sur la décision Deutsche Telekom’ (2008) 41 Droit de l’immatériel: informatique, médias, communication 17–21. 66

State Aid and Price Regulation 57 of permitting operators to challenge the decisions of domestic regulators before EU courts is altogether another kettle of fish. The European Court provided some welcome specifics in its ruling of 14 September 2016, Trajektna luka Split v Commission.68 In this case, a private operator managing the passenger terminal in the port of Split alleged that the port authorities there provided State aid to the State-owned ferry operator. According to that operator, the port authorities were capping the fees for the use of port facilities, as far as domestic traffic was concerned, at a level at least 40 per cent below those charged by other ports in Croatia, and at between 45 and 70 per cent below those charged for international traffic. As a result, the operator would have been forced to continue operating at a loss. According to the Commission, the measure in question did not involve State resources within the meaning of Article 107 paragraph 1 TFEU, to the extent that the financial resources flowed directly from one entity to another, without going through a public or private entity designated by State authorities to administer the transfer of funds. The Court endorsed the broad lines of this analysis, by pointing out that, although the port authorities set the rates, fees were directly paid to the claimant by users of the services. In these conditions, in light of the PreussenElektra judgment (see section VI below for more on that), the measure in question did not involve any transfer of State resources within the meaning of Article 107 paragraph 1 TFEU. The Court specified that, in the case at hand, it could not be considered that the amounts in question were constantly under public control owing to the mere fact that the operator was unable to set the price of its services. The Court found that the fact that the claimant was not free to dispose of its resources, because it could not set its own fares or the scope of its services, was simply due to its having accepted to deliver the services provided for by law in the maritime public domain at allegedly low rates, when it signed the concession agreement. The Court also stressed that the fact that the claimant had suffered losses was irrelevant for determining whether State resources had been employed. According to the Court, Article 107 TFEU aims to protect competition on the internal market and cannot be deviated from this end purpose, to be used as a means to question the financial terms and conditions of a concession contract that the concessionaire deems undue. Clearly, this judgment shows that the chances of challenging the decisions of the domestic regulator before EU judges are slim, and might conflict with the condition governing the use of State resources, a condition which will be addressed in the next (and last) section of this chapter.

VI. FOCUS ON THE NOTION OF STATE RESOURCES: PREUSSENELEKTRA VERSUS ESSENT THROUGH THE LENS OF VENT DE COLÈRE

The question of the implication of State resources as part of a specific public measure is one of the thorniest issues concerning State aid control in the energy sector.

68

Case T-57/15 Trajektna luka Split v Commission, EU:T:2016:470.

58 Guillaume Dezobry and Adrien de Hauteclocque This question crops up constantly, most recently, for instance, in the debate surrounding capacity mechanisms. This is a major issue when it comes to energy tariffs since, often, the State does not transfer resources directly from the public budget. As a matter of fact, resources are, more often than not, transferred among private market participants—typically— from the consumer to the producer of renewable energies, through a dedicated fund (the so-called levy). The general consensus is that the relationship between the condition of State resources and the condition of the funds has been largely resolved in the Vent de Colère ea case.69 Yet, as we shall see, the answer given by the Court is more ambiguous than it seems at first sight. Predictably, the Court confirmed in its decision that the public support mechanism for wind energy in France was imputable to the State and was financed from the State coffers. The question raised by the French Council of State, which seemed deceivingly obvious at first blush, was interesting in terms of its concrete consequences, since the support mechanism had no doubt made it possible to develop renewables in France (but at what cost?) Nevertheless, from a case law point of view, the Court passed up the opportunity to specify the concept of State intervention or the use of State resources in such cases, which has been a sensitive issue since the well-known ruling of the Court in the PreussenElektra case.70

A. Support Mechanism for Wind Energy, Main Proceedings and Preliminary Ruling Since the Court ruling highlighted the concrete functioning of the support mechanism for producers that use wind energy, it is useful, at the outset, to flesh it out. The support mechanism includes, on the one hand, an obligation to purchase, which is incumbent upon EDF and all distributors concerned and, on the other, a full compensation mechanism for any additional costs brought on by this obligation to purchase. The additional costs, calculated relative to market prices, are assessed by the ministries in charge of the economy and energy, upon a proposal from the National Energy Regulator (CRE). The funding of these additional costs is a burden that falls on end users that pay a contribution calculated pro rata to the amount of electricity consumed up to a certain ceiling. This contribution is also used to finance the operating expenses of the CDC, which are determined annually by the ministries concerned. That contribution is received by the suppliers themselves, who are thereafter supposed to pass it on to the CDC; the latter paying four times a year to the operators concerned the amounts corresponding to the collected amounts, which are regularised the following year, if they are not the same as the additional costs brought about by the obligation to purchase. Administrative sanctions are provided for failure to pay the contribution.

69 70

Case C-262/12 Association Vent de Colère! and Others, EU:C:2013:851. Case C-379/98 PreussenElektra, EU:C:2001:160.

State Aid and Price Regulation 59 In a decision dated 15 May 2012, following an action for annulment of the two rulings setting the conditions for the purchase of electricity produced by wind-driven installations, the French Council of State decided to seek a ruling from the Court on the interpretation of Article 107 paragraph 1 TFEU, and more specifically on the type of funding mechanism put in place for State intervention or for any intervention through State resources. Apparently, the Council of State (Conseil d’État) did not harbour any doubt as to the fact that the French support mechanism, not only granted a selective advantage to its beneficiaries, but was also likely to affect competition and trade. The line of reasoning of the Council of State, essentially incorporated in points 9 to 13 of the Court ruling, puts the issue confronting the Court in perspective. The Council of State, first, underlined, through its ruling of 13 March 2001, that the Court, in the PreussenElektra case, had ruled that a domestic regulation which, on the one hand, obliges distributors to buy at regulated prices the renewable energy generated in their supply area and, on the other, places on several enterprises the burden of the additional costs related to the obligation to purchase, did not constitute an advantage granted directly or indirectly through the use of State resources. It also noted that the current French system was no longer financed by contributions from the producers, suppliers and distributors mentioned in the law, but by contributions from all end consumers of electricity in France, the price of which was set by the energy ministry on a proposal from the CRE. Finally, the Council of State cited the Essent case, in which the Court had ruled that the price supplement imposed by the State on electricity buyers to finance the system was indeed a tax (the funds remaining under State control) and must be regarded as State intervention or as an intervention through State resources. As a result, according to the Council of State, the resolution of the dispute in the main proceedings hinged on whether the new mechanism—owing to changes in the way that additional costs were covered— should be viewed as State intervention or as an intervention through State resources. At first sight, the question of the Council of State seems to only deal with the legal significance of the transfer of the funding of enterprises to end consumers and, on the other hand, with the existence of a mandatory contribution, revised annually, to be paid by the latter. Most likely, the reference to the Essent71 case and, in particular, to the issue of public control over the contributions imposed on end consumers, justified the deeper analysis of the role of the CDC made by the Court.

B. The State Origin of Resources in the Vent de Colère Case In points 16 to 18, the Court dealt quickly with the issue of the measure’s imputability on the basis of the Stardust ruling.72 To the extent that the support mechanism was instituted by a law, said mechanism was imputable to the State, which was really a non-issue, unlike the question of the State origin of resources.

71 72

Case C-206/06 Essent Netwerk Noord, EU:C:2008:413. See ch 1 of this volume on the issue of imputability.

60 Guillaume Dezobry and Adrien de Hauteclocque In this regard, before examining the circumstances of the case, the Court recalled the now established case law whereby the concept of State origin of resources includes the advantages granted by a public or private body, designated or set up by that State with a view to administering the aid in question. On the other hand, it is not necessary that the financial resources in question be at all times in the possession of the public treasury to qualify as State resources; it just suffices that they be constantly under public control. The Court then proceeded in three stages. First, under points 22 to 27, it found that, based on case law, the fact that undertakings, subject to the obligation to purchase, kept the contributions received as long as these did not cover the aggregate amount of extra costs, was not likely to preclude the existence of an intervention through State resources, even though all funds did not transfer via the CDC. It recalled that, in the present case, contributions imposed on end consumers are entrusted to the CDC and determined by the minister in charge of energy, upon a proposal from the CRE, with administrative sanctions being provided for in case of non-payment of said contribution. Also, for purposes of insurance, the State must fully cover the extra costs related to the obligation to pay, if contributions collected by the CDC are insufficient, despite the threat of sanctions. Second, under points 28 to 33, the Court reviewed the role and the organisation of the CDC itself. It emphasised that the CDC, acting as intermediary in the administration of contributions, was a body governed by public law whose executive director was appointed by the Council of Ministers. Likewise, its supervisory board was composed of persons appointed by public institutions; it also had links with the CRE. Finally, the CDC was free to invest the contributions paid to it on the financial markets, its management expenses being deducted from the contributions paid by the end users of electricity. In such conditions, the Court found that the contributions managed by the CDC remained under State control. Third, in points 34 to 36, the Court strove to link its analysis to the PreussenElektra case. Essentially, the Court found that the facts of the present case differed from those having given rise to the PreussenElektra ruling, to the extent that, in the latter case, the funds had never been under public control and that there was no compensation mechanism, such as the one in question in the main proceedings, organised precisely by national legislation and by which the State ultimately guarantees the smooth functioning of the system. Given these circumstances, the Court found that Article 107, paragraph 1 TFEU should be interpreted as follows: [A] mechanism to fully cover the extra costs imposed on undertakings owing to the obligation to purchase wind energy at a price higher than the market price, the financing of which is effected by all end-consumers of electricity on the French territory, such as the one resulting from Law N° 2000-108, is tantamount to an intervention with State resources.

The Court finally refused to limit the impact of its ruling over time as requested by the French government, in view of the substantial amounts involved. On the one hand, the French government, according to the Court, could not, in good faith, overlook the ban on carrying out aid measures. On the other, the Court recalled that, by virtue of established case law, the financial consequences which might

State Aid and Price Regulation 61 result for a Member State from a preliminary ruling did not justify limiting its impact over time. C. PreussenElektra Versus Essent: The Ambiguous Contribution of Vent de Colère ea Hence, as rightly posed by the Council of State, the key question in the present case was to know whether the system for covering extra costs made the French mechanism for supporting wind energy and the mechanism ‘à la PreussenElektra’ or ‘à la Essent’ more alike. With regard to the prerogatives and the governance of the CDC and to the role as last resort guarantor of the French State, very little doubt remained as to the fact that the French mechanism belongs to the second category. The extensive analysis conducted by the Court did not, however, result in a robust analytical framework likely to clarify the rules of the game once and for all. In this type of case, two central questions arise as to the State origin of the resources. First, what is the actual burden borne by the State? This is a core issue because that type of mechanism generally aims to make up for market deficiencies by facilitating investment in non-mature or capital-intensive technologies. Second, what is the degree of control exerted by the State over money flows? As to the question of the financial burden actually borne by the State, point 26 of the Vent de Colère ea ruling found that the French State was required to fully cover the additional costs imposed on undertakings, assuming that the amount of contributions collected from end users of electricity was insufficient to cover these extra costs. In light of the preliminary question, which dealt with the financing mechanism, it was not possible to conclude that the support mechanism concerned necessarily implied the existence of State resources. Nevertheless, due to the fact that the obligation to purchase mainly concerned EDF, a quasi-monopolistic, State-controlled enterprise, there was little doubt as to the existence of State resources. As for the degree of control of State over money flows, it follows from case law that contributions need not be constantly in the possession of the State or the body set up for the purpose of managing the aid; the fact that they remain at all times under public control and therefore at the disposal of the competent domestic authorities suffices for them to be qualified as State resources. The Court, through a somehow circuitous line of reasoning, based itself on a body of evidence to answer that question. Unfortunately, it did not clarify which concrete circumstances must be considered as necessary or sufficient for the purposes of establishing ‘constant control’ over resources in this type of case. In this instance, the question of ‘constant control’ by the State over contributions was not immediately obvious, since, as is evidenced by point 27 of the ruling, companies that are subject to the obligation to purchase keep the contributions received from end consumers until they have covered their extra costs, so that part of the contributions do not transit through CDC. The Court showed surprising caution on this issue. As may be seen from points 22 to 26 of the ruling, it emphasised the concurrency of five factors: (i) the existence of a contribution imposed on end consumers, and no longer on undertakings in the sector;

62 Guillaume Dezobry and Adrien de Hauteclocque (ii) the existence of administrative sanctions for non-payment of the contribution; (iii) the existence of a third-party body; (iv) the fact that the mechanism is defined precisely, in particular as regards changes in the amount of contributions, by way of legislation and regulations; and (v) the role of the State as guarantor of last resort. Based on these five factors, the Court just concluded that existence of an intervention through State resources could not be ruled out. However, one might at once conclude that there was ‘constant control’ over these resources due to the fact that CDC and the French State must provide ongoing accounting control over the contributions and full coverage of the extra costs related to the obligation to purchase, and that, therefore, intervention had taken place through State resources. The Court seems to endorse a quite strict interpretation of the concept of ‘constant control’ of resources, instead of clarifying it fully; by contrast, one may wonder about the issue of whether, in the absence of such and such criterion, or of a specific combination thereof, it would have been possible to rule out the presence of State resources, thereby clarifying the outline of mechanisms à la PreussenElektra. Because the existence of an intervention through State resources could not be ruled out in view of the five factors above, the Court thought it necessary to assess how CDC concretely operated in order to conclude to the existence of State resources or, rather—and this is more ambiguous—to the fact that ‘[contributions] managed by the CDC Deposit Fund must be regarded as remaining under public control’ (point 33). In reading points 28 to 33 of the ruling, it seems difficult to move beyond the Essent category, when a third-party organisation is in charge of the system and when, as in this case, we are dealing with a public legal entity over which the State exerts real influence, and which keeps some degree of leeway as to the use of contributions. As we have seen, the Court has constantly held that, with respect to the condition relative to the existence of State intervention or intervention through State resources, the aid granted to public or private bodies, if they have been set up or designated by the State for this purpose, are also relevant. This does not teach us anything on how to tackle this problem in concrete cases. The Vent de Colère ea case is of interest in this regard because it highlights three dimensions of the problem: first, is the said organisation public or private?; second, what is the influence of the State over its funding, organisation and appointment of its directors?; third, what are its prerogatives (eg, can it impose sanctions?); and, above all, what is the effective control exerted by that organisation over contributions? Admittedly, it might have been desirable to establish a presumption (rebuttable?) as to the first question, by introducing a principle according to which the existence of a public organisation, through which funds may transit, is tantamount to the existence of public control over resources, as propounded by Advocate General Jääskinen in points 42 to 47 of his conclusions.73 However, this would not have altered, in the final analysis, the findings of the Court, in so far as only a part of the contributions transited through CDC.

73

Case C-262/12 Vent de Colère and Others, Opinion of AG Jääskinen, ECLI:EU:C:2013:469.

State Aid and Price Regulation 63 Interestingly, the Court, in the Vent de Colère ea ruling, never concluded on the existence of a ‘constant control’ by the CDC over contributions. It only concluded on the existence of public control, by the CDC, over the contributions transiting through it, and therefore to partial control, ie, control over only a part of the contributions. Are we, then, to conclude that the Vent de Colère ea case lays down the principle that partial control of resources may be sufficient, in particular, where such control is extensive enough? In fact, by insisting as it did on the public status of CDC and on its leeway in managing contributions, as well as on the role of the State in its governance, the Court apparently held, on the basis of the five abovementioned factors, that the existence of partial control constituted, in this case, State intervention or intervention through State resources within the meaning of Article 107, paragraph 1 TFEU. Against this backdrop, to claim that the Court might have ruled, once and for all, on the issue of the State origin of resources with respect to this type of support mechanism would be premature, at the very least. It is a safe bet to assume that Member States, in order to evade classification of the aid, will continue to devise ingenious ways to structure such mechanisms, since the present state of the law still grants them some leeway. Nevertheless, the legal issue here discussed might become moot, at least in the energy field, because a scheme à la PreussenElektra might no longer be capable of meeting the new challenges ahead. The huge investments required for achieving the new objectives of the European energy and climate policy won’t happen without the help of private operators. Yet, it is hard to imagine how the latter will accept to durably dispense with State guarantees when making investments on such scale? Finally, it must be emphasised that the Vent de Colère ea ruling is of interest, not only for all the questions it raises as well as for those it does not. Indeed, as regards qualification of the aid, the question of who benefits from the advantage remains open, as is that of the application of the Altmark74 judgment mentioned by the Advocate General in its findings. Likewise, with regard to the possible compatibility of the aid with the internal market, the impact of the guidelines on energy and the environment and the very real question of the application of Article 106 paragraph 2 TFEU, will provide a great many legal debates in the future.

D. The EEG 2012 Case Recent cases have not upended the results of the Vent de Colère ea case.75 The judgment of 10 May 2016, Germany v Commission must again be cited,76 since it has enabled the Court to clarify the scope and extent of the PreussenElektra ruling. In this case, according to the Court, the duties of managing and administrating this system can be likened, in view of their effects, to a State concession. Indeed, the funds used to operate such a system, which comprise of the extra costs passed on to 74 75 76

Case C-280/00 Altmark Trans and Regierungspräsidium Magdeburg, EU:C:2003:415. eg, Case C-275/13 Elcogás, EU:C:2014:2314. Case T-47/15 Germany v Commission, EU:T:2016:281 (under appeal).

64 Guillaume Dezobry and Adrien de Hauteclocque end users and paid by electricity suppliers to TSOs (Transmission System Operators), do not transit among independent economic operators. These funds are accounted for separately and are solely being allocated to the financing of support and compensation schemes, and for no other purpose. Therefore, the Court found that such funds remain under the dominant influence of public authorities. Next, the Court held that the amounts involved, generated by the EEG levy, did entail a State resource amounting to a tax, because electricity suppliers usually pass on the financial burden resulting from the EEG levy to end customers and because the impact on end consumers must be viewed as an anticipated consequence organised by the German lawmaker. The Court also emphasised that TSOs are controlled by the State in their task of managing the aid system for EEG power generation. As such, the role of such organisations is not that of economic entities acting freely on the market for the purpose of making profits, but their role is delineated by the legislator. TSOs are, also, under the obligation to manage the monies collected, in accordance with applicable legislation, in a specific joint account, under the supervision of public authorities.

VII. GENERAL CONCLUSION

Public intervention on energy prices is nothing new. However, its reach has undoubtedly increased since the process of liberalisation began. Member States now tend to distort prices for renewable energy, nuclear and back-up capacities (through so-called capacity mechanisms). Producers, household customers and energyintensive users almost all benefit from some sort of preferential tariffs. Twenty years after liberalisation started, Member States are interfering with the price formation mechanism all along the energy supply and consumption chain. The free market segment of the industry indeed gets narrower and we can wonder whether energy markets still deserve the label ‘liberalised’. Given the new objectives we have, in particular climate change mitigation and affordability, we cannot but conclude that public intervention on energy prices at national level is here to stay. In this context, EU State aid control remains a key tool to regulate in the wider European interest. The Union institutions have developed an important body of case law and decisions on price regulation and EU State aid law. Major difficulties nonetheless remain. Among them, we can highlight, for instance, the issue of the existence of an advantage, as it is not always easy to determine what would have been the competitive price without public intervention. Another is the issue of the involvement of State resources, which constantly resurface in new guises. In the new context, tailormade individual aid schemes for targeted customers (as opposed to producers) seem difficult to implement. Regimes favouring a category of users by exempting them from a component of the energy price should remain a common feature of the energy policy landscape. As renewable support schemes will progressively be rolled back while technologies are maturing, the key issue in the coming years will be the protection of energy-intensive users, in order to avoid carbon leakage. A careful balance will have to be reached, not only to avoid distortions in downstream markets, but also to ensure that the tariff deficit does not threaten the viability of the whole system.

3 State Aid and Taxation: Special Focus on Energy and the Environment LENA SANDBERG

I. INTRODUCTION

T

HIS CHAPTER DISCUSSES: (i) the presence of State aid through tax exemption and derogations from levies, or surcharges, as well as (ii) the compatibility of such tax measures with the internal market, in particular, pursuant to the European Commission ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG).

II. THE PRESENCE OF STATE AID

A. General Remarks As has been described in chapter one, in order to establish that State aid is involved, the following four conditions must be met: (i) the measure must confer on recipients an economic advantage which is not received under normal market conditions; (ii) the advantage must be granted by the State or through State resources; (iii) the measure must be selective by favouring certain undertakings or the production of certain goods; and (iv) the measure must distort competition and affect trade between Member States. These four conditions for the presence of State aid are cumulative. Therefore, if only one condition is not met, State aid will not be present. Normally, it is fairly easy to establish that a tax exemption involves the transfer of State resources since the State foregoes revenue it would otherwise have collected. It is usually also simple to establish that a tax exemption involves an advantage for the beneficiary who will have a lower tax burden. However, the criterion which most often gives rise to problems is selectivity. The following will therefore focus on the selectivity criterion in the context of energy taxation.

66 Lena Sandberg B. Presence of State Aid through Tax Derogations i. The Requirement of Selectivity The selectivity criterion is broadly interpreted. For the purposes of the establishment of selectivity, it is thus irrelevant how many may benefit from the tax exemption. In other words, the fact that the majority of potential taxpayers may be exempt from the full tax does not in and of itself remove the selective nature of the tax exemption. Take for example investment loan schemes exclusively available to SMEs under which they benefit from interest rates lower than market rates.1 This is a selective measure despite the fact that SMEs comprise about 99 per cent of the EU economy.2 In many cases the selectivity criterion does not pose problems. For example, if all companies are subject to a CO2 tax while energy-intensive companies are granted an exemption, it is clear that the energy-intensive companies are selectively favoured over other companies.3 However, ‘general measures’, that is to say, tax measures which apply uniformly to all undertakings, are not selective and therefore do not involve State aid. It can be difficult to determine whether a tax measure based on criteria that apply equally to all is selective. For example, the General Court has found that a complex investment and lease scheme under which investors in Spanish ships could benefit from a tax rebate on capital gains was not to be selective, despite the fact that it was conditional upon investments in the Spanish shipping sector. The General Court pointed out that the tax rebate was available under the same conditions to any potential investor.4 Finally, even if it has been determined that a tax exemption is selective, it may nevertheless be justified by what is referred to as the ‘nature and the logic’ of the (national) system. In other words, although a tax exemption may be prima facie selective, the tax exemption may be justified by the national tax system if the exempt undertakings are generally treated differently for tax purposes (eg, cooperatives versus other types of company).5 This is, however, a justification which is less and less used and is therefore of less practical relevance. a. The Reference Framework The EU courts have designed the so-called ‘reference framework’ methodology for determining whether a tax measure is selective. The methodology comprises three steps. The first step is to determine the group of undertakings covered by a tax, which is referred to as the ‘system of reference’ or ‘reference framework’. The second step is to identify whether there is any selectivity—that is to say, an exemption or reduction in the tax. The third and final step is to consider whether the tax exemption

1

SME Interest Subsidy Programme Saarland (Case N 341/2003) [2004] OJ C13/13. See Commission’s DG ‘Internal Market, Industry, Entrepreneurship and SMEs’, available at: www.ec.europa.eu/growth/smes_en. 3 Bulgaria support to energy-intensive users in the form of a reduced surcharge (Case SA.45861) [2016] OJ C425/1. 4 Joined Cases T-515/13 and T-719/13 Spain v Commission, EU:T:2015:1004. 5 Case C-78/08 Ministero dell’Economia e delle Finanze v Paint Graphos, EU:C:2011:550. 2

State Aid in the Form of Taxation and Emission Costs

67

or reduction may be justified by the nature and logic of the tax system which, in the affirmative, would remove the selective nature of the measure. The reference framework methodology was applied in the Gibraltar case. The Gibraltarian Corporate Tax Act provided for an exemption for offshore companies. However, following the Commission’s objections that this would involve State aid,6 the Gibraltarian government reformed its tax act and included a series of criteria which when fulfilled would give rise to a reduced tax burden. Nevertheless, the Commission still found that while prima facie, all companies could indeed fulfil the new criteria in the reformed tax act in order to be exempt from corporate tax, implying that the tax regime was a general measure, de facto, however, meant only offshore companies could benefit from the exemption, and the measure was thus found to be selective.7 Upon appeal, the General Court disagreed with the Commission.8 However, the CJEU decided that the Gibraltarian tax act had to be considered as a whole, and found the reformed tax act to be selective.9 In particular, the CJEU held that for selectivity purposes account should be taken not only of those initially subject to the tax, but also of those who in principle would not be subject to the tax but could be in light of its objective. Otherwise, the State aid rules could easily be circumvented by tax measures that formally apply to all undertakings while in practice they favour only certain undertakings. In other words, account should be taken of the effect of the tax measure. Therefore, in the Gibraltar judgment—which is also one of the key judgments regarding tax selectivity in energy taxation cases—the CJEU did not apply the reference framework methodology in a straightforward manner. Nevertheless, the reference framework methodology for identifying selective measures has remained a key parameter used by the Commission in its energy taxation cases.10 b. Justification by the Nature and Logic of the System As mentioned above, and according to the third step of the reference framework, a selective advantage may still be justified by the nature and logic of the tax system.11 While the assessment of whether a measure is justified by the nature of the logic of

6 Gibraltar government corporation tax reform (Case C66/22) Commission Decision 2005/261/EC [2005] OJ L85/1. 7 The reformed tax act introduced on all companies making a profit a payroll tax on employees and a business property occupation tax. However, both taxes were capped to an amount corresponding to 15% of the profits. The Commission found that the tax measure was selective insofar as it exempted non-profit companies from payroll and business property occupation taxes (although these undertakings had many employees and occupied many business premises). The Commission also found the tax measure to be selective since very profitable companies which were liable to pay both payroll and business occupation tax benefited from a reduced tax burden (by being taxed an amount corresponding only to 15% of the profits). In other words, this de facto favoured offshore companies. 8 Case T-211/04 Government of Gibraltar v Commission, EU:T:2008:595. 9 Case C-106/09 P Commission v Government of Gibraltar and UK, EU:C:2011:732. 10 eg, Modification of the CO 2 tax for quota-regulated fuel consumption in the industry (Case 41/2006) [2009] OJ L345/18. 11 eg, in the Gibraltar case the de facto tax exemption for offshore companies could not be justified by the general logic of the Gibraltarian tax system which was to collect taxes from corporations.

68 Lena Sandberg the tax system would appear to require a consideration of the objective of the tax and the exemption, this is not the case according to the Court’s ruling in the British Aggregates case. In the British Aggregates case which concerned the extraction of minerals for aggregates (primarily used in construction)12 the UK had imposed an environmental levy on minerals (including granal) that were used to produce virgin aggregates because mineral abstraction for this purpose is damaging for the environment. Recycled aggregates (which are less damaging for the environment) were exempt from the levy in order to give an incentive to use recycled aggregates.13 The Commission found that the tax exemption was prima facie selective because users of the recycled aggregate were favoured, but then found the exemption to be justified by the nature and logic of the national tax regime in light of its environmentally friendly objective. The Commission thus concluded that the tax exemption was not selective.14 Upon appeal by the British Aggregate Association (BAA), the General Court upheld the Commission’s decision. In this context the BAA had argued that if the logic of the levy regime would genuinely be that abstractions damaging the environment should be subject to the levy, then mineral abstraction for uses other than aggregates which are equally damaging to the environment should also be taxed, which was not the case. In other words, the exemption for environmentally friendly aggregates could only be justified if that logic would be coherent, which would require the taxation of all mineral extraction causing damage to the environment. However, the General Court dismissed this argument by stating that Member States have a degree of discretion for purposes of imposing environmental levies. Upon appeal by BAA to the CJEU, the latter held that the exemption from the levy did entail a selective advantage15 since State aid is defined in terms of its effect, not its objectives. In this regard, the CJEU criticised the General Court for taking an approach solely based on the environmental objective of the measure, thereby undermining the purpose of Article 107(3) TFEU. In light of the British Aggregates case, it is therefore not possible to justify a selective measure by reference to the (environmental) objective pursued by the tax (and the exemption) under the nature and logic of the national tax system. Rather, the objective of the measure is relevant only for purposes of determining whether the measure would be compatible with the internal market.16 c. Justification by the Nature and Logic of the System: CO2 Tax Exemptions In the British Aggregates case the justification of selectivity by the nature and logic of the system was refused because it was based on considering the objectives of the measure. However, justification of the nature and logic of the system has been

12 13 14 15 16

Case C-487/06 P British Aggregates v Commission, EU:C:2008:757. The levy did not apply to extraction of such minerals for uses other than aggregates. British Aggregates Levy (Case SA.34775) [2013] OJ C348/162. British Aggregates (n 12) para 79. ibid, para 92.

State Aid in the Form of Taxation and Emission Costs

69

refused for many other reasons. For example, in the Commission’s decisions on CO2 tax exemption cases, the Commission refused to consider the tax exemptions to be justified by the nature and logic of the system because the justifying measure (the Emission Trading Scheme) did not form part of the national tax system but merely part of the national legal system. Historically, the Scandinavian countries have always imposed relatively high taxes on undertakings’ CO2 emissions based on ambitious climate change agendas in these countries. However, when the Emission Trading Scheme (ETS) Directive was introduced in 2003, undertakings also had to ‘pay’ allowances for their emissions and they therefore had what the Scandinavian countries referred to as a ‘double burden’.17 As a result, these countries introduced tax exemptions for undertakings which were subject both to a CO2 tax and the ETS scheme, that is to say, the energyintensive companies. In the Commission’s decision on the Danish CO2 tax exemption, the Danish State had proposed a full CO2 tax exemption for energy-intensive undertakings, arguing that they had such a ‘double burden’ by being subject both to the ETS and the high CO2 taxes.18 The Commission considered that the tax exemption was selective. However, the Danish authorities argued that the exemption was justified by the logic of the national tax system since the CO2 tax and the ETS pursued the same environmental objective.19 In applying the reference framework methodology, the Commission found that the reference system consisted of the undertakings subject to the CO2 tax. However, while the ETS served to reduce carbon emissions, the CO2 tax served to collect revenues for the Danish State. Further, while the target groups of the two measures were almost identical there was still a slight difference. Therefore, the Commission considered that the ETS did not even form part of the tax reference system, and on that basis decided that the selective tax exemption could not be justified by reference to the nature and logic of the system.20 The Commission reached the same conclusion in the Swedish CO2 tax exemption case,21 which arose from similar facts to those in the Danish CO2 tax decision.22 In this case, the Commission did not even engage in a discussion regarding the selectivity and potential justification by the nature and logic of the national tax system. In light of those cases, it is clear that the Commission is of the opinion that the justification by the nature and logic of the tax system must stem from a reason which forms part of the tax system, not just the national legal system. Another way to look at this is that the Commission did not accept justification by the overall nature and

17 Council Directive 2003/96/EC of 27 October 2003 restructuring the Community framework for the taxation of energy products and electricity [2003] OJ L283/51. 18 Modification of the CO tax (n 10). 2 19 ibid, para 38. 20 Nevertheless, the tax exemption was, after some amendments, found to be compatible with the internal market, given that Member States would not introduce CO2 taxes in the first place if tax reductions cannot be considered compatible. 21 CO tax reduction for fuel used in installations covered by EU ETS (Case N22/2008) [2008] 2 OJ C184/3. 22 ibid, para 17.

70 Lena Sandberg logic of the legal system in those cases because the ETS and the CO2 tax pursued different objectives (the reduction of CO2 emissions, versus the collection of revenues). However, based on the British Aggregates case (where the CJEU rejected taking account of the objective of the measure), this should not even have been a possibility.

C. Presence of State Aid through Derogations from Levies or Surcharges While the preceding section has dealt with issues relevant for determining whether straightforward tax exemptions involve State aid, this section addresses how to determine whether a parafiscal levy or surcharge involves State aid. While it is usually not problematic to determine whether a levy or surcharge scheme is selective and whether they grant an advantage to the beneficiaries, these schemes often involve multiple level financing structures and, for that reason, the most challenging part is most often to determine whether the measure in question involves State resources. Therefore, the following focuses on State resources. Parafiscal levy schemes (or surcharge schemes) for renewable energy are in most cases based on the Directive on renewable energy.23 Article 27 of the Directive on renewable energy obliges Member States to produce a minimum amount of renewable energy on an annual basis.24 This obligation has been in force since the end of 2010 and has forced Member States to consider how to restructure their energy mix in order to ensure that the minimum renewable energy threshold is met. However, given that the costs for producing renewable energy by far exceeds the production costs of conventional energy, Member States had to tackle the issue of raising financing to cover the additional costs and thereby incentivise investors to take part in renewable energy projects.25 It is for that reason that several Member States established levy or surcharge schemes under which energy consumers usually foot the bill. i. The Requirement of State Resources Parafiscal levy schemes have traditionally been used in the agricultural sector where charges are levied by public or private bodies on the production or marketing of agricultural products with a view to financing activities for the benefit of the sector as a whole.

23 See Case C-677/11 Doux Elevage, EU:C:2013:348; Case C-206/06 Essent Netwerk Noord and Others, EU:C:2008:413; Case T-251/11 Austria v Commission, EU:T:2014:1060; Joined case C-393/04 and C-41/05 Air Liquide Industries Belgium, EU:C:2006:403; Case C-379/98 PreussenElektra, EU:C:2001:160; and Case C-78/76 Steinike and Weinlig, EU:C:1977:52. 24 Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources, amending and subsequently repealing Directives 2011/77/EC and 2003/30/EC [2009] OJ L140/16. 25 Although the costs of operating a renewable energy plant (eg, wind power plant) are marginal, it is the upfront investment costs for renewable energy which must be factored into the sales price of renewable energy that can cause it to be higher than the conventional energy sales price.

State Aid in the Form of Taxation and Emission Costs

71

The CJEU first ruled on parafiscal levies in the Steinike and Weinlig case.26 The case involved a German importer of citrus juices which, like other companies in the sector, was required by law to pay a contribution to a fund, established and controlled by the State. The fund used the proceeds to promote the sale and exports of German agricultural food and forestry products by organising fairs and collective advertising for the benefit of those who had contributed to the fund. The CJEU held that although the companies contributing to (and benefiting from) the fund were private, the contributions were administered by a State-controlled fund. This essentially meant that the contributions should be considered as having been paid to the State and consequently any payments made from the fund to beneficiaries was be considered as State resources. The CJEU’s position in the Steinike and Weinlig case has been consistently applied by the Commission in subsequent decisions which have been upheld by the European courts.27 In summary, the Court has established that State resources are involved when the State controls (i) the payment of contributions (by for instance making them compulsory)28 to an intermediary controlled by the State; and (ii) the State controls the intermediary’s disbursements of such funding to beneficiaries. In those circumstances the State controls both the collection of the funds and how the proceeds are spent.29 In this regard, the Court has also clarified that it is irrelevant whether the body is public or private as long as it is controlled by the State.30 The Commission has applied the theory on parafiscal levies in several energy cases involving contributions paid to an intermediary to finance the generation of renewable energy. While some cases clearly involve a situation where the funding is collected by a State-controlled body which also determines how the proceeds are spent, this is not the situation in all cases. Recently, the Commission has stretched the

26

Steinike and Weinlig (n 23). In Case C-259/85 France v Commission, EU:C:1987:478 concerning the payment of contributions by clothing and knitwear industries to a committee which used the proceeds for the promotion of textile and clothing companies through promotion campaigns and technical centres, the Court confirmed that a system of parafiscal charges levied in an industry is not sufficient to bring it outside the scope of Art 107(1) and confirmed its character of aid. 28 This is just an example and compulsory payment is not a requirement for State resources to be involved. 29 In Doux Elevage (n 23), the Court seemed to reason that there was no parafiscal levy involved as there was no State-controlled intermediary and the disbursements of the funds were not controlled by the State. In Case C-345/02 Pearle BV, EU:C:2004:448 paras 36–38, the Court pointed out that, since the costs incurred by the public body for private purposes were offset in full by the levies imposed on the undertakings benefiting therefrom, there was no advantage which would constitute an additional burden for the State or that body, thus there was no State aid. In both, Joined Cases C-72/91 and C-73/91 Sloman-Neptun, EU:C:1993:97, and Case C-189/91 Kirsammer Hack, EU:C:1993:907 the Court found that there was no parafiscal charge involved, which seems to be due to the fact that there was no intermediary involved. The Court’s doctrine on parafiscal levies has been consistently applied by the Commission, see eg, Commission Decision SA.37499 of 27 November 2013, Netherlands parafiscal levies on flower bulbs for technical support, paras 36–39. 30 See Case C-57/86 Greece v Commission, EU:C:1988:284, paras 12–13. In Case C-126/01 GEMO SA, EU:C:2003:622, para 23, the Court established that according to settled case law, it is not appropriate to distinguish between cases in which aid is granted directly by the State and those in which it is granted by a public or private body designated or established by that State. 27

72 Lena Sandberg theory on parafiscal levies to situations where there was no intermediary and concluded that the measure involved a transfer of State resources solely on the basis of the fact that the State controlled how the funding is collected and spent. ii. The Role of a State-Controlled Intermediary a. PreussenElektra The PreussenElektra case concerned a German law from 1990 which required electricity supply undertakings to purchase electricity produced within their area of supply from renewable energy producers at minimum prices (above the market price of electricity).31 An amendment of the law in 1998 introduced the possibility for the suppliers to recover the additional costs from network operators, that is, distribution system operators (DSOs). PreussenElektra, a DSO, brought an action to a German court claiming recovery for the compensation PreussenElektra had paid to an electricity supplier to cover its additional costs, caused by the purchase obligation. PreussenElektra argued that its payment amounted to non-notified State aid to the renewable energy producers. The national court referred the question to the CJEU. The CJEU clarified that the German law which imposed an obligation on electricity suppliers to purchase electricity produced from renewable energy sources at minimum prices did not constitute State aid, as there were no State resources involved. The obligation to purchase electricity at a minimum price served merely to reallocate funding from the suppliers to renewable energy producers, without which such funding passed through the State coffers. Therefore, the measure did not involve a direct or indirect transfer of State resources.32 As stated above, the German law entitled the suppliers (who were subject to the purchase obligation) to finance the purchase obligation by passing on all their costs to the DSOs, such as PreussenElektra (which was how the case arose). This is important since the Commission has subsequently distinguished the PreussenElektra case from other cases precisely by arguing that in the PreussenElektra case the suppliers had to bear some of the costs of the purchase obligation themselves and thus were not able to pass on all the costs of the purchase obligation. b. Essent The Essent case involved the collection of a surcharge on electricity customers by electricity network operators which had to transfer the funding to a designated company (SEP, a joint subsidiary of electricity producers). The surcharge was collected through an increase in the price of electricity charged to final consumers

31

PreussenElektra (n 23). ibid, para 59. The Court also pointed out that indirect effects, such as the reduction in tax revenues for the State as a result of the deterioration of the economic position of the companies due to the purchase obligation, are irrelevant. The Court held that such consequences are an inherent feature of the legislation at issue and could not be regarded as granting an advantage to the renewable energy producers at the expense of the State; ibid, para 62. 32

State Aid in the Form of Taxation and Emission Costs

73

(although certain large customers were exempt). The charge was imposed to finance stranded costs which electricity generators incurred prior to the liberalisation of the electricity market. The CJEU held that since the funding originated in the surcharge imposed by the State, the designated company was obliged to use the funding for compensating electricity producers and was required by law to have a detailed account of ingoing and outgoing sums, as it involved State resources. The CJEU distinguished this case from PreussenElektra on the basis that in that case ‘[t]he undertakings had not been appointed by the State to manage a State resource, but were bound by an obligation to purchase by means of their own financial resources’ (emphasis added).33 Yet, as stated above, in PreussenElektra the German law specifically stated that the electricity suppliers, subject to the purchase obligation could recover their costs from the DSOs, which was exactly the reason that the DSO, PreussenElektra, brought a claim to the national court.34 A more plausible reason for distinguishing the Essent case from the PreussenElektra case would appear to be that in the PreussenElektra case there was (allegedly) no intermediary involved. c. Austria’s Green Electricity Act The Commission adopted two decisions on the Austrian Green Electricity Act: (i) in the decision adopted 22 July 2009,35 the Commission considered that a purchase obligation on network operators to buy green electricity at prices above market prices, which was financed by a charge on electricity suppliers, involved State aid to the green electricity producers; (ii) in the decision adopted on 8 March 2011,36 the Commission considered that the same charge which was financed by final consumers except for certain large energy-intensive consumers involved State aid to the energy-intensive companies. In both cases electricity suppliers were required to purchase a certain volume of renewable electricity at a price above the market price (the clearing price) fixed by the Austrian authorities. The purchase obligation was financed through a so-called ‘electricity charge’ which constituted the difference between the market price and the clearing price. The electricity charge was imposed on electricity suppliers who had to pay it to a settlement centre (Abwicklungsstelle für Ökostrom AG, OeMAG). OeMAG was governed by public law and entitled to collect the electricity charge in Austria and control the distribution of the funds. Electricity suppliers were free to recuperate the costs of the electricity charge from consumers, and so the electricity charge was therefore ultimately financed by final consumers. The Commission held that, given that the electricity charge was paid to an intermediate body, OeMAG, controlled by the State, and that the State (through OeMAG)

33

Essent (n 23) para. 74. ibid; nothing indicated that the electricity suppliers could pass on only part of their costs to DSOs. See The Austrian Green Electricity Act (Case SA.26036) [2009] OJ C217/12. 36 See The Austrian Green Electricity Act (Case SA.26036) [2011] OJ L235/45, paras 60–87. It had already taken this position in its opening decision of 22 July 2009. 34 35

74 Lena Sandberg controlled the distribution of the funds, State resources were involved. This reasoning was applied in both cases. The electricity charge was considered to finance (i) the purchase obligation which benefited the renewable energy producers (decision of 22 July 2009); and (ii) the exemption which benefited the energy-intensive consumers (decision of 8 March 2011). Both groups were therefore considered to receive funding from State resources and thus State aid. The Commission’s decision of 8 March 2011 was upheld by the General Court on 11 December 2014.37 Both the Court and the Commission emphasised that the fact that the electricity charge was payable to a State-controlled intermediary that also controlled the distribution of the funds, distinguished this case from the PreussenElektra case.38 Interestingly, it was clear that just like in PreussenElektra, the suppliers in the Austrian Electricity Act could also choose to finance the costs of the electricity charge themselves. However, in this case the Commission did not consider that that feature brought it outside the scope of the State aid rules.39 This is probably because, in this case, there was actually a State-controlled intermediary involved. d. Vent de Colère The Vent de Colère case involved an obligation on DSOs to purchase electricity from wind power producers at a price above the market price. The costs were financed by a compulsory levy on final consumers which was paid to a State-controlled intermediary, Caisse des dépôts, that centralised the funding in a special account and then disbursed the funds to wind power producers. The Minister of Energy determined the amount to be paid to the beneficiaries. The CJEU held that State resources were involved as the State imposed a surcharge which went through a State-controlled intermediary to the beneficiaries.40 Indeed, the Vent de Colère case fulfilled all the conditions for the presence of State resources as it entailed a compulsory charge on final consumers paid to a State-controlled intermediary which, in turn, also controlled the amount paid to the beneficiaries. The CJEU also distinguished this case from PreussenElectra on that basis by stating that in PreussenElektra, the funds at issue could not be considered a State resource since they were not at any time under public control and there was no mechanism, such as the one at issue in the main proceedings in the present case, established and regulated by the Member State, for offsetting the additional costs arising from that obligation to purchase and through which the State offered those private operators the certain prospect that the additional costs would be covered in full (emphasis added).41

37

See Austria v Commission (n 23). ibid, paras 59, 67; See Case SA.26036, The Austrian Green Electricity Act (n 36) para 68. 39 In fact, there was also an ‘electricity charge’ in the PreussenElektra case through the obligation to purchase green electricity at minimum prices but there was no State-controlled intermediary. 40 See Case C-262/12 Vent de Colère and Others, EU:C:2013:851, paras 17–37. 41 ibid, para. 36. 38

State Aid in the Form of Taxation and Emission Costs

75

e. StromNEV In 2013, the Commission opened an in-depth investigation into the Stromnetzentgeltverordnung42 (StromNEV 2011), after receiving several complaints concerning an exemption from network charges for large electricity consumers whose energy consumption reached (i) 7000 hours of use; and (ii) 10 gigawatt hours of energy consumption.43 The exemption, estimated to amount to around €300 million in 2012, was to be financed by electricity consumers, who had to pay a special levy, ie, the so-called §19-surcharge. The financing mechanism in StromNEV 2011 required that Transmission System Operators (TSOs) first had to compensate DSOs for any losses they had incurred as a result of granting the exemption to large consumers. In parallel, the DSOs were required to collect the surcharge from the consumers and then transfer the proceeds to the TSOs.44 In the decision opening the formal investigation procedure the Commission found that the exemption to large energy consumers had been granted through State resources because: first, the State had imposed a special surcharge on electrical consumers designed to finance the exemption; second, the State had appointed an undertaking to administer the charge, namely the TSO (which functioned as an intermediary subject to State control); third, the State had established rules concerning the use and destination of the surcharge; and fourth, the State could monitor the financial flows through control mechanisms.45 In particular, the Commission pointed out that the State imposed the surcharge and fixed its amount and while the TSOs were privately owned, their handling of the funding was subject to State control and they could not use any proceeds exceeding the amount necessary for compensating the DSOs. In addition, the StromNEV 2011 required the TSOs to centralise the proceeds of the surcharge and equalise the financial burden between them, depending on their customer portfolio.46 The Commission also added that, despite the fact that the TSOs had to pre-finance the aid, neither the TSOs nor the DSOs bore the burden of financing it since they would be fully compensated through the §19-surcharge.47 When comparing the case at hand to PreussenElektra, the Commission argued that in PreussenElektra, the financial burden was borne also by the electricity suppliers since there was no compulsory surcharge to compensate them.48 However, as pointed out previously in PreussenElektra, the electricity suppliers were entitled to require compensation of

42

German Regulation on network charges for electricity. Exemption from network charges for large electricity consumers (§19 StromNEV) in Germany (Case SA.34045) [2013] OJ C128/43. 44 Moreover, an equalisation mechanism provided that the TSOs equalised the sum of their payments to DSOs and their own losses among them; they were however allowed to pass on the additional costs (stemming from the equalisation mechanism) to the final consumers. 45 See Exemption from network charges for large electricity consumers (n 43) para 49. 46 ibid, para 67. 47 ibid, para 50 f. 48 ibid, para 57. 43

76 Lena Sandberg the costs of the purchase obligation from the DSOs, which was precisely the reason that PreussenElektra brought the action in a national court. Again, it would appear as if the key issue was that in PreussenElektra there was no State-controlled intermediary, whereas there was in StromNEV. In fact, if it is voluntary to pass on the relevant costs, any rational commercial operator would, in any event, seek to pass on the financial burden to its customers. On 22 August 2013, the German State amended the StromNEV which now provides that TSOs can choose whether to pass on part of the costs (by requiring the DSOs to recuperate the costs from final consumers) and there is thus now no compulsory surcharge.49 Indeed, it is not clear why the Commission considers it to be decisive whether the funding is financed by the ultimate consumers or by the entities obliged to purchase or contribute to the costs of renewable energy as long as these entities pay the funds to a State-controlled intermediary which passes the funding on to preidentified beneficiaries. The question for the Commission is now whether it is prepared to conclude that State resources are involved in the StromNEV case merely based on the fact that the TSOs function as a State-controlled intermediary and that the State controls the destination of the funds paid by them.50 It seems that the Commission is still struggling with this question given that it is now more than three years since it opened the procedure in this case. f. The German Law on Renewable Energy Act: EEG On 25 November 2014,51 after receiving numerous complaints, the Commission opened the formal investigation procedure into the German Act from 2012 on renewable energy, the Erneuerbare Energien Gesetz (EEG-2012), which introduced a surcharge for the financing of renewable energy sources in Germany from which the energy-intensive industry was partly exempt.52 Since the EEG-2012 is a successor

49 While it is still referred to as a surcharge, the fact that TSOs are free to determine whether to impose it or not, means that it is not a surcharge in the typical sense. On 12 April 2016, the equalisation mechanisms of § 19(2) StromNEV 2011 and 2013 were declared void by order of the German federal court (Bundesgerichtshof), but on 29 July 2016 the Bundestag adopted a law which reinstalled the equalisation mechanisms of StromNEV 2013. 50 The German State also included in the StromNEV 2013 that if the TSOs decide to recuperate their losses, the TSOs must recover the costs from all consumers except with respect to large energy consumers from which a maximum of 0,05–0,025 Cent/kWh may be recovered. This is relevant for the compatibility of the aid. 51 See Support of renewable electricity and reduced EEG surcharge for energy-intensive users (Case SA.33995) [2015] OJ L250/122. 52 ibid, in its substantially amended version of 2012. The previous version of the EEG Act 2012, namely the EEG Act of 2000, was found by the Commission not to constitute aid in its Decision of 22 May 2002, Law on promotion of electricity generation from renewable energies (Case NN 27/2000) [2002] OJ C164/3, which was in line with the preceding system introduced in 1998 (based on a purchase obligation and found by the Court of Justice not to constitute State aid in PreussenElektra (n 23). In its Decision, the Commission followed the same line of reasoning and concluded that the system of support for renewable energy, which was based on a purchase obligation at minimum price, did not involve the transfer of State resources (see Press Release IP/02/739, 22 May 2002). In contrast, the Commission found in its Decision EEG 2014 (Case SA.38632) [2015] OJ C325/1, that the version of the EEG Act of 2014, involving support granted to producers of renewable electricity in the form of feed-in tariffs and market premiums did constitute aid, but was in line with the 2008 Guidelines on State aid for environmental protection.

State Aid in the Form of Taxation and Emission Costs

77

of the regime which was found not to involve State resources in PreussenElektra, this case is of particular importance. The network operators (in most cases the DSOs) were obliged to purchase electricity produced within their network area from renewable energy sources and from mine gas at prices above the market price. In parallel, such green electricity producers also had the possibility to sell their electricity directly on the market and obtain a market premium from the network operator, the amount of which was fixed by law. The TSOs were under the obligation to compensate the DSOs for the entire cost resulting from the purchase obligation and the market premium. The TSOs were entitled to ask the electricity suppliers to pay the remaining share of their financial burden in case the TSOs were unable to sell the electricity on the spot market for a price that would cover their costs of compensating DSOs. This charge, collected from the electricity suppliers, constituted the EEG-surcharge and fully compensated TSOs for their costs. However, while suppliers are generally free to recuperate their costs (of paying the TSOs) from final consumers, the EEG-Act 2012 capped the amount which may be recovered from energy-intensive users. The Commission found that there were two types of selective advantages involving State resources. First, EEG electricity producers received the additional revenues through the purchase obligation and/or the premium.53 Second, the Commission found that the benefit which the energy-intensive companies received through having a cap on the costs of the EEG-surcharge (which suppliers could recover from them) involved State resources. In this regard, the Commission relied on the Essent judgment and found that as the State imposed a surcharge (EEG-surcharge) to finance the sale of green electricity at above market prices, and had entrusted an intermediary, the TSOs, with the task of centralising and administering all financial flows obtained on the basis of the surcharge to finance the green electricity, State resources were involved. This was further substantiated by the fact that the TSOs had to keep a separate account for the EEG-surcharge and were not allowed to use the excess of the surcharge to finance other activities. Finally, the State monitored the financial flows through extensive control mechanisms.54 Upon appeal by Germany claiming that State resources were not involved, the Commission’s decision in the EEG case was upheld by the General Court on 10 May 2016.55 The Court held that (i) the TSOs were entrusted by the EEG with managing the system for supporting the production of renewable electricity; (ii) the EEG confers on the TSOs a series of obligations and rights as regards the financing mechanism, making the TSOs the central point in the financing operation; (iii) the costs of the renewable were passed on to final consumers; and (iv) the financing was

53 The EEG 2012 fixed a maximum on the amount of the EEG-surcharge which the TSOs could recuperate from the suppliers of green electricity (which were often also generators of green electricity)—even if, in practice, all electricity suppliers passed the surcharge entirely on to their customers. Perhaps due to the fact that suppliers may also be generators of green electricity, the Commission found that this was merely a feature supporting the system and did not involve State aid; see Support of renewable electricity (n 51) para 14 f. 54 See Exemption from network charges for large electricity consumers (n 43) para 96. 55 Case T-47/15 Germany v Commission, EU:T:2016:281.

78 Lena Sandberg the subject of a separate account in the TSO. The judgment of the General Court was appealed to the Court of Justice, and the appeal is still pending.56 iii. Conclusion In all the court cases deemed to involve parafiscal levies to finance electricity generators, namely Essent, Austrian Electricity Act, Vent de Colère, the State had imposed a compulsory surcharge on electricity suppliers or final consumers. However, in the case of the Austrian Electricity Act (which was considered to involve State resources), the surcharge consisted of the additional price for purchasing green electricity, which could arguably also be said to exist in the purchase obligation in the PreussenElektra case (which was considered not to involve State resources). Therefore, it seems questionable whether the compulsory surcharge is decisive for a finding of State resources. Rather, it seems that a decisive aspect is the existence of a State-controlled intermediary which administers the contributions it receives and has to spend the proceeds in accordance with State instructions. Given that in the StromNEV 2013 (replacing the StromNEV 2011, assessed by the Commission in its decision) the German authorities abolished the surcharge as a compulsory charge (and the network operators can choose whether to recover the costs from the final consumers), it is not clear whether the Commission will still pursue the case. Based on the Commission’s ‘Notice on the notion of State aid’ which also addresses parafiscal charges, it seems as if the Commission has kept all possibilities open by stating that ‘a transfer of State resources is present where the charges paid by private persons transit through a public or private entity designated to channel them to the beneficiaries’ from which it seems that there is no need for a compulsory charge.57

III. COMPATIBILITY OF THE AID WITH THE INTERNAL MARKET

A. General Remarks The ‘Guidelines of the Commission on State aid for environmental protection and energy 2014–2020’ (EEAG), together with the Block Exemption Regulation (GBER),58 constitute the main basis for approving State aid for the protection of the environment and more generally to the energy sector. The GBER is issued on the basis of Article 108(4) of the Treaty59 and Council Regulation (EC) No 994/98, empowering the Commission to exempt the grant of

56

Case C-405/16 P Germany v Commission (appeal pending). See Commission, Commission Notice on the notion of State aid as referred to in Article 107(1) TFEU [2016] OJ C261/1, paras 63–65. 58 Commission Regulation (EU) No 651/2014 of 17 June 2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty [2014] OJ L187/1. 59 On the basis of Art 108(4) TFEU, the Commission may adopt regulations to exempt aid as determined by Art 109 TFEU. 57

State Aid in the Form of Taxation and Emission Costs

79

State aid for a variety of purposes from the notification requirement. On that basis, the GBER sets out the conditions under which Member States may grant State aid to objectives covered by the GBER without notifying the Commission. If the grant of State aid for the protection of the environment or the energy sector is not covered by the GBER, Member States must notify the Commission to obtain an approval. In that case, the aid is likely to be assessed on the basis of the EEAG,60 which sets out general guidance on how the Commission may find State aid for the protection of the environment (and to the energy sector), compatible with the Common Market.61

B. Aid in the Form of Reductions or Exemptions from Environmental and Energy Taxes The Commission considers that environmental taxes increase the costs for damaging the environment and therefore has the effect of discouraging such behaviour. The Commission has also recognised that although reductions from environmental taxes may have adverse effects, without the possibility to grant such reductions, the introduction of taxes would be unacceptable for industry,62 which may leave countries where energy is cheap and where costs for protecting the environment are either not imposed or are significantly lower than in the EU. With that in mind, the Commission has observed that tax reductions may incentivise Member States to introduce a higher overall level of environmental taxes, and thereby contribute to an increased level of environmental protection. This assumes, however, that the overall objective of discouraging harmful environmental behaviour has not been undermined which requires that any tax reduction does not go down to zero, and thus that a minimum level of taxation is still applicable. For the purposes of assessing compatibility, the Commission distinguishes between ‘harmonised taxes’ and ‘non-harmonised taxes’. With that distinction in mind, the Commission’s assessment of cases can be divided into three main groups: 1. Tax reductions which concern harmonised taxes that are covered by the Energy Taxation Directive (harmonising energy taxation in the EU) and where the tax reductions respect the minimum thresholds set forth in this Directive.63 2. Tax reductions which concern harmonised taxes that are covered by another taxation harmonisation Directive (than the Energy Taxation Directive) and respect the minimum thresholds set out therein. 60 The EEAG is issued on the basis of Art 107(3)(c) TFEU, which enables the Commission to approve State aid which is compatible with the internal market to facilitate the development of certain economic activities within the EU. See Commission, ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG) [2014] OJ C200/1. 61 The Commission may also approve State aid for energy or environmental purposes directly on the basis of Art 107(3)(c) which it relies on in cases not covered by the EEAG such as in the case of Hinkley Point C (nuclear power station): Hinkley Point (Case SA.34947) [2014] OJ C69/60. 62 EEAG (n 60) para 167. 63 Council Directive 2003/96/EC on restructuring the Community Framework for the taxation of energy products and electricity (n 17).

80 Lena Sandberg 3. Tax reductions which concern non-harmonised taxes that (i) are not covered by any taxation harmonisation Directive; or (ii) are covered by a taxation harmonisation Directive, but do not respect the minimum thresholds set out in that Directive. Tax reductions falling into the first category are assessed on the basis of the GBER.64 Provided that such reductions fulfil the criteria set out in the GBER, Member States may implement them without notifying the Commission.65 Tax reductions falling into the second and the third category are assessed on the basis of the EEAG. Member States must therefore notify such tax reductions to the Commission and await its approval before implementing them. Both the criteria in the GBER and those set out in the EEAG are based on certain common principles, namely (i) the ‘necessity’ of the tax reduction (that is, whether it is necessary in light of the significance of the reduction and the nature of the tax); and (ii) the ‘proportionality’ of the tax reduction. i. Harmonised Environmental Taxes under the GBER Harmonised taxes are taxes which have been the subject of a Directive fixing a minimum tax level throughout the EU, thereby ‘harmonising’ the EU tax level. The Energy Taxation Directive 2003/96/EC (ETD) is an example of a Directive which harmonises tax rates on several energy consumption products and electricity, including natural gas, coal and mineral oil by setting out minimum tax rates which Member States must apply. The grant of reductions to taxes covered by a taxation harmonisation Directive, such as the ETD, benefit from a simplified approach for the purposes of assessing the proportionality and necessity. According to Article 44 of the GBER, the reduction of environmental taxes covered by ETD are considered to be compatible with the common market provided that (i) the beneficiaries pay at least the minimum tax level set out in the ETD, while fulfilling all other conditions of the ETD; and (ii) the choice of beneficiaries is based on objective and transparent criteria.66 If these conditions are fulfilled, the tax reductions may be introduced and remain in effect until July 2021.67 The first part of the first condition, ie, that the beneficiaries pay the minimum tax level set by the ETD, refers to Article 4 of the ETD which provides that the beneficiaries must pay the minimum level of taxation established by Annex I. According

64 Commission Regulation (EU) 651/2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty (n 58). 65 It should be recalled that according to Arts 9–11, the GBER is only applicable provided that the Member State (i) publish the relevant aid measure on a comprehensive website; and (ii) report the information of the measure to the Commission within 20 working days following its entry into force. 66 According to Art 44(3) GBER, instead of tax reductions, the Member States may grant a fixed compensation or a combination of both mechanisms. 67 The GBER expires on 31 December 2020 (Art 59 GBER), but it provides for a transitional period of six months for any aid scheme (Art 58(4) GBER).

State Aid in the Form of Taxation and Emission Costs

81

to Annex I, different taxation minimum rates apply to motor fuels, heating fuels and electricity.68 The second part of the first condition, ie, that all other conditions of the ETD must be met, refers, inter alia, to Article 17 of the ETD which provides that tax reductions may only be applied to the consumption of energy products used for heating, stationary motors, plants (and machinery) used in construction, civil engineering, public works as well as electricity. Article 17 of the ETD also provides that tax reductions may only favour particular beneficiaries, ie, beneficiaries which qualify as ‘energy-intensive companies’, namely companies (i) which purchase energy products and electricity for at least 3 per cent of the production value; or (ii) which pay an energy amounting to at least 0.5 per cent of the added value (ie, total turnover liable to VAT).69 According to Article 17 of the ETD, permitted beneficiaries also include those undertakings which conclude agreements, tradeable permit schemes or equivalent arrangements, which contribute to the environmental protection objectives or to improvements in energy efficiency. The second condition of the GBER (that the beneficiaries are selected on the basis of transparent and objective criteria) is usually always met and is therefore not discussed further. Prior to the adoption of the GBER in June 2014, tax reductions under the ETD were assessed under the EEAG. Therefore, while the following examples were assessed under the EEAG previously, they would be subject to the GBER, had they been introduced today. A good example of tax reductions covered by the ETD is the CO2 tax exemption cases in Denmark and Sweden, referred to in section II.B.i.c above. In those cases, the Commission decided to approve the proposed reductions under the EEAG70 given that (i) even after the applied tax reduction, the relevant undertakings were still paying taxes above the harmonised minimum of the ETD; (ii) the undertakings qualified as energy-intensive companies under Article 17 ETD; and (iii) the criteria for benefiting from the reduction were both transparent and objective since the conditions for benefiting from the reduction was that the undertakings participated in the ETS. In the Swedish case, the Commission noted that as aid beneficiaries were subject to the ETS, any increase in their individual CO2 consumption as a result of the tax reduction would require them to buy ETS certificates or prevent them from surrendering allowances. Hence, the tax reduction would not remove the undertakings’ incentive to decrease their CO2 emissions. Accordingly, the Commission took into

68 In fact, Art 17(2) ETD also allows for tax reductions covered by the ETD down to zero. However, since the GBER requires that the minimum thresholds in the ETD are adhered to, tax reductions down to zero must be assessed under the EEAG. 69 Art 17(1)(a) GBER provides that ‘value-added’ means the total turnover liable to VAT, including export sales minus the total purchases liable to VAT including imports. 70 See CO tax reduction for fuel used in installations covered by EU ETS (n 21) and Modification of 2 the CO2 tax (n 10).

82 Lena Sandberg account that the purpose of the tax reduction was precisely to relieve the undertakings from a financial burden, imposed to incentivise an environmental improvement, which would in any event still be achieved (through the ETS).71 In the Danish case, the Commission refused to approve CO2 tax exemptions, which went below the minimum tax level. This is discussed further in section II.B.i below. ii. Harmonised Environmental Taxes under the EEAG While the GBER addresses tax reductions that concern taxes under the ETD, the EEAG allows for tax reductions which concern taxes that are covered by any other taxation harmonisation Directive. As mentioned above, the grant of reductions to taxes covered by a taxation harmonisation Directive benefits from a simplified approach for the purposes of assessing the proportionality and necessity of such tax reductions. In this regard, the EEAG provides that the Commission will find the tax reduction compatible (i) if it respects the minimum thresholds set out in the relevant harmonisation Directive; (ii) if the choice of beneficiaries is based on objective and transparent criteria; and (iii) if the tax reduction is granted in the same way for all competitors in the same sector. iii. Non-Harmonised Environmental Taxes under the EEAG If the tax reductions concern taxes which are not covered by any harmonisation Directive, or (ii) if the tax reductions concern taxes which are covered by a harmonisation Directive, but the beneficiaries pay less than the minimum established in the relevant Directive, they are considered as non-harmonised taxes.72 For instance, a CO2 tax reduction, which goes below the minimum rate fixed in the ETD, is an example of a non-harmonised tax reduction. The Commission assess reductions in non-harmonised taxes on the basis of a stricter approach regarding both its necessity and proportionality. The EEAG provide that the Commission will consider tax reductions from nonharmonised taxes to be deemed necessary (i) if the choice of beneficiaries is based on objective and transparent criteria; (ii) without the reduction, the tax will result in a substantial increase of production costs (as there is no point in approving a reduction to a tax which does not have a significant financial impact on an undertaking); and (iii) this increase in production costs cannot be passed on to customers without a significant sales reduction (since otherwise there is no need for a tax reduction). The EEAG also provides that non-harmonised taxes are considered to be proportionate (i) if the aid beneficiaries pay a minimum of 20 per cent of the tax, or (ii) if an agreement is concluded between the authorities where the beneficiaries commit to

71

CO2 tax reduction for fuel used in installations covered by EU ETS (n 21) para 25. The Member States may decide to grant the aid either as a tax reduction, as a fixed compensation or a combination of both. EEAG (n 60) para 179. 72

State Aid in the Form of Taxation and Emission Costs

83

achieve environmental protection objectives having the same effect as if beneficiaries paid at least 20 per cent of the tax rate.73 If the Commission deems the tax reduction to be both necessary and proportionate, the tax reduction will be allowed for a maximum of 10 years, but the Commission may set a shorter time frame.74 An example of a case where the Commission accepted that agreements would achieve the same objective as the environmental tax is the Swedish electricity tax case.75 In this case, the Commission approved that the companies were fully tax exempt in return for the conclusion of an agreement between the exempt companies and the Swedish government to implement an energy management programme in order to improve energy efficiency. The five-year programme imposed various obligations on the companies in order to increase energy efficiency and thereby reduce emissions to the same level as the full electricity tax. However, since 2004 the Commission has not approved agreements which justify a full tax exemption and there is no doubt that the Commission is generally reluctant to approve full tax exemptions as they are considered to ‘circumvent’ the State aid rules. For example, the Commission approved a NOx tax reduction of 47 per cent for Danish cement producers,76 but refused to approve energy tax reductions for Dutch ceramic producers on the grounds that the reduction represented more than 80 per cent of the tax.77 Finally, as mentioned above, the Commission refused to approve the part of the Danish CO2 tax exemption which went to zero by reference to the double burden costs under ETS.78 iv. Specific Situations of Harmonised Taxes The EEAG also provides guidance for specific tax reductions, which do not fall under the classic rules on harmonised and non-harmonised taxes, such as the carbon tax levy. The rules in the EEAG on carbon taxes were especially drafted for the purposes of introducing a Carbon Price Floor in the UK. The EEAG provides that the Commission may approve a reduction in a carbon tax on energy products used for electricity production, which is directly linked to

73 Such commitments must satisfy three cumulative conditions: (i) the agreement must specify the targets and fix a time schedule to reach them; (ii) independent and timely monitoring of the commitments must be ensured; and (iii) the commitments must be periodically revised and effective penalty arrangements triggered if the commitments are not met. See EEAG (n 60) para 178. 74 The EEAG provides that the Commission may fix a timeframe of four years or less: see EEAG (n 60) para 242. 75 Full electricity tax relief for energy-intensive companies (Case N253/2004) [2005] OJ C136/36, ss 4.2, 7. 76 Denmark NOx tax reductions for large polluters and companies reducing pollution (Case N327/2008) [2010] OJ C166/1, para 55. The Commission noted that the general tax level was higher than it would have been without the reduction (and the beneficiaries still paid 53% of the tax after the proposed relief). 77 See Netherlands ceramics (Case C5/2009) [2010] OJ L186/32, para 84. The Commission considered that the Dutch authorities had not submitted data demonstrating that the tax reduction would only result in a ‘limited distortion of competition’ in order to justify a lower rate than 20%. 78 Modification of the CO tax (n 10) paras 65 and 69. 2

84 Lena Sandberg the ETS allowance price, provided that (i) the aid is only granted to sectors and subsectors listed in Annex II of the ETS Guidelines to compensate for additional indirect costs resulting from the tax; (ii) the aid intensity and maximum aid intensities are calculated as defined in points 27–30 of the ETS Guidelines; and (iii) the aid is granted as a lump sum that can be paid to the beneficiary in the year in which the costs are incurred or in the following year.79 In May 2014, the Commission assessed the Carbon Price Floor introduced by the UK under the EEAG.80 The introduction of the Carbon Price Floor entailed that energy producers using fossil fuels for energy production would be subject to a climate levy (from which they had previously been exempt). However, it was given that the energy producers would pass on the costs of a climate levy to their customers, which comprised energy-intensive companies. The UK authorities requested that the energy-intensive companies be relieved of some of the costs of the climate change levy. However, the levy was not imposed on the energy-intensive companies and thus a tax relief could not be granted directly to them. Therefore, the UK requested that the Commission should approve the grant of a tax relief in another tax (also a climate levy) which was imposed on the energyintensive companies. It was on this basis that the Commission developed the new above-mentioned rules in the EEAG, which permitted it to authorise the tax relief to the energy-intensive companies.

C. Compatibility of Aid in the Form of Levies or Surcharges to Finance Renewable Energy i. Commission Guidelines (EEAG) As mentioned above, Member States have introduced parafiscal levy schemes (or surcharge schemes) where the consumers bear the costs in order to be able to finance renewable energy. Such schemes normally ensure that payment is collected from the consumers and paid through an intermediary to the renewable energy producers. Given that the financing of renewable energy is based on a long-standing EU policy objective, the Commission has approved these types of aid schemes under (different versions of) the EEAG for several decades. However, as appears from the various cases on levy or surcharge schemes to finance renewable energy, these schemes also involve State aid to energy-intensive companies through the fact that they are often exempt from paying the levy or the surcharge. It is this type of aid which the Commission normally does not approve since it does not lead to an improvement in the environment either directly or indirectly. On the other hand, the reason that Member States exempt energy-intensive companies from the financing schemes is that electricity costs represent between

79

See EEAG (n 60) para 180. UK Carbon Price Floor (Case SA.35449) [2014] OJ C348/1, para 5. The CPF is a tax on the level of carbon emissions of the fuels used in electricity generation. It is primarily designed to attract low carbon investment by raising the price of pollution and increasing the rewards for low carbon projects. 80

State Aid in the Form of Taxation and Emission Costs

85

40 and 60 per cent of the production costs. Had they been liable to pay renewable surcharges or levies, their contributions would be so significant that many companies would have to consider relocating. In light of that, the EEAG allows Member States to grant partial compensation for these additional costs, provided there is no discrimination and no overcompensation. Annex 3 of the EEAG lists sectors which are deemed to be exposed to an international competitive risk caused by significant payments to renewable support schemes.81 According to the EEAG, the Commission may approve renewable financing schemes which reduce the contribution paid by energy-intensive companies, provided they are (i) based on objective, non-discriminatory and transparent criteria; and (ii) the energy-intensive companies pay at least 15 per cent of the additional costs. However, in practice, even this payment level was too significant for the energyintensive industries and, on the occasion of the EEG case (described below), the Commission introduced new provisions in the EEAG under which the Commission may approve renewable support schemes where energy-intensive companies pay less than 15 per cent of the costs.82 The extent of the costs for financing aid to renewable energy may be limited to 4 per cent of the gross value added of the undertaking concerned, and to 0.5 per cent of the gross value added of the undertaking concerned provided that its electro-intensity is at least 20 per cent.83 A key requirement is, however, an adjustment plan which provides that energy-intensive companies progressively pay more.84 ii. The EEG Case In the State aid Decision on the German Renewable Energy Law 2012 (EEG 2012), Germany imposed a surcharge to finance the support for renewables (EEGsurcharge). The DSOs were obliged to purchase electricity based on renewable sources at a price above the market price. The renewable electricity was transferred to the TSOs, who in turn had to compensate the DSO for the price. However, the TSOs had to sell the electricity on the spot market. If a net loss occurred, the TSOs could pass the costs on to electricity suppliers in the form of the EEG-surcharge. The electricity suppliers were, in turn, free to recuperate the financing from the electricity consumers.

81 State aid for sectors not listed in Annex 3 can only be granted if the undertaking has an electrointensity of at least 20% and belong to a sector with a trade intensity of at least 4% at Union level: see EEAG (n 60) para 186. 82 EEAG (n 60) para 189: ‘When needed, Member States have the possibility to further limit the amount of the costs resulting from financing aid to renewable energy to be paid at undertaking level to 4% of the gross value added of the undertaking concerned. For undertakings having an electro-intensity of at least 20%, Member States can limit the overall amount to be paid to 0,5% of the gross value added of the undertaking concerned’. 83 ibid. 84 ibid, paras 194–97.

86 Lena Sandberg a. Aid to Renewable Energy Producers As explained in detail in section II.C.ii above, in the EEG case, the Commission found that the obligation to purchase electricity based on renewable energy sources (which is essentially a feed-in tariff) involved State aid to the renewable energy producers. Based on the previous EEAG (applicable until 2014), the Commission considered the aid to be compatible, inter alia, because (i) the aid was granted as a premium in addition to the market price; (ii) generators would sell its electricity directly in the market; and (iii) the aid is only granted until the plant has been fully depreciated. Under the current EEAG, the Commission also requires that, in order for aid to renewable sources to be compatible (i) generators must be subject to standard balancing responsibilities; (ii) generators should not have an incentive to generate electricity under negative prices; and (iii) after 2017 the aid must be granted in a competitive bidding process. b. Aid to Energy-Intensive Companies The EEG 2012 provided for a cap on the EEG surcharge payable by energy-intensive undertakings (who therefore benefited from a reduced surcharge), which the Commission found to involve State aid to the energy-intensive companies. The Commission approved the aid by reference to the EEAG and the fact that the energy-intensive companies had an electro-intensity of at least 20 per cent and paid 0.5 per cent of the undertakings’ gross value added.85 This arrangement meant that the energyintensive companies benefited from the exemption almost in full. However, given that in 2013 and 2014 some beneficiaries effectively benefited from a more significant reduction than 0.5 per cent of the undertakings’ gross value added, Germany was required to recover the exceeding aid from the beneficiaries. However, there was one obstacle. Aid in the form of parafiscal duties cannot be found compatible with the common market if they are financed from charges on products imported from other Member States, since this constitutes discrimination based on nationality and infringes Articles 30 TFEU (on the free movement of goods) and 110 TFEU (prohibiting taxation discriminating between domestically produced goods and goods imported from other Member States).86 The Commission found that the EEG-surcharge would also be imposed on imported electricity since the purchased renewable based electricity also included imported electricity. The discrimination can only be remedied through the reimbursement of the discriminatory charge or by reinvesting the share of the surcharge imposed on imports into projects that benefit imports, thereby compensating for the discrimination. This is also what Germany did, as it committed to invest €50 million in cross-border interconnector projects. Through a complex calculation method, the Commission determined that this was also the value of the discrimination on imported electricity. It should be added, however, that although the Commission had mostly approved the EEG as compatible State aid (except for a smaller recovery amount as stated

85 86

ibid, para 189. See cases referred above (n 23).

State Aid in the Form of Taxation and Emission Costs

87

above), Germany has nonetheless appealed the decision to the General Court, arguing that the EEG-surcharge does not involve State aid in the first place.87 On 10 May 2016, the General Court dismissed Germany’s appeal in its entirety but, on 19 July 2016, Germany appealed the judgment of the General Court to the Court of Justice, asking it to set the judgment aside.88 iii. The StromNEV Case89 As stated above, in its decision opening the formal investigation procedure on StromNEV, the Commission found that the exemption to large energy consumers had been granted through State resources because: first, the State had imposed a special surcharge on electrical consumers designed to finance the exemption; second, the State had appointed an undertaking to administer the charge, namely the TSO (which functioned as an intermediary subject to State control); third, the State had established rules concerning the use and destination of the surcharge; and fourth, the State could monitor the financial flows through control mechanisms.90 The former version of §19(2) StromNEV (2011) fully exempted large electricity consumers (LEC) from paying network charges. In the current version of §19(2) StromNEV (2013), LEC are not fully exempt from paying network charges, but have to pay individual network charges ranging between minimum levels of 10 and 20 per cent of the published network charges. Although, in its decision opening the formal investigation procedure the Commission did not consider that StromNEV would entail compatible State aid, the fact that the network charges range between 10 and and 20 per cent of the total amount means that they broadly correspond to the current proportionality under the current EEAG, requiring payment of 0.5 to 20 per cent depending on the companies’ electro intensity.91 In this context, it should be recalled that the compulsory charge in StromNEV 2013 has been abolished, and thus the scheme may not involve State aid in the first place, which may be the reason that the Commission has not yet issued a final decision in the case (three years after the opening of the procedure).

D. Aid in the Form of Cost Relief Due to Indirect Emissions Cost i. Overview of the ETS The Emissions Trading Scheme Directive (ETS Directive) introduced a price on carbon by establishing a system whereby energy-intensive companies have to ‘surrender’ allowances to be allowed to emit greenhouse gases.92 The purpose of putting a

87

Case T-47/15 Germany v Commission (n 55). Case C-405/16 P Germany v Commission (n 56) (appeal pending). 89 Exemption from network charges for large electricity consumers (n 43). 90 ibid, para 49. 91 EEAG (n 60) paras 186–89. 92 Directive 2009/29/EC of the European Parliament and of the Council of 23 April 2009 amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community (ETS Directive) [2009] OJ L140/63, Art 4. 88

88 Lena Sandberg price on CO2 emissions is to reduce greenhouse gas emissions by 20 per cent by 2020 and by at least 40 per cent by 2030, compared with emissions in 1990. The ETS was introduced in 2005, and based on trading periods which permit the system to be regularly reviewed and updated. The first trading period was from 2005 to 2007; the second trading period was from 2008 to 2012; and the third trading period (which is the current trading period) is from 2013 to 2020.93 For each trading period a cap is fixed on the total number of allowances to be issued. For example, for the current trading period a Community-wide cap was fixed at the beginning of 2013 which equals the average of total allowances issued in the preceding second trading period (2008–12).94 Scarcity of allowances is created by decreasing the annual number of allowances each year (until 2020) by a linear reduction factor of 1.74 per cent.95 Under the current trading period allowances must be purchased at auctions unless an undertaking falls within an exception which entitles them to allowances free of charge.96 However, the ETS Directive also provides for important exemptions such as for undertakings, notably for district heating, cogeneration, new entrants and companies exposed to carbon leakage.97 The auction of allowances takes place on auction platforms, such as the European Energy Exchange (EEX) of the ICE. Out of the 100 per cent allowances to be auctioned, 88 per cent are distributed to all Member States in proportion to their respective verified emissions from 2005,98 while poorer Member States and Member States that have reduced their emissions receive additional allowances.99 Allowances are transferrable.100 This means that a company covered by the ETS may either buy allowances on auctions, or from another company. Companies covered by ETS must therefore either (i) buy allowances to cover their emission needs, or (ii) invest in environmentally friendly equipment to reduce its emission needs. Those that emit more than their allowances are punished by having to buy additional allowances,101 while those who invest in CO2 reducing equipment are rewarded as they can sell excess allowances on the market.

93

The fourth trading period will be from 2021 to 2030. ETS Directive (n 92) Art 9. The total amount of EEA-wide allowances for the base year 2013 was 2,084,301,856. See the Commission, ‘EU ETS Handbook’, available at www.ec.europa.eu/clima/sites/ clima/files/docs/ets_handbook_en.pdf, 22. 95 ETS Directive (n 92) Art 9. 96 ibid, Art 10. 97 ibid, Art 10a and Commission Decision of 27 April 2011 determining transitional Union-wide rules for the harmonised free allocation of emission allowances pursuant to Article 10a of Council Directive 2003/87 of the European Parliament and of the Council [2011] OJ L130/1. 98 Or the average emissions from the first trading period (2005–07), whichever is higher. ETS Directive (n 92) Art 10(2)(a). 99 Ten per cent are given to those Member States which are poorer, see ETS Directive (n 92) Art 10(2) (b). The last 2% are given to those Member States which, in 2005, were successful in reducing their emissions by at least 20% compared with the base year under the Kyoto Protocol; see ibid, Art 10(2)(c). 100 ETS Directive (n 92) Art 12(1). 101 If undertakings produce more emissions than they hold in allowances, they are subject to penalties set by Member States and/or a penalty of €100 for each tonne of emitted CO2 for which an undertaking failed to surrender allowances: ETS Directive (n 92) Art 16(3). 94

State Aid in the Form of Taxation and Emission Costs

89

ii. Carbon Leakage: Indirect Emissions Costs Under the current trading period, power plants must ‘pay’ allowances and they therefore include their direct emissions costs in the price of electricity sold through the energy exchanges (day-ahead or ‘spot’ market and forward markets).102 On the energy exchange, power plant capacity is drawn upon, starting from the lowest-cost option, until the demand is met. The offer price of the last power station needed to satisfy this demand (the ‘marginal power plant’) determines the market price for all the other power stations.103 Given that this is normally a thermal plant (based on fossil fuels),104 the price will include emissions costs.105 This are referred to as ‘indirect emissions costs’. For certain electricity customers, namely energy-intensive companies, electricity costs already represent a heavy burden given that electricity costs account for between 40 and 60 per cent of their production costs. The additional costs of indirect emissions will therefore squeeze their profits even more.106 While normally companies would pass on such costs to their customers, energy-intensive customers are unable to pass on the indirect emissions costs as they risk being outcompeted by undertakings from countries who do not have to bear such an environmental cost burden (such as companies from Asia or the Middle East), or because of sector restraints. For example, aluminium producers are barred from passing on indirect emissions costs given that their sales price is fixed on the London Metals Exchange (LME).107 In summary, energy-intensive companies must therefore both (i) obtain free allowances to cover their own ‘direct emissions’ (which is already high given their high energy consumption); and (ii) pay for the indirect emissions costs passed on to them by energy producers. Many energy-intensive companies have explained that, as a result, they will (over time) shift their production to countries outside the EU with no environmental cost burden. This will increase the net output of global CO2 emissions and thereby undermine the ETS. This phenomenon is referred to as ‘carbon leakage’.108 To counteract carbon leakage, the ETS Directive provides for preferential treatment of companies exposed to carbon leakage in two ways. First, energy-intensive companies that are deemed to be exposed to carbon leakage,109 and thus included

102

ETS Directive (n 92) Art 10a(3). Power prices are thus the result of the point at which supply and demand intersect. 104 The indirect emission cost is thus set by the marginal cost, usually through a mix of gas-fired (0.4 tonnes CO2 per MWh) and coal-fired plants (0,9 tonnes of CO2 per MWh). 105 Given the presence of interconnectors, this is also the case in markets with a high share of CO free 2 electricity (eg, hydropower or nuclear). Despite the increasingly higher share of renewable energy, fossilfuelled power plants are still expected to determine the prices for the next decade. 106 eg, in the aluminium industry, indirect emissions costs can be six to seven times higher than their direct emissions cost. 107 LME, which is the world’s largest non-ferrous metals exchange, determines the price of aluminium months ahead for contracts used worldwide. 108 The first step is the cessation of investments in existing EU production facilities, eg, since 2011 aluminium productions in Italy, Spain, the UK, the Netherlands and Germany have announced closures. 109 Companies in sectors or subsectors are deemed to be exposed to carbon leakage if: (i) direct and indirect costs induced by the implementation of the Directive would increase production cost, calculated as a proportion of the gross value added, by at least 5%; and (ii) the sector’s trade intensity with non-EU 103

90 Lena Sandberg in the so-called ‘carbon-leakage’ list, are granted free allowances.110 Second, the Commission has also indicated that it is willing to approve the grant of State aid to companies exposed to carbon leakage in order to compensate for their indirect emissions costs, provided certain conditions are met. These conditions are laid down in the ETS Guidelines,111 which set out a framework for approving State aid granted to compensate energy-intensive companies for their indirect emissions costs. Many companies in sectors and subsectors which are deemed to be exposed to carbon leakage under the ETS Directive are also eligible for State aid under Annex II of the ETS Guidelines.112 The ETS Guidelines provide a complex formula for determining the maximum aid per installation.113 Further, in order to ensure that companies always have an incentive to reduce emissions, the ETS Guidelines do not allow for the grant of State aid to compensate undertakings in full for the cost of indirect emissions and the grant of aid must be digressive over time. For example, between 2013 and 2015, Members States were entitled to grant State aid for up to 85 per cent of the indirect emissions costs, while the limit is currently 80 per cent until 2018.114 Between 2019 and 2020, compensation will be limited to 75 per cent of indirect emissions costs. The Commission has applied the ETS Guidelines in a number of cases and found the aid in such cases to be compatible since the compensation schemes were designed in accordance with the ETS Guidelines: the eligible undertakings were included in Annex II of the ETS Guidelines, the compensation was not available for the entirety of indirect emissions costs and the aid was degressive over time.115 Both the ETS Directive and the ETS Guidelines are subject to scrutiny in preparation for the new trading period starting in 2021. In this context, companies exposed to carbon leakage have argued that the prevention of full compensation does not ensure that the companies have an incentive to reduce emissions, since they have no influence over the indirect emissions costs: the indirect emissions costs refer to emissions which do not stem from their own installations but from those of the energy producer. In the same vein, energy-intensive companies have criticised the difference in treatment of their direct emissions (subject to 100 per cent guaranteed free allowances) and indirect emissions (with only the potential possibility of granting State aid compensation), in particular since the energy-intensive companies have no control of the indirect emissions but must pay for them. countries (imports and exports) is above 10%. Carbon leakage is also deemed to be present if (i) the sum of direct and indirect additional costs is at least 30%; or (ii) the non-EU trade intensity is above 30%. ETS Directive (n 92) Art 10a(14)–(17). 110

ETS Directive (n 92) Art 10a(12). Commission, ‘Guidelines on certain State aid measures in the context of the greenhouse gas emission allowance trading scheme post-2012’ (ETS Guidelines) [2012] OJ C158/4. 112 ibid, point 23 and Annex II. 113 ibid, point 27. 114 ibid, point 26. 115 Such compensation schemes include the Commissions Decisions: Environmental protection (Case SA. 36103) [2013] OJ C353/1 (Germany); Compensation for indirect EU ETS costs (Case SA.36650) [2014] OJ C17/8 (Spain); Compensation for Indirect EU ETS costs Vlaanderen (Case SA.37017) [2013] OJ C354/1; Compensation for Indirect EU ETS costs in the Netherlands (Case SA.37084) [2013] OJ C353/1; Compensation of indirect EU ETS costs in Greece (Case SA.38630) [2014] OJ C348/1; Relief from indirect CO2 costs in electricity in Lithuania (Case SA.41981) [2016] OJ C198/1; Compensation for indirect CO2 costs in Slovakia (Case SA.43506) [2016] C 142/1. 111

4 WTO Subsidy Rules: Implications for Energy ANNA-ALEXANDRA MARHOLD*

I. INTRODUCTION: THE RELEVANCE OF WTO SUBSIDY RULES FOR ENERGY

T

HE ENERGY SECTOR is one of the most heavily subsidised in the world.1 This can be explained by the fact that the energy sector is often capital intensive and closely connected to the industrial policy of a State, whether based on fossil fuels or, to the contrary, to promote a cleaner means of energy production to gain ground (eg, by means of Feed-in Tariffs (FITs)). Due to the wide presence of subsidies in the sector, the World Trade Organization (WTO) rules on subsidies automatically become relevant to it. The rules of the Agreement on Subsidies and Countervailing Measures (ASCM) influence and bind the choices of the governments of WTO Members with respect to their industrial and other policies.2 While the vocabulary and the content of WTO subsidy rules differ from European Union rules on State aid, WTO subsidy rules are no less important for the EU and its Member States.3 As WTO Members, third countries can challenge the EU and its Member States’ schemes under the SCM Agreement. Vice versa, the EU and its Members States can trigger subsidy disputes against third countries in the WTO forum.4 For this reason, it is important to consider whether support schemes that the EU and its Members States have in place, including those for energy, are designed to be consistent with WTO law. Note, however, that the SCM Agreement is generally understood to only apply to goods, not services—this is essential to keep in mind as the energy

* The author wishes to thank Petros Mavroidis, Robert Howse, Bernard Hoekman, Catherine Redgwell, Leigh Hancher, Adrien de Hauteclocque and Vasyl Chornyi for their valuable feedback in preparing this chapter. 1 H Birhanu Asmelash, ‘Energy Subsidies and WTO Dispute Settlement: Why Only Renewable Energy Subsidies Are Challenged’ (2015) 18 Journal of International Economic Law 261, 270. 2 Agreement on Subsidies and Countervailing Measures (ASCM) (SCM Agreement); Marrakesh Agreement Establishing the World Trade Organization (15 April 1994) Annex 1A, 1869 UNTS 14. 3 See generally on this issue, L Rubini, The Definition of Subsidy and State Aid—WTO and EC Law in Comparative Perspective (Oxford, Oxford University Press, 2010). 4 The EU has been a complainant in more than 23 cases against other WTO Members in subsidies disputes; as a respondent, the EU was targeted in more than 17 cases relating to subsidies and countervailing measures, see: www.wto.org.

92 Anna-Alexandra Marhold sector, for a large part, consists of activities that would fall into the category of energy services and not purely trade in (energy) goods.5 The dilemma with respect to energy subsidies seems to be that, on the one hand, subsidies on clean energy production and consumption are needed to correct market failures and to promote legitimate policy goals such as contributing to sustainable development through the scale up of clean energy, including expanding its trade.6 However, experience has shown that support schemes for clean energy, by their nature and design, make them sensitive to WTO dispute settlement. Much more harmful subsidies on fossil fuels, on the other hand, are omnipresent yet often escape being addressed in the multilateral trading system, for reasons that will be explained in this contribution.7 This chapter will unveil this discrepancy and discuss some of its implications by considering the rationales of WTO subsidy rules and how they play out in the energy sector. It will especially focus on explaining the difference in nature between fossil fuel and clean energy subsidies and how this impacts their treatment under WTO subsidy disciplines. Section II will briefly explain the rationale of subsidy disciplines in the WTO, after which it will lay out their legal coverage. Section III will subsequently zoom in on how fossil fuel and renewable subsidies have been treated in the forum so far and address some legal challenges and possible solutions in that respect. The conclusion will recap the findings in this chapter.

II. THE REGULATION OF SUBSIDIES IN WTO LAW: AN OVERVIEW

A. Subsidies in International Trade: Understanding the Rationale of the ASCM The concept of what constitutes a subsidy is not straightforward and in and of itself has been a topic of debate among economists and lawyers alike.8 For the sake of this contribution, we can agree to characterise a subsidy as a direct or indirect government transfer of funds to an entity in the private sector in order to keep the price of

5 Because the ASCM is placed under Annex 1.A of the Agreement Establishing the World Trade Organization, named ‘Multilateral Agreements on Trade in Goods’. 6 See, eg, L Rubini, ‘Rethinking International Subsidies Disciplines: Rationale and Possible Avenues for Reform’ (2015) The E15 Initiative Overview Paper (International Centre for Trade and Sustainable Development and World Economic Forum) 8; Rubini writes: ‘Among the various obstacles to green energy and its competiveness, the existence of significant support to conventional fossil fuel energy (both in terms of subsidies to production and consumption) cannot be overlooked. At the same time, often thanks to public support, green technologies are developing extremely fast. As a result, some types of clean energy, such as solar, are almost on a par with conventional energy. This means that some degree of differentiation is needed in any new disciplines providing for exemptions to subsidies (that is, certain sources need more/less protection than others’). 7 ibid. 8 AO Sykes, ‘The Economics of WTO Rules on Subsidies and Countervailing Measures’ (2003) 186 Chicago John M Olin Law and Economics Working Paper 1, 4; and P Mavriodis, The Regulation of International Trade—Volume 2: The WTO Agreements on Trade in Goods (Cambridge, MA, MIT Press, 2016) 191 ff.

WTO Subsidy Rules: Implications for Energy 93 a commodity or service low.9 Governments may utilise subsidies in their industrial policies, to boost a certain region, or to incentivise the production and consumption of a certain product, thereby increasing its market share.10 The effects of subsidies are ambiguous: in some cases, a subsidy can increase the aggregate welfare of a country by expanding trade, while in others, it may cause harm to its industry. The negative externality often associated with subsidies is the displacement of efficient producers. As mentioned above, governments usually subsidise with the goal of making a certain product cheaper and increasing its quantity on the market. This, in principle, should not decrease the aggregate welfare of an importing country; quite the contrary, if the markets in question are competitive and well functioning.11 However, more often than not perfectly competitive markets are confined to economist’s models than the everyday reality of the multilateral trading system. Subsidies can affect international trade—and therefore the energy sector where it pertains to international trade—in several ways. The starting assumption is that a country generally should set its internal policies, whose effects are confined within their borders, as it pleases.12 However, a large share of policies, whether domestic or international, do affect foreign consumers as well as corporations abroad.13 One important reason to have subsidy disciplines in WTO law is that, without such rules, WTO Members’ governments may not take into account these negative externalities behind their borders when drafting their internal policies. In the context of WTO law, the negotiators of the ASCM wanted to mitigate the negative effects that are associated with this (causing harm to foreign industry by internal subsidisation policies). It is precisely because of this that the injury to the foreign producer is central to the disciplines in the Agreement, and not inflicting harm on the aggregate welfare of another country in general terms.14 By means of the ASCM, WTO Members accordingly discourage certain policies that would cause injury to competing producer interests in importing countries (bear in mind that the producer, not consumer interest is at the heart of WTO regulation).15

9 WTO, World Trade Report 2006, Exploring the Links Between Subsidies, Trade and the WTO (WTO, 2006) 47; This is a purely economic, but not a legal definition. Legally speaking it was not an easy task to agree on the definition of a subsidy; See also generally on this, Rubini, The Definition of Subsidy and State Aid (n 3). 10 See generally J Bohanes, WTO Dispute Settlement and Industrial Policy—The E15 Initiative, Think Piece (International Centre for Trade and Sustainable Development and World Economic Forum, 2015). 11 ibid; also, see H Horn and PC Mavroidis (eds), Legal and Economic Principles of World Trade Law (Cambridge, Cambridge University Press, 2013) 58. 12 PC Mavroidis, Trade in Goods, 2nd edn (Oxford, Oxford University Press, 2012) 523; note however, that internal policies may nevertheless affect the investments of foreign companies. 13 ibid. 14 Pt V SCM Agreement (n 2) and Mavroidis, Trade in Goods (n 12) 524. 15 This is, eg, reflected in the text of Arts 12.9, 14, 15.1 and 19 of the SCM Agreement (n 2); see also Mavroidis, Trade in Goods (n 12) 525. Mavroidis explains that in the sense of political economy, countries do not always set their policy with aggregate economic welfare in mind. Rather, policies are often a compromise of competing interests, some receiving more weight than others (eg, governments give in to political pressure and tend to give more weight to producer interests rather than consumer welfare in their trade policy).

94 Anna-Alexandra Marhold The underlying thought of the remedying Countervailing Duties (CVD), is then to offset the effect of the subsidy and restore the competitive relationship as it was prior to the subsidy.16 In this sense, the objective of the ASCM is, in a sense, similar to that of the Anti-Dumping Agreement (ADA).17 Apart from subsidies that simply benefit producers, there are, understandably, subsidies that contribute to wider societal interests. These are the types of subsidy that typically correct market failures and negative externalities, such as the need to invest in cleaner energy to mitigate climate change.18 In those instances, subsidies can be seen as making up for a market failure indicating the slow development of a green economy.19 It is here that we touch upon the crux of the matter with regard to (clean) energy to be discussed further in the chapter: the rationale of a subsidy actually matters, although the SCM Agreement does not take this into account at present.20 While the ASCM in the past distinguished partially between types of subsidies (to be elaborated on below), it is unclear why the ASCM was not drafted to more strictly distinguish between different types of subsidy at the outset.21 From the WTO rules as they currently stand, it is moreover not always easy to demonstrate whether or not a subsidy actually has been granted.22 It may be simply because of the sophisticated nature of a particular scheme that confers benefit. While a subsidy may have been bestowed in the broad economic sense, it does not have to necessarily qualify as such in the legal sense of the SCM Agreement (eg, it may lack the necessary specificity within the meaning of Article 2, discussed below).

B. The Definition of a Subsidy in the SCM Agreement i. Prohibited, Actionable and Non-Actionable Subsidies Not all subsidies are illegal in the WTO context and a distinction is made between prohibited and actionable subsidies. Part II of the ASCM in Article 3 lists prohibited subsidies that a Member should refrain from altogether. These are subsidies contingent on exports (Article 3.1(a) ASCM) and local content requirements (LCRs) (Article 3.1(b) ASCM). Next, Part III lists actionable subsidies. These are allowed, but ‘no Member should cause … adverse effects to the interests of another Member’.23 Thus, if an actionable subsidy causes harm to a Member’s industry, the Member in question may take action against it. Members have two methods to seek recourse to prohibited and actionable subsidies: they can impose Countervailing Duties (CVDs)

16

Mavroidis, Trade in Goods (n 12) 525. Agreement on Anti-Dumping: Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994; Marrakesh Agreement (n 2) Annex 1A, 1868 UNTS 201. 18 RS Pindyck and DL Rubinfeld, Microeconomics, 8th edn (New York, Pearson, 2013) 667 ff. 19 Rubini, The Definition of Subsidy and State Aid (n 3) 8. 20 Mavroidis, Trade in Goods (n 12) 526. 21 Note that there is the lapsed Art 8 ASCM, which did take into account policy rationales of subsidies to a certain extent. This will be discussed in more detail in section III below. 22 Mavroidis, Trade in Goods (n 12) 527. 23 Art 5 SCM Agreement (n 2). 17

WTO Subsidy Rules: Implications for Energy 95 or they can initiate a dispute at the WTO Dispute Settlement Body (DSB).24 A third type of (expired) non-actionable subsidy is taken up in Part IV (Article 8 SCM).25 Article 8 ASCM did take into account certain legitimate policy goals and subsidies falling into this category were non-actionable. However the problem is, as will be discussed below, Article 8 ASCM has lapsed and this category of subsidies is no longer available in WTO law. Article XVI on subsidies of the General Agreement on Tariffs and Trade (GATT) was the predecessor of the ASCM.26 What this Article—among others—was missing, is one of the fundamentals needed to deal with the discipline effectively in WTO law: a definition of what constitutes a subsidy. The ASCM brought an improvement to this by incorporating definitions in Part I:27 Article 1.1(a)(1) ASCM defines a subsidy as (a) a contribution by a government, (b) through which a benefit is conferred. Article 2 of the SCM Agreement adds one more crucial element to this definition: the benefit must be conferred to a specific recipient.28 Each of these conditions will be briefly laid out below. ii. Contribution by a Government Article 1.1(a)(1) ASCM sets out an exhaustive list of conditions under which a governmental support scheme is considered a financial contribution. The term ‘government’ here entails a governmental public body, and includes its regional and local authorities, as well as State-owned enterprises.29 The Appellate Body (AB) in US—Softwood Lumber IV confirmed that this list of forms of financial contributions is exhaustive but at the same time wide ranging and includes measures such as loans and grants, government revenues that are otherwise due forgone, and government provision of goods and services.30 iii. Benefit to the Recipient To qualify as a subsidy, the financial contribution must confer a benefit (Article 1.1(b) ASCM). Demonstrating that this indeed is the case is not always a straightforward exercise. For instance, it goes without saying that it may be easier to determine that a direct transfer of a sum of money confers a benefit, as opposed a scenario in which a government purchases a good, or guarantees a set price for the good produced.31 In Canada—Measures Affecting the Export of Civilian Aircraft, the AB was of the 24

SCM Agreement (n 2) Pt V—Countervailing Measures. Art 8 (expired) SCM Agreement (n 2). 26 GATT 1947: General Agreement on Tariffs and Trade (30 October 1947) 61 Stat A-11, TIAS 1700, 55 UNTS 194. 27 SCM Agreement (n 2) Pt I—General Provisions. 28 Arts 1 and 2 SCM Agreement (n 2) and Mavroidis, Trade in Goods (n 12) 532. 29 P van den Bossche and W Zdouc, The Law and Policy of the World Trade Organization, 3nd edn (Cambridge, Cambridge University Press, 2008) 758. 30 WTO, Appellate Body Report, United States—Final Countervailing Duty Determination with Respect to Certain Softwood Lumber from Canada, WT/DS257/AB/R, adopted 17 February 2004, DSR 2004:II, 571, para 52. 31 Asmelash (n 1) 272. 25

96 Anna-Alexandra Marhold opinion that generally, to assess whether a benefit was conferred, one would have to look to whether the financial contribution left the recipient better off than it would have been without the contribution.32 Another benchmark may be to look at Article 14 of the ASCM, which deals with calculation of the subsidy.33 In what is known as the ‘private investor’-test, the Article explains that the term ‘benefit’ refers to ‘benefit to the recipient’, using the market or private investors to help determine the amount and existence of the benefit.34 For example, if the government provides equity capital, it shall not be considered to confer a benefit, lest the investment decision in question can be considered to be incompatible with the usual investment practices of a private investor in the Member State concerned.35 iv. Specificity To qualify as a subsidy within the meaning of the SCM Agreement, the subsidy in question moreover has to be deemed ‘specific’ pursuant to Article 2 of the Agreement.36 The thought behind this requirement was that only specific financial contributions can lead to trade distortions through inefficient resource allocation.37 As the Panel in US—Upland Cotton put it, a subsidy is not specific if it is ‘sufficiently broadly available throughout an economy as not to benefit a particular limited group of producers of certain products’.38 The idea is that if a subsidy is available to all producers in the country, there is not one producer or group of producers that can attract such resources at the expense of others.39 Schemes must either be de jure or de facto specific to an enterprise (Article 2.1(a) ASCM), industry (Article 2.1(b) ASCM) or particular region (Article 2.2 ASCM), to qualify as a subsidy.40 With regard to prohibited subsides in Part II (contingent on export and LCRs), the specificity requirement does not have to be met: Article 2.3 sets out that these subsidies are deemed to be specific a priori.41 Subsidies can be applied to consumers and to producers.42 Note though, that the dividing line is not always clear as producers can be simultaneously consumers. Nevertheless, we can generalise that the first type consists of intermediate (firms) and final consumers (households). The second type subsidises the producers of a certain

32 WTO, Panel Report, Canada—Measures Affecting the Export of Civilian Aircraft, WT/DS70/R, adopted 20 August 1999, upheld by Appellate Body Report WT/DS70/AB/R, DSR 1999:IV, 1443, para 157. 33 Art 14 SCM Agreement (n 2). 34 PC Mavroidis, PA Messerlin and JM Wauters, The Law and Economics of Contingent Protection in the WTO (Cheltenham, Edward Elgar Publishing, 2008) 325. 35 ibid. 36 Art 2 SCM Agreement (n 2). 37 Mavroidis, Trade in Goods (n 12) 549. 38 WTO, Panel Report, United States—Subsidies on Upland Cotton, WT/DS267/R, Add.1 to Add 3 and Corr 1, adopted 21 March 2005, as modified by Appellate Body Report WT/DS267/AB/R, DSR 2005:II, 299, para 7.1142. 39 Mavroidis, Trade in Goods (n 12) 549. 40 Art 2.1 SCM Agreement (n 2). 41 Art 2.3 SCM Agreement (n 2); Mavroidis, Trade in Goods (n 12) 549. 42 B Clements, D Coady, S Fabrizio, S Gupta, T Alleyne and C Sdralevich (eds), Energy Subsidy Reform: Lessons and Implications (Washington, International Monetary Fund, 2013) 1, 2.

WTO Subsidy Rules: Implications for Energy 97 product.43 The subsidy disciplines of the WTO are relevant for both. It is important to keep in mind, however, that it is overall much easier to establish specificity in production subsidies than in consumer subsidies.44 The reason for this is that the latter is often more general in nature. As will be elaborated upon in the following section, the distinction between producer and consumer subsidies and their respective specificity has far-reaching legal implications for the treatment between renewable energy vis-a-vis fossil fuel subsidies in WTO law.

III. WTO SUBSIDY RULES AND ENERGY REGULATION: IMPLICATIONS FOR FOSSIL FUELS AND RENEWABLES

The way WTO rules are set up can lead to intricate policy outcomes when it comes to fossil fuel subsidies versus support schemes for renewable energy. As it stands now, the consistency of fossil fuel subsidies with the SCM Agreement has not been challenged in the WTO forum directly.45 Renewable energy schemes, however, remain sensitive to WTO dispute settlement as is evidenced by a recent string of cases before the DSB.46 This section will discuss the legal-technical side of how WTO subsidy rules are set up and how this interacts with subsidy schemes for fossil fuels versus those for renewables. It will use dual energy pricing policies and Feed-in Tariff (FIT) schemes as an example here, as these two support mechanisms are omnipresent. It will examine both types and their interaction with the SCM Agreement. As alluded to above, energy subsidies can be broadly divided into producer and consumer subsidies.47 The first type of energy subsidy concerns intermediate (firms) and final consumers (households). One could think of, for example, cheaper inputs for energy-intensive industries, or lower energy bills for household consumers due to subsidised energy. The second type subsidises the producers of fuel products, coal, natural gas and electricity.48 Feed-in Tariffs, which are in place to stimulate the scale up of renewables in the energy mix, are also an example of producer subsidies.

43

ibid. Asmelash (n 1) 273. 45 Although the EU has been applying anti-dumping duties under the Anti-Dumping Agreement (ADA) in connection with dual pricing of Russian gas, which is at the centre of an ongoing dispute, see EU— Cost Adjustment Methodologies and Certain Anti-Dumping Measures on Imports from Russia, Second Complaint, WT/DS494 (Request for Consultations received 19 May 2015). 46 Examples of cases concerning renewable energy, while not all subsidies related, are, among others: Appellate Body Reports, Canada—Certain Measures Affecting the Renewable Energy Generation Sector / Canada—Measures Relating to the Feed-in Tariff Program, WT/DS412/AB/R/WT/DS426/AB/R, adopted 24 May 2013, DSR 2013:I, 7; Panel Reports, Canada—Certain Measures Affecting the Renewable Energy Generation Sector / Canada—Measures Relating to the Feed-in Tariff Program, WT/DS412/R and Add 1/WT/DS426/R and Add 1, adopted 24 May 2013, as modified by Appellate Body Reports WT/DS412/ AB/R/WT/DS426/AB/R, DSR 2013:I, 237; Panel Report, India—Certain Measures Relating to Solar Cells and Solar Modules, WT/DS456/R, circulated to WTO Members 24 February 2016 (appeal pending) and EU—Biodiesel DS473, European Union—Anti-Dumping Measures on Biodiesel from Argentina, WT/ DS473 (Panel composed on 23 June 2014). 47 Clements et al (eds) (n 42) 1, 2. 48 ibid. 44

98 Anna-Alexandra Marhold Overall, specificity in producer subsidies is easier to distinguish: clean and renewable energy programmes (such as Feed-in Tariffs), generally constitute subsidies in the form of regional or national programmes to stimulate the production of clean energy and are therefore more clearly defined as ‘specific’.49 Fossil fuel subsidies, on the contrary, are often subsidies to consumers which are moreover available throughout the whole economy, making it harder to qualify them as such.50

A. Fossil Fuel Subsidies: Harmful but Difficult to Tackle in WTO Law i. Overview ‘Fossil fuel subsidies’ is an overarching term used for subsidies that are granted by the government towards the production and consumption of energy from fossil fuels.51 There are two broad categories of such subsidies: those that are targeted at reducing the price for domestic consumers, and those that aim at increasing domestic supplies.52 Consumer subsidies are those subsidies that occur when the prices paid by consumers (including firms and households) are below the costs for supply, including the costs for transport and distribution of energy.53 And, vice versa, producer subsidies occur when the prices for energy supply are above this level.54 When fossil fuels are traded internationally, the supply costs are based on the international price.55 The two categories of consumer and producers subsidies are connected, however, for reasons of the elasticity of consumption, and it may be at times difficult to separate them in practice. Regarding consumer subsidies, these can further be divided into ‘pre-tax’ and ‘post-tax’ subsidies.56 The first occurs when the price paid by the household or firm is below the actual costs for distribution and supply, while the latter occurs when the taxes paid for are below their efficient level.57 It is estimated that fossil fuel subsidies encompass both consumer and producer subsidies.58 The actual ways in which fossil fuel subsidies are instituted remain opaque and unclear, not least because an international standardised system monitoring

49 Asmelash (n 1) 274 points out that the ‘renewable’ energy market as such could already been seen as specific vis-a-vis the energy market as a whole. 50 ibid. 51 The International Energy Agency in 2006 stated that one of the biggest barriers concerning energy subsidies in the OECD countries is a lack of up-to-date empirical data and analysis. Studies that have been undertaken on energy subsidies in OECD countries show results with remarkably large variance, due to (the different) methodologies used and the variety of definitions of energy subsidy incorporated. 52 R Steenblik, ‘Subsidies in the Traditional Energy Sector’ in J Pauwelyn (ed), Global Challenges at the Intersection of Trade, Energy and Environment (Geneva, Centre for Trade and Economic Integration, the Graduate Institute Geneva, 2010) 186. 53 D Coady, S Fabrizio, M Hussain, B Shang and Y Zouhar, ‘Defining and Measuring Energy Subsidies’ in Energy Subsidy Reform: Lessons and Implications (Washington, International Monetary Fund, 2013) 5. 54 ibid. 55 ibid. 56 ibid. 57 ibid. 58 ibid, 7.

WTO Subsidy Rules: Implications for Energy 99 them is absent; this is one of the biggest problems in addressing such subsidies comprehensively.59 The situation is such that even in data-rich countries, a lot of disagreement exists as to how much a particular energy sector is subsidised and the way this is calculated, but their estimated amount is vast.60 In countries that lack clear data, it is even more challenging to discern the nature and amount of such subsidies. Sometimes, the only available means of measurement is to look at energy prices, for example, if there are strikingly low energy prices for consumers (ie, far below the international market price).61 Subsidies on fossil fuels are more often consumer subsidies (generally available in the whole economy), than subsidies for production (usually targeted and specific).62 Aside from being harmful for the environment (fossil fuels encourage wasteful consumption), fossil fuel subsidies can additionally greatly affect the terms of international trade.63 They impact the energy industries themselves and the industries that use energy as an immediate input (one can especially think of heavy industries such as steel, glass, cement etc).64 The so-called practice of ‘energy dual pricing’ illustrates this well. Energy dual pricing is a widely utilised form of fossil fuel subsidisation. It has been subject to much opposition and debate in the past, not least by the EU and the US.65 Through dual pricing, governments of resource-rich countries sell their energy abroad at much higher prices than domestically. They can do so, because they often have a monopoly over the resource in question.66 In international energy trade, dual pricing policies tend to be instituted by means of export taxes (not regulated in WTO law). By means of energy dual pricing, it could be said that household consumers in the exporting countries are indirectly subsidised, at least in the economic sense, as they pay a lower overall price for their energy. And intermediate consumers (firms) can be said to be subsidised as well, as they have lower input prices for their industries.67 For this reason, it comes as no surprise that a governmentally instituted dual pricing

59 ibid, 5; and Steenblik (n 52) 184, referring to G-20, Leaders’ Statement, The Pittsburgh Summit, 24–25 September 2009, available at: www.g20.org/Documents/pittsburgh_summit_leaders_statement_ 250909.pdf. 60 For instance, in the US, they amounted to around $600 billion in the year 2011, Coady et al (n 53) 5: around $600 billion in the US in 2011. 61 Steenblik (n 52) 184. 62 Asmelash (n 1) 267. 63 V Pogoretskyy, ‘Energy Dual Pricing in International Trade: Subsidies and Anti-Dumping Perspectives’ in Y Selivanova (ed), Regulation of Energy in International Trade Law: WTO, NAFTA and Energy Charter (Alphen a/d Rijn, Wolters Kluwer, 2011) 247 stating: ‘Indeed, it leads to inefficient and artificially stimulated consumption of energy resources, especially, with respect to coal and coal-based electricity’. 64 Steenblik (n 52) 187. 65 The European Union and the United States have been opposed to dual pricing policies since the 1970s oil crises, although the US stance on the matter may have changed since the discovery of large shale gas supplies on US territory. 66 Especially OPEC Members (Saudi Arabia) and Russia are notorious for implementing dual energy pricing policies, R Quick, ‘Dual Pricing’ in J Pauwelyn (ed), Global Challenges at the Intersection of Trade, Energy and Environment (Geneva, Centre for Trade and Economic Integration, the Graduate Institute Geneva, 2010) 194. 67 WTO, World Trade Report 2010, Trade in Natural Resources (Geneva, World Trade Organization, 2010) 173 ff.

100 Anna-Alexandra Marhold programme could be seen as equivalent to the ‘provision of goods or services’ under Article 1.1(a)(1)(iii) ASCM.68 For instance, dual pricing schemes have led to capitalintensive investments in the Saudi petrochemical sector.69 This is all well and good for the country applying such schemes. In fact, countries that utilise energy dual pricing themselves usually justify these policies by using the argument that it is a form of increasing their development, ie, a developmental tool.70 However, there are negative externalities for industries of other WTO Members: it affects the competitive balance between energy-intensive industries and the inputs for products of such industries.71 ii. Fossil Fuel Subsidies and WTO Law: Legal Difficulties under the ASCM Despite the negative effects fossil fuel subsidies may have on the environment and the economy, it is questionable if such subsidies in the form of dual pricing policies would be deemed to fit the definition of Article 1.1(a)(1)(iii) ASCM in practice. In the WTO US—Export Restraints case, dual pricing policies, although not on fossil fuels, were discussed before a WTO Panel.72 Canada challenged the US approach that treated its export restraints as ‘financial contributions’ in countervailing duty investigations to suspected subsidised imports.73 The US treated these restraints as financial contributions, in the sense of government entrusted or directed provision of goods by a private body in the sense of Article 1.1(a)(1)(iv) SCM Agreement. Yet, the Panel disagreed with the US and concluded that the treatment of export restraints cannot qualify as a financial contribution, therefore making it inconsistent with Article 1.1(a) of the SCM Agreement.74 The Panel did emphasise, however, that its findings were limited to this instance only.75 This therefore does not exclude that a Panel may judge differently if the question on dual pricing policies in the fossil fuel sector arises before it. It moreover begs the question whether cheaper inputs through dual pricing always confer a ‘benefit’ in the sense of Article 14(d) ASCM.76 This Article prescribes that in order to confer a benefit, the good in question has to be provided at a ‘less than adequate remuneration’. In such instances, one must look at the prevailing market conditions to determine the adequate remuneration for a good. The problem with this approach is that in many of the countries where dual pricing policies are implemented, the government is often the predominant provider of the good.

68

Art 1.1(a)(1)(iii) SCM Agreement (n 2) and World Trade Report 2010 (n 67) 173 ff. Quick (n 66) 195. 70 World Trade Report 2010 (n 67) 174. 71 Pogoretskyy (n 63) 247. 72 WTO, Panel Report, United States—Measures Treating Exports Restraints as Subsidies, WT/DS194/R and Corr 2, adopted 23 August 2001, DSR 2001:XI, 5767. 73 World Trade Report 2010 (n 67) 173. 74 See Panel in US—Export Restraints (n 72) paras 8.19, 8.75 and 8.76. 75 ibid. 76 Art 14(d) SCM Agreement (n 2) (Calculation of the Amount of Subsidy in Terms of the Benefit to the Recipient). 69

WTO Subsidy Rules: Implications for Energy 101 Consequently, it becomes difficult, if not impossible, to assess what the objective market conditions would be. In the case of Saudi Arabia, for instance, it is the Stateowned Saudi Aramco company that is in control of fossil fuel extraction and trade. And the standing timber at the centre of the Canada—Softwood Lumber IV subsidies dispute was supplied chiefly by the provincial governments.77 Consequently, in cases where the government is the main provider of the good, the WTO DSB may have to look at practices in other markets in order to be able to assess whether a benefit was conferred.78 But the main problem concerning fossil fuel subsidies is that, no matter how tradedistortive they may be, they usually confer a general benefit to a country (in terms of greater income from sales of energy for a higher price abroad, ie, increased aggregate welfare), rather than being a specific subsidy in the sense of Articles 1.1 and 2 ASCM.79 While fossil fuel subsidies may constitute subsidies in the purely economic sense, this type of behaviour is difficult to place in the definition of a subsidy as it stands in WTO law at present, as it would likely fall short of the specificity requirement.80 Because of their nature, fossil fuel subsidies tend not to be ‘specific to an enterprise or industry or group of enterprises or industries’ within the meaning of Article 2.1 SCM.81 The reason for this is that the revenues from fossil fuel subsidies, if maintained on a governmental level, are usually collected by the State and used for boosting its general budget, rather than using the returns for specific, targeted programmes. While fossil fuel subsidies may be ‘specific’ to the energy sector in a given country, it is not the energy sector as such which is the intended beneficiary of such subsidies. Rather, it is the other way around: revenues from the traditional energy sector are used to subsidise the general budget and national industries that use energy as an input. Even if fossil fuel subsidies may be de facto subsidies through the provision of discounted goods, this would not necessarily result in a de jure specific subsidy, because in most cases, the low-priced product will be generally available within the economy of the subsidising government, ie, available to all users, and there may be difficulty in meeting the ‘specificity’ threshold set out in the SCM Agreement.82

77 In the Canada—Softwood Lumber IV Case, Appellate Body Report, United States—Final Countervailing Duty Determination with Respect to Certain Softwood Lumber from Canada, WT/DS257/AB/R, adopted 17 February 2004, DSR 2004:II, 571 the Appellate Body stated, for example, that: ‘[I]t is likely that (the government) can affect through its own pricing strategy the prices of private providers … inducing (those providers) to align their prices to the point where there may be little difference, if any, between the government price and the private prices’, paras 101, 103. 78 World Trade Report 2010 (n 67) 174. 79 Quick (n 66) 195. 80 Steenblik (n 52) explains this well: while subsidies will show in the governmental budget, market transfers do not. 81 Art 2.1 SCM Agreement (n 2). 82 See G Marceau, ‘The WTO in the Emerging Energy Governance Debate’ in J Pauwelyn (ed), Global Challenges at the Intersection of Trade, Energy and Environment (Geneva, Centre for Trade and Economic Integration, the Graduate Institute Geneva, 2010); and World Trade Report 2010 (n 67) 36.

102 Anna-Alexandra Marhold Article 2.1(c) does, however, mention four factors that can be taken into consideration when assessing whether a de jure non-specific subsidy may nevertheless be de facto specific: 1. The use of a subsidy programme by a limited number of certain enterprises. 2. The predominant use of such a programme by certain enterprises. 3. The granting of disproportionately large amounts of subsidy to certain enterprises. 4. The manner in which discretion has been exercised by a granting authority in the decision to allow a subsidy.83 Whether certain fossil fuel subsidies would rhyme with these factors would have to be assessed on a case-by-case basis. Nevertheless, it is highly likely that revenues from, for example, dual pricing programmes and the ways in which they are channelled into the national budget, and in turn spent by the State, may still be considered too general to comply with any of the factors set out in Article 2.1(c). For fossil fuel subsidy schemes to fit the definition of a subsidy as set out by the SCM Agreement, WTO subsidies regulation would have to add one crucial legal element that is currently missing: there is no provision in the Agreement that disciplines the transfer between consumers and producers through overall government policies that raise or lower the price of a product compared with what it would have been in the absence of the subsidy.84 Such transfers would generally be based on an import/ export tariff or other trade barriers.85 In the case of energy dual pricing, for instance, an export tax on crude petroleum may drive the price of the international market up. The government will collect the export tax and redistribute it, thereby benefiting domestic producers and consumers.86 There may have been good reasons for not including these types of transfer in the rules of the ASCM. One explanation is that you cannot show specificity in these instances, without having to include any form of governmental income support in it. As we have seen, this missing element would come very close to the concept of contributing to the aggregate welfare of a country which for several reasons is not at the heart of the SCM Agreement, as discussed in section II. If this element had been been added to WTO subsidy disciplines, however, the result might have caused an undesirable externality. Paramount to the WTO Dispute Settlement System is that a dispute must be brought by one WTO Member challenging the fossil fuel subsidy schemes of another Member to begin with. This could lead to unwanted consequences: Members initiating subsidy disputes against one another based on very general governmental policies that do not target one specific industry or recipient. These would not only lack merit, they would be harmful for the state of the multilateral trading system as such. Instead of the SCM Agreement providing for any legal provisions for this particular type of transfer through overall governmental

83 84 85 86

Art 2.1(c) SCM Agreement (n 2) and World Trade Report 2010 (n 67) 174. Steenblik (n 52) 185. ibid. ibid.

WTO Subsidy Rules: Implications for Energy 103 policies, the Agreement focuses on subsidisation in other forms. The outcome for fossil fuel subsidies, however, is that it is problematic to tackle them under current WTO subsidies disciplines. It seems difficult at present to redesign the Agreement in a way that would include these subsidies in its scope, yet ensure that general governmental policies do not fall within this definition. In conclusion, it proves hard to fit fossil fuel subsidies into Articles 1.1 and 2 of the ASCM, making the WTO legal framework less than an optimal forum to address these harmful subsidies effectively. Nevertheless, it worth noting that former WTO Director General (DG) Pascal Lamy said the following with respect to fossil fuel subsidies: ‘discussion on the reform of fossil-fuel subsidies has largely bypassed the WTO. This is a missed opportunity’.87 Current DG Roberto Azevedo has also touched upon the issue.88 While it is a good idea to rethink fossil fuel subsidies in the WTO context, the question is to what extent the ASCM can play a role in reducing fossil fuel subsidies or whether other avenues should be explored. Instead, as the section will argue, it may be more fruitful to see in which ways subsidies for renewable energy could be permissible under the Agreement.

B. Renewable Energy Subsidies: Legitimate Policy Goals but Sensitive to WTO Dispute Settlement i. Overview Due to their policy objectives, support schemes for clean energy generally enjoy broad support around the world.89 Their main goal is—or in any case should be— not to be protectionist in nature but rather to mitigate climate change caused by CO2 emissions, which is traditionally associated with fossil fuel combustion for electricity generation and heat production.90 This wide-ranging consensus about the need to promote universal access to sustainable energy through the deployment of renewable energy was confirmed in the Preamble to the United Framework Convention on Climate Change (UNFCCC) COP21 Paris Agreement in late 2015.91 We have seen above that fossil fuel subsidies usually come in the form of pre-tax consumer subsidies. The opposite is true when it comes to support programmes for clean energy. Producer, as opposed to consumer, subsidies are much more common

87 See T Meyer, ‘Energy Subsidies and the World Trade Organization’ (2013) 17(22) ASIL Insights (10 September 2013); and P Lamy, Director-General, WTO, Remarks to the Workshop on the Role of Intergovernmental Agreements in Energy Policy (29 April 2013), available at: www.wto.org/audio/ wks24042013_dgpl.mp3 (audio recording). 88 See WTO website: ‘Azevedo seeks further cooperation with Energy Charter to support access to energy’ www.wto.org/english/news_e/spra_e/spra55_e.htm. 89 See in this context also generally E Barrett Lydgate, ‘Biofuels, Sustainability and Trade-Related Regulatory Chill’ (2012) 15 Journal of International Economic Law 157. 90 International Energy Agency, CO Emissions from Fuel Combustion: Highlights (Paris, International 2 Energy Agency, 2013). 91 COP21 Paris Agreement: United Framework Convention on Climate Change (UNFCCC), UN Doc FCCC/CP/2015/L.9/Rev.1 ‘Adoption of the Paris Agreement’ (12 December 2015) Preamble.

104 Anna-Alexandra Marhold to support their scale up.92 Since the share of clean energy in the overall energy market is growing but still small, governments—under pressure to meet their climate targets—are often eager to design support schemes for the production of clean energy due to its positive effects on lowering greenhouse emissions.93 We can assert that without government intervention, many of the markets for clean and renewable energy would not exist in the first place. An important element with regard to renewable energy subsidies here is that the government through its policies may sometimes be seen as ‘creating’ the market for renewable energy. It can be complex to determine the amount of the subsidy and find the proper ‘market benchmark’ to test against, in particular if that market was non-existent prior to governmental involvement. This was one of the issues in the Canada—Renewable Energy/Feed in Tariff case.94 According to the WTO AB in that case, a distinction should be made between instances where the government intervenes in an existing market and when the government actually creates a market through its interventions.95 From an economic perspective, many subsidies for clean energy are set up as investment subsidies to expand renewable energy capacity.96 They usually come in a variety of forms such as FITs, power purchase agreements, capital grants and soft loans, favourable tax treatment, and funding for R&D.97 In the US, a clear example of a renewable energy investment subsidy is the Federal Production Tax Credit, which gives a 2.3 cent incentive per kilowatt-hour for the first 10 years of the operation of a renewable energy facility.98 Then there are renewable portfolio standards which subsidise clean energy implicitly in the form of green certificate values, through the mandate to buy a certain percentage of renewable energy.99 The FIT is arguably the most popular financing mechanism at present and is used as an example in this study.100 Through a FIT, the government guarantees to pay a

92

Steenblik (n 52) 186. See, eg, the 2030 Energy Strategy by the European Commission, available at:www.ec.europa.eu/ energy/en/topics/energy-strategy/2030-energy-strategy and COP21 Paris Agreement (n 91). 94 Canada—Renewable Energy / Canada—Feed-in Tariff Program (n 46). 95 Asmelash (n 1) 273 and Appellate Body Reports Canada—Renewable Energy / Canada—Feed-in Tariff Program (n 46) para 5.118. and 5.169: In this instance, the Appellate Body found that the relevant market for solar and wind power electricity are the competitive markets for wind and solar-generated electricity that results from the specific energy supply-mix set by the government, but not the single market for electricity generated from all sources of energy. See ibid, para 5.174. where a government creates a market, it cannot be said that the government intervention distorts the market, as there would not be a market if the government had not created it. 96 S Charnovitz and C Fischer, ‘Canada—Renewable Energy: Implications for WTO Law on Green and Not-So-Green Subsidies’ (2014) EUI Working Papers, RSCAS 2014/09, 4. 97 Asmelash (n 1) 8 quoting A Ghosh with H Gangania, Governing Clean Energy Subsidies: What, Why, and How Legal? (Geneva, International Centre for Trade and Sustainable Development, 2012); FITs are a guaranteed price for clean energy by producers. 98 Charnovitz and Fischer (n 96) 4. 99 ibid. 100 International Renewable Energy Agency, Renewable Energy Auctions in Developing Countries (Abu Dhabi, IRENA, 2013) 6; International Renewable Energy Agency, Renewable Energy Auctions: A Guide (Abu Dhabi, IRENA, 2015) 13; and generally United Nations Environmental Programme, Feed-in Tariffs as a Policy Instrument for Promoting Renewable Energies and Green Economies in Developing Countries (Geneva, United Nations Environmental Programme, 2012). 93

WTO Subsidy Rules: Implications for Energy 105 certain (above-market) price per kilowatt-hour to the producers of renewable energy ‘to feed it into the national energy grid’. FITs are targeted at future investments through offering new producers of clean energy long-term contracts for this elevated price.101 To make a FIT programme effective, prices must be set high and be stable enough to provide enough incentives for those investments. In turn, suppliers of electricity are then required to buy electricity generated from these clean energy sources. These programmes may be designed in a way that the amount of the subsidy gradually decreases over the years, as the renewable energy in question becomes more profitable and gains more market share in the economy.102 Note that FITs may or may not be transferred by the government directly, or borne by the consumers by an add-on to their energy bill. There are several reasons why FIT programmes have been a successful tool in the scale up of clean energy. First, these programmes usually have a long time frame and are therefore accompanied by long-term price guarantees, meaning that they provide a lot of stability for investors.103 Another advantage is that the programme design of FIT schemes is often flexible and therefore could and should be adapted to the economic needs of the country in question, as well as to the changing market conditions and advances in technology.104 Additionally, FITs may be beneficial for the development of the local production of clean energy.105 However, this proves to be a particularly sensitive element of FIT policies and largely depends on how they are designed, as will be discussed below. ii. Clean Energy Subsidies and the SCM Agreement: Specificity and No Recourse to Exceptions With respect to WTO subsidy rules, support schemes for renewable energy such as FITs, have a better chance of qualifying as a subsidy in the sense of the SCM Agreement. This understandably depends on the design of the scheme in question, and to what extent a government is involved in financing the scheme. But the crucial difference with fossil fuel subsidies is that renewable energy subsidies are quite clearly specifically designed and targeted to boost the production of clean energy; it is their inherent policy goal. Therefore, such subsidies will likely be limited to ‘certain enterprises’, namely the producers of clean energy and meet the ‘specificity’ requirement of Article 2.1 SCM Agreement, thereby making it easier to fit the

101

Charnovitz and Fischer (n 96) 5. So-called digression mechanisms. This was for instance the case in Germany: see European Commission, ‘Press release, State aid: Commission Approves German Renewable Energy Law EEG 2014’, Brussels, July 2014, available at: www.europa.eu/rapid/press-release_IP-14-867_en.htm; see also, European Commission, ‘Press release, State Aid: Commission Approves German Aid Scheme For Renewable Energy (EEG 2012)’; and European Commission, ‘Press release, State aid: Commission Approves German Renewable Energy Law (EEG 2014) for Railway Sector, European Commission’, Brussels, available at: www.europa.eu/rapid/press-release_IP-14-2123_en.htm. 103 UNEP–WTO Report, Trade and Climate Change (Geneva, United Nations Environmental Programme and World Trade Organization, 2009) 114 ff. 104 ibid. 105 ibid, 115. 102

106 Anna-Alexandra Marhold definition of a subsidy into Article 1 ASCM. This, in turn, would make such schemes actionable subsidies in the sense of Article 5 ASCM by means of Article 1.2. That in and of itself may be not be a major problem, if the actionable subsidies in question do not cause injury to another Member and are not challenged before a WTO Panel. Nevertheless, they can have cross-border effects, especially when the subsidised energy is traded abroad over a grid, or when it is competing domestically with imported energy. Consequently, once such subsidies do cause harm to the industry of another WTO Member, a dispute may be triggered in the Dispute Settlement System. This leads to the conclusion that the fact that these schemes fall more quickly into the actionable subsidies category makes them immediately more sensitive to WTO dispute settlement proceedings. The Appellate Body in Canada—Renewable Energy/Feed in Tariff must have been aware of this policy paradox, realising that labelling the FIT in question as such would have adverse implications for renewable energy schemes around the world. Perhaps for that reason, it strategically decided that it was ‘unable to complete the analysis’ whether the FIT in question was a subsidy or not.106 As previously mentioned, FIT schemes may be also beneficial for the development of local production of clean energy. Yet, it is this element that is particularly problematic in WTO subsidy legislation: in WTO law, LCRs are contrary to the SCM Agreement by means of Article 3.1(b) and Article 2.1 of the Agreement on TradeRelated Investment Measures (TRIMS).107 The consequence is that under WTO law, FIT policies may not include any local content requirement elements. LCRs were another central issue in the WTO Canada—Renewable Energy/FeedIn Tariffs case.108 In that instance, the Canadian Province of Ontario instituted an FIT programme implemented by the Ontario Power Authority (OPA) in 2009. In its rules, the Authority in Article 2.1 stated that ‘the FIT Contract will require that wind-power Projects and solar Projects achieve a Minimum Required Domestic Content Level’.109 The law demanded a minimum of component parts and services from producers in Ontario. It is interesting to note that while it was Canada that was challenged in the WTO, it was thus a regional, and not national, policy that was challenged in this instance. The involvement of Ontario was extensive, meaning for instance that Ontario Power Generation was responsible for the majority supply of energy. Additionally, Ontario almost completely owned the high voltage transmission system, as well as the Independent Electricity System Operator. Through its FIT programme, Ontario not only wanted to replace coal with cleaner options through adding wind and solar energy to the mix, but it simultaneously intended to provide incentives to enable new green industries and investments in the production of clean energy technologies.110 To this end, Ontario did not limit itself in utilising

106 ie, whether or not the conferred a benefit in the sense of Art 1.1(b) ASCM, AB report para 5.246 of the Canada—Renewable Energy / Canada—Feed-in Tariff Program case (n 46). 107 Art 3.1(b) SCM Agreement (n 2) and Art 2.1 Agreement on Trade-Related Investment Measures (TRIMS Agreement); Marrakesh Agreement (n 2) Annex 1A, 1868 UNTS 186. 108 Canada—Renewable Energy / Canada—Feed-in Tariff Program (n 46) paras 5.78–79. 109 Ontario Feed-In Tariff Program Rules, Version 1.3.2, 29 October 2010, s 6.4(a). 110 Charnovitz and Fischer (n 96) 2.

WTO Subsidy Rules: Implications for Energy 107 a FIT scheme only, but added another policy instrument to the mix, the LRC.111 The EU and Japan, the Members that challenged Canada before a WTO panel in this instance, did not directly attack the FIT scheme as a violation of Article 5 ASCM (Actionable Subsidies). Instead, they based their claim against Canada on discrimination due to the local content requirement in the Rules of the Ontario Authority, invoking Article 2.1 of TRIMS (which is by its nature inconsistent with GATT, Article III(4)). If local content requirements in FIT schemes are considered traderelated investment measures, they become prohibited subsidies in the sense of Article 3.1(b) SCM Agreement per se, leading to a de facto trade barrier.112 There was also a stand-alone claim on the basis of GATT, Article III(4), as well as a claim of violation of Articles 3.1(b) and 3.2 of the SCM Agreement, depending on whether the FIT could qualify as a subsidy in the first place.113 As we have seen in the preceding paragraph, the AB managed to duck that question, in a way that some refer to as ‘legal acrobatics’.114 Returning to the issue of local content requirements, the WTO AB decided that Canada’s LCRs were indeed a domestic requirement in the sense of Article 2.1 of TRIMS and thus automatically a violation of GATT Article III(4).115 This again goes to show that while FIT schemes, even if they may qualify as an actionable subsidy, may be relatively unproblematic, if they are not challenged by another Member under Article 5 ASCM. However, as soon as additional elements such as local content requirements are added to such schemes, the matter becomes more challenging as they easily fall into the prohibited subsidies category of Article 3 SCM Agreement. This does not mean that WTO Members, while knowing that LCRs are contrary to WTO law, shy away from them. There are examples of countries that impose LCRs and domestic assistance on renewable energy nevertheless.116 Another negative externality of this state of play is that renewable energy policies are often instituted on regional and local levels, and not on a national level (as was indeed the case in Canada—Renewable Energy).117 Because of this, national governments, while responsible for the actions of regional governments, may not always be in a position to oversee the design of regional subsidy programmes.118 Consequently, climate friendly subsidies are also at greater risk of being targeted at the WTO for this reason, as local policymakers may be less aware of the design of WTO rules.

111

ibid. See A Cosbey and L Rubini, Does It FIT? An Assessment of the Effectiveness of Renewable Energy Measures and of the Implications of the Canada-Renewable Energy/FIT Disputes (The E15 Initiative, Think Piece) (ICTSD and WEF, Geneva 2013). 113 ibid. 114 See A Cosbey and PC Mavroidis, ‘A Turquoise Mess: Green Subsidies, Blue Industrial Policy and Renewable Energy: The Case for Redrafting the Subsidies Agreement of the WTO’ EUI Working Papers, RSCAS 2014/17; Robert Schuman Centre for Advanced Studies, Global Governance Programme—82 and Canada—Renewable Energy / Canada—Feed-in Tariff Program (n 46), Report of the AB, para 5.246. 115 AB in Canada—Renewable Energy / Canada—Feed-in Tariff Program (n 114) para 5.33. 116 Charnovitz and Fischer (n 96) 4; and OECD Joint Working Party on Trade and Environment, Domestic Incentive Measures for Renewable Energy with Possible Trade Implications (Paris, Organisation for Economic Cooperation and Development, 2011) 46. 117 Meyer (n 87). 118 ibid. 112

108 Anna-Alexandra Marhold This is in stark contrast to fossil fuel policies, which are often national policies and thereby inevitably drafted with more awareness of WTO disciplines. More successful FIT policies avoid using risky LCRs. In Germany, for instance, the government has issued a purchase obligation for all electricity network operators to buy energy from renewable sources at a minimum price. In that instance, the costs of the programme are divided between electricity supply undertakings, buying clean energy, and private electricity network operators.119 The network operators additionally manage the implementation of the of the FIT programme by means of supplier contracts.120 As opposed to the Canadian example, the government, rather than being directly connected with the generation and supply of energy itself, has a mere regulatory role in Germany’s case. Under current WTO subsidy rules, governments wanting to put in place such schemes thus only have one hope left in case such subsidies are indeed deemed actionable under the SCM Agreement: they trust that they will not cause injury to another Member. And in case they do, that the affected Member in question will abstain from taking action against them in WTO Dispute Settlement. The difficulty is, however, that actionable subsidies (if they are not contingent on export or containing LCRs), may be the only legally available policy instrument for governments to promote the production and increase the market share of clean energy at present. Originally, a third type of non-actionable subsidies was taken up in Part IV (Article 8 SCM).121 Article 8 ASCM actually did take into account certain legitimate policy goals for subsidisation and they were deemed non-actionable of the basis of this Article, even if they caused harm to another Member’s industry. Thus, there was an attempt to a certain extent to take into account the underlying wider policy obectives for subsidisation in the Agreement. It concerned three types of subsidy: 1. Those for research and development (Article 8.2(a) ASCM). 2. Regional aid within the territory of a Member (Article 8.2(b) ASCM). 3. Last but not least, environmental subsidies (‘assistance to promote adaptation of existing facilities to new environmental requirements imposed by law and/or regulations which result in greater constraints and financial burden on firms’ (Article 8.2(c) ASCM)).122 119 M Wilke, Feed-In Tariffs for Renewable Energy and WTO Subsidy Rules: An Initial Legal Review (Geneva, International Centre for Trade and Sustainable Development, 2011) 6. 120 ibid 121 Art 8 (expired) SCM Agreement (n 2). 122 ibid, Art 8.2(c): Notwithstanding the provisions of Parts III and V, the following subsidies shall be non-actionable: (c) assistance to promote adaptation of existing facilities to new environmental requirements imposed by law and/or regulations which result in greater constraints and financial burden on firms, provided that the assistance:

(i) is a one-time non-recurring measure; and (ii) is limited to 20 per cent of the cost of adaptation; and (iii) does not cover the cost of replacing and operating the assisted investment, which must be fully borne by firms; and (iv) is directly linked to and proportionate to a firm’s planned reduction of nuisances and pollution, and does not cover any manufacturing cost savings which may be achieved; and (v) is available to all firms which can adopt the new equipment and/or production processes.

WTO Subsidy Rules: Implications for Energy 109 The category of environmental subsidies was and remains especially relevant for the scale up in clean and renewable energy, as one could easily see how such subsidies would fit into this category. The problem is that the non-actionable subsidies in Part IV were to last for an initial period of five years only, subject to renewal.123 Due to a lack of agreement on that renewal among WTO Members, this category of subsidies under the ASCM ceased to exist as of 2000.124 The result is that environmental subsidies that originally fell in this non-actionable category are today either actionable or prohibited subsidies in the sense of the SCM Agreement. This is much to the discontent of many proponents of such subsidies in the context of climate change mitigation.125 Both Rubini and Howse, for instance, argue that in case non-actionable subsidies are not reinstated in the ASCM, GATT Article XX (General Exceptions) should be available to justify them, although this proposal is not without problems itself.126 While this scenario would offer an alternative solution in case Article 8 ASCM is not reinstated, there are also opponents to this view. Mavroidis, for instance, is of the opinion that the whole idea of Article 8 SCM Agreement was that no recourse to GATT Article XX exceptions would be necessary.127 In his view, this was the underlying reason for negotiating Article 8 SCM in the first place.128 Although there is a certain overlap between the lists of GATT Article XX and Article 8 ASCM (‘green’ subsidies could be placed under subparagraphs (b) and (g)), ASCM negotiating history indicates that the idea was to deal with ‘green’ subsidies in a selfcontained manner in the SCM Agreement context.129 Moreover, Mavroidis argues that if GATT Article XX exceptions to the SCM Agreement were to be allowed, such exceptions would not meet the chapeau of the Article, since the chapeau calls for absence of discrimination.130 To add to this, nowhere does the ASCM establish a link with GATT and not one Panel has accepted this view so far. Nevertheless, the frustration of those who advocate in favour of reinstating the expired Article 8 SCM or alternatively seek recourse in GATT Article XX Exceptions can be well understood from looking at the initial problem that the SCM Agreement does not take into account the (policy) reason and the context of a subsidy.131

123

Mavroidis, Trade in Goods (n 12) 566. Pursuant to Art 31 SCM Agreement (n 2); Mavroidis, Trade in Goods (n 12) 566. 125 See notably Rubini, ‘Rethinking International Subsidies Disciplines’ (n 6); and R Howse, Climate Mitigation Subsidies and the WTO Legal Framework: A Policy Analysis (Geneva, International Institute for Sustainable Development, 2010). 126 ibid; however, Mavroidis, Trade in Goods (n 12) 365 believes allowing this would be complicated. 127 See Mavroidis, Trade in Goods (n 12) 365 ff, arguing that allowing GATT Art XX to function as an exception to the SCM Agreement would put into question the idea of establishing a so-called ‘trichotomy’ between prohibited, actionable and non-actionable subsidies. 128 See PC Mavroidis, The Regulation of International Trade—Volume 1: The GATT (Cambridge, MA, MIT Press, 2015) 476–77. 129 See Mavroidis, Trade in Goods (n 12) 365 ff. 130 ibid. 131 PC Mavroidis, G Bermann and M Wu, The Law of the World Trade Organization: Documents, Cases and Analysis (New York, West Publishing, 2010) 567. 124

110 Anna-Alexandra Marhold IV. CONCLUSION: ENERGY SUBSIDIES IN THE WTO—THE NEED TO BALANCE POLICY GOALS AND LEGAL CONSTRAINTS

Because of the nature of subsidies on fossil fuels versus those on renewables, the disciplines of the SCM Agreement currently make it much easier to trigger disputes against renewable energy subsidies schemes, while addressing fossil fuel subsidies in the WTO system proves more difficult. The situation as it stands thus leads to the following ‘policy paradox’: Members are unlikely to challenge harmful and trade distorting fossil fuel subsidies in the WTO, inter alia, because of their presumed incapacity to meet the specificity requirement under Article 2 ASCM. Clean and renewable energy subsidies, on the other hand, by way of their design, are much more ‘sensitive’, so to speak, to being the subject of a trade dispute in the WTO Dispute Settlement System. And the evidence is there: a whole string of clean energy subsidy disputes is being litigated in the WTO, while no such disputes under the SCM Agreement have taken place against fossil fuel subsidies. The policy paradox is caused by two important elements in WTO subsidy disciplines as they stand now: (1) the lack of specificity in fossil fuel subsidies pursuant to Article 2 SCM Agreement, a condition that needs to be met to make the subsidies in question at least actionable; and (2) the fact that support schemes towards renewable energy are often producer subsidies and specific per se. This latter element is aggravated by the reality that a category for non-actionable subsidies is no longer available in the SCM Agreement and straightforward exceptions are not present either (although the applicability of Article XX GATT has so far not been tested in practice). While it is very difficult to tackle fossil fuel subsidies in WTO law at present, there should be, at minimum, avenues in WTO law to legitimise subsidies for the scale up of clean energy. The most straightforward way to do this would be to reinstate the expired ASCM Article 8. While this is easier said than done, the WTO Membership should look beyond direct obstacles and remind themselves what they committed to in the Paris Agreement in late 2015.132 In the absence of this, the applicability of GATT Article XX defences and their applicability to the ASCM could be put to the test in dispute settlement. Without a doubt, there can be a fine line between furthering justified policy goals and maintaining a similar measure which constitutes disguised protectionism.133 Nevertheless, governments should have options available to them to support clean and renewable energy, without running the risk of being challenged for this purpose only.134 This contribution therefore concludes that it proves challenging to address fossil fuel subsidies in the multilateral trading system at present. Nevertheless, this fact only stresses the need to give countries more policy space in the WTO to incentivise the scale up of clean energy.

132

COP21 Paris Agreement (n 91). See generally D Coppens, WTO Disciplines on Subsidies and Countervailing Measures-Balancing Policy Space and Legal Constraints (Cambridge, Cambridge University Press, 2014). 134 See on this notably T Moerenhout and L Casier, WTO Members, Not the Appellate Body, Need to Clarify Boundaries in Renewable Energy Support (Geneva, International Institute for Sustainable Development, 2013). 133

5 Compatibility of RES: A Legal and Economic Approach FRANCESCO MARIA SALERNO, SÉBASTIEN DOUGUET AND VINCENT RIOUS

I. INTRODUCTION

T

HE 2014 GUIDELINES are the most recent Guidelines to have been adopted after the 2012 Communication on State aid modernisation (SAM). As such, they represent the culmination of a process, which started in 2005, to inject more economic analysis into the State aid compatibility assessment (both under Article 107(3) and Article 106(2))—also dubbed a ‘quiet revolution’.1 The 2014 Guidelines do have some novel features which take their cue directly from economic analysis, namely the use of auctions to allocate funds for green generators and the phase out of feed-in tariffs. The introduction of these novel features has not been without criticism and is currently subject to appeal in Luxembourg.2 This chapter shows that the injection of more economic analysis into the compatibility assessment is part of a trajectory of development which results from the legal structure of Article 107 TFEU. This point is buttressed with a benchmark comparison with the structure of Article 101 TFEU, which displays a similar framework. There are also undeniable differences in the structure of Articles 101 and 107 TFEU, and so the speed of evolution of the role of economic analysis is different between antitrust and State aid. But, aside from this difference in pace, just as was the case with antitrust, the ascent of economic analysis in the State aid compatibility assessment seems a well-established trend, and one that is likely to continue. This chapter provides a combined approach to the 2014 Guidelines provision on RES generation, joining a legal and an economic perspective. Section II starts with the legal analysis of the structure of Article 107 TFEU, and compares it with that

1 See L Hancher, ‘Towards an Economic Analysis of State Aids’ (2005) 4(3) European State Aid Quarterly 425, 431. 2 In September 2014 EREF, the European Renewable Energies Federation, lodged an action for the annulment of the 2014 Guidelines. In November 2015, the Court declared the action inadmissible by order. See Case T-694/14 EREF v Commission, EU:T:2015:915.

114 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious of Article 101 TFEU, laying down the foundations for the method of a benchmark comparison and the findings about the direction of the trajectory. Section III carries out the economic analysis of the 2014 Guidelines provisions on aid for RES generation, and in particular the introduction of feed-in premiums and auctions. Section IV concludes by highlighting areas in the support to RES where an economic analysis could play a useful role in which it currently does not.

II. WHAT ROLE CAN AN ECONOMIC ANALYSIS PLAY IN THE COMPATIBILITY ASSESSMENT? INSIGHTS FROM THE EXPERIENCE WITH ARTICLE 101 TFEU

The structures of Article 107 TFEU and Article 101 TFEU display a number of similarities. These similarities justify using the trajectory of an economic analysis in antitrust as a benchmark for comparing the trajectory of economic analysis in State aid. This comparison sheds light on two—common—features: (i) the tendency towards formulating ex ante rules to administer the power to grant exemptions from the prohibitions laid down in the first sub-paragraphs of both Articles; and (ii) the use of ‘white’ and ‘black’ clauses in formulating these ex ante rules, for which an economic analysis provides the underpinning rationale. The comparison also shows that there are differences in the structure of the provisions, mainly because Article 107 TFEU is addressed to States, while Article 101 TFEU tackles private behaviour. These differences explain why, compared with Article 101 TFEU, the use of an economic analysis in Article 107 TFEU is still in development.

A. The Structure of Article 101 TFEU and its Similarities with that of Article 107 TFEU Several commentators often argue—with good reason—that, within the realm of the competition provisions in the TFEU, State aid is markedly distinct from antitrust.3 However, if one looks purely at the structure of the provisions a number of similarities emerge, which can warrant using the path of economic analysis in one set of provisions as a benchmark for the other. More specifically, the structure of Article 101 TFEU displays a number of similarities with that of Article 107. The table below (Table 1) reproduces the wording of Article 101, sub-paragraphs 1 and 3, and juxtaposes it with the wording of Article 107, sub-paragraphs 1 and 3.

3 F de Cecco, State Aid and the European Economic Constitution (Oxford, Hart Publishing, 2013) ch 3 ‘The Distinctive Nature of State Aid Law’.

Compatibility of RES 115 Table 1: Comparing the Wording of Articles 101 TFEU and 107 TFEU Article 101 1

The following shall be prohibited as incompatible with the internal market: all agreements between undertakings, decisions by associations of undertakings and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the internal market …

2 3

… The provisions of paragraph 1 may, however, be declared inapplicable in the case of: —

any agreement or category of agreements between undertakings, — any decision or category of decisions by associations of undertakings, — any concerted practice or category of concerted practices, which contributes to improving the production or distribution of goods or to promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit, and which does not: (a) impose on the undertakings concerned restrictions which are not indispensable to the attainment of these objectives; (b) afford such undertakings the possibility of eliminating competition in respect of a substantial part of the products in question.

Article 107 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, in so far as it affects trade between Member States, be incompatible with the internal market.

… The following may be considered to be compatible with the internal market: (a) aid to promote the economic development of areas where the standard of living is abnormally low or where there is serious underemployment, and of the regions referred to in Article 349, in view of their structural, economic and social situation; (b) aid to promote the execution of an important project of common European interest or to remedy a serious disturbance in the economy of a Member State; (c) aid to facilitate the development of certain economic activities or of certain economic areas, where such aid does not adversely affect trading conditions to an extent contrary to the common interest; (d) aid to promote culture and heritage conservation where such aid does not affect trading conditions and competition in the Union to an extent that is contrary to the common interest; (e) such other categories of aid as may be specified by decision of the Council on a proposal from the Commission.

The comparison illustrates that there are a number of similarities—as well as differences—in the structure of the provisions, which are summarised in the table below (Table 2).

116 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious Table 2: Comparing the Structure of Articles 101 TFEU and 107 TFEU Article 101

Article 107

Bifurcated structure: prohibition/ authorisation





Need for prior authorisation





Commission’s exclusive competence on authorisation





Sanctions





Private parties as addressees







First, both Article 101 TFEU and Article 107 TFEU have a ‘bifurcated’ structure, in the sense that both are composed of two parts, where the first is a broad prohibition, and the second sets out an exemption.4 — Second, the exemption is granted in the form of a prior authorisation enshrined in a Commission decision. More specifically, for Article 101 TFEU, the notification procedure and the need for a prior authorisation was set out in Regulation 17/62.5 For State aid, its basis is directly in the Treaty: Article 108(3) TFEU (the so-called standstill clause).6 — Finally, in both cases the Commission has exclusive competence to grant the exemption. An assessment of the structure of the provisions, however, also reveals differences: —



While Article 101 TFEU is directed at undertakings, Article 107 TFEU is addressed to Member States. The implication is that the type of acts which are covered are different: in the former case, it is private law agreements, in the latter, it is public law measures stemming from a public policy. Breaching Article 101 TFEU is an offence which carries a pecuniary fine for the undertaking(s). In State aid law, a Member State granting incompatible aid is

4 See, eg, B van Houtte, ‘A Standard of Reason in EEC Antitrust Law: Some Comments on the Application of Parts 1 and 3 of Article 85’ (1982–83) 4 Northwestern Journal of International Law & Business 497; M Siragusa, ‘Rethinking Article 85: Problems and Challenges in the Design and Enforcement of the EC Competition Rules’ in BE Hawk (ed), International Antitrust Law and Policy 1997: Annual Proceedings of the Fordham Corporate Law Institute (New York, Juris Publishing, 1998); R Busscher, M Herz and H Vedder, ‘A Commentary on Article 101 TFEU’ in H-J Blanke and S Mangiameli (eds), The Treaty on the Functioning of the European Union (TFEU): A Commentary (forthcoming). The version of 4 April 2016 is available at: SSRN www. ssrn.com/abstract=2758711. 5 Council Regulation (EEC) 17/62 of 6 February 1962 First Regulation implementing Articles 85 and 86 of the Treaty [1962] OJ P13/2004. 6 The standstill obligation implies that new aid must not be granted or paid to the beneficiary before the Commission has been notified of the matter, has had time to consider it, and has agreed that such aid can be paid.

Compatibility of RES 117 not ‘fined’. The obligation to recover is not a punitive act, but rather ‘the logical consequence’ of a finding of incompatible aid.7 These similarities indicate that the trajectory of economic analysis in antitrust can be a benchmark to compare the trajectory of economic analysis in State aid. Section II.B thus sketches the trajectory of the use of economics in Article 101 TFEU. Section II.C then compares the trajectory of economic analysis in Article 101 TFEU with that of Article 107 TFEU, pointing out similarities and differences, and linking them to the structures of the provisions.

B. The Role of Economic Analysis in the Application of Article 101 TFEU The Commission was often criticised for applying Article 101(1) TFEU with too little regard for an economics-oriented approach and the application of legal criteria to determine what constitutes a restriction of competition.8 This criticism was especially pronounced as regards the treatment of supply and distribution agreements, often called ‘vertical’ agreements to underscore the fact that they are entered into between companies operating at different levels of the production or distribution chain. Typical examples are distribution agreements between manufacturers and wholesalers or retailers. This rigorous treatment also caused a divergence with the more lenient treatment of these agreement in US antitrust law.9 What was thus perceived as a ‘legalistic’ approach to the notion of restriction of competition gave prominence to exemption provisions in Article 101(3) TFEU, and so did the Commission role in handing down exemptions under this provision. But the Commission’s monopoly power to grant exemptions caused what commentators dubbed a ‘system failure’,10 ie, the grinding to a halt of a procedural mechanism

7 See Commission, ‘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, para 13; see also Case C-305/89 Italy v Commission, EU:C:1991:142; Case C-183/91 Commission v Greece, EU:C:1993:233; Case T-324/00 CDA Datenträger Albrechts v Commission, EU:T:2005:364. In fact, the obligation to recover can bring benefits to the Member State in so far as it entails recovering significant sums for the public budget. The recent decision of the Commission concerning €13 billion in forfeited taxation that Ireland has allegedly granted to Apple (Apple (Case SA. 38373), not yet published in OJ) is a case in point as it could have represented for Ireland an occasion to claw back a significant amount of what the Commission believes are unpaid taxes. 8 R Joilet, The Rule of Reason in Antitrust Law. American, German and Common Market Laws in Comparative Perspective (The Hague, Nijhoff, 1967) 77–106, 117 to the end; V Korah, ‘The Rise and Fall of Provisional Validity—The Need for a Rule of Reason in EEC Antitrust’ (1981) 3 Northwestern Journal of International Law & Business 320; MC Schechter, ‘The Rule of Reason in European Competition Law’ (1982) 9 Legal Issues of Economic Integration 2, 1; I Forrester and C Norral, ‘The Laicization of Community Law: Self-help and the Rule of Reason (1984) 21 CML Rev 11; V Korah, ‘EEC Competition Policy—Legal Form or Economic Efficiency’ (1986) 39 Current Legal Problems 85; J Venit, ‘Pronuptia: Ancillary Restraints on Unholy Alliances?’ (1986) 11 EL Rev 213; DL Holley, ‘EEC Competition Practice: A Thirty Years Retroperspective’ (1992–93) 16 Fordham International Law Journal 342. 9 R Wish and B Sufrin, ‘Article 85 and the Rule of Reason’ (1987) 7 Yearbook of European Law 1, 4–8. 10 BE Hawk, ‘System Failure: Vertical Restraints and EC Competition Law’ (1995) 32 CML Rev 973.

118 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious which could not cope with the sheer volume of notifications, while businesses and parties waited in a limbo of legal uncertainty. Before the adoption of Regulation 1/2003, some commentators argued that a possible way out was to inject more economic analysis into the application of Article 101(1) TFEU, so that fewer agreements would be caught, hence less need for the Commission’s use of its powers under Article 101(3) TFEU.11 Spurred by the Court, the Commission did endeavour to factor in a more effectsbased approach, so that fewer agreements would be caught by Article 101(1) TFEU.12 But the main solution to deal with the backlog of notifications piling up was to adopt block exemptions. This solution was possible because Article 101(3) TFEU refers to ‘categories’ of agreement which can be granted an exemption. Following empowerment from the Council, the Commission was thus authorised to ‘block exempt’ several types of agreement, including exclusive distribution and purchasing agreements, specialisation agreements etc. The block exemptions addressed the system failure because agreements within the terms of the block exemption did not need to be notified to the Commission. Block exemptions consist mainly of a list of ‘white’ and ‘black’ clauses. White clauses list restrictions that, although falling within the prohibition of Article 101(1) TFEU, are exempted. Black clauses, by contrast, are restrictions which, if included in an agreement, will remove it from the benefit of the block exemption. In some cases, white (or black) clauses can be traced directly to economic analysis. For instance, the ‘white’ clause ex Article 2(1) of Regulation 1983/83 on exclusive distribution,13 that ‘no restriction on competition shall be imposed on the supplier other than the obligation not to supply the contract goods to users in the contract territory’ finds its rationale in the economic consideration that a distributor requires immunity from intra-brand competition as an incentive for promoting the product. Thus, the lack of economic analysis in Article 101(1) TFEU spurred the development of block exemptions as an instrument to make the best use of the Commission’s monopoly over Article 101(3) TFEU. Block exemptions are based on ‘white’ and ‘black’ clauses. In some cases, economic analysis explains the rationale behind the characterisation of the clauses (ie, white versus black). Against this background, the next section addresses the evolution of the role of economic analysis in State aid to show that a similar approach to Article 107(1) TFEU has also spurred the role of block exemptions and guidelines in State aid as instruments to best administer the authorisations of ex Article 107(3) TFEU.14 These guidelines/block exemptions also display provisions akin to white/black clauses, and some of these clauses find their rationale in economic analysis.

11

See Siragusa (n 4). Case C-56/65 Société Technique Minière v Maschinenbau Ulm, EU:C:1966:38. 13 Commission Regulation (EEC) 1983/83 of 22 June 1983 on the application of Article 85(3) of the Treaty to categories of exclusive distribution agreements [1983] OJ L173/1. 14 For simplicity, the reference is to Art 107(3) TFEU. The same reasoning obtains with Art 106(2) TFEU as a basis for assessing compatibility. 12

Compatibility of RES 119 C. Economic Analysis in Article 107 TFEU compared with Article 101 TFEU Even with the above cursory look at the role of economic analysis in Article 101 TFEU, it is already possible to discern a number of similarities and a similar trajectory. i. Expansionist Tendency in Defining what is Aid After several decades, the notion of aid still raises controversy and is far from settled.15 One discernible trend is a tendency to enlarge the scope of this concept. One of the best examples probably comes from the energy sector, with the debate on State resources. The Court’s restrictive holding in PreussenElektra16 was first eroded by a number of other judgments,17 and then utterly overturned by a Commission decision upheld by the Court.18 The by-product of this expansionist tendency to interpret the notion of aid is, just like in Article 101 TFEU, the broadening of the prohibition in Article 107(1) TFEU and, in turn, the expansion of the sphere of the exemption provisions. ii. Increased Use of Block Exemptions Coupled with Guidelines As the Commission enjoys a monopoly on the exemption from the prohibition of ex Article 107(1) TFEU, it has found itself in much the same situation as it was with the monopoly over Article 101(3) TFEU. The solution has also been similar: the adoption of block exemptions having the same effect as those adopted for Article 101(3) TFEU: processing in an ex ante fashion categories of aid eliminates the need for notification and legal certainty. According to the Commission: In the future, 3/4 of today’s aid measures and about 2/3 of total aid amounts granted by Member States could be covered by the new GBER. This could even extend to up to 90% of all aid measures, if Member States use the GBER to the full extent by designing their measures in order to fit its requirements. This means that only the cases with the biggest potential to distort competition will remain for ex-ante assessment by the Commission before they can be granted to beneficiaries (notification).19

The Commission has also adopted guidelines in the State aid field with essentially the same function as block exemptions, albeit measures that do not fulfil the criteria in the block exemption need to be notified, even if they meet the criteria in the guidelines.

15 Commission, ‘Commission Notice on the notion of State aid as referred to in Article107(1) of the Treaty on the Functioning of the European Union’ [2016] OJ C262/1. 16 Case C-379/98 PreussenElektra, EU:C:2001:160. 17 Case C-262/12 Vent de Colère, EU:C:2013:851; Case C-206/06 Essent Netwerk Noord EU:C:2008:413. 18 Case T-47/15 Germany v Commission, EU:T:2016:281. 19 See the short memo accompanying the publication of the General Block Exemption in 2014, available at: www.europa.eu/rapid/press-release_MEMO-14-369_en.htm.

120 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious iii. Economic Analysis in the Exemption Provisions under Article 107(3) TFEU In 2005, the Commission adopted the State Aid Action Plan (SAAP) which marked the beginning of what was called ‘a more refined economic approach’ to State aid, ie, the implementation of a more economics, effect-based approach to State aid control.20 As will be explained below, in essence, this approach involved balancing between the positive and negative economic effects of aid (the so-called Balancing Test). In May 2012, the drive for injecting more economic analysis into State aid resulted in the Communication on State aid modernisation (SAM).21 The SAM called for a common approach in the revision of the different guidelines and frameworks, which have translated in assessment the principles for compatibility that are common throughout different sectors. In conclusion, the lack of economic analysis in the application of Article 107(1) TFEU is one of the main causes for the expansion of the sphere of Article 107(3) TFEU and that, much as it was the case with Article 101 TFEU, the Commission’s response has been to use ex ante rules in the form of block exemptions and/or guidelines to make the best use of its monopoly in administering the authorisation ex Article 107(3) TFEU. The other similarity in the path of economic analysis in Article 101 TFEU and in Article 107 TFEU is that block exemptions/guidelines are also drafted using ‘white’ and ‘black’ clauses, and that economic analysis explains their rationale. However, the structure of the provisions also bears undeniable differences. The most relevant is that Article 101 TFEU is directed at private law agreements, while Article 107 TFEU addresses public law measures stemming from a public policy. The implication is that, while under Article 101(3) TFEU, economic considerations are the sole measuring rod by which an agreement is ultimately assessed for exemption,22 under Article 107(3) TFEU, economic analysis —



provides a way to carry out the quantification exercise that is necessary for the balancing test in the compatibility assessment (for instance, by calculating the eligible costs which can be covered by an aid measure); and even becomes a requirement for compatibility, in itself, in certain cases.

However, Member States continue to enjoy a margin of discretion to adopt nonefficient aid measures, which nevertheless serve their policy goals. The Court has given support to this view when holding that the economic efficiency of an undertaking in supplying the SGEI and, in particular, the question of whether an undertaking responsible for the SGEI may fulfil its public service

20 See HW Friederiszick, LH Röller and V Verouden ‘European State Aid Control: An Economic Framework’ in P Buccirossi (ed), Handbook of Antitrust Economics (Cambridge, MA, MIT Press, 2008). 21 Commission, ‘EU State Aid Modernisation (SAM)’ (Communication) COM(2012) 209 final. 22 The Guidelines on the application of Art 101(3) TFEU make clear that the balancing exercise conducted in the field of antitrust exclusively takes into account efficiency. As para 42 of the Guidelines states ‘[g]oals pursued by other Treaty provisions can be taken into account to the extent that they can be subsumed under the four conditions of Article 81(3)’.

Compatibility of RES 121 obligations at lower cost is irrelevant for assessing the compatibility of State aid in the light of Article 86(2) EC. Through the assessment of the proportionality of the aid, Article 86(2) EC seeks only to prevent the operator responsible for the SGEI benefiting from funding which exceeds the net costs of the public service (M6 and TF1 v Commission, paragraphs 140 and 141). Therefore, in the absence of Community rules, the Commission is not entitled to rule on the basis of public service tasks assigned to the public operator, such as the level of costs linked to that service, or the expediency of the political choices made in this regard by the national authorities, or the economic efficiency of the public operator.23

The possibility to implement non-efficient aid measures is also borne out by the absence of a general obligation24 to choose beneficiaries pursuant to tender procedures as a condition for compatibility. As long as a Member State can do without a competitive selection process to select aid beneficiaries, this means that aid will be granted to firms with different levels of efficiency and, therefore, the goal for which the aid is disbursed will not be met at the least possible cost.25 The current limitation in the role economic analysis plays in State aid explains some of the unintended consequences and distortions stemming from aid measures. Section IV discusses this issue further with regard to aid for RES generation and how it could be improved by injecting more economic analysis, in particular to avoid the risk of having to disburse aid to correct the negative—and unforeseen— consequences of other aid measures.

D. How Economic Analysis Frames EU State Aid Control: Comparing Potential Distortion with the Value for the Common Interest The compatibility assessment of State aid now relies on common assessment criteria (see Table 3). These criteria are themselves based on the ‘balancing test’, which was 23 Case T-137/10 CBI v Commission, EU:T:2012:584, paras 293–94. Similarly, in Case T-57/11 Castelnou Energía v Commission (T-57/11), EU:T:2014:102, para 170 (quoting judgment in Case C-159/94 Commission v France, EU:C:1997:501, para 101), in reply to arguments that a certain policy goal could be achieved at a lower cost by using alternative measures to the ones the Member State was proposing, the General Court held that a Member State must ‘set out in detail the reasons for which, in the event of elimination of the aid measure in question, the performance of the tasks of general economic interest under economically acceptable conditions would be jeopardised’; conversely, they do not have a duty ‘to prove, positively, that no other conceivable measure, which by definition would be hypothetical, could enable those tasks to be performed under the same conditions’. Both judgments were rendered in connection with Art 106(2) TFEU, but the same considerations would apply to measures assessed under Art 107(3) TFEU. 24 As discussed in section III, the 2014 Guidelines do provide for an obligation to use auctions to choose beneficiaries, thus showing that economic analysis is making strides as a condition for compatibility in and of itself. 25 For SGEI, see Commission, ‘Guide to the application of the European Union rules on state aid, public procurement and the internal market to services of general economic interest, and in particular to social services of general interest’ (Staff Working Document) SWD (2013) 53 final; reply to question 110. See also the reply to question 113: ‘Does the Decision require selection of the least expensive undertaking for the provision of SGEIs? No, this is not required by the Decision. Member States are responsible for defining the SGEIs they want and, in particular, the quality of these services. Where the quality is higher, the costs of providing the service may be higher and the compensation can cover all the costs actually incurred by the company’.

122 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious announced by the Commission in its 2005 State Aid Action Plan as the core of a ‘refined economic approach’ for assessing aid compatibility. According to this framework, an aid measure is compatible with European law if: 1. It is aimed at a well-defined objective of common interest or of market failure correction. 2. It is well designed to design to achieve this objective (ie, if it is efficient and effective).26 3. The trade and competition distortions it might introduce are limited and are offset by its positive effects.

Table 3: Criteria for State Aid Compatibility as detailed in EEAG 2014 Section 3.1(27) of the Guidelines on State aid for environmental protection and energy 2014–2020: The Communication on State aid modernisation of 8 May 2012 called for the identification and definition of common principles applicable to the assessment of compatibility of all aid measures carried out by the Commission. For that purpose, the Commission will consider a State aid measure compatible with the internal market only if it satisfies each of the following criteria: a) b)

c) d)

e)

f)

g)

contribution to a well-defined objective of common interest: a State aid measure aims at an objective of common interest in accordance with Article 107(3) of the Treaty; need for State intervention: the State aid measure is targeted towards a situation where aid can bring about a material improvement that the market alone cannot deliver, for example by remedying a well-defined market failure; appropriateness of the aid measure: the proposed aid measure is an appropriate policy instrument to address the objective of common interest; incentive effect: the aid changes the behaviour of the undertaking(s) concerned in such a way that it engages in additional activity which it would not carry out without the aid or which it would carry out in a restricted or different manner; proportionality of the aid (aid kept to the minimum): the aid amount is limited to the minimum needed to incentivise the additional investment or activity in the area concerned; avoidance of undue negative effects on competition and trade between Member States: the negative effects of aid are sufficiently limited, so that the overall balance of the measure is positive; transparency of aid: Member States, the Commission, economic operators, and the public, have easy access to all relevant acts and to pertinent information about the aid awarded thereunder.

Source: [2014] OJ C200/01.

26 In a simplified way, an aid measure qualified as State aid is effective if it enables the achievement of the objective (ie, to correct the market failure or to achieve the objective of common interest). It is efficient if it achieves the objective at the least cost.

Compatibility of RES 123 In practice, the Commission uses the balancing test in a simplified way by incorporating its criteria in streamlined assessments and decision processes, as well as in sectorial guidelines for aid design. On the one hand, the gradual assessment methodology used by the European Commission (see Table 4) ensures that aid measures that would qualify as State aid but present a limited (or no) magnitude of distortions can be exempted from State aid notification as long as they respect a simplified set of conditions (eg, the amount of the aid, objective of the aid), set out in block exemptions. In such cases, the potential negative economic impact of the aid measure is presumed low enough that the policy objectives prevail and that no further analysis is needed. On the other hand, the European Commission provides for a standard assessment in a series of sectors (energy, environment, agriculture and forestry, transport etc) for which European objectives of common interest exist. In those cases, the Commission publishes guidelines for aid design, whose rules reflect the balancing test. Much like the case with ‘white’ clauses in antitrust block exemptions, aid measures which comply with these guidelines are presumed to be compatible aid ex ante because their design relies on economically based criteria and tools. Table 4: Criteria for State Aid Compatibility as detailed in EEAG 2014 1

No aid—De minimis

2

General Block Exemption Regulation (GBER)

3

Standard Assessment

4

Detailed Assessment

Source: DG Competition (2008)27.

III. HOW ECONOMIC ANALYSIS EXPLAINS THE EVOLUTION OF RES SUPPORT SCHEMES

The development of electricity generation from renewable energy sources (RES-E) has been promoted by Member States and the European Commission as part of the means to ensure a competitive, sustainable and secure energy market. Specific and regularly strengthened targets have been set, such as a 40 per cent reduction in greenhouse gas emissions for 2030 compared with 1990 levels and EU-wide binding targets for renewable energy of at least 27 per cent in 2030.28

27 Commisison, ‘Vademecum. Community law on State aid’ (2008), available at: www.ec.europa.eu/ competition/state_aid/studies_reports/vademecum_on_rules_09_2008_en.pdf. 28 See Commission, ‘A policy framework for climate and energy in the period from 2020 to 2030’ (Communication) COM(2014) 15 final. The previous 2020 climate and energy package set the 2020 targets at 20% for the cut in greenhouse gas emissions and at 20% for EU energy from renewables. See Commission, ‘20 by 2020—Europe’s climate change opportunity’ (Communication) COM(2008) 30 final. The latest 2015 progress report showed that the projected share of renewable energy was 15.3% in 2014, see Commission, ‘Renewable energy progress report’ (Report) SWD (2015) 117 final.

124 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious To reach these objectives, a series of support mechanisms have been designed and implemented at Member State or European level as liberalised energy markets appeared to be unable to integrate RES due to their low competitiveness as emerging technologies, and because existing market failures seem to prevent the optimal development of RES-E generation. From an economic point of view, several types of externalities are thus identified and can account for under-investment in the sector.29 In particular, the externality in terms of ‘learning effects’ is the main justifier for a continuous support of RES technologies.30 It assumes that conventional technologies (gas-fired, nuclear etc. have benefited over the years from a continuous increase in competitiveness and a decrease in costs and that there is now a ‘lock-in effect’ which ensures that they remain more competitive than any other new technology that may launch today. As a result, an investor in an RES project today faces very high costs, as the technology is still immature, and is not able to compete with conventional power plants, thus losing money. But investing in the project would actually accelerate the ‘learning-by-doing’ process which enables a progressive decrease in costs and the greater competitiveness of RES in the long term, then benefiting the whole system (with lower costs of energy compared with a situation with only conventional technologies). Since the investor does not capture these future benefits in terms of an energy cost decrease, they will not invest today, or not enough, without support. In that case, economies of learning are never activated and RES generation will not able to become competitive. Aid measures are justified in this regard. In response to the market failures identified, the support mechanisms have been aimed at restoring sustainable investment conditions for RES-E generation within or outside the energy markets, in particular through the instauration of long-term contracts (eg, feed-in tariffs) or quota schemes. Thanks to these measures RES-E generation has benefited from considerable growth over the last decade, with installed capacity in the EU-28 reaching 260 GW in 2015.31 But these support mechanisms, however successful, also imply market and competition distortions at EU level which can prove contrary to the common interest. In the aftermath of Directive 2009/28 with its binding RES targets, as the scale of support offered by Member States grew, more and more schemes have been subject to State aid scrutiny. This trend was also spurred by an expansionist approach in the interpretation of the notion of aid, as noted earlier. As a consequence, most new

29 In economic theory, externalities happen when some economic parties do not bear the cost of their action (negative externalities) or when some economic parties do not capture the social benefits of their actions (positive externalities). This results in under- or over-investment in RES capacity and overall non-optimal outcomes in the energy market. 30 Other major externalities that are identified and justify support for RES are: (1) the negative externality in terms of reduction in CO2 emissions, as the CO2 price is currently too low to enable RES-E generation to capture the benefit directly from the energy price (in theory, if well priced, the energy price integrates the cost of CO2 emissions); and (2) the positive externality in terms of research and development, as investments in R&D can never be capitalised by the investor and also benefit the new entrants. 31 ENTSO-E, Statistical factsheet 2015 (2016).

Compatibility of RES 125 support schemes are now concerned by State aid control and their design should comply with the criteria and balancing test conditions enforced by the European Commission. Compared with the former context, economic analysis thus plays a reinforced role with regard to the validation and design of RES-E support schemes. It should verify that market failures impacting incentives to invest in RES generation are still present and still justify renewed support. It should also ensure that the positive effects of the new schemes for the common interest are greater than the distortions generated on EU trade and competition.

A. Evolution of the EU Guidelines for RES-E Support Schemes With renewable energy being a strategic sector with regard to EU policy objectives, the aid measures benefit from the simplified standardised approach for State aid control. Since 2001, they are subject to guidelines which set the conditions for the ex ante compatibility of aid. Accordingly, these guidelines encompass the findings and recommendations of economic analysis, but the simplified approach brings risks regarding the efficiency of compliant aid measures. On the one hand, economic criteria could be put aside or subdued in favour of general and more political criteria, although the issue has not directly affected RES-E generation. On the other hand, the application of simplified and quickly outdated criteria can lead to the adoption of support schemes whose design is not (or no longer) relevant from an economic point of view. Those latter issues were particularly critical in the case of the 2008 Environmental Aid Guidelines. Indeed, over the last decade disruptions and evolutions in the energy sector have fundamentally altered the economics of renewable energy sources and the conditions for their integration in the energy market. RES-E generation has gone from a measured development to massive deployment levels, leading to high distortions of the energy system and new competition issues that did not exist or were relatively limited when Environmental Aid Guidelines were first implemented. Meanwhile, renewable energy sources have become more competitive and their level of maturity has improved enough that direct participation in the energy market can be envisaged. Correlated issues have been identified, such as how to limit the cost of support to mature technologies while encouraging innovation and the pursuit of RES cost decreases. As a result, by 2014 the 2008 Environmental Aid Guidelines had become outdated and were at risk of encouraging the development of inefficient and outdated support mechanisms for RES-E generation, all the more so as more aid measures began to be notified as State aid. The new Energy and Environmental State Aid Guidelines (EEAG) implemented by the European Commission in 2014 have sought to solve these issues. They introduce a major reform in the design of State aid by reinforcing the place of economic rationale in compliance criteria. Their objectives are rearticulated towards a better design of environmental and energy support schemes, which should be designed by taking into account the following four dimensions: (1) technological neutrality, (2) competition, (3) market integration and (4) the efficiency of the mechanism.

126 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious Concerning RES-E generation in particular, the EEAG focus on minimising the potential distortions that support schemes might generate. But they also prepare for the gradual phase-out of RES-E support schemes by taking into account the improved competitiveness and maturity of some technologies and by providing rules for integration in the energy market. Within the principles of the new Guidelines therefore lies the idea that all RES-E generation will gradually become competitive and will only be developed through market-based mechanisms by 2030.

B. Analysing the EEAG Reform from an Economic Point of View The EEAG aim to improve the way support schemes are rolled out by focusing especially on the level of market distortion and on the selection process for attributing aid to new RES projects. In theory, this should reflect a repositioning of the economic balancing test at the core of the Guidelines, and should thus ensure that new aid measures will be more effective and more efficient in reaching RES penetration targets, and that risks linked with RES penetration will be dealt with. Such a preliminary appraisal can be broadened by actually applying the rationale of the balancing test to the changes introduced by the EEAG. This can be done by testing the new rules according to the following criteria inspired by the balancing test: —



The new support schemes should enable only the development of RES-E projects that face externalities and market failures and cannot be developed through the energy market. The Guidelines should therefore deal with the improved competitiveness of RES-E generation and the near-maturity of some technologies. The new support schemes should be more efficient, ie, they should enable the development of RES-E projects at the least cost for society. This means ensuring that the new measures lead to fewer distortions of competition and trade and/ or that they are less expensive. At an upper level, the provisions of the EEAG should promote the integration of RES into the market.

This approach also enables a test of the findings in section II, ie, that guidelines in State aid share many similarities with block exemptions in antitrust and that they are articulated around ‘white’ (and/or ‘black’) clauses, reflecting an ex ante economic appraisal. i. Summary of the New Rules of EEAG Applying to Electricity Generation from Renewable Energy Sources The EEAG introduce the following key design features for new RES-E aid measures: —

The new aid measures should be based on feed-in premium market-based mechanisms that enhances RES integration into the market. Under these mechanisms, the generators will sell their electricity directly onto the market and will receive a premium on top of the energy price. Current schemes such as feed-in tariffs (generators benefit from long-term purchase agreement and are

Compatibility of RES 127







completely insensitive to price signals) and green certificates (RES quota obligations are given to electricity suppliers) are to be phased out along with the most severe distortive characteristics of existing aid measures. For example, RES-E generators will no longer be incentivised to produce at negative prices, which led to severe market failures under the previous schemes.32 The integration of RES-E generation will be further reinforced by the end of exemption from market obligations and responsibilities for generators. In particular, the scope of balancing responsibilities should now include electricity from renewable energy sources. New aid should be granted through competitive, technology-neutral bidding procedures. This aims to ensure that new RES capacities are selected and supported through competitive mechanisms, which would enable a reduction in the cost of aid measures and to improve the transparency of the costs and technological features of RES-E generation. Some exemptions from the previous rules are provided for small technologies as well as demonstration projects in order to set a level playing field between mature and innovative technologies and between large and small capacities, the latter being disadvantaged by auction procedures (cf section III.B.iii below).

Hereafter the two main aspects of the new Guidelines are analysed from an economic point of view, namely the adoption of feed-in premiums as the main support tool and the implementation of competitive tenders for the selection and support of most new RES capacities. ii. Analysis of the Mandatory Path to a Feed-In Premium The European Commission introduced the systematic application of feed-in premiums for new RES support schemes along with the phase-out of other mechanisms such as feed-in tariffs or green certificates. Economic literature has already studied, in depth, the efficiency and effectiveness of support schemes to renewables and its findings enable a better understanding of the decision of the European Commission. In particular, green certificates have

32 The phenomenon of negative prices is linked to the technical constraints of conventional generation units. Under a certain generation threshold, these units need to stop, thus leading to an additional cost for generators to stop and then start again. To avoid this cost, generators are ready to offer low and even negative prices on the market to ensure that their bids are selected and that they can produce over the threshold. In such a situation, the product of negative price and sold volume must remain lower that the cost of stopping and starting. In itself, the phenomenon does not characterise an inefficient market. However, it is amplified to a severe level due to support schemes to renewables, that often entail the priority of injection for RES generators. Due to priority of injection and feed-in tariffs, the RES generators are indifferent to prices and are ensured to produce first, even at negative prices. As a result, conventional generators have to bid at more extreme negative prices to ensure they can produce enough. All of this leads to the higher frequency and magnitude of negative prices in the market, and a high financial impact on conventional generation.

128 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious already been proven to be less efficient, riskier and costlier than other mechanisms,33 and its phase-out had already been recommended prior to the 2014 reform. The prevalence of feed-in premiums over feed-in tariffs is, however, more debated and appears to depend mostly on the economic characteristics of RES-E deployment. The choice between the two mechanisms indeed implies a trade-off between the costs of exposure to market risk and the benefits of market integration.34 —







Feed-in tariffs are often favoured because they isolate RES-E producers from risks, hence enabling much a lower cost of capital and giving a greater incentive to develop new projects. Developers are thus exposed neither to the risk linked with support fixation and project selection (contrary to tender-based mechanisms or green certificate obligations) nor to the volatility of prices on the energy market. In addition to the lower cost of capital, the feed-in tariffs also seem to enable less costly support levels as regulators compute the level of support ex ante for each technology and with localisation and/or capacity specifications. Provided that asymmetry of information remains low, this leads to better control of future costs and avoids risks linked with deadweight losses, windfall profits or market power. Nevertheless, feed-in tariffs also cause negative effects and economic losses which are contrary to the goals that the Commission has set for the energy market. The mechanism indeed isolates RES-E generators from the market. As a result, those parties are not responsive to market signals and their decisions often bring private gains but losses for the system as a whole. For instance, feed-in tariffs are often associated with priority of injection, which completes the RES-E generation purchase agreement and ensures that their electricity will be sold in priority on the market. This isolates the RES-E generators from the demand risk, as they are guaranteed to get paid for the electricity they produce at almost every demand level, independent of the marginal price on the market. However, this leads to severe market distortions, whose main symptom is the increased frequency of negative prices. Feed-in premiums are mostly contrasted with feed-in tariffs in terms of both advantages and limits. They enable the integration of renewable electricity generation into the market and thus limit the number of distortions that could be caused by other mechanisms, especially in the case where a priority of injection is not possible. Market integration should bring more competition between RES generators and conventional generators, a better transmission of market signals and a better optimisation of the energy markets through market-based mechanisms, thus ensuring a better match of supply and demand. RES generators should be more incentivised to participate in the markets in an economically efficient way.

33 See C Hiroux and M Saguan, ‘Large-Scale Wind Power in European Electricity Markets: Time for Revisiting Support Schemes and Electricity Market Designs?’ (2010) 38 Energy Policy 3135; D Finon and Y Perez, ‘The social efficiency of instruments of promotion of renewable energies: A transaction-cost perspective’ (2007) 62 Ecological Economics 77. 34 A Held, M Ragwitz, M Gephart, E de Visser and C Klessmann, Design features of support schemes for renewable electricity (Ecofys Report, 2014).

Compatibility of RES 129 —

On the other hand, feed-in premiums are based on market principles, which implies a higher cost of capital for RES generators (which are not isolated from market risks in contrast to feed-in tariffs) and higher risks of market power or windfall profits. Economic theory hence expects this mechanism to cost more for the final consumer to support RES-E generation, and all the more so if safeguards are not implemented.35

In the current economic context of RES deployment, the European Commission estimates distortions caused by feed-in tariffs and priority of injection have reached a critical level and justify the transition to potentially costlier feed-in premiums. As already explained in the previous section, an increasing RES-E deployment means that stakes are higher and that the impact of RES penetration on the markets is tangible enough to be addressed. RES-E technologies have also become more competitive and more mature, which lowers the issues raised in terms of the cost of capital and market risk. The risk premium should also be mitigated thanks to the exemption measures allowed by the European Commission, which ensure that the smallest capacities and demonstration projects can still benefit from feed-in tariffs and thus from enabling investment conditions. Furthermore, the coupling of feed-in premiums and competitive bidding should lead to improved efficiency and lower support prices in the support mechanism. iii. Analysis of the Path to Competitive Bidding The new Guidelines associate the feed-in premiums with competitive technologyneutral bidding processes, which will ensure that the selection and the support of beneficiary projects is based on market-based mechanisms. From a theoretical point of view, the passage to competitive tenders should improve the selection of the most efficient and effective projects to reach the related objectives of common interest. These positive effects should be especially reinforced when considering that auctions are designed as technology neutral for larger capacities and more mature technologies, thus driving competition within the whole RES sector and incentivising developers to lower their costs and to innovate. Compared with a situation without such tenders, the global cost of RES support should thus logically decrease and should be better correlated with the actual investment and generation costs. Generators should be more transparent as they bid for the feed-in premium, which should help reduce the asymmetry of information in the sector and will support the strategies of legislators and regulators to reduce the cost of supporting RES. The regulators will, for example, be more informed as to the evolution of technology costs, enabling them to adapt the tendering rules over time to better take into account the evolution of the RES economic context, and therefore to propose a faster decrease in support levels.

35 J Schallenberg-Rodriguez and R Haas, ‘Fixed Feed-in Tariff Versus Premium: A Review of the Current Spanish System’ (2012) 16 Renewable & Sustainable Energy Reviews 293.

130 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious The regulators should also benefit from easier control of the support scheme. Indeed, their ability to fix the maximum procured capacity and maximum support price will enable them to better control the volume of new RES capacities and the related costs. Table 5: Recommended Principles of a Tender for a Feed-In Premium The experiences in competitive tenders for a feed-in premium (eg, in the UK,36 Germany)37 appear to follow common principles in line with the recommendations of the economic theory and the State aid rules. In particular, a competitive tender for a single type of technology should work as follows: The regulator or the legislator fixes both technological and financial limits of the support scheme. For instance, identifying the precise technology (e., solar PV on roof), setting a capacity range per project (eg, capacity between 150 kW and 500 kW). The regulator can also set minimum and maximum levels of cumulated capacities as limits of the tender. A minimum level is necessary to assess the level under which the auction becomes ineffective. A maximum capacity level may be necessary to comply with the grid’s technical limits (congestions, flexibility etc). More often than not, a maximum capacity limit will be dictated by the budget available per year or over the whole period. Operators are expected to bid their strike price, ie, the market price under which they shall be receive a premium.38 In almost all FiP schemes, the regulator also sets a maximum acceptable strike price (ceiling or reserve price) which limits what an operator can bid. This regulatory ceiling is a cap to the outcome of the auction and is fixed by the regulator in a manner that is similar to the old feed-in tariffs: it must incentivise a minimum development of new capacities, but must be kept at the minimum necessary. The administrative strike price is then calculated based on the levelised cost trend of the RES technology and the objectives towards future cost reduction. (continued)

36 See, eg, Department of Energy and Climate Change, Contract for Difference: Final Allocation Framework for the October 2014 Allocation Round (2014). 37 Bundesministerium für Wirtschaft und Energie, ‘EEG-Novelle 2016. Fortgeschriebenes Eckpunktepapier zum Vorschlag des BMWi für das neue EEG’ (2016). 38 When the market price is below the strike price, the RES producer would benefit from a premium that would bring its level of remuneration to the strike price (or in between, in case of a sliding/fixed FiP). When the market price is above the strike price, the RES producer would not benefit from a support any more, and could even be forced to reimburse the difference between the energy price and the strike price. This is the case of the Contract-for-Difference scheme adopted in the UK.

Compatibility of RES 131 Table 5: (Continued) The regulator can also stipulate caps and floors to the premium to be received (independent from the strike price) as well as specific ceilings or rules by technology or capacity. Operators bids also take into account a certain volume of capacity, so that the bid is actually composed of two dimensions (i) volume and (ii) strike price. The bids are then ordered by increasing the strike price and the auction is finalised. In a simplified way, there are two opposing auction designs. Under the pay-as-bid design, every accepted bidder receives its bid and not more. Under the pay-as-clear design, conversely, the last accepted bidder (at which the maximum budget or maximum capacity is reached) sets the strike price that all accepted bids will receive. The auction designs are, of course, more complex and incorporate many other features to deal with the risks of collusion or an ineffective auction. For instance, a pay-as-clear design can stipulate that the ceiling price is received by all bidders in the case of insufficient bid capacity, in order to incentivise more bidders to participate.39 Otherwise, the final strike price is set at the price asked by the last accepted bid. From there, the level of support per generator will eventually depend on the other design features of the auction (eg, pay-as-bid, average or marginal price, eventual cap and floors to the premium, specificities per technology, size. See Ecofys, 2014). The situation is quite similar in the case of technology-neutral tenders, although there can be different administrative strike prices and different capacity constraints for each technology. In such a case, while the last accepted bid (in case the sum of all bids exceeds the overall budget) still sets the final strike price and will effectively receive it, all other accepted bids will receive the maximum between this final strike price and the administrative strike price that was set for their specific technologies.

In practice, however, competitive tenders for RES technologies have often proven to be ill-adjusted and economically inefficient.40 This is due to the fact that tender designs are often simplified and that potential limits are overlooked by the regulators. The main identified limits that drive the inefficiencies of tender schemes are linked to the market failures that tend to characterise market functioning but can be exacerbated in the case of specific auctions: (1) potentially high transaction cost;41 (2) the risks of market power and strategic behaviour; (3) barriers to new entrants with limited size and/or less mature technologies, which have higher costs and cannot cope with the higher risk linked with competitive tenders (the risk of not

39 Del Rio et Linares, ‘Back to the future? Rethinking Auctions for Renewable Electricity Support’ (2014) 35 Renewable & Sustainable Energy Reviews 42. 40 C Batlle, P Linares, M Klobasa, J Winkler and A Ortner, Review report on interactions between RES-E support instruments and electricity markets (Beyond2020, 2012); Held et al (n 34); Del Rio et Linares (n 39). 41 The transaction cost is the cost incurred in an exchange. It includes information cost, administrative cost, monitoring cost, etc. In the specific case of an auction, it is, in particular, influenced by the cost of designing bids (on the generators’ side) and of setting price and volume objectives (on the auction’s administration side).

132 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious recovering bidding costs, penalty risk etc); and (4) global low effectiveness characterised by low participation, low innovation and low technological variety, and leading to the dynamic inefficiency of the mechanism. The auction schemes are also costlier because they are a one-off instance, and are thus exposed at each new iteration to political decisions and low social acceptability. Yet, the reform launched by the European Commission should, in theory, prove to be efficient enough as it responds directly to these potential issues by taking them into account in the Guidelines. Thus, the competition bidding can actually be avoided when the Member State can prove that the procedure would generate negative effects such as low participation, the risk of market power or higher support costs. Besides, exemptions are possible for installations below a certain size (eg, installed capacity of less than 1 MW for solar PV, of less than 6 MW for wind energy) and for demonstration projects and should ensure continuous innovation and limit the risk of barriers to new entrants, which can still benefit from easier support conditions. Finally, the limits and costs of auction processes can be addressed ex ante by Member States through careful design and implementation and thanks to the use of elementary tools and theory.42 In this respect, safeguards such as competition authorities and DG Competition will have a new role to play to ensure that RES competitive bidding complies with economic and legal recommendations. iv. Conclusion on the New Guidelines As with the earlier 2008 EAG Guidelines, the 2014 Energy and Environmental State Aid Guidelines aim to apply simplified rules for the compatibility assessment of RES support schemes. They recommend design features for new support schemes which are based on economic criteria and on the new context of RES-E development. In particular, the Guidelines look to ensure that new support schemes internalise the improving maturity and lower need for support of current RES technologies as well as the higher risk of distortions of the energy market. Just like ‘white’ clauses in the antitrust block exemptions, the main new features enforced by the 2014 Guidelines, namely the path to feed-in premiums and the implementation of competitive tendering, are borne by the balancing test and should at least lead to lower support costs and better integration of renewables into the energy market. Efficiency of the support schemes is finally addressed, as are the overall risks of overcompensating mature technologies or under-incentivising further innovation in the sector. It now remains to be seen if the support schemes that develop under the 2014 Guidelines will effectively apply these new features and actually lead to the desired outcomes (see Table 6 box below for a contextual overview of support schemes in 2017).

42 P Klemperer, ‘What Really Matters in Auction Design’ (2001) 16 Journal of Economic Perspectives 1, 169.

Compatibility of RES 133 Table 6: Recent Changes in RES Support Schemes in the EU As the deadline for complete implementation of the new requirements was fixed at January 2017 in the 2014 EEAG Guidelines, more and more European support schemes have been switching to a combination of feed-in premiums and auctions that comply with the Guidelines’ criteria for operating aid.43 The UK, Italy, Germany, France, Poland and Greece have hence launched or are preparing to launch new, modernised schemes and are joining the Member States which already complied with the new criteria, such as the Netherlands and Lithuania. Still, these new support models show a wide variety of features and options, which translates the degree of interpretation of the current Guidelines and the flexibility that Member States still benefit from. Regarding feed-in premiums, some schemes are still applying a ‘FiT-style’ model, with a sliding premium that ensures a guaranteed remuneration independent of the energy price (eg, the UK), while some other schemes are more original and potentially risky (eg, in Italy the RES generator can benefit from energy prices higher than the guaranteed remuneration). Regarding auctions, two categories of scheme are clearly emerging, with divergent views on technology neutrality. On the one hand, some schemes clearly adopt the new paradigm that seems to be defended by DG Competition, with a technology-neutral competitive process for higher capacities or more mature technologies (eg, in the UK or in Poland). On the other hand, some other countries are still pushing for technology-specific tenders, such as France, Italy and Germany, with their choice relying on the possibility of derogating from the general rule in case of a risk of auction failure. The latest German scheme was thus approved by the European Commission on the basis that technology neutrality would not have ensured the most cost-efficient results, given the specificities of the German market.44

IV. WHAT ECONOMIC ANALYSIS CURRENTLY DOES NOT DO, BUT SHOULD: UNINTENDED CONSEQUENCES

Since investments in RES-E generation have a multi-decade lifespan, support schemes pushing for the development of RES-E have long-lasting economic consequences on the power system. An economically relevant implementation of aid in the sector should thus consider not only their immediate impact (in terms of economic efficiency and of distortions on the national and European power systems) but also the impact in the long run arising from the cumulative amount of RES-E integrated in the national and European power systems. Therefore, the modernisation of RES State aid design should take into account dynamic aspects of RES penetration and support as well as the required evolution of related mechanisms in the energy sector. In particular, an adaptation of market design is needed to keep it efficient with the increasing penetration of RES in the energy market.

43 Note that other Member States such as Spain comply with the Guidelines while not relying on operating aid. In their stead, investment aid is allocated based on competitive, technology neutral tenders. 44 European Commission press release of 20 December 2016: ‘State aid: Commission approves auction scheme for electricity from renewable sources in Germany’, available at: europa.eu/rapid/ press-release_IP-16-4471_en.htm.

134 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious Economic analysis has a further role to play in this respect: whether to design efficient State aid and a market design in a dynamic way, or to avoid building aid on top of aid without any logic (eg, exemption of RES support for some categories of consumers), and to push for the further harmonisation of rules and designs in Europe beyond the scope of the EEAG Guidelines. What should be avoided is a design that could lead through cumulative effects to a structural distortion of the fundamentals of competition, and which would be impossible to redress. This section is organised as follows. First, we investigate how economics could help regarding the question of value deflation coming from the decrease of the energy market price resulting from greater RES integration. We then question how economics could help deal with the impact of RES on the overall energy mix. With a greater share of RES, it will be more and more necessary to consider RES as a mainstream technology and shift from support to a process of market integration of RES, in particular with the provision of services to the network. The final two subsections study the question of sharing efficiently the burden of RES support on a national and European scale.

A. Long-Term Impact of State Aid for RES on Value Deflation and Possible Solutions While the 2014 EEAG Guidelines tackle directly the issue of coherence between RES support and market integration, renewable policy has yet to fully grasp the longterm impact of current support schemes and growing RES penetration. Economic analysis can hence be used to further anticipate the impact of RES State aid in the long run and to suggest improvements to current designs. In this respect, two main types of long-term impact of the support for and integration of RES can be identified. We tackle the first one here and the second one in section IV.B. First, the continuation of support for developing RES-E generation is likely to lead to a paradoxical situation where support to investment might not cease and where RES-E is never fully integrated into the market. This effect, called ‘value deflation’,45 stems from the non-alignment between the decrease in RES investment cost, the decrease in RES marginal value for the system and the decrease in energy price. In a simplified way, the increase in the amount of RES integrated in the power system leads to lower average market prices due to the addition of zero-marginal cost energy resources to the market. All things being equal, this could lead to an increase in support levels when those are calculated as the difference between the total RES cost (or LCOE, levelised cost of energy) and the energy price.46 It can be illustrated as in Figure 1.

45 R Green and TO Léautier, ‘Do costs fall faster than revenues? Dynamics of renewables entry into electricity markets’ (2015) Toulouse School of Economics Working Paper 591. 46 Green and Léautier also show that this effect is reinforced by a reduction of (elastic) demand resulting from the support for RES adding to the energy price.

Compatibility of RES 135

Figure 1: Impact of the Integration of RES on the Power Price and the Required Level of Subsidy

This effect is partly compensated by the decrease in the cost of new RES-E generation and the progress along the learning cost. Nevertheless, empirical studies show that the decrease in cost is not enough to catch up with the decrease in energy prices.47 As a result, unless there is a major drop in RES costs or a stabilisation of energy prices (eg, due to a change in the energy mix), the support of RES may not end and State aid will have failed in getting RES-E generation to a level of competitiveness and direct participation in the market. The issue is finally worsened by a decrease in the actual value of new RES for the system: as RES penetration increases, the marginal value of new RES generation in terms of flexibility, reduction in CO2 emissions or response to externalities gets lower and might even raise doubts as to the rationality of continuing support. State aid design can address the value deflation issue by preventing an escalation of effects. It should enable a further decrease in subsidies and better market integration and it should aim at lowering the cost of future RES. Above all, the key is to avoid future subsidies by pushing now for a technological breakthrough. New RES scheme designs, but also future guidelines for RES State aid (expected for 2021), should thus build on the 2014 EEAG and focus even more on the harmonisation of support schemes and the need to differentiate support depending on technological advancement and maturity. There is indeed a continuum between the different types of support for new technologies depending on their maturity. Foxon and others illustrated the diffusion of innovation and the most efficient support with a so-called S curve.48 47 V Sivaram and S Khan, ‘Solar Power Needs a More Ambitious Cost Target’ (2016) 1 Nature Energy 4; J Jenkins, ‘Can market fixes overcome the declining value of wind and solar at high market shares?’ (Energy Collective, 14 August 2015), available at: www.theenergycollective.com/jessejenkins/2259541/ can-market-fixes-overcome-declining-value-wind-and-solar-high-market-shares. 48 TJ Foxon, R Gross, A Chase, J Howes, A Arnall and D Anderson, ‘UK Innovation Systems for New and Renewable Energy Technologies: Drivers, Barriers and Systems Failures’ (2005) 33 Energy Policy 2123.

136 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious

Market Pull Product/Technology Push More market oriented instrument

Market penetration (indicative)

Self sufficient given environmental taxation, regulations or trading

RD subsidy

Subsidy on production Subsidy on investment

R&D programmes and grants

R&D

ROCs, fiscal incentives such as fuel duty relief

Statutory obligations + grants or capital expenditure–fiscal Public incentives, public procurement, demo procurement grants, demo programmes

Demonstration

Pre-commercial

Supported commercial

Fully commercial

Technology maturity by ‘stage’

Figure 2: The Typical Pattern for the Adequacy between the Types of Support Schemes for RES and the Level of Maturity Source: Foxton et al (2005) 33 Energy Policy 2123.

In our view and following Foxon and others,49 the S curve diffusion model can be divided into 5 stages: (1) invention, (2) the applied R&D phase, (3) the demonstration phase, (4) the pre-commercial release and (5) the commercial release.50 In this representation, the diffusion of technology follows three phases: an initial one with the take-off of technology (stages 1 and 2); a second one with the acceleration of development under the effect of increasing returns from adoption and cost reductions (stages 3 and 4); and a third one with the slowdown of development when the technology approaches commercial maturity (stages 4 and 5). For an efficient support of technological innovations, the support schemes must be adapted to each of these stages. Figure 2 borrowed from Foxon and others provides a summary of the tools that can be implemented at each stage in the development of innovative technology. A policymaker should then build an efficient overall balance of support to innovative generation considering that a portfolio of different technologies (for instance, different chemical combinations or forms of PV panels) exists with different technological maturities. Support should then be proportionate to the innovativeness of technology and the expectations of cost reduction. 49 50

ibid. D Finon, ‘Photovoltaïque: les défauts du tarif d’achat’ (2009) 588 Revue de l’Energie 85.

Compatibility of RES 137 Above all, one should avoid betting only on close-to-maturity technologies, which would result in the effect demonstrated by Green and Léautier with a never-ending support to RES if the cost of new investment does not sufficiently decrease. In this respect, the present 2014 EEAG Guidelines go in the right direction, as they generalise a uniform support method for most mature technologies while enabling more relaxed aid schemes for demonstration projects. However, the efficiency of such a binary system compared with a more linear path of different support designs remains to be seen. Going back to Figure 2, it notably seems that the current designs only consider the beginning of the technological progress of RES technologies. Future improvements should focus in this regard on enlarging the differentiation in the final stages of advancement. Conditions for access to tenders should be tightened, which would give incentives for further innovation while reserving new tools (fiscal incentives etc) to the most mature technologies.

B. Long-Term Impact of State Aid for RES on Energy Mix and Possible Solutions Another long-term consequence of the integration of RES is the risk of overcapacity arising from a rapid addition of new capacities to the market, and which is particularly present in the current context marked by a stable load. Mechanically, the issue of overcapacity is due to the decrease in energy prices that follows the addition of new RES capacities. The power plants, in particular the peak-load generators, see their revenues decrease and are incentivised to reduce their capacity, either through mothballing or shutdown. In theory (cf Figure 3), at the equilibrium, the integration of RES changes the efficient energy mix as it impacts the overall power load. The optimal distribution between baseload, semi-baseload and peak-load power plants naturally changes, with more peak generators and fewer semi-baseload and baseload generators. But, in practice, the generators tend to wait for as long as possible before shutting down their power plants, as they aim to benefit from the shutdown of others. Indeed, when a generator decides to shut down one of its power plants, the price increases. This specific generator then benefits from the price increase as the rest of its capacity is still running, but its competitors also benefit from the price increase without changing their capacity. It is hence through the arbitrage between the capacity reduction, the price increase and the impact on its position relative to the other competitors, that a generator will make its decision. The result of this strategic behaviour is a slowing of the process of price increases and of renewal of the energy mix, which is all the more pronounced given that the decisions to shut down can be implemented only very slowly. Besides, once investment is committed the avoidable cost at the prospect of closing a power plant (mainly maintenance) is quite small, which justifies generators waiting a long time before taking the decision to close their power plant despite a low energy price. The distortions from the ideal energy mix are then immediate: overcapacity is maintained, in particular in semi-load, and peak capacity shuts down first instead of increasing, in theory.

138 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious

Figure 3: Impact of the Integration of RES on the Energy Mix of Conventional Generators

In order to avoid overcapacity and inefficient distortion of the conventional energy mix, State aid for RES should be complemented by a specific market redesign. Indeed, the massive integration of RES and associated supports and State aid induce overall overcapacity and inefficient incentives to shutting down power plants.51 Mechanisms should then be designed to encourage early retirement of the most expensive conventional capacities in excess. The energy price would then recover with two consequences. First, the remaining conventional capacity is better off. Second, the level of required subsidy is lower. As a result, the transition to the new energy mix could be accelerated.52 Although measures explicitly taking into account the impact of RES on the energy mix have not accompanied the roll-out of support schemes, several new solutions have already been developed and could prove to be well adapted to complement additional RES integration. The Commission expects to unveil a legislative proposal on a new market design by the end of 2016. In parallel, Capacity Remuneration Mechanisms can also play a crucial role in particular capacity markets.53 First, it sends a volume signal that eases the

51 B Caldecott and J McDaniels, ‘Stranded generation assets: Implications for European capacity mechanisms, energy markets and climate policy’ (2014) Stranded Assets Programme Working Paper. 52 Agora Energiewende, The Power Market Pentagon: A Pragmatic Power Market Design for Europe’s Energy Transition (2016). 53 L De Vries and P Heijnen, ‘The Impact of Electricity Market Design Upon Investment Under Uncertainty: The Effectiveness of Capacity Mechanisms’ (2008) 16(3) Utilities Policy 215; D Finon and V Pignon, ‘Electricité et sécurité de fourniture de long terme. La Recherche D’instruments Règlementaires Respectueux du Marché Electrique’ (2006) 10 Revue ISMEA—Economie et Société—série Energie; M Hasani and SH Hosseini, ‘Dynamic Assessment of Capacity Investment in Electricity Market Considering Complementary Capacity Mechanisms’ (2011) 36 Energy 277; N Hary, V Rious and M Saguan, ‘The Electricity Generation Adequacy Problem: Assessing Dynamic Effects of Capacity Remuneration Mechanisms’ (2016) 91 Energy Policy 113.

Compatibility of RES 139 coordination of all the generators. Second, it selects the generators with the lowest long-term cost and forces the generators with the highest long-term cost to get out of the market and eventually shut down. More precisely, generators compete on the basis of their avoidable costs or required revenue to balance the net present value of an investment project over the time horizon considered by the capacity market. Indeed, once they invest, there is no further economic rationale to consider their investment cost in their decision. It is only the costs that they can still control that they must rationally consider, that is to say, the avoidable costs over the time frame of the capacity market. Conversely, if they are about to invest, they will bid the revenue they need in the capacity market.54

C. A More Active Role of RES in the Market We have seen previously that the higher the amount of integrated RES, the greater the need for the market friendliness of RES-E generation in the energy market. It is also true when considering the other dimensions of energy as a commercial product, that is to say, reserves and participation in network management. As RES-E generation becomes increasingly market friendly, it should provide more and more services to the TSO, as conventional generation currently does, and economics can also help in this respect towards building an efficient and adapted market design. It is well known that when network constraints are created only by RES, the only way to manage them efficiently from a certain level of congestion (that is to say, after efficient network investments have been realised) is to strongly limit the production level of the RES generators. When the RES production level of these generators is low, they suffer no limitation. When the RES production level of these generators is high, their production level is in practice restricted to respect the network constraints.55 The idea is now also considered for reserves.56 In some situations, the production level of RES is so high that conventional generators and, in particular thermal ones, are not required to run to supply energy. Nevertheless, to cope with uncertainties (unexpected load or generation changes or generators disruption), reserves are required. Relying on RES to provide all or part of the reserves would avoid the cost of must-run generators when the RES production level is high enough to provide energy and reserves. The power system would hence run more efficiently. This additional revenue would also reduce the required level of support for RES, even if only to a limited extent. The provision rules of reserves must then be adapted, for instance as it is currently being discussed in Spain.57 54

J Wilson, ‘Forward Capacity Market CONEfusion’ (2010) 23(9) The Electricity Journal 25. V Rious, ‘Le développement du réseau de transport dans un système électrique libéralisé, un problème de coordination avec la production’ (PhD dissertation, University Paris-Sud 11, 2007). 56 Note that the impact of RES on the level of required reserves should be taken into account in the assessment of costs imposed by RES in the State aid balancing test. 57 I De La Fuente, ‘Ancillary Services in Spain: dealing with High Penetration of RES’ (RED, slides 2015). 55

140 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious

Figure 4: Benefits of RES from their Provision of Reserves in Cases of Situations of High Production Level

Until recently, reserves were provided only by conventional generators (thermal, hydro and nuclear generators) and more recently by consumers. Due to their uncertain production level, RES are generally not allowed to provide reserves. Denmark and Germany are exceptions.58 As a consequence, thermal generators are required to run at minimal load levels (as are so-called must-run generators) in order to be able to provide reserves in case of need; however, this is uneconomical because of the oversupply of RES (see Figure 4).

D. Efficient Sharing of the Economic Burden of RES Support It is not only the integration of RES that disturbs the merit order of the power system, but also the burden subsidy which impacts the behaviour of consumers. Indeed, adding to the market price of electricity, the cost of subsidising RES induces a reduction of demand and thus distorts even further the market equilibrium. Assimilating the RES support cost to a fixed cost (ie, paid upfront before generation units are put into service and generate value for the system), it is possible to use economic theory to demonstrate that it is possible to limit such distortion sharing. Ramsey and Boiteux have shown that the distortion from the recovery of a fixed cost can be optimally limited if it is allocated proportionally to the inverse of consumers’ elasticity.59 The idea is to allocate the smallest share of fixed cost to

58 P Sorknaes, AN Andersen, J Tang and S Strom, ‘Market Integration of Wind Power in Electricity System Balancing’ (2013) 1(3) Energy Strategy Reviews 174; L Hirth and I Ziegenhagen, ‘Balancing Power and Variable Renewables: Three Links’ (2015) 50 Renewable & Sustainable Energy Reviews 1035. 59 FP Ramsey, ‘A Contribution to the Theory of Taxation’ (1927) 145 The Economic Journal 47; M Boiteux, ‘Sur la gestion des Monopoles Publics astreints à l’équilibre budgétaire’ (1956) 24 Econometrica 1, 22.

Compatibility of RES 141 the consumers driving the market equilibrium, that is to say, those with the higher elasticity/price-responsiveness. The same rationale can apply to the cost of subsidising RES, allocating it less to the most elastic/price-responsive consumers.60 In practice, the debate has focused on the exemption of big energy consumers from RES support as they face international competition on their downstream market (eg, metals, steel, chemistry, paper). The idea was to reproduce the subsidies implemented by other non-European countries in order to reduce energy prices and help the development of these industries. However, this rationale may be erroneous. Indeed, it implicitly assumes that big energy consumers are more price-sensitive than other types of consumer like residential or commercial consumers, because they seem able to close their factories and move to more welcoming countries in terms of energy prices. But in fact, this not does not seem to be generally true. For instance, in France, in the case of electricity, the industry (with or without big electricity consumers) is less elastic (respectively –0.8 and –0.5) than commercial or even residential consumers (respectively –1.3 and –1.0).61 Consequently, the exemptions of RES support benefiting big energy consumers can have detrimental effects if they do not follow the Ramsey–Boiteux pricing. It is thus possible to go further and use economic analysis to better align the effects of the RES support burden with the solutions. In particular, it is crucial that the solutions to alleviate the burden on consumers do not lead to a vicious cycle of State aid to combat State aid. The inclusion in the 2014 EEAG of restrictive provisions for the exemption of big energy consumers goes in the right direction, but much work remains in terms of designing an aid scheme for which the cost allocation method is economically efficient (in the sense that it does not drive demand down or lead to competition distortions between countries) and sustainable (in that it enables the prolonged support of renewables over the course of several decades without lasting impact on the attractiveness of electricity as an energy means).

E. Cooperation for Cheaper Development of RES The new Guidelines have brought a major improvement by stating a unique type of support (feed-in premium and competitive tenders) for most new aid schemes. However, the support for RES is still organised at national level while the energy market is going to be European and power flows ignore borders from a technical point of view. Even with harmonised structures, the use of different aid schemes in each Member State is likely to result in competition distortions between countries and in the inefficient RES investment. This might be the case of a country with little RES potential subsidising, at a high price, local RES generation instead of authorising foreign capacities to bid in its auctions.

60 V Rious, ‘L’analyse économique des tarifs préférentiels pour les grands consommateurs d’électricité’ (2014) 1 Concurrences 17. 61 Through a meta-analysis, Labandeira et al show that it is also true for all types of energy sources. See X Labandeira, J Labeaga and X López-Otero, ‘A Meta-analysis on the Price Elasticity of Energy Demand’ (2015) 4 Economics for Energy Working Paper.

142 Francesco Maria Salerno, Sébastien Douguet and Vincent Rious In response to such a distortion risk, it is necessary to further push the possibility of cooperation mechanisms between Member States, which would enable the development of RES through statistical transfers of renewable production, joint production projects or joint support schemes. The idea, first introduced in the 2009 Renewable Energy Directive but not really followed since then, is to help some countries meet their target for RES integration in a more cost-effective way thanks to cheaper resources in other Member States. The European Commission has hence estimated the benefit of cooperation mechanisms up to 10 per cent of the support cost, that is to say, around €31 billion over the whole period up to 2020 at EU level.62 This would make the benefit from cooperation for RES development far higher than benefits from transmission cooperation.63 However, what we notice in reality is that transmission cooperation measures have become more concrete (in particular with the ENTSO-E Ten-Year Network Development Plan) while RES cooperation mechanisms are rare.64 Only Norway and Sweden have agreed on a common market for green certificates but Norway already plans to leave from 2020.65 To date, no other cooperation mechanisms haven been implemented to our knowledge, even when hypothetical cases were studied.66 The situation is however beginning to change slowly, along with the new guidelines and the global modernisation and harmonisation of rules that they entail. The UK and Germany, in their new schemes, hence plan a progressive opening of their schemes to foreign capacities. While this is a very small step, it could galvanise a whole set of new attempts at cooperation. The reasoning should not however stop at cooperation, as new requirements of harmonisation would be raised consequently.67 Indeed, administration can stand for 20 per cent of the operations and maintenance cost.68 As a result, any change in national rules can largely impact the overall cost of RES. Consequently, developing cooperation mechanisms without harmonisation of the most impactful rules for the cost of RES would only trigger a competition between national rules for integrating

62 C Klessmann, E De Visser, F Wigand, M Gephart, G Resch and S Busch, Cooperation between EU Member States under the RES Directive (Ecofys Report, 2014). 63 M Saguan and L Meeus, ‘Impact of the Regulatory Framework for Transmission Investments on the Cost of Renewable Energy in the EU’ (2014) 43 Energy Economics 185. 64 In Case C-573/12 Ålands Vindkraft, EU:C:2014:2037, the Court of Justice took a careful approach in a case that could have otherwise resulted in a judge’s attempt at creating a European green certificate market. This does not mean, however, that such a market cannot be created by the European legislature. 65 See Bloomberg, ‘Norway Seeks to Quit Joint Renewable Subsidy System with Sweden’ (15 April 2016). 66 S Busch, L Liebmann, G Resch, J Nysten and M Gephart, Cooperation under the RES Directive Case studies: Joint Support Schemes (Ecofys Report, 2014); N Cusumano, A Lorenzoni, J Nysten and M Gephart, Cooperation under the RES Directive Case Study: Joint Projects/Statistical Transfer between Malta and Italy (Ecofys Report, 2014); M Gephart, C Klessmann and J Nysten, Cooperation under the RES Directive Case Study: Joint Projects between the Netherlands and Portugal (Ecofys Report, 2014); M Ten Donkelaar, L Kitzing, J Nysten, M Gephart and C Klessmann, Cooperation under the RES Directive Case study: Statistical Transfer between Estonia and Luxembourg (Ecofys Report, 2014). 67 O Sartor, ‘What can EU policy do to support renewable electricity in France?’ (2016) 6 IDDRI Working Papers. 68 Deloitte, Establishing the investment case. Wind power (2014).

Compatibility of RES 143 RES and not between the quality of renewable resources. Rather than pushing for cooperation mechanisms it would be more efficient to first study whether the harmonisation of national rules would be more beneficial to ensure a level playing field.

V. CONCLUSION

Economic analysis has guided the development of the State aid compatibility analysis. The legal structure of TFEU Article 107 coupled with the increased use of block exemptions coupled with guidelines have accompanied a progressively more refined application of economic analysis that intends to address previous and currently observed limitations. The enforcement of compatibility criteria and the design of new sectorial guidelines are then becoming more advanced, allowing issues caused by previous aid schemes to be tackled while anticipating the risks linked with new aid thanks to a whole new set of economic rationale. The refinement of the place of economic analysis is particularly interesting in the case of State aid for renewable energy sources. The expansionist approach in the State aid definition coupled with the massive RES deployment now observed have led to a need for new aid designs which respect the requirements of State aid compatibility criteria and which rely more closely on economic principles. The EEAG 2014 Guidelines are a major step forward as they enforce market-based tools for new support schemes such as feed-in premiums and selection by competitive tenders. This should ensure a better market integration of RES and help ensure the progressive path to competitiveness and the end of support. Of course, this supposes that the new aid schemes are successful in their objectives and thus that design recommendations are followed. In particular, the application of technology-neutral solutions and the treatment of specific renewable technologies such as offshore wind will have to be regarded with caution. It also assumes further efforts that could be developed in the future guidelines, for example, with regard to the long-term impact of RES or the sharing of the support’s economic burden. But the efforts of the European Commission to develop and integrate RES are not limited to the scope of Article TFEU 107 and to the design of support schemes. The evolution of market design and infrastructure regulation and the achievement of a successful Energy Union require thorough work on the impact of RES and the way to align market conditions. There also, economic analysis has been at the core of the modernisation of the market and regulations and new topics are still emerging due to the progressive deployment of RES. Last is the modernisation of guarantees of origin and the standardisation of an EU-wide electricity production tracking system. The next renewable energy directive REDII is expected to address the issue and to give more power to the consumer to access information and to actively participate in energy markets. Nonetheless, as in State aid, the state of reflection is still a work in progress with many changes ahead.

6 Capacity Mechanisms and Auctions LEIGH HANCHER AND CHRISTOPH RIECHMANN

I. INTRODUCTION

A. The Recent Debate on Capacity Mechanisms

E

LECTRICITY GENERATION AND wholesale markets, if only modestly regulated, have tended to pay generators and traders for the energy (the MWh) that they produce, but not the capacity (the MW, or the potential to produce). The policy debate regarding capacity remuneration mechanisms (CRM) centres on the question of whether some further public intervention is required in the market, so that capacity is more explicitly rewarded. As we discuss below there is broad agreement by stakeholders that clear regulatory rules (balancing arrangements, in particular) are required to ensure sustainable investment in the sector. However, there is disagreement over whether this should go as far as explicitly mandating markets for capacity or only mandating rules, so that incentives are created for market driven capacity products. The European Commission (EC) continues to be sceptical of the need for and benefit of explicit CRMs. At the same time, several Member States (and the European regulator ACER) have been more favourable to CRMs. CRMs are not a recent invention. Between 1990 and 2001, for example, the electricity market in England and Wales included a capacity payment as a separate element from the electricity wholesale price. Ireland, Italy and Spain have made capacity payments to electricity generators for many years and in Sweden strategic reserves have existed since 2003. Nevertheless, in recent years an increased interest in capacity mechanisms has led to the planning and introduction of a large number of new schemes in the EU and although not the subject of this chapter, in many other countries including the US and Latin America. The renewed interest in such mechanisms has been accompanied by a number of policy initiatives at EU level, culminating in the launch by DG Competition (DG COMP) of a sector-wide inquiry1 under the recently amended State aid rules and the launching of a new ‘market design’

1 Communication from the Commission, ‘Launching the public consultation process on a new energy market design’ COM(2015) 340 final; Commission, ‘Final Report of the Sector Inquiry on Capacity Mechanisms’ COM(2016) 752 final (Final Report).

146 Leigh Hancher and Christoph Riechmann initiative (MDI) led by DG Energy (DG ENER) which has been reflected in the draft legal package on ‘Clean Energy for all Europeans’.2

B. Aim and Scope of the Chapter The Commission announced at an early stage that it would use the findings of the sector inquiry not only to assess State aid measures but also to develop legislative proposals on a revised electricity market design.3 While both initiatives take different perspectives they are also interrelated: —

Ensuring State aid compliance: The sector inquiry takes a competition law perspective. The aim is to ensure that no beneficiary of any CRM, where implemented, gains an unfair advantage over competitors (those who also benefit or ones who may not benefit). From this perspective the necessity and proportionality of any CRM measures is important and any CRM should address market deficiencies that have been identified locally. Causes for market deficiencies will differ between Member States. If CRMs were then to address the specific national issue this could well lead to very different CRMs being applied in different Member States. State aid compliance also requires the possibility for market players to participate in capacity mechanisms of neighbouring Member States. — Harmonising the internal energy market: One of the concerns of DG ENER in its market design initiative is to ensure the effective functioning of the internal energy market. As individual Member States introduce and develop different CRM designs, there is an increasing concern that the level of market integration that has been achieved over recent years fades as market designs may become less compatible between countries. This concern may require a degree of harmonisation between CRMs where they are applied. — Security of Supply (SoS): DG ENER is also concerned that an uncoordinated evolution of national market designs (some with and some without CRMs) may endanger the reliability and adequacy of generation on the system. As we explore in this chapter, there are therefore some inherent tensions between these perspectives. DG COMP, as the guardian of EU competition law, may favour minimalist CRMs that focus on specific national issues and, as demonstrated by the Final Report on the Sector Inquiry on CRMs, it is concerned that CRMs may increase the market power of incumbents and restrict new entry.4 The advocates of

2 This package was published on 30 November 2016. Also known as the ‘Winter Package’ or ‘Clean Energy Package’, it consists of eight legislative measures to facilitate the transition to a clean energy economy. The overall objectives of each proposed measure are briefly outlined in the Commission Communication ‘Clean Energy for all Europeans’ COM(2016) 860 final. 3 Commission, ‘Interim Report of the Sector Inquiry on Capacity Mechanisms’ C(2016) 2107 final (Interim Report). 4 Commission, ‘SWD accompanying the Final Report of the Sector Inquiry on Capacity Mechanisms’ SWD (2016) 385 final, s 6.4 (SWD accompanying the Final Report).

Capacity Mechanisms and Auctions 147 the internal market may favour a more uniform approach over a nationally targeted CRM approach. Some proponents of a strong SoS policy may argue that CRMs should be mandated, possibly even in Member States that currently still have reservations. A possible way of thinking about the two recent policy initiatives (MDI and Sector Inquiry) is to regard the MDI as a precursory step which helps define, from an energy industry perspective, the common interest and the possible factual need for CRMs. The assessment by DG ENER as part of the MDI also helps define the appropriateness of any mechanisms (eg, by pointing out a ‘no regrets’ measure that should be implemented before evaluating the need for a CRM). As we discuss below, both policy initiatives have also converged on some fundamental principles which are of importance to the ongoing application of the EU State aid rules: —

Adequacy assessment: The need for a CRM will need to be explored through an EU-wide generation adequacy assessment: This assessment needs to take into account the interaction between the electricity systems of different Member States. — Basic principles for market design: The implementation of ‘no regret’ measures should be prioritised over the introduction of CRMs. CRMs should distort the market as little as possible. Market reforms may reduce concerns about security of supply or even remove the need for CRMs altogether. No CRM should be a substitute for market reform. — Cross-border participation: Any CRM needs to be open to participation from other Member States.

C. Structure of the Chapter In this chapter, we consider the evolution of EU policy in light of divergent national CRMs (section II). We then highlight the material issues in the debate as initially reflected in the literature (section III) and subsequently addressed in the market design initiative (section IV) and then in the Final Report of the Sector Inquiry (section V). Section VI traces the impact of the Commission’s developing policy on CRMs and its method of assessment on the application of the EEAG 2014 to individual national cases. Section VII concludes with some observations on the application of the EEAG in light of the new market design proposals.

II. THE EVOLUTION OF EU POLICY

A. The EU Perspective In today’s electricity markets, CRMs can have a dual rationale: —

First, they are claimed to be necessary for generation adequacy reasons and to attract investment in new capacity, often considered necessary to meet the challenge that a large number of conventional plants will have to be closed down as a consequence of EU climate policy.

148 Leigh Hancher and Christoph Riechmann —

Second, capacity payments secure generation flexibility.5 They incentivise existing thermal capacity to stay in the system and provide the backup service to fully integrate renewable energy sources (RES) into electricity systems, even when the run-times of these thermal plants are much lower than originally conceived.

As this latter rationale became more prominent, on 22 May 2013 the European Council called for particular priority to be given by the Commission to provide guidance on capacity payments. The Commission met this call with a Communication of 5 November 2013 on delivering the internal electricity market and making the most of public intervention,6 by assessing the main features of public intervention and providing guidance on how they can be designed or adapted in order to increase their effectiveness.7 In its earlier Consultation Paper opening a public consultation on generation adequacy,8 the Commission acknowledged the significant impact of RES capacity on energy-only markets. According to the resulting November 2013 Communication, capacity mechanisms threaten market integration.9 The final version of the Energy and Environmental Aid Guidelines (EEAG), which entered into force on 1 July 2014 devotes a separate section to State aid for generation adequacy.10

B. National Initiatives How far national CRM designs may differ already becomes clear, when considering some design classifications for regimes that are in use in Europe (See also Table 1): —

Centralised versus decentralised contracting of capacity: In some regimes a central authority, for example, the TSO (Sweden, Finland, UK) underwrites

5 P Cramton and S Stoft, ‘The Convergence of Market Designs for Adequate Generating Capacity with Special Attention to the CAISO’s Resource Adequacy Problem’, White Paper for the Electricity Oversight Board (April 2006); PL Joskow, ‘Energy Markets and Investment in New Generating Capacity’ in D Helm (ed), The New Energy Paradigm (Oxford University Press, 2007); P Cramton, A Ockenfels and S Stoft ‘Capacity Market Fundamentals’ (2013) 2 Economics of Energy & Environmental Policy 27. 6 Commmunication from the Commission, ‘Delivering the internal electricity market and making the most of public intervention’ COM(2013) 7243 final. 7 The Commission accompanied the November 2013 Communication with five Staff Working Documents, one of which deals with generation adequacy in the internal electricity market. Commission, ‘Generation Adequacy in the internal electricity market—guidance on public interventions’ SWD (2013) 438 final. 8 Commission, ‘Consultation Paper on generation adequacy, capacity mechanisms and the internal market in electricity’, available at: www.ec.europa.eu/energy/sites/ener/files/documents/20130207_ generation_adequacy_consultation_document.pdf. 9 This approach had permeated the Commission’s ‘objective justification’ discussion within the Generation Adequacy SWD. The Commission’s early misgivings towards capacity payments also influenced the proposed proportionality assessment, as illustrated by the extensive 13 separate recommendations listed in the ‘recommendations to avoid distortion of internal electricity market’. See FE Gonzales ‘EU Policy on Capacity Mechanisms’ in L Hancher, A de Hauteclocque and M Sadowska (eds), Capacity Mechanisms in the EU Energy Market: Law, Policy, and Economics (Oxford, Oxford University Press, 2015). 10 Communication from the Commission, ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG) [2014] OJ C200/1.

Capacity Mechanisms and Auctions 149





capacity contracts, whereas in other regimes decentralised players, such as retailers, face a capacity obligation and they contract a plant to cover their individual obligations (eg, in the new French regime). Separated from the energy only market or closely linked with it: A plant that benefits from a CRM may be prohibited from also participating in the energy market (eg, in the Strategic Reserve regimes of Sweden and Finland) or they may be allowed to also sell energy through the general energy market (eg, in the UK or under the new French regime). Targeted versus market-wide: The CRM may cover only a small part of capacity (eg, in the strategic reserve regimes of Sweden and Finland) or it may extend to all capacity that is deemed necessary on the system (eg, in the UK or under the new French regime).

Table 1: Capacity Mechanisms in the Sector Inquiry Tender for new capacity

Strategic reserve

Targeted capacity payment

Belgium**

Belgium

Italy

France

Denmark**

Poland

Ireland**

Germany***

Portugal***

Poland

Spain***

Sweden Germany (Interruptibility Scheme) Ireland (Interruptibility Scheme) Italy (Interruptibility Scheme)*** Poland (Interruptibility Scheme) Portugal (Interruptibility Scheme) Spain (Interruptibility Scheme) Central buyer Ireland*

De-central obligation France*

Market-wide cap. payment Ireland

Italy* Source: European Commission based on replies to the Sector Inquiry.11 * Planned Mechanism (or being implemented). ** Past Mechanism (or never implemented). *** Multiple capacity mechanisms of the same type.

11

Interim Report (n 3).

150 Leigh Hancher and Christoph Riechmann III. MATERIAL ISSUES IN THE DEBATE

In this section, we discuss the material issues raised by CRMs in the context of the Sector Inquiry and the MDI. We start out with a short review of the underlying economic issues as discussed in the international literature and consider how DG ENER views the challenges and possible solutions from a broad (rather than a narrower State aid) perspective. A debate on the need for CRMs has not only been held in the EU, but also in other regions of the world and, most prominently, in North and South America. The debate has produced a mainstream framework for thinking about CRMs.12 At the same time, authors will not fully agree both in their assessment on the severity of some of the identified issues and therefore also on the conclusions for policymaking. Furthermore, different national conditions can lead to variations in assessment per country. The main insights from this global debate are outlined below. A. CRMs are a Means to Remedying Certain Market and Policy Failures The important implication of this is that CRMs are not a means in themselves and, from a policy perspective, they need to be justified by a market or policy failure. An assessment of such failures, and also an investigation of whether the failures can be addressed at the root already should therefore precede the design and implementation of any CRM. In other words, there is no default presumption that all countries need CRMs (unlike eg, there is today a presumption that for practical purposes all countries or regions need a central bank to manage monetary policy).13 B. There is a Core Set of Potential Market or Policy Failures that Need to be Explored In essence, all these failures may cause capacity to be insufficiently remunerated. This, in turn, would lead to under-investment and an inefficiently low level of generation adequacy. While different authors may classify market and policy failures differently, a broad classification encompasses:14 —

Market/policy failure I—externalities/public goods nature of security of supply given inelastic demand: First, consumers cannot or do not express their

12 See, eg, Cramton and Stoft, ‘The Convergence of Market Designs’ (n 5); D Finon and V Pignon ‘Electricity and Long-term Capacity Adequacy: The Quest for Regulatory Mechanism Compatible with Electricity Market’ (2008) 16(3) Utilities Policy 143; or WW Hogan, ‘On an “Energy Only” Electricity Market Design for Resource Adequacy’ (2005) Working Paper Harvard University; P Rodilla and C Batlle ‘Security of Electricity Supply at the Generation Level: Problem Analysis’ (2012) 40 Energy Policy 177; D Newbery ‘Missing Money and Missing Markets: Reliability, Capacity Auctions and Interconnectors’ (2015) EPRG Working Paper 1508, Cambridge Working Paper in Economics 1513. 13 See further Final Report (n 1) s 3. 14 See also J Perner and C Riechmann ‘Energy Market Design with Capacity Mechanisms’ in L Hancher, A de Hauteclocque, M Sadowska (eds), Capacity Mechanisms in the EU Energy Market: Law, Policy, and Economics (Oxford, Oxford University Press, 2015).

Capacity Mechanisms and Auctions 151 willingness to pay and respond insufficiently to price signals that otherwise signal scarcity (consumers are price-inelastic). Secondly, in extreme stress situations, with partial supply disruption (brown-out), the prevailing energy price may not reflect the social value of electricity to consumers; some demand with high willingness to pay may be curtailed and the market price has to be defined administratively: — Adverse effect on investment decisions: If these prices earned by generators turn out not to reflect high scarcity values, producers will reflect this lower valuation in their investment (and disinvestment) decisions. As a result, less generation capacity may be held on the system than is economically and socially desirable. The provision of (reserve) capacity therefore has characteristics of a public good: the market provides for less capacity than is socially optimal. — Adverse effects on consumers: Vis-a-vis consumers, there may be inefficient rationing of demand in case of power supply failures. When demand responds insufficiently to high electricity prices in periods of a capacity shortage, random rationing of demand can occur (black-outs or brownouts). Some consumers, who value reliable supply more than other customers, will not have the opportunity to effectively signal their higher valuation, and they may be interrupted along with other consumers in their network area. — Market/policy failure II—market volatility which may create prohibitive price risk: If plant investors depend on price spikes in situations of generation scarcity in order to amortise their investment cost, they will consider investments as particularly risky. With uncertainty about the frequency and the level of price spikes, risk-averse investors will become more cautious and will require higher risk premia, either to invest or to continue operation. This would eventually lead to higher prices for consumers (or in cases where the public is not prepared to afford such premia, a lack of investment). This may be particularly relevant where the volatility of electricity wholesale prices rises with the increasing penetration of wind and solar generation. The basic argument here is one of a market failure in financial markets (to efficiently fund the continued operation of power stations). Such instances may justify the introduction of CRMs, so that part of the financial risk is shifted from plant investors (who receive some form of remuneration for capacity) to the collective of consumers (who eventually pay) and where this outweighs the detrimental effects of the government intervention, for example, increased regulatory and policy risk within a CRM. — Market/policy failure III—the threat of or the actual political intervention in price formation (‘missing money’): In practice, authorities that are wary of price spikes in the short-term energy market may introduce explicit or implicit price caps, which lead to lower revenues for generators to cover their fixed and investment costs. This is known as the ‘missing money’ problem. This could justify the introduction of a capacity mechanism in order to compensate for expected lower revenues. Even the mere threat of price caps may lower investors’ return expectations and this may need to be addressed, if only by

152 Leigh Hancher and Christoph Riechmann



policymakers reassuring the market that prices would not be capped. The key question, therefore, is whether a government or authority can credibly commit to not interfering in the energy market. Authorities with low credibility may be driven to employ CRMs to overcome the missing money problem, even if they would ideally like to avoid the introduction of a CRM. Market/policy failure IV—an issue specific to the EU is that of failures due to lacking coordination in the market design between highly interconnected electricity markets: For example, if country A introduces a CRM, then this could have ambiguous effects on neighbouring country B. On the one hand, country B may benefit from the creation of extra capacity in country A. On the other hand, some investment may be lured away from B to A. The net effect on system reliability in country B needs to be established considering the precise circumstances. However, a risk to B may not be ruled out entirely and this needs to be addressed (by better coordination between countries or potentially by also introducing a CRM in country B).15

C. Some Failures may be Addressed More Directly at the Root Without the Need for CRMs If the key concern is that electricity demand is not sufficiently responsive to price signals, then emphasis could be laid on making consumers more responsive, for example, by strengthening the balancing responsibilities of retailers and end users and by facilitating the dissemination of metering and settlement technology that enables the dynamic pricing of electricity.16 Depending on national circumstances some countries may benefit from CRMs, while others may not. The important implication is that the situation needs to be assessed for each country or wider region. It should consequently be no surprise if some countries opt for CRMs while others do not. The recent debate in Europe has also helped to clarify that CRMs are by no means the only solution to resolving market and policy failures. There are often leaner and less invasive measures that help improve the performance of electricity generation and wholesale markets, for example, a tightening of the responsibility of electricity retailers to maintain their individual energy balances. This is an important lesson that has been a key influence on the debate at EU level and is confirmed in the Sector Inquiry and MD initiative. There is a presumption and requirement that certain leaner market design reforms are designed (and implemented) before assessing the need for CRMs.17

15

See also Final Report (n 1) s 3. The development of demand-side response (DSR) and the reform of short-term and balancing markets are also addressed in the new package, in particular in the recast Regulation and recast Directive on the internal electricity market. 17 See also Final Report (n 1) s 3. 16

Capacity Mechanisms and Auctions 153 The assessment of the costs and benefits of using CRMs also needs to consider the practical complications in implementing CRMs in terms of: —

Capacity not coming on stream as planned—there is growing evidence that the desired and contracted capacities do not necessarily materialise, rendering a CRM less effective than intended.18 — CRMs taking time to bed in and, from experience, need to be reformed and refined as time passes.19

IV. DG ENER’S ‘MARKET DESIGN INITIATIVE’(MDI)

A. Issues and Concerns in Relation to the Market Design Initiative The MDI focuses on two main policy objectives:20 1. The reliability of electric energy supply: Reliable supplies require, among other things, generation adequacy. Consequently, the EC is concerned with a market design that creates an environment for efficient plant investment and operation. 2. The affordability of energy supply: The reliable operation should be achieved at affordable prices. This requires a market design that drives efficient investment. Consequently, CRMs can and should be used where they are part of a cost-efficient solution for reliable supplies, but not where they unduly impose a cost burden without significant added benefit to the system and ultimately to consumers. In its assessment, DG ENER takes a much wider perspective and we summarise below its key findings, focusing on those aspects that are most relevant for the assessment of CRM. The Staff Working Document accompanying the legal package of the Clean Energy Package identified four main areas of concern:21 —

Problem Area I: The current market design is not fit for an increasing share of variable decentralised generation and technological developments. Short-term markets, as well as balancing markets, are often not efficiently organised to deal with volatile renewable in feed. Moreover, as part of the promotion of renewable and decentralised energy supplies, there have been exemptions from fundamental market principles (eg, priority dispatch for wind and solar), which cause

18 Susanna Twidale, ‘Blow to 800 million pound Trafford gas plant as Britain withdraws subsidy’ Reuters (5 July 2016). 19 eg, see the discussion around design issues with the capacity mechanism in Colombia: D Harbord, ‘CREG Expert Panel on Colombian Energy Market Reform’ (4 October 2016); SD McRae and FA Wolak, ‘Diagnosing the Causes of the Recent El Niño Event and Recommendations for Reform’ (2016), available at: www.web.stanford.edu/group/fwolak/cgi-bin/sites/default/files/diagnosing-el-nino_mcrae_wolak.pdf. 20 See, eg, Commission, ‘Impact Assessment of the Market Design Initiative—Staff Working Document’ SWD (2016) 410 final. 21 ibid, 30.

154 Leigh Hancher and Christoph Riechmann increasing distortions in the energy market with rising renewable penetration. Consumers do not actively engage in the market and demand response potential remains largely untapped. In addition, distribution system operators (DSOs) do not have the required incentives to actively manage loads within their networks. — Problem Area II: Uncertainty about sufficient future generation investments and uncoordinated capacity markets. There is a lack of adequate investment signals due to regulatory failures and imperfections in the electricity market (market/ policy failures I to III above). The MDI further identifies uncoordinated State interventions to deal with real or perceived capacity problems (market/policy failure IV above). — Problem Area III: Member States do not take sufficient account of what happens across their borders when preparing for and managing electricity crisis situations (market/policy failure IV, above). Specific issues include too much focus on plans in the national context, the lack of information-sharing and transparency between Member States, and the lack of a common approach to identifying and assessing risks. While DG ENER here mainly has issues with coordinating the management of networks in mind, there is also a link to enhanced coordination on generation adequacy and its assessment. — Problem Area IV: The slow deployment of new services, low levels of service and questionable market performance on retail markets. Problem Area II is the key area to which CRMs could be the solution. However, it is important to also consider the solutions that DG ENER proposes in the other areas, particularly for Problem Area I (but also III) as the proposed reforms have a bearing on the need for and the appropriate design of CRMs. The key recommendations in relation to CRMs are: —

EU-wide adequacy assessment: There should be one common EU-wide assessment of generation adequacy, undertaken by European Network of Transmission System Operators for Electricity (ENTSO-E) or groups of regional TSOs. This would replace regional or national assessments or introduce an adequacy assessment. A European harmonisation would thereby extend to adequacy criteria that are employed (eg, loss of load probability, or energy not supplied), the methodology and model for the assessment and the data used in the analysis. — CRMs that comply with design rules can be admissible: DG ENER now acknowledges the possible need to adopt CRMs but advises against the comprehensive use of CRMs in all Members States, let alone the introduction of one uniform regime in all Member States. Rather, CRMs should only be used where necessary, be designed to address a specific issue and follow certain basic principles (to be established at Union level).22

22 See also Final Report (n 1) s 5.2 (‘Which capacity mechanisms for which problems?’) and s 6 (‘Getting the design Right’).

Capacity Mechanisms and Auctions 155 —

EU framework for cross-border participation: The EC requires that explicit participation from abroad is possible in any CRM. The EC will develop a common framework for such cross-border participation. The EC will be more prescriptive as to the design of cross-border participation than it will be in relation to the national CRMs. I

II

Variable decentral generation Inefficient short-term markets Preferential market access for renewables Poor consumer engagement Untapped demand response

Uncertain future generation investment Lack of adequate investment signals due to regulatory and market failures Uncoordinated state interventions to deal with capacity problems

Create level playing field among resources

EU wide adequacy assessment

Strengthen shortterm markets

CRM to comply with design principles

Pull DRM and distributed resource into market

EU framework for x- border participation

III

Ignoring neighbouring systems Crisis plans and actions national in focus Lack of information sharing and transparency No common approach to identifying and assessing risks

Minimum EU rules for prevention & crisis management Regional co-operation Systematic monitoring

IV

Inefficient downstream markets Weak competition in retail markets Conflicts of interest regarding management and handling of data Poor consumer engagement

Standard defintion of energy poverty Phasing out of retail price regulation EU data management rules Facilitate customer switching

Harmonisation and coordination within the EU

Figure 1: Market Design Initiative—Conclusions from the Impact Assessment Source: Based on EC DG ENER—Market Design Initiative—Impact Assessment.23

B. Towards a Harmonised Approach between the Market Design Initiative and Sector Inquiry? The MDI confirmed the need for further harmonisation measures.24 Illustrative here is the proposed approach to the management of crisis situations (Problem Area III).25 Although some degree of coordination has resulted from voluntary agreements as well as network codes, cross-border assistance in times of crisis could be hampered by a lack of common principles and rules governing cooperation, assistance and cost compensation. Risks as such are assessed and addressed on the basis of very different methods and from a national perspective only.26 A mandatory risk preparedness plan

23

‘Impact Assessment’ (n 20). In general, stakeholders seem to favour common minimum EU rules for prevention and crisis management and the use of a common methodology to be followed for short-term risk assessment. Deeper harmonisation is not contemplated. 25 See also AF Mercados, E-Bridge and REF-E, ‘Identification of Appropriate Generation and System Adequacy Standards for the Internal Electricity Market. Final Report’ (2016) Report for DirectorateGeneral for Energy Internal Energy Market. 26 See ‘Impact Assessment’ (n 20) 97. 24

156 Leigh Hancher and Christoph Riechmann would include ex ante arrangements between Member States, in particular regions to address cross-border issues such as joint or simultaneous crisis situations. The Commission has proposed a draft Regulation on Risk Preparedness in the electricity sector.27 ENTSO-E shall develop a methodology for assessing short-term adequacy, namely seasonal adequacy as well as week-ahead to intraday generation adequacy forecasts. Once this methodology has been approved by ACER, it should be used by Member States and ENTSO-E in their short-term assessments. That assessment complements the long-term resource adequacy assessment proposed in the recast Electricity Regulation.28 The Commission also proposes the creation of Regional Operating Centres (ROCs), effectively joint ventures of regional TSOs. ROCs would take binding decisions, for example, to coordinate security analysis, determine the regional generation reserve requirement and streamline monitoring.29 Crisis prevention and management would be made more effective by requiring Member States to cooperate effectively and to put in place tools to monitor the security of supply via the Electricity Coordination Group. The notion of ROCs is contentious in some Members States and ultimately some alternative arrangements to ROCs may be made. When we refer to ROCs in the following we mean those institutions that ultimately carry those responsibilities which the EC foresees the ROCs holding. DG ENER’s findings in relation to the ‘market design initiative’ are likely to also have a bearing on the application of the EEAG up to its expiry as well as its eventual revision after 2020. Broadly, the findings in DG ENER’s final report (on the MDI) confirm the approach previously adopted in the EEAG while not providing significant new detail. As such, it provides high-level guidance and a confirmation of approaches already set out in the EEAG. DG COMP in its findings on the Sector Inquiry also considers the three dimensions explored by DG ENER in the MDI (Figure 2): — A generation adequacy assessment.30 — National CRMs, in particular their need and design. — Cross-border participation.31

27 Commission, ‘Proposal for a Regulation of the European Parliament and of the Council on riskpreparedness in the electricity sector and repealing Directive 2005/89/EC’ (Draft Electricity Regulation) COM(2016) 862 final. 28 ibid, Art 19. 29 ibid, Art 38. 30 The Commission states that ‘rigorous adequacy assessment against a well-defined economic reliability standard is crucial for identifying risks to the security of supply and for determining the necessary size of any capacity mechanism’. Final Report (n 1) 17. 31 The Commission concludes that ‘market wide capacity mechanisms must be open to explicit crossborder participation in order to minimise distortions to cross-border competition and trade, ensure incentives for continued investment in interconnection and reduce the long-term costs of European security of supply’. ibid, 18.

Capacity Mechanisms and Auctions 157

Figure 2: Principles of State Aid Legislation and Conclusions of the Sector Inquiry

C. Adequacy Assessment In its Final Report of the Sector Inquiry as well as in the Report on the Market Design initiative, the Commission requires that a need for a CRM is identified. We consider that this need could then represent a common interest. A necessary condition would be one where a threat to resource adequacy is identified, which would pose a threat to the reliability of supplies. If such threats exist, its remedy would be in the public or common interest. The identification of an adequacy concern may, however, not be a sufficient condition to legitimise the introduction of a CRM. If leaner, less intrusive reforms are available and these would already address the adequacy concern, then these may need to be favoured over more intrusive CRMs.32 Member States are obliged to identify any regulatory distortion that causes an adequacy concern and shall publish a timeline for adopting measures to eliminate it33 (eg, enabling scarcity pricing, DSR, energy efficiency and interconnection). A sufficient condition for the use of CRMs would therefore be one where the cause of the adequacy issue is identified and where, even after removing regulatory obstacles and failures and other less intrusive reforms (eg, those in relation to Problem Area I above; namely, the strengthening of balancing arrangements and short-term markets, the mobilisation of demand response etc), only CRMs will serve to remedy the identified failure. DG ENER appears to hold little trust in individual Member States conducting such adequacy reviews in an effective and unbiased manner. DG ENER therefore

32 The Commission concludes that ‘electricity market reforms are indispensable since they help to address concerns about inadequate security of supply. However, most Member States have yet to implement appropriate reforms’. ibid, 17. See also Draft Electricity Regulation (n 27). 33 See also Draft Electricity Regulation (n 27) Art 18.

158 Leigh Hancher and Christoph Riechmann intends to assign the role to the ENTSO-E, in designing and undertaking such analysis. The resource adequacy assessment shall:34 — —

— — — —

Be undertaken on bidding zone level and cover at least all Member States. Reflect appropriate scenarios of projected demand and supply including plant retirement and newly built, and gauge energy efficiency measures and sensitivities on wholesale prices and carbon price developments. Take account of all resources including existing and future generation, energy storage, demand response, and import and export possibilities. Consider the removal of any market distortions. Include scenarios without existing or planned capacity mechanisms. Be based on a market model using, where applicable, the flow-based approach and apply probabilistic calculations.

In its methodology, ENTSO-E is asked to incorporate the value of lost load, the ‘cost of new entry’ for generation, or demand response; and a reliability standard expressed as ‘expected energy not served’ and the ‘loss of load expectation’.35

D. Market Design Principles Moreover, DG ENER recommends a number of actions that would tend to reduce the need for CRMs (‘no regret’ measures): —



34 35

The further development of demand-side participation in the electricity market: This measure serves several purposes. First, it helps to mobilise demand-side response in the electricity market, thus better enabling the sector to respond flexibly and with market mechanisms to generation shortfalls. This reduces the physical risks from technical strains. Secondly, this measure better allows for turning the public good of supply security into a private good, so that those with less willingness to pay can be excluded from supply in stress situations. If this approach can be successfully implemented, then most of the market and policy failure concerns could be addressed, removing the need to introduce CRMs in the first place: Security of supply would become a private rather than a public good (Failure I); electricity prices would likely be less volatile if demand response helps dampen extreme stress situations (Failure II); and, ideally, there would then be less need for policymakers to intervene and explicitly or implicitly cap prices (Failure III). The strengthening of short-term and balancing markets (in response to Problem Area I) would similarly have the effect of enhancing the resilience of the physical system and to financially rewarding those market players that are capable of providing respective services.

ibid, Art 19(4). ibid, Art 19(5).

Capacity Mechanisms and Auctions 159 Regarding design principles for CRMs, the Commission develops a clear position in its Final Report of the Sector Inquiry and we discuss this in the respective subsection V.B below.

E. Cross-Border Participation in CRMs A particular concern under State aid rules will be whether generators in other Member States or interconnectors are allowed to participate in the domestic CRM. Cross-border participation adds a further layer of complexity and some critical questions need to be addressed, for example, whether foreign generators, interconnectors or both should be allowed to participate in the CRM. The EC appears to fear that leaving the approach to cross-border participation to the discretion of Member States risks undesired outcomes. Through the Electricity Regulation (recast,) the EC has already clarified that it favours:36 —

Direct ‘explicit cross-border participation’ by foreign players (rather than an implicit and abstract recognition in any national regime that some foreign transmission and generation capacity will be available to support system reliability in other Member States). — Ensuring that the rewards from foreign participation are channelled to that party that contributes to removing the bottleneck. For example, if the shortage in the region is generation capacity, while transmission capacity is ample, then the ultimate reward should reach generators. Conversely, if the issue is a lack of transmission capacity, while there is sufficient generation capacity in the region, then funds should ultimately be channelled toward the transmission. This principle does not require that generators or TSOs are the immediate recipients of capacity payments. It does require though, that there are mechanisms to allow for an appropriate redistribution of the funds.37 — The participation by market participants in the CRMs of multiple jurisdictions should be allowed in principle. In practice though, it will need to be ensured that those jurisdictions are unlikely to experience stress situations during the same hours or days. Ensuring effective cross-border participation will involve a number of institutions. ENTSO-E and ACER will cooperate in devising a methodology to be applied by the

36

Similarly also in the Final Sector Inquiry report. See Final Report (n 1). eg, a regime is conceivable where interconnectors participate in a CRM. In case there is ample generation capacity, a TSO could rely on sufficient generation being available in stress situations. The focus here must be that the interconnector capacity is available in stress situations. It would then be appropriate for the IC owner to retain any capacity remuneration. If, by contrast, generation capacity was short and interconnector capacity ample, then the interconnector who participates in a CRM would likely want to contract back-up generation (ignoring the complexities of unbundling for the moment) to ensure that the interconnector could actually flow power in a crisis situation. In this instance, funds would be redistributed from the interconnector (who was the immediate beneficiary in the CRM) to generators in neighbouring countries. 37

160 Leigh Hancher and Christoph Riechmann proposed new regional operational centres (ROCs), which would produce annual calculations on the maximum entry capacity available.38 The TSO where the foreign capacity is located must establish that the interested capacity providers can provide the requisite technical performance as required by the CRM in which the provider intends to participate and must register (with ENTSO-E) the capacity provider in the ENTSO-Registry as an eligible capacity provider. NRAs shall ensure that crossborder participation is organised in an effective and non-discriminatory manner, and in particular provide for the enforcement of non-availability payments across borders. ENTSO-E and ACER will also be required to establish a methodology for sharing the difference in costs arising from cross-border participation. As part of the harmonised approach, several aspects would need to be further clarified:39 —









38

Interconnectors or generators—should interconnectors and generators both be allowed to participate in the CRM of a neighbouring country at the same time, or should only one type of player be allowed? In theory both types of player could participate at the same time, although this would increase the complexity of the system. Participation by interconnectors may have the advantage that the TSO in the country procuring capacity only needs to deal with a small number of known (and regulated) entities. The downside would be that the operator of the interconnector may need to sign contracts with back-up generation to ensure that generation capacity is actually available behind the interconnector. This would raise concerns about the unbundling of transmission and generation. Proof of interconnector (IC) availability—what is the criterion for determining whether, in a stress situation, the capacity obligation held by a party that accesses the local CRM through a cross-border interconnector has actually been fulfilled? For example, is it required that power actually flows on the interconnector or is it sufficient to demonstrate that power could have flown, had this ultimately been required? Derating of IC capacity—to what extent should the interconnector capacity be considered in a country’s CRM and in particular how should the capacity be derated when accounting for its contribution to local generation adequacy? Transitional exemptions (for pragmatic reasons)—should there be transitional exemptions from cross-border participation, for example, in the phase when a national CRM is initially set up? Participation in multiple CRMs—how to design the option to participate in multiple regimes while ensuring an effective contribution to system reliability (and avoiding abuse of the mechanism).

Established pursuant to Art 32 of the Electricity Regulation (recast). Frontier Economics, ‘Participation of interconnected capacity in the GB capacity market’ (2014) Report prepared for the UK Department of Energy and Climate Change (DECC); Frontier Economics, ‘Cross border participation in CRMs—Presentation materials’ (2015) 3rd DG COMP EEAG working group on capacity mechanisms. 39

Capacity Mechanisms and Auctions 161 V. THE SECTOR INQUIRY ON CRMS

The Sector Inquiry’s proclaimed aim is to contribute to the Commission’s Energy Union agenda and the development of a new market design that is ‘fit for the future’ by assessing to what extent capacity mechanisms are appropriate instruments to ensure sufficient electricity supply40 while at the same time minimising the distortion of competition or trade in the Internal Electricity Market.41

A. The Sector Inquiry: Interim Report The 11 Member States under assessment had opted for the introduction of one or more capacity mechanisms to address perceived residual market failures,42 and 28 mechanisms were identified in these 11 countries.43 The designs of the mechanisms vary widely, but all have in common the underlying principle of enabling revenues for capacity providers and thus they may fall within the category of State aid measures. Neither the Interim nor the Final Report assessed whether the existing or planned capacity mechanisms in the individual Member States complied with EU State aid rules. In its Final Report the Commission states that it will use the findings of the inquiry to assess capacity mechanisms notified in relation to State aid.44 The interim conclusions, published in April 2016 already offered some clues as to how the Commission viewed the interplay with the proposed MDI. In its discussion on reliability of electricity supply it considered that as regional and European-wide methodologies for assessing generation adequacy mature and become more reliable, they would be used as a basis for assessing the necessity of introducing capacity mechanisms. In the meantime, Member States should ensure they

40 Recital 6 of the decision to launch the inquiry states that ‘[t]he Commission aims to gain a better understanding of the existence and functioning of capacity mechanisms through a sector inquiry’. See Commission Decision of 29 April 2015 initiating an inquiry on capacity mechanisms in the electricity sector C(2015) 2814 final. 41 In most of the 11 Member States covered by the inquiry the generation mix now consists of substantial shares of variable renewable energy sources (RES). Wind and solar generation technologies have achieved the largest shares of installed capacity in Denmark (40%), Germany (38%), Spain (28%) and Portugal (25%). The shares of variable RES are expected to increase further, in particular as some Member States are still making progress and are increasing the share of renewables in their country in order to reach their 2020 targets. 42 It should be noted that Croatia was included in this list but the Commission now appears to doubt whether the national measures in question are financed by State resources. 43 See Table 1, above section II. 44 The inquiry is an interesting exercise in its own right given that this is the first occasion on which the Commission has made use of its new powers under Art 25(1) of the revised Procedural Regulation 2015/1589: ‘Where the information available substantiates a reasonable suspicion that State aid measures in a particular sector or based on a particular aid instrument may materially restrict or distort competition within the internal market in several Member States, or that existing aid measures in a particular sector in several Member States are not, or no longer, compatible with the internal market, the Commission may conduct an inquiry across various Member States into the sector of the economy or the use of the aid instrument concerned’ (emphasis added).

162 Leigh Hancher and Christoph Riechmann undertake thorough national adequacy assessments following emerging best practice, and compare the situation without intervention against an economic reliability standard, before intervening in their markets.45 It is here that State aid control comes into its own.

B. The Final Sector Inquiry Report The final Sector Inquiry Report46 builds on the analytical considerations developed in the literature. Unfocused CRMs are a potential threat to the efficient operation of the Internal Energy Market—which is essentially an energy (ie, MWh) market, not a capacity (ie, MW) market. The key conclusions of the report can be summarised as follows: —





There is a need to reform the energy markets with a particular focus on effectively working balancing markets and imbalance prices as well as the removal of implicit and explicit price caps in the energy market. Failure to implement such reforms should not serve as an excuse to implement CRMs. Such reforms may not necessarily overcome all market and regulatory failures. Lacking mobilisation of demand-side response and localised adequacy issues may be of particular concern. Where such concerns prevail, CRMs may be appropriate. Any CRM must be justified by an assessment of the market failures and an adequacy assessment that quantitatively highlights the reliability concern. A quantitative and model-based assessment, for example, following the probabilistic modelling approached developed by ENTSO-E is appropriate.47

Where CRMs are employed, they need to apply competitive processes48—for example, auctions—and focus on and address the identified problem: —





For long-term risks, market-wide capacity mechanisms are the most appropriate instruments—alongside market reforms to limit the role of the capacity mechanism. For temporary risks, a strategic reserve is likely to be a more appropriate solution while the market is reformed to deliver security of supply in the longer term. The reserve must be held outside the market. For a local adequacy issue, the problem should be solved by better grid connections or more appropriate bidding zones, but various mechanisms may be appropriate transitional tools.

45 Commission, ‘SWD accompanying the Interim Report of the Sector Inquiry on Capacity Mechanisms’ (SWD accompanying the Interim Report) SWD (2016) 119 final, 62. 46 Final Report (n 1). 47 The EC points to the Market Design Initiative and proposals therein as a suitable benchmark. 48 ‘The sector inquiry demonstrates that capacity mechanisms should be open to all types of potential capacity providers and feature a competitive price-setting process to ensure that competition minimises the price paid for capacity’. Final Report (n 1) 16.

Capacity Mechanisms and Auctions 163 —

To develop a flexible demand side, an interruptibility scheme may be an appropriate solution. — Administrative capacity payments are not likely to be appropriate because the lack of competitive process means a high risk of failing to achieve the objective, or of overcompensating. — CRMs should generally be open to different types of offer, ie, generation and demand. However, one positive technology discrimination contemplated is the special treatment of demand-side measures. There can be several motivations for such special treatment. First, stimulating demand-side participation is the remaining challenge to the efficient working of electricity markets. If demand is sufficiently responsive and price sensitive, then generation adequacy would not pose an issue. In such a world a tightening generation reserve margin would result in high electricity prices and consumers would—in response—lower their demand until the generation reserve was sufficient to ensure supply security. Secondly, the technology to boost demand response is lagging behind and may need some support. The EC also contemplates49 a second instance of discrimination: it proposes that plants with emissions in excess of 550g CO2/kWh be excluded by CRMs, this would practically exclude all coal and oil plants and older gas fired plants. — Uninterrupted trade of electricity and capacity is a particular concern. CRMs must be open to cross-border participation. Moreover, CRMs should be designed such that they do not unnecessarily suppress price spikes that would typically incentivise the cross-border exchange of energy. The EC prefers capacity signals to come from price spikes in the internal energy market as opposed to penalties for non-compliance, as a means to enhance market integration.

VI. THE STATE AID GUIDELINES AND THE IMPLICATIONS FOR CAPACITY MECHANISMS

The application of the EU State aid rules to national generation adequacy support mechanisms is not an alternative to or substitute for a harmonised approach to dealing with generation adequacy at the European level (as considered under the MD initiative). Rather these rules have a vital complementary function. In its individual decisions, the Commission considers each national measure on its own merits. It is therefore unlikely, absent an agreed preferred market design, that the application of the State aid rules alone will result in the adoption of a uniform market design or CRM approach across the EU. Indeed, the interim and final reports of the Sector Inquiry suggest this might not be a valid objective. Each country and/or region should be considered on its own merits and CRMs—where applied—should address the specific regional or local issue.

49

Electricity Regulation (recast) Art 23(4).

164 Leigh Hancher and Christoph Riechmann At the same time, it is important to consider the criteria developed by the Commission in the EEAG of 2014 and how these have been applied prior to the launch of the Sector Inquiry as well as subsequent to its conclusion. The Final Report of the Sector Inquiry did not propose the adoption of a renewed set of guidelines. The results of the inquiry undoubtedly allow the Commission to monitor the evolution of CRMs more closely and to refine the application of its guidance.50

A. The EEAG of 2014 and their Recent Application The EEAG51 contains specific rules for the assessment of ‘aid for generation adequacy’. If the proposed support scheme falls within the scope of Article 107(1) TFEU it is for the Member State to demonstrate that the measure is compatible aid and complies with the common assessment principles52 as further tailored to capacity mechanisms in the specific conditions set out in section 3.9.53 We first summarise the conditions and examine their application to cases adopted under the EEAG prior to the launch of the Sector Inquiry as well as those adopted during and subsequent to the inquiry process. Section 3.9.1—contribution to an ‘objective of common interest’: it is for the Member State to demonstrate that it faces a genuine generation adequacy issue. The compatibility criteria appear to leave ample scope to Member States to design the form of investment or operating aid it intends to confer. It allows Member States to pursue the objectives of addressing both short-term concerns relating to system adequacy caused by the lack of flexible generation and/or to define targets for generation adequacy ‘which Member States wish to ensure regardless of short term considerations’. Member States are encouraged but not obliged to avoid achieving generation adequacy by subsidising environmentally or economically harmful forms of generation, and to consider alternative strategies such as demand-side response and increasing interconnection capacity. Further, the precise objective of the measures should be clearly defined, and the identification of a generation adequacy problem should be consistent with the analysis carried out regularly by ENTSO-E.54 Section 3.9.2—the necessity of the State aid measure: a clear analysis and quantification of the nature and causes of the generation adequacy problem is required (market/regulatory failures). The reasons why the market cannot deliver adequate capacity in the absence of intervention must be demonstrated:55 the Member State

50

See SWD accompanying the Final Report (n 4) 168. EEAG (n 10). 52 ibid, ss 3.1 and 3.2. 53 Where the aid is below €15 million per project per undertaking, then it is not notifiable and will be considered compatible aid as long as it complies with the General Block Exemption Regulation (GBER) 2014. 54 EEAG (n 10) para 221. 55 See ibid, para 224(d), the Commission will assess other factors that might cause or exacerbate the generation adequacy problem, such as regulatory or market failures, including, for example, caps on wholesale prices. 51

Capacity Mechanisms and Auctions 165 must show that alternative means to address generation adequacy cannot meet the required targets.56 Should the Commission consider that the cause of the problem lies in regulated wholesale pricing—which is no longer allowed in accordance with the EU’s internal electricity market legislation—it may cast doubt on the necessity of the proposed aid measure. Section 3.9.3—the appropriateness of aid: the measures should only remunerate the service of pure availability and not for the sale of electricity, and should be open and provide adequate incentives to both existing and future generators as well as substitutable technologies, such as DSR and storage solutions. Interconnection capacity as a potential solution should be taken into account. Section 3.9.4—incentive effect: the aid must induce the beneficiary to change its behaviour to improve the functioning of a secure, affordable and sustainable energy market, a change in behaviour which it would not occur without the aid. Section 3.9.5 requires that the overall amount of aid should allow beneficiaries a reasonable rate of return. A competitive bidding procedure that is based on clear, transparent and non-discriminatory criteria is considered as leading to reasonable rates of return. Overcompensation must be avoided and the measure should include a mechanism to avoid windfall profits. The price paid for the availability must automatically tend to zero when the level of capacity supplied is expected to be adequate to meet the level of capacity demanded.57 Section 3.9.6 deals with the balancing test—the avoidance of negative effects on competition and trade. Positive elements include a requirement to apply competitive bidding and technological neutrality58 when designing CRMs, the inclusion of DSR, interconnectors and storage in the design of the mechanism, as well as the guarantee of participation by a sufficient number of generators to establish a competitive price for the capacity. The measure should not reduce incentives to invest in interconnection capacity, or undermine market coupling, including balancing markets. It should not ‘unduly strengthen market dominance’—especially that of the historic incumbent—and should give preference to low-carbon generators in case of equivalent technical and economic parameters. Importantly, the participation of operators from other Member States should be taken into account but only ‘in so far that participation is physically possible and the capacity can be physically provided to the Member State implementing the measure and the obligations set out in the measure can be enforced’.59 This approach allows

56 eg, the development of market coupling, intraday markets, balancing and ancillary service markets and storage. 57 Unlike many other categories of aid covered by the EEAG, aid for generation adequacy is not subject to additional requirements on aid intensity and so it may be assumed that the maximum of 100% coverage is allowed. 58 It is worth noting, though, that the EC contradicts itself when proposing that in the medium term only plants with a specific emission below 550gCO2/KWh_el (Art 23(4) Electricity Regulation (recast)) can be remunerated within a CRM. Such a rule contradicts technological neutrality, especially considering that EC climate policy is steered via the European Union Emission Trading Scheme (ETS). The emission threshold derives from a policy which the EIB has given itself, where it will not support lending for power projects with emissions in excess of 550gCO2/KWh_el. 59 EEAG (n 10) para 232(B). At fn 97 it is however provided that ‘Schemes should be adjusted in the event that common arrangements are adopted to facilitate cross-border participation in such schemes’.

166 Leigh Hancher and Christoph Riechmann the possible exclusion of operators from other Member States if the notifying government can establish that either sufficient capacity is not available on existing interconnectors, or that they have no means of sanctioning a beneficiary who subsequently reneges on its obligations to provide availability. Ensuring adequate cross-border participation in national schemes has proved to be complex both legally as well as technically.60

B. The UK Capacity Market The UK capacity mechanism was cleared on 24 July 2014.61 The mechanism is based on market-wide capacity auctions and was adopted following extensive consultations led by the then Department of Energy & Climate Change.62 The UK system involves centralised auctions held four years ahead of the year in which capacity must be delivered. The UK government identified two market failures that prevent the market from bringing the necessary capacity to meet the established generation adequacy standard. Further, the UK claimed that reliability is a public good.63 The Commission dealt with the objective of common interest and necessity of the aid together.64 It considered that the UK had adequately demonstrated that capacity adequacy could reach critical levels as of 2018/19. In respect of the first capacity auction held on 16 December 2014, the UK government had followed the national TSO (National Grid) and taken a conservative estimate of the contribution from interconnector flows. Similarly, the potential contribution of DSR would be kept under review and transitional auction arrangements would be made to support the growth of DSR.65 The Commission found that the UK had sufficiently explored ways to mitigate the negative impacts that the notified measure may have on the objective of phasing out environmentally harmful subsidies, including the award for contracts for difference (CfDs) to support low-carbon generators.66 As existing generators and DSR operators would not be faced with the same investment costs as new plant, longer contracts (15 years) were justified for new investment, and hence the measure was viewed as technology neutral. The measure was deemed to be appropriate. The UK capacity mechanism would support the development of an active demand side and will support increased liquidity and competition in both the capacity and the electricity markets. Smaller generators and DSR participants and

60

See further F Roques et al, ch 7 in this volume. GB Capacity Mechanism (Case SA. 35980) [2014] OJ C348/1. 62 See for an early assessment, N Evans, ‘Security of Electricy Supply: is a Capacity Market the answer?’ (2013) 2 Letters and Notes on Regulation (published by the Regulatory Policy Institute). See for a detailed discussion of the evolution the regime, P Willis, ‘United Kingdom’ in L Hancher, A de Hauteclocque and M Sadowska (eds), Capacity Mechanisms in the EU Energy Market: Law, Policy, and Economics (Oxford, Oxford University Press, 2015). 63 GB Capacity Mechanism (n 61) paras 83–86 and Table 1, section II. 64 GB Capacity Mechanism (n 61) para 119. 65 ibid, para 122. 66 ibid, para 128. 61

Capacity Mechanisms and Auctions 167 suppliers have a clear route to market and receive a fair value for the capacity they provide. The measure remunerates only the service of pure availability of capacity. Despite a complaint that the scheme would result in over capacity, the Commission concluded that the number of new plants would be limited to the minimum necessary as competitive existing plants would be likely to bid at lower prices.67 Interconnectors would only be allowed to bid directly in the auction process, as if they were generators, once the scheme was adjusted in time for the second auction.68 The Commission was satisfied with this approach, as it was similarly convinced that the differential treatment of existing and new plants was not discriminatory, and indeed certain requirements imposed on existing as opposed to new plants were justified in order to mitigate market power.69 The scheme was also proportional: capacity providers would receive fair compensation, allowing them to earn a reasonable rate of return, while avoiding the risk of windfall profits. When performing the balancing test, the Commission found that the positive effects outweighed any potential negative outcomes. In particular, the business case for future interconnection would not be undermined, the scheme would not depress electricity prices in the energy market and new investment will be encouraged, countering the risk of market dominance from existing plant as price takers. The measure would also give preference to low-carbon generators.70 In fact, the auctions resulted in substantial capacity being provided from diesel generators, although this may be of lesser environmental concern if the respective units truly serve as a reserve and are rarely called upon.71 In the meantime, the decision has been challenged— in Tempus Energy and Tempus Energy Technology v Commission and in MPF Holdings v Commission.72 Capacity Auction 2016 Market conditions in Great Britain have changed considerably since 2014, and in particular wholesale prices have decreased significantly. An unexpectedly higher

67

ibid, para 147. See SWD accompanying the Interim Report (n 45) 78. 69 GB Capacity Mechanism (n 61) paras 130–39. 70 ibid, para 153. 71 Low utilisation of such diesel units is a likely scenario given their high operating cost. 72 Case T-793/14 Tempus Energy and Tempus Energy Technology v Commission and Case T-788/14 MPF Holdings v Commission. In the former case, the applicants have claimed that, by failing to open the formal investigation procedure, the Commission violated Art 108(2) TFEU, the principles of nondiscrimination, proportionality and legitimate expectation and made a wrong assessment of the facts. In particular, the applicants submit that the Commission failed to properly assess the potential role of Demand-Side Response (DSR) in the UK capacity market; the restrictions on the duration of DSR contracts under the capacity market violate the principles of legitimate expectation and non-discrimination. The applicants further claim that the requirement for DSR operators to choose between transitional and enduring market auctions violates the principles of legitimate expectation and non-discrimination and the capacity market’s failure to provide for additional remuneration for savings in transmission and distribution losses from DSR violates the principles of non-discrimination and legitimate expectation. 68

168 Leigh Hancher and Christoph Riechmann number of capacity providers risk closing. To address this, the UK planned a supplementary capacity auction (SCA) in January 2017, under the same conditions as discussed above. This means that the ‘Great Britain Capacity Market’ will de facto become operational one year earlier than initially proposed. Both the SCA and the Great Britain Capacity Market aim to ensure that sufficient electricity capacity is available by providing power plants and demand response operators with remuneration on top of the income obtained from selling electricity on the market. The UK has taken specific measures to ensure that the SCA will be competitive, for instance by encouraging new build generating capacity to accelerate completion and by attracting as many demand response operators as possible to the supplementary capacity auction. The Commission was therefore satisfied that the risk of overcompensating participating capacity providers has been reduced to the minimum.73 C. Greece74 On 31 March 2016 the Commission published its decision on the compensation awarded in the form of a fixed capacity premium to certain electricity generators (predominately Combined Cycle Gas Turbine (CCGT)) for the provision of flexibility services (ramping services) to the Greek TSO. This decision is perhaps atypical in that the scheme lasted for one year, and in the course of the procedure the Greek authorities had committed to extensive reforms to its electricity market to implement the EU target model (for the wider electricity wholesale market) and to reduce the shares of the incumbent, PPC.75 Although the measure was an administrative payment scheme for all capacity providers (hydro, CHP or gas), the Commission considered that it addressed a market or regulatory failure that could not be more appropriately addressed in the near term. Given the various restrictions in the Greek system on the availability of interconnected capacity and demand-side response, the measure was appropriate as only gas and hydro plants could provide the necessary flexibility. In the absence of the measure there was a high risk that some gas-fired plants would exit the market, albeit that the capacity payments provided would be lower than the total fixed costs. Hence the measure was proportionate. The Commission found that even although the incumbent, PPC, owned the majority of eligible capacity and would receive the ‘majority of payments’, the measure would not distort competition given that the Greek authorities had made commitments to increase competition in generation and retail (see paragraph 103).76

73

Supplementary Capacity Auction in GB (Case SA.44475) [2017] OJ C20/1. Transitory electricity flexibility remuneration mechanism (FRM) (Case SA. 38968) [2016] OJ C241/1. 75 See ibid, para 25 and fn 12. 76 See ibid, para 103. 74

Capacity Mechanisms and Auctions 169 D. France77 The formal investigation into the French decentralised capacity regime,78 based on a system of tradable certificates was commenced by an ex officio investigation by the Commission in late November 2014.79 Given that the French authorities took the position that the regime either (i) does not constitute State aid within the meaning of Article 107(1) TFEU or (ii) in the alternative is compensation for the performance of a Public Service Obligation (PSO) in accordance with the so-called Altmark criteria, it had not notified the regime for State aid clearance. In April 2010, a working group issued a report that stressed two main market failures in the French market (the Sido-Poignant Report).80 First, it confirmed the lack of investment in new capacity. Second it highlighted the absence of mechanisms to develop demand response.81 Subsequently the government adopted the New Electricity Market Organisation (the NEMO Law). Article 6 of the NEMO Law defines the key principles of a capacity mechanism, to be implemented by a governmental decree.82 The capacity mechanism imposes obligations on both electricity suppliers and operators of generation and/or demand response capacities.83 It is based on a decentralised certificate of origin model. Electricity suppliers are required to contribute to the security of supply in proportion to the electricity consumption of their customers and are required by law to hold, for each given period,84 a specific amount of capacity certificates determined by the TSO, demonstrating that they are able to cover the needs of their customer portfolio in peak periods. Capacity certificates are exchangeable and transferrable. Capacity certificates offered by operators of generation and/or demand response capacity are matched with bids by electricity

77 Two measures were under investigation: French country wide capacity mechanism (Case SA. 39621) and Brittany CCGT (Case SA.40454). 78 French country wide capacity mechanism (Case SA. 39621) Commission Decision 2017/503 [2017] OJ L83/116. 79 France had notified the regime to DG Energy as a form of PSO, pursuant to Art 3(15) of Directive 2009/72 on the internal electricity market [2009] OJ L111/59. 80 The objective of the study was to identify peak demand for electricity and study technical solutions to reduce it through demand response. Further, the group was asked to investigate ways to increase demand response and make proposals promoting demand response measures, rather than the construction of new power plants. 81 With respect to the latter, the report noted that enhancing demand response participation in the energy-only market is doomed to failure because even if energy-only markets can theoretically be profitable for peak generation capacity (and symmetrically for demand response measures) the visibility they offer is not sufficient. Price spikes are too random in frequency and level and the risk is too great for an investor. No supplier has an interest to assume the risk of such an investment, to the extent that any failure would not necessarily be of his responsibility and does not necessarily lead to unbearable losses. 82 These stipulations have since been codified in 2011 in Arts L.335-1 to L.335-6 of the French Energy Code. 83 A decree implementing the capacity mechanism was adopted on 14 December 2012 (the Capacity Decree). According to this decree, the French Minister of Ecology, Sustainable Development and Energy (energy minister) shall adopt technical rules for the capacity mechanism. A proposal shall be prepared by the TSO, following an opinion of the ommission de Régulation de l’Energie (CRE), the energy regulator. 84 The capacity obligation reflects the supplier’s consumption during peak demand periods (PP1) as adjusted plus additional commitments for a further 10 days (PP2).

170 Leigh Hancher and Christoph Riechmann suppliers (willing to purchase these certificates) via a market-based mechanism (certificate market). The resulting price of capacity certificates is expected to reflect the cost of the capacity necessary to ensure the safety of the system.85 As discussed in chapter fifteen the Commission concluded that the French regime falls within the definition of a State aid given that the certificates are granted free of charge by the French authorities so that State resources are foregone.86 It has also held that the Altmark criteria are not met.87 As to the application of sections 3.9.1 and 3.9.2 of the EEAG, the Commission had doubts in its opening decision that the measure was necessary or served an objective of common interest, as the most recent ENTSO-E scenario outlook and adequacy forecast casts doubt on France’s own calculations and forecasts.88 There appears to be no generation adequacy threat before 2025 yet the capacity regime should be operational as of 2017.89 Nor is the mechanism an appropriate one given that it establishes a distinction between explicit and implicit demand response,90 excludes the direct participation of foreign interconnectors and/or capacity,91 and as initially designed, appears unable to attract new investment.92 Finally, the Commission doubted its proportionality, as it has a ‘strong tendency’ to strengthen the incumbent supplier, EDF’s, dominant position.93 In order to alleviate these concerns, France agreed to modify the scheme so that new capacities can obtain certificates with a seven-year duration instead of the standard one-year duration, if they are shown to be more competitive than existing capacities. The longer contract duration is thought to give sufficient investment certainty for new projects and facilitate the entry of new market players. To give new projects enough time to be developed, new capacities will be contracted through an organised public auction in four years’ time (2021). The French capacity mechanism will also be open to capacity providers, both generators and demand response operators, located in neighbouring Member States, subject to the expected capacity of the interconnector at peak times (around 7 GW in total). It is the first mechanism to explicitly include and remunerate foreign capacities. Moreover, France will introduce a series of measures to prevent possible market manipulation. In particular, capacity declarations of providers will be compared with historical benchmarks, to prevent providers from under-certifying their capacities

85 [Technical r—developed by RTE after extensive consultation and critical but positive decision of the French regulator (CRE) in 22 Jan 2015.] 86 The Commission’s preliminary assessment is published in French in [2016] OJ C45/35. See paras 103–18. Moreover, the Commission considers that even if the certificates are funded by final consumers the system differs from that at issue in PreussenElektra—see paras 119–25. 87 ibid, paras 126–37, and in particular para 131. 88 ibid, paras 155–58. 89 The French competition authority has itself cast doubt on whether there is a genuine ‘missing money’ problem in the French system. See ibid, para 159. 90 ibid, para 171. 91 ibid, para 172. 92 ibid, paras 174–83. 93 ibid, para 205.

Capacity Mechanisms and Auctions 171 to artificially drive up capacity prices. Also, large capacity providers will have to offer specified minimum numbers of certificates during the organised auctions, to increase liquidity on the capacity market. The French energy regulator will review the mechanism annually, to assess if the mechanism is still appropriate. The Commission confirmed that the modifications proposed by the French authorities adequately address its concerns. Brittany94 This measure—a call for a targeted tender for new CCGT plant—is intended to secure electricity supply in Brittany and was notified to the Commission in January 2015. The French authorities have claimed that the measure is a PSO. As the Commission had doubts that the ‘Altmark’ tests are met,95 and in particular that there is indeed a supply threat in the region, it opened a formal investigation to assess its compatibility under section 3.9 of the EEAG. At this stage, the Commission did not consider the measure could contribute to a common objective, or was necessary.96 Security of supply could have been achieved by less distortive measures—the measure is not open to existing generators or to operators using substitutable technologies. Despite a risk of overcompensation, there is no clawback mechanism.97 Finally, given that it might strengthen the position of EDF,98 and is limited to a particular type of technology, it could have an adverse impact on competition and trade. The Commission doubted that the proposed tender would deal with the ‘missing money’ problem as financial support for new plant would have adverse effects on existing capacity and might aggravate the ‘missing money’ problem faced by existing plant.99 In other words, the measure risked creating a subsidy dependent market so that investors would only develop projects on the basis of public tenders backed by subsidies. Obviously, this creates a vicious circle. Following a detailed, formal investigation the Commission declared the aid would be compatible aid when the French authorities have taken the steps necessary to ensure that for the entire duration of the aid the beneficiary of the aid does not provide energy from the plant to an operator which holds more than 40 % of the electricity generation capacity on the French market, whether under a tolling agreement or under a long-term contract of sale for the energy produced by the plant at a price equal to 95 % of the market price.100

94 95 96 97 98 99 100

Brittany CCGT (Case SA.40454) [2016] OJ C46/69. See ibid, paras 72–85. ibid, paras 111–21. ibid, para 98. ibid, para 137. ibid, para 83. para 226 of the final decision [2017] OJ L235/1.

172 Leigh Hancher and Christoph Riechmann E. The German ABLAV Scheme: The Interim Report of the Sector Inquiry Applied101 The German authorities notified the proposed ABLAV scheme on 16 April 2016 for reasons of legal certainty, arguing that it did not create a selective advantage nor did it involve State resources or affect trade between Member States. Under this scheme, which involves compensation to eligible consumers who have signed interruptibility contracts, the German TSOs will organise weekly auctions to define the support for the development of a more flexible demand side. The scheme aims at unlocking and contracting 1.5 GW demand-side flexibility which will become increasingly necessary as the share of intermittent renewables continues to increase. The selection of offers and the thresholds of remuneration for tendered services are set out in a State ordinance—the ‘ABLAV Ordinance’. The Commission found that the scheme would confer a selective advantage which was paid for by a compulsory levy on final consumers, collected and managed by the TSOs and transferred to a select group (mainly energy-intensive users) of beneficiaries in accordance with the Ordinance. The beneficiaries were active on a variety of markets open to competition and on which there is intra-EU trade. Furthermore, the Commission noted that there could not be a strict separation between the interruptibility scheme and the wholesale electricity market. The TSOs can use the loads as if they were capacities contracted as primary balancing reserve, while the consumers could decide not to make the loads available to the TSO for a certain time and become active on the energy market instead.102 The Commission, relying quite extensively on its interim findings in the Sector Inquiry, classified the ABLAV scheme as a CRM as it is a type of ancillary service which TSOs procured at the request of a government, and is used to ensure capacity is available to balance the system over longer periods so that they have the same effect as a capacity mechanism, and had to be assessed under section 3.9 of the EEAG.103 A central question in the assessment was the existence of a market failure preventing German demand-side providers from becoming active on the market to provide both short-term flexibility to the TSO and long-term flexibility of the demand side to enhance security of supply. The Commission analysed other ways of facilitating demand-side participation including the removal of legal or other practical barriers which prohibited or limited the participation of demand response in wholesale, balancing and/or ancillary services market. Alternatively, the market could be reformed in order to enable price signals that would activate interruptible loads spontaneously.104 The German authorities had in the meantime introduced significant reforms, of which the revised ABLAV scheme forms part, but maintained that even if demand

101

Interruptibility scheme AbLaV (Case SA.43735) [2016] OJ C471/1. ibid, paras 44–46. 103 ibid, para 52 referring to the Interim Report (n 3) 39. 104 ibid, paras 66–68. The Commission relied on a report produced by its Joint Research Centre in July 2016 which found the German market to be unique in that it has opened all markets to demand response while at the same time making actual participation impossible. See ibid, para 71 and fn 23. 102

Capacity Mechanisms and Auctions 173 response operators no longer faced legal barriers to participating on the balancing market, they remain at a competitive disadvantage. A tailor-made approach to reward availability as provided by the ABLAV scheme increases demand response participation. Hence, the Commission concluded that the measure was necessary for the period for which the measure would apply—ie, until 2022 by which time the market reforms already underway might make it superfluous.105 The Commission went on to find that that scheme was appropriate even although it was not technologically neutral—a differential approach was necessary to pursue the general policy objective of flexibilisation of the demand side.106 The aid was proportionate as it was designed to prevent over-procurement and overcompensation to the energyintensive sectors. The Commission’s Interim Report to the Sector Inquiry had highlighted the risk that interruptibility schemes could lead to indirect subsidies for this sector.107 Finally, the German authorities’ arguments that the risk of distortion to competition is very small given that the hours during which the loads are allowed to participate on the wholesale market alongside non-subsidised power is limited to rare occasions when the price on the day ahead market is exceptionally high, were accepted.108

F. The Network Reserve in Germany As underlined in the Commission’s Final Report on the Sector Inquiry, strategic reserves are generally unsuited to address the causes of a generation adequacy problem. They may be effective in responding to shortage situations, but they do not remedy the causes of the shortages, which may lie in a general missing funding problem, the absence of market reforms or ill-defined bidding zones. Strategic reserves can only be acceptable transitional instruments, for example, to address a temporary capacity shortage by ensuring that sufficient backup capacity remains available provided they are accompanied by measures that address the cause of the adequacy problem (eg, market reforms). The Network Reserve can only be an appropriate measure if it is transitional and accompanied by measures and market reforms that address the underlying causes of the adequacy concern. The Commission considers the Network Reserve appropriate only up to and including the winter of 2019/20, ie, until 30 June 2020. The Network Reserve aims to ensure that the German generation fleet is adequately equipped to ensure the balance of electricity demand and supply at all times and in all parts of the network and ensures that generation capacity remains adequate in regions that cannot be properly supplied due to the lack of sufficient transmission capacity, by preventing plants relevant for maintaining system security from closing down. The Network Reserve has a particular, regional function, given that capacity

105 106 107 108

ibid, para 76. ibid, paras 81–84. ibid, paras 88–92. ibid, para 97.

174 Leigh Hancher and Christoph Riechmann is held outside the market, kept on standby and at the disposal of the TSO to ramp up and generate in order to address shortage situations.109 As Germany has put in place the Network Reserve with the objective of ensuring a secure electricity supply in particular in southern Germany, by ensuring system integrity also at times of high generation in the north and high consumption in the south, it was therefore deemed to be targeted at the general common objective of ensuring a secure energy supply. The Network Reserve did not undermine the objective of environmental protection. It was necessary to allow for a rapid roll-out of renewable energy sources and in parallel with operating the Network Reserve, Germany promotes the development of more demand-side management through the ABLAV interruptibility scheme. Finally, Germany is in the process of strengthening its transmission grid in order to reduce and ultimately remove the Network Reserve altogether. As a result, the Network Reserve is targeted at and contributes to a well-defined objective of common interest, namely that of security of supply in southern Germany until sufficient transmission capacity is built. Although the assessment of generation adequacy deviated from that carried out by ENTSO-E here the question is whether that generation, even if sufficient in absolute terms, will be located in the right places to enable the TSO to continue to manage grid congestions. Market forces are currently not able to ensure generation adequacy in southern Germany so that there is a need for State intervention. As regards taking into account the contribution of interconnectors, Germany’s system analysis takes into account the contribution of interconnectors and sizes the Network Reserve in accordance with the established NTC-values, albeit in a passive manner, namely by simply taking over the NTC-values established by the TSOs. The present decision is therefore without prejudice to an assessment by the Commission, using its regulatory and/or competition powers, of the current TSO practice with regard to the calculation and determination of cross-border capacity. With regard to the participation of foreign generators to capacity mechanisms, the Network Reserve provides for their inclusion of foreign generators, but only as an option of last resort, namely when domestic participants are not sufficient. The Commission recognised that for strategic reserves in general, the participation of foreign capacities may be difficult to implement. Strategic reserves are usually dispatched to cover an imminent shortage of capacity once all other possibilities including imports are exhausted. If part of the reserve is located abroad an equivalent amount of interconnection capacity would have to be available or even reserved, to ensure the power generated actually reaches the shortage area. The explicit participation of foreign generators in the German Network Reserve does not however require the reservation of interconnection capacity, partly because they are located in the same bidding zone (Austria) and partly because they reduce the need for Germany to export power so that sufficient power remains in Germany’s consumption centres. It was deemed important that foreign capacity is considered

109

German network reserve (Case SA. 42955) [2017] OJ C68/1, para 44.

Capacity Mechanisms and Auctions 175 as a potential participant in this case. To ensure that the Network Reserve indeed remains a temporary measure, Germany committed to four measures that will be taken in parallel with the ongoing construction works expanding the grid.110 Although the scheme does not include a competitive allocation procedure, it was held to be appropriate given that there are not enough system relevant plants that have been prevented from closing down to fill the Network Reserve. A competitive allocation procedure would therefore risk leading to a non-competitive outcome with excessive remuneration. Commitments First, Germany has committed to pursuing an agreement with Austria enabling German TSOs to make use of market-based re-dispatch by Austrian generators. While this does not solve the underlying congestion issue, it would significantly downsize the Network Reserve and hence reduce the need for State aid. Second, Germany has committed to assessing the possible implementation of measures to increase the efficiency of its market-based re-dispatch and load management through, inter alia, improved cooperation between TSOs and better integration of RES and CHP. Third, Germany has committed to taking measures aimed at making increased use of interruptible loads and thus making the demand side more flexible, inter alia, through the ABLAV interruptibility scheme, but also through the opening up of the balancing market for interruptible loads. Fourth, Germany has committed to constructively supporting the formal launch by ACER of ENTSO-E’s review process of bidding zones as foreseen in Commission Regulation (EU) 2015/1222. It will also support the objective and factual evaluation of a plurality of bidding zones within Member States and commits to constructively consider the outcome of this process. The Commission deems this fourth commitment of particular importance in view of the significant and structural detrimental effects that an improper definition of a bidding zone can have, as described in recitals 79 and 80. Sizing bidding zones appropriately is the most constructive way of addressing network constraints. The Commission also notes that the envisaged split of the bidding zone on the Austrian border is likely to alleviate the congestion at this border, but is unlikely to remove intra-German congestion entirely.

VII. CONCLUSIONS

The adoption of the ‘Clean Energy Package’ in November 2016 has confirmed the need for new market design proposals to deal with both market and regulatory failures and to trigger investment, but the Commission is not always optimistic that this strategy will produce solutions in the short or near term. It will obviously take time

110

As well as the other measures described in s 3.3.3.1 of this Decision.

176 Leigh Hancher and Christoph Riechmann to negotiate the 2016 ‘Clean Energy Package’ into final, binding measures, many of which are currently forecast only to enter into force in 2020. The Final Report of the Sector Inquiry confirms that certain market and regulatory failures can be difficult to remedy so that even if enforcing scarcity prices would reduce the missing money problem, investors may still lack trust in the market framework (at least in selected countries) and market power may still need to be controlled. It is an open question if electricity market reforms can address these issues, so that residual market failures may well persist—and would have to be tackled through political intervention in market design and State aid control. As we illustrated with reference to the French and German cases, when evaluating proposed national CRMs and presuming they are justified in principle following the adequacy assessment, DG COMP will be further concerned whether the regime in question is compliant with the State aid rules and can — —

address the residual concerns that cannot be eliminated by simpler means (eg, by removing regulatory distortions such as price caps); and avoid unnecessary market distortions or limit cross-border trade.

The assessment of the costs and benefits of using CRMs also need to consider the practical complications in implementing CRMs in terms of capacity possibly not coming on stream as planned and CRMs taking time to bed in and be reformed and refined as time passes. In conclusion, it is apparent that the Commission is not endorsing CRMs per se and remains sceptical of their need. At the same time, it acknowledges that CRMs may be applied under certain circumstances. In some cases, the Commission considers that the need for CRMs may be transitional and in such cases their use should also be transitional. The need for further measures will need to be justified by a formal adequacy assessment. DG ENER appears to hold little trust in individual Member States conducting such adequacy reviews in an effective and unbiased manner and foresees an enhanced role for ENTSO-E and ACER, in designing and undertaking such analysis. CRMs are not accepted as a remedy to a poor design of short-term markets. The Commission favours and prioritises the reform of shorter-term markets (eg, intraday and balancing markets) and it considers such reforms as a no-regrets strategy. Equally, electricity markets are unlikely to function effectively if the design is not sufficiently flexible to integrate demand response and to place balancing responsibilities on volatile (renewable) generators. Economic considerations suggest that such reforms may in certain cases be sufficient to alleviate generation adequacy concerns, thereby potentially removing the need for CRMs. Where CRMs are used they will need to include an option for the cross-border participation of foreign capacities based on common principles. Given the significant complexities in electricity market design, State aid rules may need to be applied flexibly but as the latest decisions on the various CRMs in Germany indicate, also firmly. Depending on the market failures and security of supply situation as identified in the adequacy assessment,

Capacity Mechanisms and Auctions 177 different types of capacity mechanisms may constitute an appropriate response. For instance, where structural, long-term adequacy concerns are identified, it is likely that the best way to address these is by way of a market-wide capacity mechanism. Where on the other hand, there are good reasons to assume that the market does not (yet) deliver appropriate exit signals, or where market exit needs to be managed in an orderly and gradual way to prevent too many closures leading to temporary shortages, a temporary strategic reserve … may be a more appropriate response … [I]n market areas where market reforms are still in the early stages of their implementation and market participants are hesitant to invest on the basis of price signals alone, a strategic reserve can provide an effective transitional measure on the road to market-based new investment inspired by market reforms.111

111 SA. 45852 German Capacity Reserve, para 80, available at: ec.europa.eu/competition/state_aid/cases/ 269083/269083_1890897_10_2.pdf.

7 Cross-Border Participation in Capacity Mechanisms: Legal and Economic Issues FABIEN ROQUES, CHARLES VERHAEGHE AND GUILLAUME DEZOBRY

I. INTRODUCTION

T

HE EUROPEAN POLICIES on energy and Europe’s Energy Union strategy aim to ensure secure, clean and affordable energy supplies to European consumers.1 Among these objectives, security of supply has been a growing concern in several Member States in the context of the energy transition and the large deployment of intermittent renewable energy sources. In the absence of a common vision on how to address these concerns, a number of governments and/or regulators have taken steps or are debating the introduction of capacity mechanisms in order to guarantee the security of electricity supplies. These mechanisms aim to ensure capacity resource adequacy by providing a specific remuneration to generators and other capacity providers, such as storage or demand-side response, for their contribution in periods of potential system stress. The European Commission (EC) has identified up to 35 capacity mechanisms in place in 11 Member States. The rationale for the introduction of such capacity mechanisms and the different approaches are detailed in chapter six.2 These different capacity mechanisms are subject to the European State aid regulations. In this respect, the EC launched a sector inquiry and developed a set of guidelines and best practices for the design of capacity mechanisms.3 In the Sector Inquiry Report, the EC has taken a clear stance on the issue of participation in cross-border capacities: ‘mechanisms should be open to any capacity, including capacity located in other Member States which can effectively contribute to meeting the required generation adequacy standard and security of supply’.4

1 Commission, ‘A Framework Strategy for a Resilient Energy Union with a Forward-Looking Climate Change Policy’ (Communication) COM(2015) 80. 2 Commission, ‘Final Report of the Sector Inquiry on Capacity Mechanisms’ (Final Report on CM) COM(2016) 752 final. 3 ibid. 4 Commission, ‘Generation Adequacy in the Internal Electricity Market—Guidance on Public Interventions’ SWD(2013) 438 final, 29.

180 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry Moreover, for the EC, it is ‘a minimum requirement … for capacity mechanisms to be open to capacity abroad’5 and ‘[a]ny territorial restriction may be incompatible with the Union acquis’. By forcing the introduction of cross-border participation in capacity mechanisms, the EC wants the value of interconnection in terms of security of supply to be properly reflected and thus to avoid distortions in investment incentives and market functioning. A number of scholars have studied the potential benefits of cross-border participation in capacity mechanisms. Cepeda and Finon compared capacity expansion and generation adequacy in two neighbouring markets with capacity contracts implemented in only one of them.6 They pointed out that differences between regulatory policies and market designs may result in a free-riding problem and distort the normal functioning of the neighbouring markets, as well as the reliability of supply. Furthermore, Finon considers that a capacity remuneration mechanism which excludes cross border participants would also have distortive effects on long-term competition, as it doesn’t capture the advantages of multi-system competition.7 More recent works such as, among others, Hawker et al, and Ringler et al, have supported these arguments.8 They suggest that the design of capacity mechanisms needs to be coordinated with the possibility of cross-border participation in order to promote cross-EU trade and competition in electricity generation with investment signals for new generation capacity and interconnection coming from zonal electricity prices that reflect scarcity value. The potential benefits of cross-border participation materialise in several aspects which have been confirmed by various scholars.9 On the one hand, there will be additional available capacity committed to a capacity mechanism, further ensuring cross-border security of supply. On the other hand, as enhanced generation adequacy reduces the occurrences of loss-of-load incidents, cross-border participation in a capacity mechanism can further reduce overall costs, and therefore improve consumer and social welfare. Nonetheless, cross-border participation in capacity mechanisms raises a number of issues from both an economic and a legal point of view, which could cause doubt as to the best approach to interconnections in capacity mechanisms, and whether forcing the explicit cross-border participation of foreign entities is proportionate and efficient.

5 Commission, ‘Progress Towards Completing the Internal Energy Market’ (Communication) COM(2014) 634 final, 14. 6 M Cepeda and D Finon, ‘Generation Capacity Adequacy in Interdependent Electricity Markets’ (2011) 39 Energy Policy 3128. 7 D Finon, ‘Capacity Mechanisms and Cross-Border Participation: The EU Wide Approach in Question’ (2014) CEEM Working Paper, 6. 8 G Hawker, K Bell and S Gill, ‘Electricity Security in the European Union—The Conflict between National Capacity Mechanisms and the Single Market’ (2017) 24 Energy Research & Social Science 51; P Ringler, D Keles and W Fichtner, ‘How to Benefit from a Common European Electricity Market Design (2017) 101 Energy Policy 629; R Meyer and O Gore, ‘Cross-Border Effects of Capacity Mechanisms: Do Uncoordinated Market Design Changes Contradict the Goals of the European Market Integration?’ (2015) 51 Energy Economics 9. 9 K Neuhoff, J Diekmann, F Kunz, S Rüster, WP Schill and S Schwenen, ‘A Coordinated Strategic Reserve to Safeguard the European Energy Transition’ (2016) 41 Utilities Policy 252.

Cross-Border Participation in CRMs 181 This chapter complements the existing literature by providing a legal and economic analysis of the key issues associated with the implementation of cross-border participation. The next section starts by discussing the legal grounds to impose crossborder participation in capacity mechanisms. The third section focuses on the key legal and economic issues associated with the implementation of cross-border participation. The final section concludes on the necessity of cross-border participation in capacity mechanisms to comply with State aid regulation.

II. THE LEGAL GROUNDS TO IMPOSE CROSS-BORDER PARTICIPATION

When it comes to legal issues, the question is quite straightforward: does EU law oblige Member States to open their capacity mechanisms to participation in crossborder capacities? The answer requires a two-step analysis: first, it must be assessed whether such a requirement is provided by sector-specific legislation (II.A); and secondly, if the answer to this previous question is negative, it must be verified if a territorial restriction such as refusing to open a capacity mechanism to capacities located abroad would be incompatible with the common market principles (II.B).

A. No Clear Positive Obligation Deriving from Sector-Specific Regulation For the time being, no piece of EU legislation—neither primary nor secondary— clearly provides that when setting up a capacity mechanism, it must be open to participate in cross-border capacities.10 i. No Clear Obligation Deriving from EU Primary Legislation It is worth recalling that Article 194 TFEU only states that ‘Union policy on energy shall aim, in a spirit of solidarity between Member States, to: … (b) ensure security of energy supply in the Union; … (d) promote the interconnection of energy networks’. Moreover, according to Article 194 paragraph 2, each Member State has the ‘right to determine the conditions for exploiting its energy resources, its choice between different energy sources and the general structure of its energy supply’. Two remarks. First, completion of the internal energy market is one of the main concerns of the Commission for whom the Energy Union must become a reality. In this respect, the fragmentation of energy markets, as well as closed capacity markets, can be seen as impediments to reaching this goal. Secondly, the level at which security of supply must be assessed reveals the tension between Member States and the European Union (EU). Should there be a shift from Member States to the Union? If the Commission does not go that far, it states

10 Such a finding could evolve in the short to medium term with the adoption of the Commission’s proposal for a regulation on the internal market for electricity (see below).

182 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry that the ‘integration of markets also implies that security of supply, including generation adequacy, is increasingly difficult to ensure on a purely national basis’.11 To some extent, what is being done within the gas sector—favouring a regional approach—could be duplicated in the electricity sector, even though the problems are very different. ii. No Clear Obligation Deriving from EU Secondary Legislation As regards secondary legislation, no clear obligation to take into account crossborder participation can be inferred from the Electricity Directive12 or from the Security of Supply (SoS) Directive.13 Neither Article 8 of the Electricity Directive nor Articles 3 and 4 of the SoS Directive provide that a Capacity Remuneration Mechanism (CRM) must be open to cross-border participation. Moreover, Regulation 714/2009 defining the conditions for access to the network for crossborder exchanges does not provide any obligation related to capacity mechanisms.14 However, on 30 November 2016 the EC presented a new package of measures, which includes amendments to the Regulation 714/2009 on the internal market for electricity regarding capacity mechanisms. Additional Articles to the Regulation set out design principles for capacity mechanisms (Article 21) and guidelines on how cross-border participation must be introduced in capacity mechanisms (Article 23). At the moment, such a document remains a proposal and has no binding effect. Concerning Article 21(1) of the proposal of the Commission, it is worth noting that the Commission limits direct participation of cross-border capacities where ‘there is a network connection between that Member State and the bidding zone applying the mechanism’. One could ask whether such a limitation is compatible with EU law principles. However, for the time being it appears that no clear obligation to take into account cross-border capacities within national capacity mechanisms derives from sectorspecific legislation.

B. Application of the Common Market Principles We have seen that no positive obligation to open a capacity mechanism to participation in cross-border capacities can be identified in EU law. It must then be assessed whether refusing participation in a cross-border capacity mechanism could be a

11

Commission, ‘Generation Adequacy’ (n 4). Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC (Electricity Directive) [2009] OJ L211/29. 13 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment (SoS Directive) [2006] OJ L33/22. 14 Regulation (EC) 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003 (Regulation 714/2009) [2009] OJ L211/15. 12

Cross-Border Participation in CRMs 183 breach of the principles governing the internal market. More accurately, it must be seen whether such a refusal would qualify as an incompatible State aid or as a nonjustified restriction to the free movement of goods or services.15 i. Assessment under State Aid Rules As Article 107(1) TFEU states, 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, in so far as it affects trade between Member States, be incompatible with the internal market.

The Treaty prohibits State aid unless the measure is justified and entrusts the assessment of its compatibility to the Commission.16 Once a measure qualifies as an aid within the meaning of Article 107 TFEU,17 it must be notified to the Commission for the EU institution to check whether or not it is compatible. In that respect, the Commission benefits from strong discretionary and decisionmaking power. However, in order to offer predictability to undertakings and Member States the Commission may publicise its approach and the criteria used to assess the compatibility of notified measures in adopting guidelines. In the matter of aids for generation adequacy, the criteria under which the Commission will assess the compatibility of the measures are expressed in its ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG).18 As regards the specific issue of cross-border capacity participation, the Commission provides two elements. First, the Commission states that in order for the measure to be in accordance with the meaning of the guidelines, ‘it should take into account to what extent interconnection capacity could remedy any possible problem of generation adequacy’.19 Secondly, in order to avoid undue negative effects on competition and trade, the measure should be designed in a way so as to make it possible for any capacity which can effectively contribute to addressing the generation adequacy problem to participate in the measure, in particular, taking into account … the participation of operators from other Member States where such participation is physically possible in particular in the regional context, that is to say, where the capacity can be physically provided to the Member State implementing the measure and the obligations set out in the measure can be enforced.20

It is interesting to note that the position of the Commission is less straightforward than the one shown previously. Indeed, in its EEAG, if cross-border participation is

15

It is not obvious whether this issue raises antitrust concerns. Art 108, para 3 TFEU. A measure will qualify as an aid if the following conditions are met: (i) it involves a transfer of State resources; (ii) it entails an economic advantage for undertakings, (iii) it distorts competition by selectively favouring certain beneficiaries; and (iv) it produces an effect on intra-Community trade. 18 Commission, ‘Guidelines on state aid for environmental protection and energy 2014–2020’ [2014] OJ C200/1 (EEAG). 19 ibid, para 226. 20 ibid, para 232. 16 17

184 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry clearly encouraged, it does not appear to be an unconditional requirement. Besides, it must be kept in mind that the provisions of the EEAG only apply if the measure qualifies as an aid within the meaning of Article 107 TFEU. In other words, they may not be applicable to market-based capacity mechanisms.21 ii. Assessment under Free Movement Principles Cross-border capacity participation raises the question of inter-State trade and must therefore be reviewed under free movement principles. Even with a State aid approach, it is worth noting that respect for the free movement principles is being assessed by the Commission. Indeed, as the Commission states in the EEAG, ‘if a State aid measure or the conditions attached to it … entail a non-severable violation of Union law, the aid cannot be declared compatible with the internal market’.22 First, it should be noted that the set of rules that shall apply—free movement of goods (Articles 34–36 TFEU) or free movement of services (Articles 56–57 TFEU)— is not straightforward. Indeed, the reasoning with regard to qualifying capacity as a good within the meaning of the Treaty because capacity relates to electricity, which has been recognised as a good, is not fully satisfactory.23 Capacity and energy are different by nature. The Commission points out that: [T]he aid should remunerate solely the service of pure availability provided by the generator, that is to say, the commitment of being available to deliver electricity and the corresponding compensation for it, for example, in terms of remuneration per MW of capacity being made available. The aid should not include any remuneration for the sale of electricity, that is to say, remuneration per MWh sold.24

This difference is particularly clear for capacity mechanisms relying on availability and not on delivery. Those mechanisms provide some kind of insurance services to the system and seem quite far from the definition of a ‘good’ within the meaning of the Treaty. By analogy to another issue related to electricity, it is interesting to point out that with regard to demand response, the French competition authority decided that this economic activity should qualify as a service and not as a good. It states that: La proposition d’une offre d’effacement à RTE par un opérateur d’effacement sur le mécanisme d’ajustement semble assimilable à un service, dans la mesure où l’activité de l’opérateur d’effacement consiste à mettre en œuvre une baisse de la consommation d’électricité de certains consommateurs en échange d’une rémunération par RTE.25

21 See ongoing debate related to the French mechanism and the question of its qualification as an aid within the meaning of Art 107 TFEU. 22 EEAG (n 18) para 29. 23 According to settled case law, electricity is a good within the meaning of Art 34 of the Treaty (eg, Case C-393/92 Gemeente Almelo, EU:C:1994:171). 24 EEAG (n 18) para 225. 25 Opinion No 12-A-19 of 26 July 2012 on demand response in the electricity sector, para 66. (Free translation: ‘An offer made by a Demand-Response operator on the balancing market should be likened to a service as, in such a case, the action of the DR operator leads certain consumers to lower their electricity consumption in exchange of a price paid by the TSO’.)

Cross-Border Participation in CRMs 185 Even the question of ‘capacity certificates’ in decentralised market-based mechanisms is not entirely clear. It could be assumed that the certificate in itself must be seen as a good within the meaning of the Treaty. However, the question is more complex than it seems. It is worth recalling the debate in the Essent Belgium case26 on the question of whether or not guarantees of origin were goods within the meaning of the Treaty. The Court of Justice chose not to decide this point.27 Nevertheless, what is at stake—the qualification of a capacity as a good or a service within the meaning of the Treaty—is not decisive for the following,28 so we will assume that capacity qualifies as a good.

III. KEY ISSUES TO IMPLEMENT CROSS-BORDER PARTICIPATION

A. Introduction Most stakeholders across Europe recognise the contribution of interconnection to the security of supply and agree on the importance of enabling cross-border participation in capacity mechanisms. The EC is willing to ensure cross-border participation in capacity mechanisms (CMs) and encourages Member States to adapt their mechanisms to make it possible, even though, at the same time, it recognises the difficulties of implementing it in practice.29 In theory, different approaches could be envisaged to take into account the contribution of interconnection to the security of supply in the capacity mechanism country: —





26

Implicit or indirect cross-border participation. The dimension of the capacity contracted in the capacity mechanism takes into account the contribution of foreign capacities and/or interconnectors, but the foreign capacities and/or interconnectors are not certified directly and not remunerated. Explicit or direct cross-border participation of interconnectors. Interconnectors, but not generation or demand-side response (DSR) capacities located in other countries (hereafter ‘foreign capacities’), are certified in the capacity mechanism and are remunerated for their participation. Explicit or direct cross-border participation of foreign capacities. Foreign capacities, but not interconnectors, are certified in the capacity mechanism and are remunerated for their participation. A specific mechanism may allocate

See Case C-204/12 to C-208/12 Essent Belgium, EU:C:2013:294, Opinion of AG Bot, para 61. At para 81 of its decision, the Court noted: ‘it is not necessary to rule definitely on [this] question’. Moreover, we know, for instance, that the European Court of Justice has ruled that lottery tickets were not goods but services because behind the ticket lies the participation in a game of chance with the hope of winning (see Case C-275/92 Schindler, EU:C:1994:119). 28 Regarding free movement principles, case law is very similar between goods or services. However, the question of the qualification of a capacity will be important, for instance, to identify the regime applicable to public tenders or public procurements. 29 Commission, ‘Commission Staff Working Document accompanying the interim report of the sector inquiry on capacity mechanisms’ (SWD accompanying the Interim Report) SWD(2016) 119 final, 135. 27

186 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry interconnection capacity to foreign capacities and thus allow remuneration to be shared between interconnectors and foreign capacities. In practice, looking at the European state of play as of the end of 2016, there are only few examples of cross-border participation in capacity mechanisms. The French capacity mechanism takes into account the contribution of interconnection through a reduction of the capacity to be procured by obliged parties. It is expected that direct cross-border participation will be in place from 2019. In Great Britain (GB), the capacity mechanism has allowed, since its second year of operation, for interconnections to participate directly in the auction. The German grid reserve allows for Austrian generators to participate in the Grid Reserve. This is made possible insofar as there is no commercial congestion between Germany and Austria. Even though this is the privileged approach of the EC, direct cross-border participation, and most specifically in foreign capacities, encounters a number of difficulties from a technical, economic and legal perspective. These include, but are not limited to, the following questions: — —





The de-rating of interconnection: The evaluation of the (maximum) contribution of a given border is a tricky task, common to all participation models. Eligible entities: There is a question about which entity actually brings the value—whether it is the interconnector itself or foreign generators—and therefore about which entity should be entitled to participate. The design of products and obligations: The terms and conditions for interconnectors or foreign capacities to participate in the CM will define whether the capacities contracted in the CM effectively contribute to the security of supply of the CM country to a comparable extent as the domestic capacities. At the same time, they should not be discriminatory and put unmanageable obligations on these entities. Access to interconnection capacities: In cases of foreign capacities’ participation, modalities to access interconnection capacities would likely be necessary to limit cross-border participation to the maximum admissible level, allocate access rights and also allow remuneration to be shared between interconnection capacity and foreign capacities according to their actual value.

In the following subsections, we will address these aspects in turn. B. Evaluation of the Cross-Border Contribution The ‘de-rating’ of the interconnection is a first and necessary step to account for the contribution of imports to security of supply, whether considered to be implicit or explicit participation. It is also a crucial step insofar as it will, by deduction, determine how much capacity will be procured domestically. For instance, if the de-rated capacity of interconnections is set too low, more capacity will be procured domestically, which may lead to overcapacities. Conversely, if the de-rated capacity of interconnections is set too high, interconnections may not be able to provide so much capacity during scarcity situations, leading to a higher risk of power interruptions and a lower level of security of supply than desired.

Cross-Border Participation in CRMs 187 As illustrated in Figure 1 below, the evaluation of the cross-border contribution mainly depends on whether: —

The interconnection is available. Technical constraints on the cross-border transmission capacity may limit the import of electricity coming into the CM country at times of system stress. A precondition for the interconnection to contribute to the CM country’s security of supply is therefore to be available at times of system stress.30 — The interconnection is importing at times of scarcity (market risk). In an event of scarcity, the interconnection will not reduce the risk of load shedding if it is not importing, even if it is available. Provided the market works efficiently, this situation would occur if a scarcity situation is occurring simultaneously in the CM country and in the neighbouring country, such that there is no sufficient surplus of energy or capacity margin in this country to allow for import flows towards the CM country (‘simultaneous scarcity risk’).

Figure 1: Assessing Cross-Border Contribution in a CM Source: FTI-CL Energy.

30 For a direct current (DC) link, the risk of unavailability mainly depends on the occurrence of a technical fault, but for an alternative current (AC) link in a meshed network, the available capacity may be varying depending on specific conditions. The complexity increases even further with a flow-based capacity calculation.

188 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry The evaluation of the cross-border contribution thus requires not only forecasting the availability of the cross-border capacity but also forecasting generation adequacy in neighbouring countries and assessing the risk of simultaneous scarcity events. As such, it is not an easy task. A priori, the same methodology could apply for implicit and explicit participation. However, one may envisage not accounting for the market risk in cases of explicit participation under specific conditions. Indeed, in cases of implicit contribution, nobody participates explicitly in the capacity mechanism and is responsible to provide capacity. On the other hand, in cases of explicit participation, foreign capacity providers or interconnectors are participating in the capacity mechanism and are subject to the terms and conditions. If these terms and conditions require them to deliver electricity in the capacity mechanism, even in cases of scarcity in the neighbouring country and if operational rules and agreements, in practice, enable these capacity providers to schedule such trades in situations of simultaneous scarcity, then one may envisage to de-rate the interconnection capacity only to account for its reliable availability. Foreign capacity providers would handle the risk of coincident scarcity themselves. The implications of such requirements in the terms and conditions will be discussed in section III.D.ii.c. Nevertheless, except under specific rules aiming to guarantee the delivery of electricity in the capacity mechanism country at times of scarcity, even under simultaneous scarcity situations, the cross-border contribution should be similar in the implicit or explicit participation models. A study recently commissioned by the EC confirms that view.31

C. Eligibility Conditions i. Who Brings the Value? The contribution of foreign entities to the security of supply stems from the combination of capacity margins in the neighbouring country and of available interconnection capacity. In economic theory, the remuneration should go to the scarce good to provide fair remuneration and incentives. If there is limited cross-border capacity connecting the country with a capacity mechanism to a country in which there is significant excess capacity, at least in comparison with the cross-border capacity, then it is the interconnection which brings the capacity value and which should receive remuneration. Conversely, if there is high cross-border capacity but limited excess capacity in the neighbouring country, in comparison to the cross-border capacity, then it is the foreign generation or DSR capacities which should receive remuneration. However, in practice situations are not as simple and clear-cut and the value may be shared between these two types of entity. If it is possible to identify the situation

31 European Commission, ‘Framework for cross-border participation in capacity mechansms—Final Report’ (December 2016).

Cross-Border Participation in CRMs 189 ex ante, fewer complex models may be possible, especially if the interconnector is the limiting factor. The direct participation of interconnections would give all the capacity remuneration to the interconnections. From a purely economic and distributive point of view, it would be efficient only if the probability of the foreign capacity is the limiting factor, and is zero or sufficiently low. Otherwise, the model for direct cross-border participation should be able to allocate the value efficiently between the interconnection and the foreign capacity. In cases where the foreign capacity is the limiting factor, one could also wonder whether this value is provided by the overall generation fleet and DSR or by contracted individual capacity providers. In practice, unless there is a case of specific arrangements to secure the contribution of individual foreign capacities (to be discussed in section III.D.ii.c), the contribution of the neighbouring country depends on the availability of the whole generation fleet and its capacity margins when the capacity mechanism country is facing tight periods. This raises the question of whether individual foreign capacities should participate in the capacity mechanism in cases of direct cross-border participation, or whether they should participate in an aggregated form. ii. Legal Compatibility of the Interconnectors’ Participation Model The explicit participation of interconnectors raises two legal concerns: is it compatible with unbundling principles (section III.C.ii.a) and with competition law principles (section III.C.ii.b)? a. Interconnectors’ Participation and Unbundling Principles Regulated Interconnectors Interconnections are, in principle, part of the transmission network, and therefore are regulated infrastructure. According to the provisions of the Energy Directive, transmission system operators are natural or legal persons responsible for operating, ensuring the maintenance of and, if necessary, developing the transmission system in a given area and, where applicable, its interconnections with other systems, and for ensuring the long-term ability of the system to meet reasonable demands for the transmission of electricity.32

Therefore, the unbundling rules provided by Directive 2009/72 should apply to interconnectors.33 Three unbundling regimes are proposed—OU, ISO or ITO: [E]ach of them is expected to be effective in removing any conflict of interests between producers, suppliers and transmission system operators. This means that they should remove the incentive for vertically integrated undertakings to discriminate against competitors as regards access to the network, as regards access to commercially relevant information and as regards investments in the network.34 32

Electricity Directive (n 12) Art 2(4). ibid, Art 9. Commission, ‘Interpretative note on Directive 2009/72/EC concerning common rules for the internal market in electricity and Directive 2009/73/EC concerning common rules for the internal market in natural gas—The unbundling regime’ (Commission Staff Working Paper) 22 January 2010, 4. 33 34

190 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry In its Final Report of the Sector Inquiry on Capacity mechanisms, the Commission pointed out that the Third Energy Package and EU Network Codes require that interconnectors are treated as transmission capacity, and fully unbundled, and that the flow of energy across borders is determined solely by electricity price differences. Member States are however considering explicit participation designs that enable the direct participations of interconnector operators, foreign capacity, or a combination of the two.35

However, it did not further elaborate on its reasoning. As the purpose of capacity mechanisms is to remunerate the capacity provider, there may be some kind of incompatibility for regulated infrastructure such as interconnectors to participate in a capacity mechanism. However, if we assume that participation is feasible and does not breach unbundling requirements, another question then arises and is related to the use of the remuneration of the regulated interconnector as a capacity provider. We know that when structural congestion appears, TSOs are allowed to regulate the access of interconnections through management methods and earn a congestion rent.36 But, under the provisions of Regulation 714/2009, the use of the congestion rent is strictly limited.37 As regards the capacity remuneration of interconnectors, it remains unclear whether or not the same rules shall apply.38 Non-Regulated/Exempted Interconnectors If the participation of regulated interconnectors is questionable, participation of non-regulated/exempted interconnectors in CRM appears to be much easier to implement. It is important to note that Article 17 of Regulation 714/2009 provides that new current interconnections can be exempted from the unbundling regimes and from the rules governing the allocation of congestion rent. Therefore, nothing seems to prevent exempted interconnectors from participating in capacity mechanisms. b. Interconnectors’ Participation and Competition Law Principles As often ruled by the ECJ, ‘a system of undistorted competition, as laid down in the Treaty, can be guaranteed only if equality of opportunities is secured as between the various economic operators’.39 The principle of ‘undistorted competition’ prevents Member States from setting up a mechanism that is run by an operator that could in the meantime compete with other participants. This principle has been applied by

35 Commission, ‘Commission Staff Working Document accompanying the final report of the sector inquiry on capacity mechanisms’ (SWD accompanying the final report) SWD(2016)385 final, 188. 36 See, ‘Guidelines on the management and allocation of available transfer capacity of interconnections between national systems’ [2009] OJ L211/29. 37 Regulation 714/2009 (n 14) Art 16(6). 38 fn 143 of the SWD accompanying the Interim Report (n 29) is not clear: ‘Just as for congestion rents earned where electricity prices differ in neighbouring interconnected markets. For regulated interconnectors, any capacity congestion rents earned would need to be appropriately regulated (eg refunded to consumers in the connected markets if the interconnectors’ revenues—including the capacity revenues— are above its regulated cap). See Regulation 714/2009 Arts 16 and 17’ (emphasis added). 39 Case C-49/07 MOTOE, EU:C:2008:376, para 51.

Cross-Border Participation in CRMs 191 the French competition authority to prevent the TSO from competing on the market for demand response.40 The same reasoning could apply to capacity mechanisms and prevent interconnectors from participating in it if the interconnector and the TSO are the same legal entity. D. Design of Products and Obligations i. General Requirements The terms and conditions for the participation in the capacity mechanism define the requirements and obligations for participating entities, including the design of the products and the penalties or settlement regimes. Their definition is important to ensure that all participating entities—domestic, but also foreign where applicable—may participate as much as possible on equal grounds and provide comparable ‘services’ to the capacity mechanism consumers. It is also fundamental because it determines the incentives for foreign entities to participate in the capacity mechanism, to bid in the procurement process and to effectively contribute to the security of supply in the capacity mechanism country. Two options are generally considered: —



Availability obligation: A capacity participating in the capacity mechanism commits to be available during certain critical periods for the security of supply of the capacity mechanism country, ie, when this capacity is likely necessary to ensure the security of supply (eg, peak hours in winter). This is the case in the French capacity mechanism, for instance. Delivery obligation: A capacity participating in the capacity mechanism commits to deliver electricity during certain periods, when the system is in a stressed situation. This is the case in the British capacity mechanism as well as in strategic reserves (Belgium, Germany etc).

In both cases, these products and obligations are designed to secure supplies from these capacities at times of scarcity. Usually, in cases of an availability obligation, safeguards are in place to secure the physical contribution of the concerned capacity at times of scarcity. For instance, TSOs may activate the capacity—if it is not generating—in cases of need, whether it is because central dispatch is in place (in US markets) or capacity providers have to bid in the balancing market, as in the French capacity mechanism.41 Should these capacities not respect their commitments, they may face penalties. 40 Opinion No 13-A-25 of 20 December 2013, para 114. It states that: ‘L’attribution des deux missions précitées d’agrément et de certification à RTE paraît s’opposer à ce que celui-ci puisse intervenir directement en tant qu’acteur sur le marché de l’effacement. A défaut, en effet, RTE pourrait être à la fois le concurrent des opérateurs d’effacement et l’entité chargée de contrôler leur activité, ex ante par la délivrance de l’agrément et ex post par la certification des volumes d’effacement réalisés’. (Free translation: ‘the allocation of the two abovementionned tasks of agreements and certifications by RTE should prevent RTE from acting on the Demand-Response market. Otherwise, RTE would be simultaneously a competitor on the DR market and the body in charge of controling DR activity by providing agreements and certifications to the volume of curtailed load’.) 41 A ‘normative’ contribution based on the forecast statistical contribution of the participating entities also exists, for instance for variable renewable energy sources.

192 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry ii. Application to Cross-Border Participation a. Differences and Difficulties in Cases of Cross-Border Participation In cases of implicit participation, the TSO evaluates the contribution of interconnections and foreign entities but these entities are not responsible, in practice, to effectively contribute and their implicit contribution is not remunerated. When it comes to direct participation, foreign capacity providers and/or interconnectors are remunerated. As such, the guarantees provided by these participants should a priori be comparable to domestic capacities: both the consumers who pay for the capacity and the policymakers are expecting equivalent guarantees on the reliability of the contribution. However, the adaptation of the product definition to interconnection capacity may raise issues. Guaranteeing the contribution across a border raises specific questions and difficulties. The interconnection does not decide on the flows across the border: the outcome of the market sets the exchanges of electricity depending on the prices: —



In the case of scarcity only in the capacity mechanism country, the energy price should be higher in the capacity mechanism country than in the neighbouring country and the interconnection should be importing towards the capacity mechanism country up to its maximum capacity. In cases where both countries are facing a scarcity situation simultaneously, the interconnection might not be importing towards the capacity mechanism country up to its maximum capacity, despite being available. The interconnection may not be contributing to the security of supply in the capacity mechanism country during such a period of scarcity.

This situation is illustrated in Figure 2 below.

Figure 2: Illustration of the Market Outcome in terms of Cross-Border Flows in Cases of Non-Simultaneous and Simultaneous Scarcity Source: FTI-CL Energy.

Cross-Border Participation in CRMs 193 Even though the market outcome may be an export towards the capacity mechanism country, there is the possibility of political intervention in such situations: it may be unacceptable for the other country to be exporting while it has to carry out selective load shedding to maintain the balance in its country. In this regard, it is important to point out that the legal framework does not ensure that the commitment of a capacity provider located in another Member State to deliver electricity during scarcity periods could be enforced. Indeed, in emergency situations or cases of force majeure, EU secondary legislation does not prevent a Member State or a TSO from curtailing cross-border supply.42 To conclude, if the product is availability-based with no activation means (which is a very theoretical situation), the interconnection and/or foreign capacity would be able to meet its obligation, although it might not be contributing to security of supply in the capacity mechanism country in a situation of simultaneous scarcity. If one wants to guarantee the physical contribution at the border, then one could impose a delivery obligation or an availability obligation coupled with an obligation to bid in the balancing market or to be ready to be activated. b. Participation in Interconnection Capacity Participation in interconnection capacity would be mostly relevant in cases where the risk of coincident scarcity is non-existent or very low. In such a situation, the value relies on the interconnection. The interconnection owner could be ready to take on the risk of non-delivery when scarcity occurs simultaneously on both sides. It would then bear the financial risk of penalty or imbalance settlement. Obviously, the acceptability of this option would depend on the materiality of the risk of simultaneous scarcity, the definition of the exact obligation43 and the level of the penalty or of the imbalance settlement.44 It could also envisage securing delivery even in such situations to hedge against this risk. In order not to be in breach and not to face the penalty, the interconnection may contract with generators or DSR operators, which would be kept as back-up capacity (‘sterilisation of capacity’). Should the interconnection not flow towards the capacity mechanism country during times of commitments in the capacity mechanism country when delivery was required, the interconnection would activate these contracts—the contracted generators would then start generating or DSR operators would curtail load—and reschedule the cross-border flow to meet its obligation. This option, however, raises the question of the impact on the market

42 More precisely, neither Art 4 of the SOS Directive (n 13), nor Art 42 of the Electricity Directive 2009/72/EC, Art 16 of Regulation 714/2009 (n 14) or Art 72 of the Guideline on CACM prevent the intervention of a Member State or of the TSO in cases of force majeure or emergency. See Commission Regulation (EU) 2015/1222 of 24 July 2015 establishing a guideline on capacity allocation and congestion management (Guideline on CACM) [2015] OJ L197/24. 43 An obligation based on delivery in rare events might be riskier than an availability-based obligation over a longer period with few cases of activations. 44 It is worth mentioning that this is the approach chosen in the UK, where the obligation is deliverybased and both IFA and Britned, two of the existing interconnections, agreed to take on the financial risk.

194 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry and the acceptability of the approach.45 Such an approach would likely need political back-up as the interconnector operator would ultimately be able to schedule exports from the neighbouring country whereas it is in a scarcity situation and risks load shedding. Under such assumptions, the direct participation of the interconnection may improve the contribution of the interconnection to the capacity mechanism country—in the short term—compared with the model of indirect participation of the interconnection capacity. Indeed, the risk of simultaneous scarcity would not need to be taken into account in the de-rating. Conversely, in the absence of such guarantees, a similar de-rating/contribution would be applied in the indirect participation or in the direct participation in interconnection capacity. c. Participation of the Foreign Capacity However, if the risk of coincident scarcity is not negligible, the value is a priori shared between interconnectors and foreign capacity. The products and obligations for foreign capacity should then be designed in order to allow for their efficient participation. In cases of scarcity only in the capacity mechanism country, the energy price should be at the price cap and it should therefore be higher in the capacity mechanism country than in the neighbouring country. Therefore, if the interconnection is available, it should be fully importing towards the capacity mechanism country and would thus contribute to the security of supply at times of stress. In such a situation, obliging the foreign generation/DSR capacity provider to generate could create undesirable impacts on the energy market, especially if its short-run marginal cost is above the market price. In cases where both countries are facing a scarcity situation, nonetheless, the interconnection might not be fully importing towards the capacity mechanism country, despite its being available and despite the foreign capacity being available and probably generating. The foreign entities would not be contributing to the security of supply in the capacity mechanism country during such a period of simultaneous scarcity. Nonetheless, the foreign capacity may respect an availability obligation or a delivery obligation defined as generating, but it would not contribute directly to the security of supply of the capacity mechanism country. To guarantee the physical contribution to the security of supply of the capacity mechanism country by the foreign capacity at times of stress, one possibility could be to define the commitment as an obligation to effectively contribute to security of supply of the capacity mechanism country (export obligation). To be able to export to the capacity mechanism country despite a situation of simultaneous scarcity TSOs would need to put in place arrangements to reschedule export flows to the capacity mechanism country up to the capacity contracted in the capacity mechanism country. However, these arrangements would imply that exports are scheduled outside the usual functioning of the market while greater curtailment of consumers may

45 In addition, in cases of flow-based capacity calculation, there may no longer be the possibility to schedule exports to the CM country despite the capacity contract.

Cross-Border Participation in CRMs 195 occur in the country. Such a counter-trading contract would need to be backed up by Member States’ political agreements but it would be unlikely to be acceptable for the neighbouring country, especially if it trusts the energy-only market to give the adequate price signals for dispatch and security of supply to schedule cross-border flows. In the absence of such politically backed arrangements, the de-rating of the interconnection cannot exceed the value calculated in cases of statistical contribution.

E. Access to Interconnection Capacity i. Options for Interconnection Capacity Access In energy markets, as cross-border capacity between Member States is limited, congestion management mechanisms need to be put in place to allocate interconnection capacity between market participants in the most efficient manner. Similarly, in the case of direct participation of foreign generators in capacity mechanisms, interconnection capacity allocation should be implemented, to allow for the cross-border trading of capacity. The analysis of the different options leads us to keep the acquisition of specific interconnection products for capacity trading (‘tickets’) as the most suitable one. Indeed, the reservation of cross-border capacity would withdraw capacity from the market especially when additional imports would be valuable and necessary in the capacity mechanism country, to allow for out-of-market exchanges. Such an approach would have a detrimental impact on the energy market, on security of supply and on overall efficiency, while there are similar concerns as to the acceptability of scheduling exports to the capacity mechanism country when the neighbouring country is in scarcity situations. Moreover, the legality of this approach could be questioned, as TSOs are supposed to maximise the interconnection capacity which is made available to the market.46 The acquisition of transmission rights, allocated by TSOs for the cross-border trading of energy, does not appear to be an efficient solution either. In practice, it would have a limited impact on effective cross-border flows because, close to realtime, flows are scheduled based on market price differences, and so it would not guarantee the effective contribution of foreign capacity, even if the nomination of these rights was compulsory. Conversely, TSOs could allocate specific products, totally independent of the wholesale energy market, avoiding any interference with the energy market. Interconnections would get additional revenues, which should in theory reflect the split of value between foreign generation/DSR capacity and interconnection capacity. Conceptually, the tickets should be priced at the difference between the (marginal) capacity price in the capacity mechanism country and the (marginal) capacity price in the neighbouring country.

46

Regulation 714/2009 (n 14) Art 16.

196 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry The allocation of these tickets for capacity trading could follow similar principles as the allocation of transmission rights for energy trading. They could be allocated through: —



The explicit auction of tickets: Foreign capacity providers would first participate in an auction to acquire tickets. The revenue of the ticket auction would accrue the total revenues of the interconnection. Successful bidders would then participate in the foreign capacity market. The implicit auction of tickets: Foreign capacity providers would participate directly in the capacity mechanism auction but, when the market clears, they will get the marginal capacity price of the foreign generation/DSR capacity.47 The price difference between this price and the clearing price of the capacity market would be attributed to the interconnection.

ii. Issues Linked to the Definition of Products in Interconnection Capacity Access In theory, the auctioning of tickets should be an efficient way of allocating interconnection capacity to the efficient foreign capacity providers and to share the capacity value between the foreign capacity and the interconnection capacity. Nonetheless, in practice the definition of the product and the associated penalty has a strong impact on the efficiency of the approach. The analysis of the product definition options led to the conclusion that foreign capacity providers would bear limited specific constraints. The obligation to be available or even to generate in periods of scarcity would not be likely to create additional costs for most generators, which should already be incentivised to be available during these periods. In the absence of capacity mechanisms in the neighbouring country, the most competitive capacity providers are therefore likely to bid at a very low price, representing the risk of being unavailable during these periods. This would particularly be the case if, generally, the generating capacity is largely higher than interconnection capacity. Conversely, power plants which specifically require additional remuneration to stay on line would have less of a chance to be successful in the ticket/capacity mechanism auction. iii. In Cases of Cross-Border Participation between Two Capacity Markets The previous analysis would change if both countries have implemented marketwide capacity mechanisms. Indeed, if the needs and the products are sufficiently aligned and if capacity providers are not allowed to participate in both mechanisms, they would have an opportunity cost in participating in one rather than the other. A capacity provider deciding to bid in the neighbouring capacity mechanism would bid at least the capacity price it expects to get in its domestic capacity mechanism. This opportunity cost would be reflected in its bid. Thus, the price for interconnection

47 In cases where foreign capacity providers from different neighbouring countries participate, there are likely to be different marginal prices for foreign capacity.

Cross-Border Participation in CRMs 197 capacity tickets would tend towards the capacity price difference between the two capacity mechanisms. This is illustrated in Figure 3 below. We consider two countries A and B. The plain lines represent supply in both countries’ capacity mechanism when there is no cross-border participation. At the equilibrium, the price in country B is much lower than the price in country A. If we now allow for cross-border participation, part of country B’s capacity is now supplied in country A’s capacity mechanism in response to price differentials. Precisely, as compared with the previous situation, supply in country B’s capacity mechanism decreases while it increases in country A. The phenomenon stops when prices in the two countries are the same

Figure 3: Cross-Border Capacity Exchange between Two CMs Source: FTI-CL Energy.

198 Fabien Roques, Charles Verhaeghe and Guillaume Dezobry or when cross-border transmission capacity reaches its limit. The latter situation is reproduced in Figure 3. At the new equilibrium, new supply curves are represented by dash lines. The difference in country A’s and country B’s capacity mechanism prices defines the interconnection ticket price. The allocation of tickets, whether explicit or implicit, could therefore more efficiently share the value between interconnection capacity and foreign generation/ DSR capacity, than in the case of cross-border participation from capacities located in a country with an energy-only market. Textbox 1: Focus on the Compulsory Participation in a Capacity Mechanism Is a CRM that makes it compulsory for any capacity provider located in a Member State to participate in the national CRM a measure with the equivalent effect of an export restriction within the meaning of Article 35 TFEU?48 It could be argued that compulsory participation in the local CRM makes it more difficult for the capacity provider, and less economically interesting, to participate in another CRM and therefore would qualify as a restriction within the meaning of Article 35 TFEU, which would need to be justified under the provisions of Article 36 TFEU. There is no clear-cut answer to this question. Therefore, it could be cautious to provide an ‘opt-out’ provision as the Commission recommended: just as the possibility should also exist for capacity located elsewhere to participate in a mechanism, it should also be possible for capacity to ‘opt out’ of its national scheme, in order to instead participate in a mechanism established elsewhere.49

IV. CONCLUSION ON THE NECESSITY OF CROSS-BORDER PARTICIPATION TO COMPLY WITH STATE AID RULES

Accounting for the contribution of interconnections in the design of a capacity mechanism is important in order to avoid over-investment and distortions. As such, the implicit model, where this contribution is evaluated and subtracted from the capacity target, appear as a no-regret and possibly transitory option. If the cross-border contribution is adequately estimated, this approach provides already substantial benefits, avoiding over-investment or over-procurement in the capacity mechanism country and limiting negative impacts on the functioning of the energy market and cross-border trading. Moreover, the implicit model seems compatible with the principle of EU law. Indeed, such an approach should be compatible with the free movement principles. Indeed, as capacity mechanisms help to strengthen the security of supply, they are dealt with as measures linked to public security issues within the meaning of Articles 34 and 36 TFEU.50 Therefore, even if a capacity mechanism closed to cross-border capacity participation infringes Article 34 TFEU, the purpose of the 48 Art 35 TFEU states that: ‘Quantitative restrictions on exports, and all measures having equivalent effect, shall be prohibited between the Member States’. 49 Commission, ‘Generation Adequacy’ (n 4). 50 In a landmark case, the Court has linked security of energy supply and public security. See Case C-72/83 Campus Oil, EU:C:1984:256.

Cross-Border Participation in CRMs 199 measure should enable it to benefit from the exemption provided by Article 36.51 In addition, the implicit participation of interconnection should limit the risk of overremuneration of domestic capacities, and therefore avoid the risk of State aid. The Commission recognises that, ‘[implicit participation] reduces the risk of domestic over-procurement and recognises the value to security of supply of connections with the Single Market for energy’.52 On the other hand, the explicit participation of foreign entities raises substantial implementation challenges, in particular to allow foreign generators to participate. Some trade-offs appear in the design of the mechanism, such as the complexity of implementation, the willingness to secure physical contribution during (coincident) scarcity situations and the impact on the market outcome and the likely acceptance of the consequences at the political level. Consequently, the choice of approach should be carefully evaluated in terms of costs and benefits. Two specific situations may facilitate the implementation of explicit cross-border participation. First, the explicit participation of interconnectors could be an adequate solution in cases where the interconnection capacity is the limiting factor in cross-border contributions and if the remuneration of interconnection capacity is likely to provide incentives to interconnectors’ operators and investors to maximise capacities during scarcity situations and build new cross-border lines. However, even though the EC considers that the remuneration of interconnection capacity through its participation in capacity mechanisms could be an incentive for new investments in interconnection,53 there has been a debate on whether such incentives could actually materialise. For instance, the revenues for interconnectors participating in the capacity mechanism are seen as a driver for investment, but this is due to specific reasons: (i) the participation of interconnection capacity is made easier by the fact that all UK interconnectors (at least so far) are DC cables; and (ii) and probably more importantly, investments in new interconnection capacity are mostly merchant or incentivised based on revenues (eg, ‘cap and floor’ regimes). Conversely, within meshed areas, building new interconnectors on a merchant basis is more complex as the network is highly meshed and interconnection capacities may be allocated through a flow-based methodology. Moreover, TSOs investing in a regulated framework are supposed to undertake investments on the basis of their overall social welfare benefits, including benefits in terms of security of supply, rather than based on potential revenues. Second, cross-border participation could take place between two (market-wide) capacity mechanisms. In the presence of similar capacity mechanisms, it is possible to introduce modalities for cross-border participation which provide efficient incentives for generators to participate. This option does not however address potential concerns in cases of coincident scarcity situations: the Member State where exchanged capacity is located may benefit more from these capacities in such cases by refusing exports to the neighbouring country. Bilateral agreements could, in theory, address such situations, but they would be difficult to reach. 51 52 53

Provided that the measure is also proportionate. Final Report on CM (n 2) 183. See, eg, SWD accompanying the Interim Report (n 29) 88.

8 Aid to Nuclear and Coal LEIGH HANCHER AND MAX KLASSE

S

ECTION I OF this chapter will focus on the relation between the EU State aid rules and the Euratom Treaty which governs most aspects of the nuclear sector. The application of the EU State aid rules to the coal sector is dealt with in the second section of this chapter.

I. NUCLEAR

A. Introduction: The EU’s Nuclear Sector—Facts and Figures Nuclear energy currently generates approximately one-third of all electricity produced in the EU. There are 129 nuclear power reactors in operation in 14 Member States with a total capacity of 120GWE. While some Member States are proceeding with new construction or the licensing of new projects, others have either decided to phase out existing nuclear plants or to reduce the share of nuclear in their energy mix. Several countries have adopted moratoria or bans on the construction of nuclear reactors in their national territory.1 The latest Commission Nuclear Illustrative Programme (PINC) of April 2016 predicts a decline in nuclear capacity up to 2025 but also that this trend could be reversed by 2030 when new reactors come on stream. By 2050, nuclear capacity is predicted to remain stable at between 95 to 105 GWE. The Commission estimates that investments in the nuclear fuel cycle between 2015 and 2050 could total between €650 to €760 billion. This breaks down into between €349–€455 billion for new build, €123 billion for decommissioning and €203 billion for waste management, including final disposal. A further €47 billion is calculated as necessary to cover long-term operations.2

B. The Relationship between the Euratom Treaty and the TFEU The Euratom Treaty, unlike the EU and (now defunct) European Coal and Steel Community (ECSC) Treaties does not contain any specific provisions on State aid 1 2

See, eg, T Mann, ‘The Legal Status of Nuclear Power in Germany’ (2014) 2 Nuclear Law Bulletin 43. Commission, ‘Nuclear Illustrative Programme’ COM(2016) 177 final.

202 Leigh Hancher and Max Klasse control. At the same time, the Euratom Treaty (EA) is a ‘lex specialis’. Nuclear energy may remain a special case in that the EU Treaty rules cannot always be straightforwardly and directly applied to evaluate aid to this sector. Indeed, the exact relationship between the two Treaties has been the subject of some controversy in legal literature.3 i. The Euratom Treaty in Brief The Euratom Treaty’s main goal at the time of its adoption was to create the conditions necessary for the speedy establishment and growth of nuclear industries (Article 1 EA). As a result, the Treaty entrusts the Commission and Member States with a duty to promote the development and use of nuclear energy (Articles 2 and 4 EA)4 as well as the safe management of spent fuel and radioactive waste (Article 2(b) EA). Pending further judicial guidance, this duty of promotion is not absolute. It may be observed that the Euratom Treaty, even if it does not contain a specific State aid regime, does not appear to immunise all national measures from Commission supervision or from the impact of the TFEU competition rules. Article 5 EA, for example, gives the Commission considerable powers to coordinate the nuclear research undertaken by Member States. The Commission may issue a specific or general request to Member States, persons or undertakings to communicate to it their programmes relating to the research, which it specifies in the request. After giving those concerned a full opportunity to comment, the Commission may deliver a reasoned opinion on each of the programmes communicated to it. These opinions enable the Commission to discourage unnecessary duplication and to direct research towards sectors which are insufficiently explored. Article 6 EA provides that to encourage the carrying out of research programmes communicated to it the Commission may, inter alia, promote the joint financing by the Member States, persons or undertakings concerned (Article 6(d) EA). Similar duties of notification of various aspects of investment in nuclear power facilities are imposed on Member States by way of Articles 34 and 37 and in chapter 4 (Articles 40–44) of the Euratom Treaty. Furthermore, the promotion of nuclear safety is a Union competence which must be linked to the protection

3 See R Ptasekaite, ‘Competition Law and Nuclear Regulation: A European Perspective’ in B Delvaux et al (eds), EU Energy Law and Policy Issues, Volume 4 (Cambridge, Intersentia, 2014); A Södersten, ‘Euratom at the Crossroads’ (PhD thesis, European University Institute, Florence, 2014) 197; M Sousa Ferro, ‘Competition Law and the Nuclear Sector: An EU Outlook’ (2010) 2 Nuclear Law Bulletin 13, 15; L Garzaniti, ‘Competition Law in the Nuclear Sector’ in INLA, Nuclear Inter Jura 2007, Proceedings 1–4 October 2007 (Brussels, Bruylant, 2008) 1211; A Bouquet, ‘Which Competition Rules for Nuclear Energy in a Liberalised European Market Environment’ in INLA, Nuclear Inter Jura 2007, Proceedings 1–4 October 2007 (Brussels, Bruylant, 2008) 1165; TF Cusack, ‘A Tale of Two Treaties: An Assessment of the Euratom Treaty in Relation to the EC Treaty’ (2003) 40 CML Rev 1, 117, 130; J Grunwald, Das Energierecht der Europäischen Gemeinschaften (Berlin, De Gruyter Recht, 2003) 234; M Pechstein, ‘Elektrizitätsbinnenmarkt und Beihilfenkontrolle im Anwendungsbereich des Euratom-Vertrags’ [2001] Europäische Zeitschrift für Wirtschaftsrecht 307, 311. 4 Art 197 EA which defines such terms as special fissile materials, source materials and ores, does not appear to apply to waste materials. Furthermore, the activities covered in Annex 1 to the Treaty would not seem to apply to decommissioning or storage as such—but only to certain forms of research.

Aid to Nuclear and Coal 203 against the dangers arising from radiation, laid down in Article 30, chapter 3, relating to health and safety.5 In addition, the rules in chapter 6 of the Euratom Treaty governing nuclear fuel supplies and the operation of the Euratom Supply Agency can be taken to support the conclusion that it was the intention of the framers of the Treaty that various forms of unfair competitive practices, including ‘all practices designed to secure a privileged position for certain users (of ores, source materials and special fissile materials) shall be prohibited’. In this context, Article 67 EA stipulates that: Prices shall be determined as a result of balancing supply against demand … the national regulations of the Member States shall not contravene such provisions. Pricing practices designed to secure a privileged position for certain users in violation of the principle of equal access … shall be prohibited.

As to radioactive waste the EU legislative framework applicable to its management is grounded on two fundamental principles: the Member States have ultimate responsibility for the safe management and disposal of spent fuel and radioactive waste and operators of nuclear installations have prime responsibility for the safety of spent fuel and radioactive waste management and must bear the costs from generation to disposal of all such products—ie, the polluter pays principle.6 Finally, Member States are, in accordance with Article 192 EA, under a general duty to ensure the fulfilment of the Treaty objectives and abstain from any measure which could jeopardise the attainment of those objectives. Recital 3 of the Euratom Treaty also confirms as one of those objectives, the guarantee of security of supply. Thus, unlike the ECSC and the TFEU Treaty, while there is no general prohibition on State aid, and there may well be legal grounds to conclude that in specific cases subsidies are to be encouraged, it cannot be concluded that the nuclear sector falls outside the TFEU rules on competition. In any event, the Commission has applied Articles 101 and 102 TFEU to the nuclear sector from an early date.7 ii. The Relationship between the Euratom Treaty and the State Aid Rules of the TFEU Article 305(2) EC (now repealed) stated that the provisions of the EC Treaty should not derogate from the rules establishing the Euratom Treaty. This provision was worded slightly differently from Article 305(1) EC (also repealed), which stated that the EC Treaty shall not affect the provisions of the ECSC Treaty. As the ECSC Treaty contained specific provisions on State aid to the coal and steel sectors, there was

5

This is confirmed in Case C-29/99 Commission v Council, EU:C:2002:734. See also Commission Recommendation 2006/851/Euratom of 24 October 2006 on the management of financial resources for the decommissioning of nuclear installations, spent fuel and radioactive waste [2006] OJ L330/31; and Council Directive 2011/70/Euratom of 19 July 2011 establishing a Community framework for the responsible and safe management of spent fuel and radioactive waste [2011] OJ L199/48, and especially Art 4(3). Special arrangements have however been agreed on accession in exchange for agreement to close nuclear plants. See, eg, Ignalina Nuclear Power Plant Tax Exemptions (Case N337/2005) [2006] OJ C162/9. 7 United Reprocessors GmbH (Case IV/26.940/a) Commission Decision 76/248/EEC [1976] OJ L51/7. 6

204 Leigh Hancher and Max Klasse no doubt that the relevant provisions applied to the exclusion of Articles 107 and 108 TFEU. Article 106a EA now regulates the relationship between the EA and the TFEU. Article 106a(3) EA replicates the text of Article 305(2) EC—the provisions of the TEU and the TFEU shall not derogate from the provisions of the Euratom Treaty. The legal relationship between the Euratom Treaty and the EU State aid rules has not yet been fully clarified by the courts however, and this has led to considerable debate as to whether the EU State aid regime may also apply to the nuclear sector.8 In his Opinion in Joined Cases 188–190/88, Advocate General Reischl was of the opinion that Articles 107–09 TFEU should be applied to regulate State aid to the nuclear sector. This was not precluded by the application of Article 305(2) EC. The Court did not address the issue directly, however.9 The duty of promotion referred to in Article 2 EA does not appear to be openended nor, given that the Euratom Treaty—unlike the ECSC Treaty—is unlimited in its duration, should it be seen as indefinite. To exclude a sector—which is, after all, now a ‘mature’ sector of the economy of a number of Member States—from any form of State aid discipline whatsoever would seem to furnish that sector with a considerable advantage over and above other competing fuel forms which are subject to the full discipline of the EU Treaty rules as well as secondary legislation on the internal energy market. Indeed, to accord a completely separate or privileged status to the State funding of nuclear energy would seem to make something of a nonsense of the concept of a single European electricity market, a concept which is predicated on competition between all forms of electricity, irrespective of the source of their generation. In Case C-490/10 EP v Council10 the Court held that the Commission had been wrong to take Article 187 EA as a legal basis for its Council Regulation (EU, Euratom) No 617/2010 of 24 June 2010 concerning the notification to the Commission of investment projects in energy infrastructure within the European Union. The Court held that: The information relating to the nuclear infrastructure is thus only a component of all the relevant information concerning the energy system of the European Union as a whole which the Commission must possess in order … to carry out an overall assessment of energy demand and supply with the aim, inter alia, of guaranteeing security of energy supply in the European Union. In that respect, it must therefore be noted that the contested regulation … does not fall within the scope of the objective of promoting or coordinating investments in the nuclear field provided for in Articles 40 EA to 44 EA, which specifically relate to the communication by undertakings engaged in the nuclear sector of all individual investment projects in that field relating to new installations and also to all replacements or conversions of a

8

See Cusack (n 3) 117, 131. The Court did not accept the submission of the French government that the so-called Transparency Directive derogated from the provisions of the Euratom Treaty as the French government had not established this point. Joined Cases C-188-190/80 France, Italy and the UK v Commission (Transparency Directive), EU:C:1982:257, para 29. 10 Case C-490/10 Parliament v Council, EU:C:2012:525. 9

Aid to Nuclear and Coal 205 certain size. By contrast, the contested regulation concerns the notification by all Member States of the aggregated data and information relating to all investment projects in energy infrastructure.11

It follows that the starting point should be that the EU Treaty rules apply to the nuclear sector provided that those rules are not at variance with those of the Euratom Treaty. Further, even where for certain matters, the provisions of the Euratom Treaty as a lex specialis can take precedence over the provisions of the EU Treaty as a lex generalis, in the interpretation of the lex specialis the Court may call on the lex generalis. In Ruling 1/78 the Court held that, in the light of the EC Treaty, the provisions of the Euratom Treaty on the nuclear internal market ‘appear to be nothing other than the application, in a highly specialised field, of the legal conceptions which form the basis of the structure of the general common market’.12 This interpretation was again confirmed in the recent Case C-5/1413 Kernkraftwerke Lippe-Ems. In this case, the referring court asked whether the fiscal duty introduced by the relevant legislation in question would be contrary to (1) the attainment of the EAEC’s objective of establishing the conditions necessary for the speedy establishment and growth of nuclear industries, set out in the second paragraph of Article 1 EA, and (2) the fulfilment of the EAEC’s duty to ensure that all its users receive a regular and equitable supply of ores and nuclear fuels, laid down in Article 2(d) EA. The Court held that: It should be noted, in that regard, first, that the combined application of the second paragraph of Article 192 EA and the second paragraph of Article 1 EA does not have the effect of requiring Member States to maintain or increase their level of use of nuclear fuel or of preventing them from taxing such use, which would make such use more costly and, therefore, less attractive.14

C. Applying Article 107 TFEU to the Nuclear Sector To date, the Commission has assessed a number of proposed national aid measures to the nuclear sector under the EU Treaty State aid rules. In those decisions where the measures in question fulfil the cumulative conditions of Article 107(1) TFEU, the Commission has in turn applied Article 107(3) TFEU to assess their compatibility. As State aid to the nuclear sector does not fall under the scope of the current guidelines on energy aid, nor did it fall under any previous sets of environmental guidelines, the Commission assesses the aid in question either directly under Article 107(3)(c) TFEU or, where relevant and in several cases, under its Rescue and Restructuring Guidelines. As we shall explain further below, in its assessment of the application of Article 107(1) TFEU to various State measures targeted at the nuclear

11 12 13 14

ibid, paras 83–84. Ruling 1/78 Physical Protecion of Nuclear Materials, EU:C:1978:202, 2172. Case C-5/14 Kernkraftwerke Lippe-Ems, EU:C:2015:354. ibid, para 102.

206 Leigh Hancher and Max Klasse sector, the Commission does not appear to have modified the way in which the various cumulative conditions are to be applied to the nuclear sector. It is however in the assessment of the compatibility of the measures at issue that the interrelationship between the EA and the State aid regime become more complex. Following the adoption of the State Aid Modernisation (SAM) exercise in 2012,15 compatibility of State aid to the nuclear sector is based on the same common assessment principles that are now applied to all types of State aid—including rescue and restructuring aid.16 Section I.C.i below will consider the application of Article 107(1) TFEU and the following section I.C.ii on compatibility will distinguish between those decisions adopted prior to 2014 and those adopted thereafter. i. Article 107(1) TFEU As this chapter will explain, the Commission has not deviated from its standard approach to applying Article 107(1) TFEU. Although the Commission has adopted a number of decisions relating to State aid in the nuclear sector, no direct challenges against a Commission decision have been lodged before the courts. The challenges lodged by a Member State to the recent Commission decision on State aid granted by the UK government for the operation of the nuclear reactor at Hinkley Point may offer an opportunity for the courts to provide definitive guidance on the relationship between the Euratom and the TFEU State aid rules.17 a. Economic Advantage As with the application of Article 107(1) TFEU to any sector of the economy or any undertaking engaged in an economic activity, it must be determined whether the potential beneficiary would benefit from an economic advantage that it would not receive in the normal course of business. In its recent decision on the UK Waste Transfer Contract for New Nuclear Plants18 the Commission assessed various measures concerning the funding of certain waste disposal services for spent fuel and intermediate level waste. The UK argued that the Waste Transfer Contracts would not provide any advantage to operators given that the service is priced in a way which, in the absence of a market in higher activity radioactive waste disposal, provides a proxy commercial service and given that the operators pay a risk fee for the cap that is equivalent to an insurance premium. The UK had therefore notified the measures for the purpose of legal certainty. The Commission did not accept this argument. It noted that the State was in effect offering three different services under the Contracts. First, a disposal service (Service A) as such, the price to be paid for the disposal of the waste and spent fuel would be determined some 25 to 30 years after a new power plant starts operation, so that the

15

See also ch 5. See, eg, Paks II (Case SA.38454) [2016] OJ C8/2 and Hinkley Point (Case SA.34947) Commission Decision 2015/658 [2015] OJ L109/44, discussed below. 17 Hinkley Point (n 16). 18 Waste Contract for New Nuclear Power Stations (Case SA.34962) [2016] OJ C161/1. 16

Aid to Nuclear and Coal 207 exact costs of the disposal services are, to a certain extent, unknown. That uncertainty is compensated by a risk premium included in the Waste Transfer Price. This latter price is in turn capped at the outset and the risk-hedging service is remunerated by a risk fee (Service B). Finally, the Contracts provide for the possibility of a so-called Early Transfer (Service C). This means that liability for the waste is transferred at a point in time when that waste is kept in interim storage. The price for this service is not capped as interim storage costs that are under the control of operators are known to them. With respect to Service A, the Commission concluded that the Waste Transfer Price does not confer any advantage on operators of new nuclear power plants as it will cover all variable costs and a portion of the fixed costs that is in proportion to the use of the storage capacity (GDF). The calculation of the costs was such that the Waste Transfer Price will include adequate remuneration for the UK for taking on the costs being higher than expected and therefore it would not involve State aid.19 As to Service C, the Commission concluded that the Early Transfer Payment would not include any advantages to operators of new nuclear plants given that the operator will have to make a lump sum payment that must entirely cover all costs linked to the interim storage services. A risk premium would be paid to fully compensate any risk of cost escalation. However, the Commission considered that Service B would confer an advantage on investors as the proposed methodology would mean that the UK government would offer an insurance against all costs realisations exceeding the cap (and the maximum costs) whereas the insurance fee covers only cost realisations between the cap and the maximum cost realisation. The Commission concluded that it could not exclude that cost realisations would exceed the maximum costs calculated under the model.20 In the case of Belgian support to the long-term operation of three nuclear power plants,21 the Commission concluded that investment guarantees in two agreements with Engie-Electrabel and EDF Belgium conferred an advantage on these companies. The agreements envisaged prolonging the operational lifetime of the nuclear reactors Doel 1 and Doel 2 (owned by Engie-Electrabel) and Tihange (owned by Engie-Electrabel together with EDF Belgium). Under the agreements, the companies have committed to invest around €1.3 billion in exchange for authorisation to run the plants for another 10 years. The companies would receive financial compensation, if Belgium decides to close the reactors before the end of the 10-year period, modifies the level of nuclear tax to be paid by the owners or changes other economic parameters of the agreements. The Commission found that the investment guarantees provide an economic advantage to Engie-Electrabel and EDF, which goes beyond what they would have been entitled to under general Belgian law, and thus entail State aid elements.

19

ibid, paras 109–11. ibid, paras 117–21. 21 Commission, ‘Commission clears Belgian support to long-term operation of three nuclear power reactors Tihange 1, Doel 1 and Doel 2’ (SA.39487, Press Release of 17 March 2017) IP/17/662. 20

208 Leigh Hancher and Max Klasse b. State Resources and Imputability Generally, the Commission has had no difficulty in establishing imputability to the State, as the majority of measures it has assessed in the nuclear sector have been directly attributable to the State.22 In British Energy,23 the Commission examined measures to cover the reprocessing, storage and ultimate disposal of spent fuel and for the decommissioning of nuclear power stations which would be assumed by the UK government, as well as measures concerning the renegotiation of nuclear fuel supply agreements concluded by British Energy (BE) with BNFL, a publicly owned company. BNFL would reduce its prices to BE by some £1000 million. The Commission took the view that a normal private creditor would not have acted in this manner and therefore the measure had to be imputed to the State. In Case C-5/14, the CJEU dealt with the question of whether Directive 2003/96 precluded national legislation, which levies a duty on the use of nuclear fuel for the commercial production of electricity. The Court held that Article 14 of Directive 2003/96 sets out an exhaustive list of the exemptions which the Member States must apply in connection with the taxation of energy products and electricity and inasmuch as it imposes on Member States the obligation not to impose taxation under the Directive on ‘energy products and electricity used to produce electricity and electricity used to maintain the ability to produce electricity’. In that regard the Court recalled that Article 2(1) of Directive 2003/96 defines ‘energy products’ for the purposes of that Directive by drawing up an exhaustive list of the products covered by that definition by reference to the codes of the combined nomenclature. As it does not appear on that list, the nuclear fuel that was the subject of the contested German legislation does not constitute an ‘energy product’ for the purposes of Directive 2003/96 and it is not, therefore, covered by the exemption laid down in Article 14(1)(a) of that Directive.24 In Hinkley Point, the UK government notified the Commission of its plans to establish a price support mechanism (a contract for difference or CfD) to ensure that the operator of the new Hinkley Point nuclear plant (EDF) will receive a stable revenue for a period of 35 years despite the volatility of the wholesale electricity price. Under this arrangement, when the market price at which the electricity is sold is lower than the determined ‘strike price’, the government will pay the difference between the strike price and the market price, but if the market price is higher than the strike price, the operator will be obliged to pay the difference to the government. The Commission reasoned that ‘the CfD will be funded through a levy on suppliers

22 See however its decision in relation to the use of international decomissioning funds allocated to Lithuania. The IIDSF had been set up in order to pool contributions from international donors to provide support to Lithuania for the decommissioning of the Ignalina nuclear power plant and to set up the new power generation capacities—Construction of a 400MW Combined Cycle Gas Turbine (CCGT) Plant at AB Lietuvos Elektrine (Case N764/2007) [2008] OJ C170/1. Support from the fund for new generation projects could not be attributed to Lithuanian authorities. 23 British Energy (Case C52/2003) Commission Decision 2005/407/EC [2005] OJ L142/26. 24 Kernkraftwerke Lippe-Ems (n 13) paras 45–48.

Aid to Nuclear and Coal 209 and under such circumstances it must be concluded that any advantages paid under the CfD are imputable to the State and are also financed from resources under the control of the State’.25 In the UK Waste decision, the Commission reasoned that as the UK had renounced the right to factor in the risk that costs might be higher than the modelled maximum in the risk fee, it had forgone resources.26 In Slovakia—decommissioning27 concerning the decommissioning of two already shut down nuclear plants (A1 and V1), the Commission considered that through the transfer of V1 and A1 from Slovenské elektrárne (SE) to a State-owned company, SE has been relieved from its obligations in respect of decommissioning and the costs relating to it. Therefore, by agreeing to this transfer in the privatisation contract signed between the Slovak government and ENEL, now the majority shareholder of SE, Slovakia has mitigated the charges which are normally included in the budget of an undertaking operating a nuclear power plant in Slovakia, and has therefore granted an advantage to SE.28 c. The Application of the MEO in the Nuclear Sector In a number of decisions, the Commission has considered the application of the Market Economy Operator (MEO) principle to the proposed measures. Despite the magnitude of the risks involved and the extensive time periods to which governments may have to commit, the Commission has confirmed that the State has acted as a normal private investor in a number of its decisions, albeit that it has not done so in every case. In TVO the Commission approved a guarantee provided by the French government to the Finish electricity producer TVO to cover a loan to purchase part of its nuclear plant from the French company AREVA NP. The Commission had opened a formal investigation following two complaints to verify whether the guarantee had been given on market terms. The Commission found that TVO would have been able to finance the entire operation without State intervention as TVO had a good credit rating and was not in difficulty. TVO had also already raised substantial capital on the international markets without the guarantee and finally, the fee paid was not below the costs of the loans concluded at the same time without State guarantees. Finally, the measure did not confer an advantage on AREVA NP as it had been selected by TVO before the guarantee was granted.29 In Hinkley Point, as described above, the UK authorities notified several measures to support the construction of a new nuclear reactor at Hinkley Point to the

25

Hinkley Point (n 16) para 329. Waste Contract for New Nuclear Power Stations (n 18) para 123. Partial financing of decommissioning of two already shut down nuclear plants (A1 and V1) (Case SA 31860) [2013] OJ 2013 C162/3. 28 ibid, para 73. 29 Commission, ‘Commission concludes that French State guarantee for Finnish nuclear power plant operator TVO does not constitute aid’ (Press Release of 26 September 2007) IP/07/1400. Garantie Coface dans le cadre de la construction par Framatome ANP d’une centrale nucléaire pour Teollisuuden Voyma Oy (Case C45/2006) Commission Decision 2008/281/EC [2008] OJ L89/15. 26 27

210 Leigh Hancher and Max Klasse Commission, but not the Credit Guarantee.30 The UK authorities argued that the Credit Guarantee would not confer an advantage on an undertaking funded through State resources since it will be offered on commercial terms in accordance with the MEO principle. The UK Government considered the Credit Guarantee and the terms of the CfD serve different purposes. The purpose of the CfD would be to provide a long-term contractual arrangement to reduce uncertainty in wholesale market prices subject to the performance of the underlying asset. The Credit Guarantee, as with commercial Credit Guarantees from financial insurers, would facilitate wider access to the long-term debt capital markets. The pricing and approval of the Credit Guarantee critically depends on the risk within the whole underlying project including the terms of the CfD. However, the reverse would not be true: the presence of a guarantee reallocates the risk profile between debt investors and the guarantor rather than altering the project risk profile. The Commission, on the other hand, found that the Credit Guarantee is the backbone of the financing of the project which has an unparalleled value. The existence of the Credit Guarantee is also essential for the project to attract outside credit. There are no examples of similar guarantees for similar projects on the market as none are being provided. Given the unprecedented nature of the project, of the financing and of the Guarantee for which there are no precisely comparable benchmarks, even if it were to consider that the remuneration minimises the support, the Commission considers that the price paid by the project company for the Credit Guarantee cannot be considered a market price, since the market does not and would not provide a similar facility to hedge the risk.31 Paks II: in May 2015, the Hungarian government notified the Commission of certain measures which it intends to implement to realise the construction of the Paks II nuclear plant.32 The Hungarian authorities claimed that the measures are market conform and are therefore not State aid. In its final decision, the Commission found that the Hungarian State will accept a lower return on its investment than a private investor would do. Hungary’s financial support for the construction of Paks II would therefore not be in line with the MEO principle and entail State aid.33 Areva Restructuring Aid34 concerned the restructuring plan that aims to restore the State-controlled Areva group’s long-term viability, refocusing the group on fuel cycle activities, and isolating Areva’s most ‘toxic’ liabilities from the sounder parts of the group. As part of the package, the electricity company EDF would take a majority stake in Areva NP. To enable the sale, the lossmaking OL3 project

30

Hinkley Point (n 16) para 183 et seq. ibid, para 338; see N Robins and T Chakma, ‘State Aid in Energy under the Spotlight: The Implications of the Hinkley Point Decision’ (2016) 15(2) European State Aid Law Quarterly 247, 253. 32 Paks II (n 16). At the same time, the Commission launched an infringement procedure against Hungary concerning the compatibility of the measures with EU public procurement rules (Directives 2004/17/EC and 2004/18/EC). The Commission argued that the Hungarian government had directly awarded the construction without a transparent procedure. 33 Commission, ‘Commission clears investment in construction of Paks II nuclear power plant in Hungary’ (Press Release of 6 March 2017) IP/17/464. 34 Restructuring Aid to Areva (Case SA.44727) [2016] OJ C301/2. Applicaton for annulment filed on 13 September 2017 in Case T-620/17 TVO v Commission. 31

Aid to Nuclear and Coal 211 (the Finnish project examined in the earlier TVO decision) would be carved out from other businesses.35 This sale remains conditional on the successful completion of safety tests for another new reactor project in France. The Commission considered whether this transaction would be in conformity with the so-called ‘private seller’ test36—a recognised variant of the MEO. Normally a private seller would establish the value of the assets to be sold by way of open tender or on the basis of an independent evaluation based on established principles.37 The French authorities had provided the Commission with sufficient evidence that an independent valuation had been carried out for Areva NP’s five distinct business units and that the transaction would not therefore amount to aid to either the seller or the buyer in this case.38 d. SGEI Compensation: The ‘Altmark’ Conditions39 In Hinkley Point, the UK government argued that the notified measure—the CfD— was compensation for the provision of a service of general economic interest (SGEI) and would therefore be exempted from State aid scrutiny because the ‘Altmark’ criteria had been fulfilled. In particular, the UK authorities argued that by contributing significantly towards the UK’s security of supply of low-carbon electricity on a long-term basis, the investment in new nuclear generation capacity to be delivered and operated within a specific timeframe and its operation within the framework of the CfD is directed towards achieving a general or public interest that is capable of being designated an SGEI.40

The Commission did not accept the UK’s arguments. It held that although the UK authorities mentioned that the aim of the measure is to incentivise or unlock investments into low-carbon generation, in particular new nuclear, this policy aim is commensurate with a common interest objective for which State aid can be granted rather than with the entrustment of an SGEI. Furthermore, the UK authorities could not enforce any specific obligations other than through termination of the contract. Given that the Commission did not consider that the generator had been entrusted with an SGEI it did not have to investigate the remaining three conditions in any detail. For the same reason, the UK could not invoke Article 106(2) TFEU as a justification.41 e. Selectivity In order to fall within the scope of Article 107(1) TFEU, the State measure in question must have the effect of conferring a selective benefit or advantage. General

35 The sale was concluded in November 2016. See, ‘France’s EDF throws Areva a lifeline with reactor deal’ (Reuters, 16 November 2016). 36 See Restructuring Aid to Areva (n 34) paras 116–17. 37 See further, Commission Notice on the notion of State aid as referred to in Article 107(1) of the Treaty on the Functioning of the European Union (Commission Notice on the Notion of Aid) [2016] OJ C262/1, para 97. 38 See Restructuring Aid to Areva (n 34) paras 118–30. 39 For a full discussion of these conditions see ch 10. 40 Hinkley Point (n 16) para 107. 41 ibid, paras 308–16.

212 Leigh Hancher and Max Klasse measures which can benefit the entire economy, such as a lowering of tax rates or interest rates do not create a selective benefit. In order to determine whether a measure is selective, the applicable system of reference must be defined. In Case C-5/14 Kernkraftwerke Lippe-Emst the ECJ ruled that an excise duty imposed on nuclear fuel as opposed to other fuels was not selective as only nuclear fuel generates radioactive waste.42 The Commission in its various decisions on the nuclear sector appears to take different positions on the issue of selectivity. In the German decommissioning decision of 2001 on the application of the EU State aid rules to German tax treatment of reserves for the decommissioning of nuclear power plants and the safe disposal of nuclear waste, the Commission held that the rules in question did not constitute State aid. The Commission found that the tax treatment in question was equally applicable to all undertakings that have to constitute similar reserves. It did not therefore confer a ‘selective benefit’ on a particular sector and hence could not be caught by Article 107(1) TFEU.43 It was noted in conclusion that ‘the Commission’s decision does not affect the application of the Euratom Treaty and is only valid to the extent that the EC Treaty is applicable’. The Commission’s decision was upheld on appeal: the General Court confirmed that the measures in question were general taxation rules, and therefore not aid within the meaning of Article 107(1) TFEU.44 In the UK Waste decision, and in so far as it concerns the so-called risk premium measures (Service B), the Commission found that the measure is selective as it will only be offered to operators of new nuclear power stations for disposal of their waste and spent fuels. The Commission found that similar provisions for disposal services will not be offered to other producers of electricity.45 In Hinkley Point,46 the Commission also considered a package of measures. First, the Commission held that the CfD protected the beneficiary from any price volatility in the electricity market as it receives always the predefined SP when selling at prices that are below this level. This ensures the beneficiary a steady stream of revenues that other operators not benefiting from a CfD do not receive. Therefore, the CfD entailed a selective advantage to the generator of nuclear power. Second, the UK intends to grant compensation in case the HPC plant were to be shut down for reasons not directly imputable to its operations, and in particular due to changes in government policy (the so-called Secretary of State Agreement). The Commission acknowledged that it could be argued that nuclear energy might incur higher risks of political shutdown than other technologies; however, other nuclear power plants in the UK appear not to benefit from similar indemnification agreements. Although the general principles underpinning UK and EU law give rise

42

Kernkraftwerke Lippe-Ems (n 13). Reserves for nuclear power station waste management and decommissioning (Case NN137/2001) [2002] OJ C77/25. 44 Case T-92/02 Stadtwerke Schwäbisch Hall ao v Commission, EU:T:2006:26. Upheld on appeal because the applicant lacked standing; Case C-176/06P Stadtwerke Schwäbisch Hall ao v Commission, EU:C:2007:730. 45 Waste Contract for New Nuclear Power Stations (n 18) para 122. 46 Hinkley Point (n 16). 43

Aid to Nuclear and Coal 213 to a right to compensation where there has been deprivation of a property right, however, a special agreement safeguarding a certain company from such risk in a specific manner appears to relieve such company of any spent fees and time lost in the enforcement of its rights deriving from general principles under UK and EU law in court or out of court. The Commission found: ‘Underpinning a legal right with a specific contractual right appears to bring an advantage to the entity enjoying such right especially since it appears to be the only one in this situation’. Therefore, the Commission considered that the Secretary of State Agreement entails certain selective advantages to the nuclear plant operator.47 Third, on the Credit Guarantee, the Commission rejected the UK’s submission that this measure was market conform (see above) and concluded that it was in fact a selective economic advantage. ii. Article 107(3) TFEU: Compatibility Assessment a. Decisions up to the Adoption of the SAM and the Common Principles of Assessment Prior to the adoption of the SAM exercise, the Commission did not always adopt a systematic or very detailed assessment of the compatibility under Article 107(3)(c) TFEU of the support measures notified to it, unless the measures in question fell under a specific set of guidelines, such as the rescue and restructuring guidelines. Following the introduction of the more refined economic approach as a result of the State Aid Action Plan (SAAP) of 2005,48 the Commission did however apply a ‘balancing test’ in cases such as the Slovakian decommissioning decision discussed below. UK Privatisation: In 1989, the UK authorities notified a package of State aid measures destined for the nuclear industry in England and Wales and in Scotland just prior to privatisation of the sector in 1989. The Commission authorised: (1) a guarantee of up to £2.5 billion to cover present and future liabilities in the nuclear sector, including the decommissioning of existing power stations at the end of their lives; (2) use of the proceeds of a levy (the non-fossil fuel levy) on consumers to bring down the higher costs of nuclear produced power and to promote renewable energy sources; and (3) the debt write-off provision of up to £1.4 billion for Scottish nuclear production. The UK authorities undertook to provide to the Commission a detailed plan outlining the future of the aided nuclear sector. The debt write-down was cleared on the basis that it was necessary to ensure that the new company was viable and did not run the risk of going bankrupt.49 In relation to the non-fossil fuel levy, it is noteworthy that the Commission approved the measure which was expected to average

47

ibid, paras 317–23. Commission, ‘State Aid Action Plan’ COM(2005) 107 final; see ch 5. Further details of the approved aid are only scantly described in the press notice issued on the decision: see Commission, ‘Commission authorises aid for UK nuclear electricity sector’ (Press Release) IP/90/267; Commission Decision in State aid Case N34/1990, SG(90)D/2049. 48 49

214 Leigh Hancher and Max Klasse £1.150 million per year (€1.55 billion) over eight years, and would be used, inter alia, to cover the difference between the price negotiated in the contracts between distribution companies and nuclear suppliers and the market price of electricity from other suppliers. The levy was to be phased out in 1998. The Commission reasoned that the plants had to be kept in operation for a certain time in order to earn sufficient profits to meet liability for decommissioning. With regard to the provision of up to £2.5 billion, the Commission reasoned that this would be used for storage and reprocessing of nuclear fuel, treatment or storage for disposal of waste and decommissioning of nuclear installations. It places the costs accruing from the past directly on the government rather than the consumer—a cost which would have to have been borne by the government in any event if the present system of supply (ie, public sector) had been continued. In its decision C31/2002—Transitional regime for the Belgian electricity marke— the Commission concluded that the proposed compensation of €377 million to Electrabel and SPE to meet the costs of dismantling experimental nuclear sites would be compatible with Article 107(3)(c) TFEU, but provided no further reasons for this finding.50 British Energy: An attempt to resolve the interplay between the Euratom Treaty and the EU Treaty State aid rules is more clearly evident in the Commission’s decision C52/03 to open procedures against the restructuring aid granted by the UK government to British Energy (BE) in 2003.51 This important case concerned a package of seven measures, not all of which were deemed to be aid but for those falling within the scope of Article 107(1) TFEU, the Commission observed that several of the measures in question concerned issues covered by the Euratom Treaty and therefore have to be assessed accordingly. However, to the extent that they are not necessary for or go beyond the objectives of the Euratom Treaty or distort or threaten to distort competition in the internal market, they have to be assessed under the EC Treaty.52

The UK government had contended that all the measures should be assessed under the Euratom Treaty, as the aims of the measures were to preserve the safety of nuclear power stations, to ensure the safe management of nuclear liabilities and to enhance security of supply by maintaining diversity of fuel sources in Great Britain as well as to avoid carbon dioxide emissions. In this final decision,53 the Commission approved the restructuring plan for BE on the basis of its then applicable guidelines for rescue and restructuring aid, subject to important conditions. BE agreed to cap its production capacity and not to extend its fossil fuel activities outside the United Kingdom and would also refrain from acquiring hydro capacity from its UK competitors. The funds received under the approved plan would be ring-fenced and BE would create three separate businesses, each with

50

Transitional regime for the Belgian electricity market (Case C31/2002) [2002] OJ C222/2. Restructuring aid granted by the UK government to British Energy (British Energy) (Case NN45/2003) [2003] OJ C180/5. 52 ibid, Pt III, 6. 53 British Energy (n 23) and Commission, ‘Commission approves restructuring of British Energy press release’ (Press Release of 22 September 2004) IP/04/1125. 51

Aid to Nuclear and Coal 215 separate accounts. Furthermore, the Commission imposed the condition that BE would not be allowed to undercut prices charged by its non-aided competitors on the market for supply to large users was to be monitored by an independent trustee. NDA: Subsequently, in December 2004 the Commission launched a formal investigation into the UK government’s plans to transfer BNFL’s nuclear assets to a Statebacked Nuclear Decommissioning Authority (NDA).54 In its final decision,55 the Commission authorised the transfer of nuclear assets from BNFL to the NDA as compatible aid on the basis of Article 107(3)(c) TFEU but subject to conditions. The Commission assessed the measures in view of the objectives of the Euratom Treaty and came to the conclusion that the measures were fully in line with the objectives of the Euratom Treaty. In relation to the NDA, the Commission observed that the measure would provide an unlimited guarantee to the NDA as the State would cover all the NDA’s expenses to the extent that these could not be covered by the NDA’s revenues from commercial activities or by financial assets transferred to it. The Commission authorised the measure on the grounds that the aid in question was in line with the objectives of the Euratom Treaty and did not affect competition to an extent contrary to the common interest, which meant that both the necessity and proportionality test for a measure to be deemed compatible with the internal market under Article 107(3)(c) TFEU were fulfilled. The Commission noted that the NDA would continue to operate some of the assets and thus would not comply with the ‘polluter pays’ principle. However, because the NDA would operate the power plants for a residual time, the Commission gave an authorisation to the measure imposing similar conditions on the NDA to those imposed on BE. The Commission also assessed whether BNFL was also a recipient of aid. The Commission found that since BNFL complied with the ‘polluter pays’ principle, the measure did not include aid to BNFL. In these decisions the Commission took account of the objectives of the Euratom Treaty when assessing the necessity of a measure found to constitute State aid. Where the impact on competition and trading conditions was likely to be negative, the Commission authorised the measure, imposing a number of conditions to ensure that the negative impact is minimised. TVO/Finland: In the course of 2005, and following two complaints, the Commission considered the legality of various export credit arrangements granted by the French government via the French agency for export credit to the Finnish electricity producer TVO for the purchase of equipment from Areva/Framatome. A formal procedure was opened to investigate, inter alia, if the total costs of the loans conformed to market practice. The OECD Arrangement56 was considered to offer some guidance but was not conclusive in this respect—compliance with the Arrangement is a necessary but not a sufficient condition under the EU State aid regime. As the

54 Commission, ‘Commission opens formal investigation into UK Nuclear Decommissioning Authority’ (Press Release of 1 December 2004) IP/04/1430. 55 Nuclear Decommissioning Authority (C39/2004) Commission Decision 2006/643/EC [2006] OJ L268/37. 56 The provision of long-term risk credit insurance for exports is covered by the OECD Arrangement.

216 Leigh Hancher and Max Klasse Commission cleared the transaction as market conform, it did not examine further the interaction between the OECD framework and the EU State aid rules.57 In Slovakia—decommissioning,58 the Commission considered the measures—ie, the transfer of the plant to a State-owned company and the subsequent financing of the safe decommissioning of obselete facilities and for the storage of spent fuel and waste by a State fund (NFF) to be an object of common interest. Although in principle the polluter should pay, due attention had to be given to cases where special solutions were most appropriate given the historical reasons. As one plant would have remained open but for commitments made in accession negotiations, the State had assumed the financial burden for early shut down. Distortion of competition was unlikely given that the plants were already closed and the aid was considered to be strictly limited to what was necessary to cover historic costs, and would not have a protective effect on national electricity production. b. Decisions After 2014 As noted above, the current EEAG59 does not apply to nuclear energy. However, following the SAM exercise, the Commission now applies the so-called common principles of assessment to aid measures in support of nuclear energy. These are: (a) the measure must make a contribution to a well-defined objective of common interest; (b) the need for State intervention; (c) the appropriateness of the aid measure; (d) the incentive effect; (e) the proportionality of the aid (aid kept to the minimum); (f) the avoidance of undue negative effects on competition and trade between Member States; and (g) the transparency of aid. In the case of Belgian support to the long-term operation of three nuclear power plants,60 the Commission cleared the investment guarantees on the basis that they would avoid undue distortions of the Belgian energy market. Among other things, the Commission took note of the obligation that Engie-Electrabel as the major player on Belgian electricity markets to sell on regulated electricity markets each year a volume equivalent to Engie-Electrabel’s share of the annual production of the power plants in question. This would ensure liquidity on Belgian electricity markets and help increase competition between electricity suppliers. In Hinkley Point,61 the Commission adopted a detailed opening decision in 2013, but in its final decision of October 2014 found that the UK authorities had demonstrated that the planned support would address a genuine market failure and that State aid was, therefore, justified. In particular, the promoters of the project would not be able to obtain the necessary financing due to its unprecedented nature and scale. With regard to the Euratom Treaty the Commission held that: As recognised in past Commission decisions, the Euratom Treaty aims at creating the ‘conditions necessary for the development of a powerful nuclear industry which will provide

57 Garantie Coface dans le cadre de la construction par Framatome ANP d’une centrale nucléaire pour Teollisuuden Voyma Oy (n 29). 58 Partial financing of decommissioning of two already shut down nuclear plants (n 27). 59 See ch 5 for further discussion. 60 Press Release IP/17/662 (n 21). 61 Hinkley Point (n 16).

Aid to Nuclear and Coal 217 extensive energy sources’. This objective is further reiterated in art 1 of the Euratom Treaty, which establishes that ‘it shall be the task of the Community to contribute to the raising of the standard of living in the Member States … by creating the conditions necessary for the speedy establishment and growth of nuclear industries’. On this basis, the Euratom Treaty establishes the Euratom Community, foreseeing the necessary instruments and attribution of responsibilities to achieve these objectives. The Commission must ensure that the provisions of this Treaty are applied … Art 2(c) of the Euratom Treaty provides that Member States shall ‘facilitate investment and ensure, particularly by encouraging ventures on the part of undertakings, the establishment of the basic installations necessary for the development of nuclear energy in the Community’.

It should also be noted that in the course of the investigation, the UK authorities agreed to modify the terms of the proposed State financing to ensure that the State aid provided is proportionate and that distortions of competition are minimised. In relation to the State guarantee, the Commission found that the initial guarantee fee which the operator would have paid to the UK Treasury was too low for a project with this risk profile. The guarantee fee was therefore raised. This increase will reduce the subsidy by more than £1 billion (€1.3 billion) and procure the UK Treasury an equivalent gain. The Commission concluded that this would reduce any distortions of competition created by the aid and would also benefit UK taxpayers. In relation to the CfD, as a result of modifications, the gains generated by the project would be better shared with UK consumers: as soon as the operator’s overall profits (return on equity) exceed the rate estimated at the time of the decision, any gain will be shared with the public entity granting the public support; and there is a new second, higher threshold above which the public entity will obtain more than half of the gains.62 These gains will be shared with UK consumers by a decrease in the price paid by the public entity to the operator (the strike price). This gain-share mechanism will be in place not only for the 35-year support duration as initially envisaged but, at the request of the Commission, for the entire 60-year lifetime of the project. In addition to the equity gain-sharing mechanism, an operating cost gainsharing mechanism and a construction cost gain-sharing mechanism (in the event that construction costs were lower than forecast) were introduced to mitigate the risk of overcompensation over the duration of the project. The then Vice President Almunia, in announcing this decision, emphasised that the Hinkley decision will not create any kind of precedents. Possible new cases regarding nuclear energy investments would be assessed on their own merits. In the meantime, in Case T-356/15 Austria v Commission, the Austrian government has claimed, inter alia, that the promotion of nuclear is not an objective of common interest.63 This claim was also supported in a separate appeal lodged by Greenpeace International and a number of German and Austrian renewable energy producers.64 The General

62 63 64

See also Robins and Chakma (n 31) 254. Case T-356/15 Austria v Commission, application on 6 July 2015. Case T-382/15 Greenpeace Energy and Others v Commission, EU:T:2016:589.

218 Leigh Hancher and Max Klasse Court has however held this latter action for annulment to be inadmissible. Neither Greenpeace nor the energy producers had established that they were individually and directly concerned by the decision. In Paks II, the Commission eventually cleared the measures in question under Article 107(3) TFEU.65 It found that Hungary had demonstrated that the financial support avoids undue distortions of the Hungarian energy market by making the following commitments: (1) Any potential profits earned by Paks II will either be used to pay back the Hungarian State for its investment or to cover normal costs for the operation of Paks II; (2) Paks II will be functionally and legally separated from the operator of the power plant to avoid market concentration and (3) Paks II will sell at least 30 per cent of its total electricity output on the open power exchange. In UK Waste, the Commission came to the conclusion that the measures were in line with Article 107(3) TFEU: as the Waste Transfer Contracts aimed at ensuring the safe management of spent fuel and waste and enable prospective operators to make investment decisions to build new nuclear plants, they were considered to further both the objectives of nuclear safety and the development of nuclear energy in line with Article 2 EA and are therefore also objectives of common interest within the meaning of Article 107(3)(c) TFEU. Also, the contracts met a well-defined market failure—the absence of suitable hedging services in the market—and the Cap and Risk Fee were appropriate instruments to address the UK’s objectives. They would have an incentive effect for new investment. As the amount of State aid would in fact be unknown until suitable disposal sites had been constructed, the assessment of the proportionality of the aid amount was based on complex modelling by external consultants. The scheme did not relieve operators from their costs—in line with the polluter pays principle—but is limited to a free insurance if those costs turn out to be higher than the modelled maximum. The chances that this will occur were considered to be low as the CAP and Risk Fee incentivise the State to safely and efficiently carry forward the disposal project so as to keep costs below the Cap and the Risk Fee. In terms of waste disposal services, distortion of competition would be very limited as it was unlikely that other operators might offer such services. c. Rescue and Restructuring Aid In principle, this form of aid is typically granted in a situation of acute financial distress where State funding is often the last resort to keep the beneficiary from going into administration. At the same time, rescue and restructuring aid is considered one of the most distortive forms of State aid as it covers the ongoing operating expenses of a company and as such it will always have a severe impact on competitors and intra-Community trade. The UK government’s rescue of the bankrupt British Energy is a case in point. In the British Energy case, the United Kingdom put the rescue package into place in September 2002 and the Commission took a decision based on the then applicable

65

Press Release IP/17/464 (n 33).

Aid to Nuclear and Coal 219 1999 Guidelines not to raise objections on the case on 27 November 2003.66 The Commission approved rescue aid in the form of loan and deposits by the UK government (which served as collateral for British Energy’s trading and non-trading (regulatory) counterparties). Restructuring plans require complex financial and business plans which, if prepared by the beneficiary, should also be reviewed by the Member State with the assistance of external experts. In British Energy, the Commission retained its own experts to review the restructuring plan. The Commission took into account that the measures also pursued the objectives of the Euratom Treaty, ie, to preserve the safety of nuclear power stations, ensure the safe management of nuclear liabilities and enhance security of supply.67 In the Areva case, the Commission eventually cleared plans by the French State to grant a capital injection of €4.5 billion to Areva under the 2014 Guidelines on State aid for restructuring of non-financial undertakings in difficulty.68 The restructuring plan to restore Areva’s competitiveness provides for various divestments including of the group’s nuclear reactor business to EDF. The French government claimed that the proposed restructuring meets an objective of common interest, particularly as the eventual bankruptcy of Areva would have grave consequences for nuclear security and safety throughout the EU and would be manifestly contrary to the objectives set forth in Articles 1, 2 and 52 EA.69 According to the Commission, Areva’s withdrawal from the nuclear reactor business will allow the group to focus on a clear and profitable business in the nuclear fuel cycle. The complete divestiture of Areva’s reactor business would reduce the group’s activities in the nuclear sector and thereby limit the distortions of competition brought about by the State support. Furthermore, the Commission stressed that a competitive Areva will also contribute to ensuring Europe’s security of uranium supply. The Commission also found that Areva will finance a significant part of the restructuring costs with proceeds from planned asset sales, including the divestiture of Areva’s reactor business. d. Stranded Costs and Nuclear Power Plants Article 24 of Directive 96/92/EC provided that Member States could apply to the Commission for a derogation from certain provisions of the Directive if this was necessary in the light of long-term commitments or guarantees that could no longer be honoured on account of the entry into force of Directive 96/92/EC. The Directive provided a strict timetable for the notification of the request for derogation.70

66

British Energy (n 23). ibid, para 326. 68 Commission, ‘Commission approves restructuring plan of French Areva group’ (Press Release of 10 January 2017) IP/17/36. 69 Restructuring Aid to Areva (n 34) paras 50–56. The Commission endorses this position (see paras 140–42). 70 Directive of the European Parliament and of the Council 96/92/EC concerning common rules for the internal market in electricity [1997] OJ L27/20. See also Case C–17/03 VEMW, EU:C:2005:362, in which the Court held that Art 24 was the only possible derogation procedure for long-term contracts reserving priority access to certain companies. 67

220 Leigh Hancher and Max Klasse In order to qualify as stranded costs, commitments or guarantees must become uneconomical because of the effect of Directive 96/92/EC on energy prices, and must significantly affect the competitiveness of the undertaking.71 The Stranded Cost Methodology is now primarily of interest for historic reasons. The basic principle of the methodology is that compensation for stranded costs should be limited in time and extent. They should not exceed the costs actually borne by the undertakings, directly caused by the liberalisation of the market, and resulting in losses. Hence, no compensation could be permitted if it only contributed towards falling profitability. The compensation must be fixed ex ante and should also include an ex post adaptation mechanism which would take into account the real evolution of the market, and in particular the actual evolution of electricity market prices. In the Belgian stranded costs case,72 the envisaged State aid contained three elements. First, compensation for the dismantling of nuclear plants throughout Belgium; second, financing the costs of a new pension regime in the electricity sector; and third, promoting renewable energy sources. The Commission found that the compensation of the costs related to the dismantling of nuclear plants, insofar as it would qualify as State aid, was in compliance with the methodology, and the Commission concluded, inter alia, that the size of these stranded costs would significantly weaken these companies’ competitive positions. The dismantling of the plants would not result in profits for the companies and the assets involved were not owned by these companies. In relation to Slovenia, the Commission authorised compensation for stranded costs for three Slovenian electricity generators: TE Šoštanj, NE Krško and TE Trbovlje.73 The measure aimed at mitigating the difficulties faced by a number of power plants during the process of liberalisation of the electricity sector in Slovenia. The Commission found the measure to be State aid, but concluded that the scheme was in line with the methodology. In particular, the Commission concluded that the State guarantee for the Krško nuclear plant was awarded before accession and thus qualified as existing aid. Moreover, the Commission found the aid for closing the gas fired part of the Trbovlje plant compatible with the criteria of the methodology. In relation to the Šoštanj plant, the compensation was authorised under Article 106(2) TFEU because although it did not meet the methodology criteria, it still qualified for exemption as compensation for a service of general economic interest as regards the security of electricity supply.74

71 ibid, para 3. It should be noted that Directive 1998/30 on the internal gas market did not contain an equivalent provision given that a separate procedure was provided in that Directive for dealing with long-term take-or-pay contracts entered into before the adoption of that Directive, see Directive of the European Parliament and of the Council 1998/30/EC concerning common rules for the internal market in natural gas [1998] OJ L204/1. 72 Commission, ‘Commission gives partial go-ahead to compensation for the Belgian electricity sector’ (Press Release of 24 April 2002) IP/02/605. 73 Slovenian Electricity Tariffs (Case C7/2005) Commission Decision 2007/580/EC [2007] OJ L219/9. 74 Commission, ‘Commission takes two decisions concerning Slovenian electricity sector’ (Press Release of 2 February 2005) IP/05/126; Commission, ‘Commission endorses support for green electricity and for security of electricity supply in Slovenia’ (Press Release of 25 April 2007) IP/07/559; Slovenian Electricity Tariffs (n 73).

Aid to Nuclear and Coal 221 iii. The Nuclear Sector and Article 106(2) TFEU The application of Article 106(2) TFEU to assess the compatibility of State support in the energy sector has arisen in a number of cases. In Hinkley Point, the UK government failed to convince the Commission that the generator had been entrusted with an SGEI and should be considered compatible with the Commission’s SGEI Framework,75 and therefore the Commission did not consider that Article 106(2) TFEU could be relied on as a legal basis for assessing the comptability of State aid to HPC.76

II. COAL

The European Coal and Steel Community (ECSC) Treaty expired on 23 July 2002 and, since that date, the general State aid rules provided for in the TFEU now apply to the coal sector, subject to certain transitional provisions provided for in secondary law.

A. Introduction: The EU’s Coal Sector—Facts and Figures Coal currently accounts for about one-quarter of all EU electricity production and it is also an important fuel for industrial processes like steel production. At the time of the adoption of the current State aid regime in 2010 (see below), the sector still employed around 100,000 people in Europe, of which 42,000 were in the coal sector itself and over 55,000 in related industries.77 The mines that rely on operating subsidies are located mostly, but not only, in the Ruhr region, in Germany, in north-west Spain and in the Jiu Valley in Romania.78 The largest coal deposits in Europe are in Poland, Germany, the United Kingdom and the Czech Republic.79 In 2015, the largest coal producers in the EU were Poland (54 Mtoe)80 and Germany (43 Mtoe), followed by the Czech Republic (16 Mtoe), Greece and Bulgaria (both 6 Mtoe).81 Owing to unfavourable geological conditions, most EU mines are not competitive in comparison with imported coal, which led to significant State support to coal production. According to a 2014 report commissioned by the Commission, coal still received roughly €10 billion in subsidies across

75

See for the UK’s reasoning Hinkley Point (n 16) para 198 et seq. ibid, paras 315 and 382. 77 See, eg, R Bajczuk, ‘The Uncertain Future of the Coal Energy Industry in Germany’ (2015) 4 OSW Commentary 188. 78 See Commission, ‘Commission proposes Council Regulation on State aid to close uncompetitive coal mines’ (Press Release of 20 July 2010) IP/10/984. 79 German Federal Ministry of Economic Affairs and Energy, ‘Energiedaten 2016’, Coal Resources, chart 42. 80 Mtoe = million tonnes of oil equivalent. 81 BP, ‘BP Statistical Review of World Energy 2016’ (2016), available at www.bp.com/content/dam/ bp/pdf/energy-economics/statistical-review-2016/bp-statistical-review-of-world-energy-2016-full-report. pdf, 32. 76

222 Leigh Hancher and Max Klasse the 28 Member States in 2012, a record low, however, if compared with historic subsidies and previous years.82 The report estimates that over 70 per cent of the cumulative support in the years from 1970 to 2012 was provided in Germany.83 The share of imported coal in the total consumption of the European Union has doubled over the last decades and is currently at 46 per cent.84 Thus, for coal, the EU is dependent on imports from Russia, Colombia and the United States. Since 2011, the market price for coal has been declining significantly.85 However, as a cheaper and more readily available alternative to other fossil fuels, coal remains a key component in the energy mix of many EU countries. With regard to the largest consumers in the EU—Germany and Poland—the total consumption of coal in both Member States has been relatively stable for several years (in Germany around 80 Mtoe, equivalent to 25 per cent in the energy mix (including power from lignitefired plants) and in Poland around 50 Mtoe, which corresponds to 50 per cent in the energy mix).86 At the same time, the EU is working towards the reduction of carbon emissions through EU regulations and co-funding programmes. Part of this strategy includes the implementation of clean coal technologies such as carbon capture and storage (CCS) or carbon pricing (via the ETS scheme). Against these developments, and due to the low profitability of many conventional power plants, investment projects have been put on hold, and existing plants across Europe are being mothballed or decommissioned.87

B. State Aid before and after the Expiry of the ECSC Treaty i. The ECSC Treaty Article 4(c) of the (now defunct) ECSC Treaty placed an absolute prohibition on all subsidies or aid granted by Member States, or special charges in any form whatsoever.88 This prohibition applied irrespective of whether an inter-State trade effect could have been established. Unlike the TFEU regime, the ECSC Treaty did not provide for an exemption from the prohibition. In practice, however, this absolute ban proved unworkable and the Commission availed itself of Article 95 ECSC

82 Ecofys, Subsidies and Costs of EU Energy. Final report (2014). The report also shows historic subsidy levels and compares the overall costs of the different power generation technologies. 83 ibid, 31. The report had sparked strong criticism from environmental organisations as it also came to the conclusion that the costs of producing one MWh of electricity from coal are significantly lower than from any other power generation technology. 84 Commission, ‘EU Energy in Figures: Statistical Pocketbook 2016’ (2016) 24, available at www.ec.europa.eu/energy/sites/ener/files/documents/pocketbook_energy-2016_web-final_final.pdf. 85 German Federal Ministry of Economic Affairs and Energy, ‘Energiedaten 2016’, Energy Prices, chart 26; for worldwide data see BP Statistical Review (n 81). 86 BP Statistical Review (n 81) 33. 87 European Parliament, ‘European Energy Industry Investments’ (European Parliament Study, 2017) 71. 88 See, eg, W Obwexer, ‘Das Ende der Europäischen Gemeinschaft für Kohle und Stahl’ (2002) 17 Europäische Zeitschrift für Wirtschaftsrecht 517; T Lübbig, ‘Änderungen des europäischen Kartellrechts nach Auslaufen des EGKS-Vertrages’ (2002) 122 Stahl und Eisen 59.

Aid to Nuclear and Coal 223 Treaty to enact a series of special decisions for the Community coal sector. These specific sets of rules dealt with both direct and indirect forms of aid. The Commission first adopted rules for State aid to the coal industry in 1986. Further rules appeared at regular intervals, culminating in Decision 3623/93/ECSC89 implemented by Decision 341/94/ECSC which expired on 23 July 2002 on the expiry of the ECSC Treaty.90 Despite the different approach between the two Treaties with respect to State aid, the decisions establishing rules for aid to the coal industry served to minimise any real divergence between the two Treaty regimes. Decision 341/94/ ECSC, inter alia, allowed for operating aid, aid for the reduction of activity, for modernisation and rationalisation of coal mining activities. All aid was subject to a notification requirement. In practice, these rules provided the framework for billions of State aid to the coal sector, mainly with the aim of guaranteeing a basic supply of energy from indigenous primary energy sources in order to strengthen the EU’s security of energy supply. Outside the specific scope of the ECSC Treaty, the Commission in the past applied Articles 107 and 108 TFEU to all aspects of a national measure not directly governed by the ECSC rules given that the European courts had affirmed that the provisions of the ECSC Treaty must be interpreted as meaning that in so far as matters are not the subject of provisions in the ECSC Treaty or rules adopted on the basis thereof, the EEC Treaty and the provisions adopted for its implementation can apply to products covered by the ECSC Treaty.91

ii. EC Council Regulation 1407/02 Since the expiry of the ECSC Treaty in 2002, aid to the coal sector falls under the EC Treaty State aid rules and, until 2010, under the then applicable Council Regulation (EC) 1407/02 (often referred to as the ‘Coal Regulation’).92 This Regulation entered into force on 23 July 2002 and was due to expire on 31 December 2010. Regulation 1407/02 laid down rules specifying the maximum intensity of aid in relation to the costs incurred and the phasing out over time of aid to the coal sector in the countries still producing coal in the Community, while making allowance for social and regional factors. The purpose of these rules was to encourage the continued efforts to restructure and modernise European coal production with the aim of guaranteeing a basic supply of energy in the European Union. The regime was based on a minimum production of coal in order to maintain a proportion of indigenous primary energy sources with a view to strengthening the EU’s security of energy supply.93 Regulation 1407/02 still allowed for investment and operation aid where this would

89 Commission Decision 3632/93/ECSC of 28 December 1993 establishing Community rules for State aid to the coal industry [1993] OJ L329/12. 90 Commission Decision 341/94/ECSC of 8 February 1994 implementing Decision 3632/93/ECSC establishing Community rules for State aid to the coal industry [1994] OJ L49/1. 91 Case 328/85 Deutsche Babcock v Hauptzollamt Lübeck-Ost, EU:C:1987:548, para 10. 92 Council Regulation (EC) 1407/2002 of 23 July 2002 on State aid to the coal industry [2002] OJ L205/8. 93 Commission, XXXIInd Report on Competition Policy 2002, para 474.

224 Leigh Hancher and Max Klasse contribute to the objective of energy security from indigenous energy sources. The support must not lead to prices lower than those for imports from third countries, and it must not lead to a distortion of competition on other electricity markets in the EU. Regulation 1407/02 also required the submission of detailed plans and related information, depending on the category of aid, the details of which and relevant forms were laid down in a Commission decision.94 The Commission took a number of decisions on the basis of Regulation 1407/2002. For instance, the Commission approved aid to the coal sector in Poland,95 Hungary,96 Germany97 and Spain.98 These aid schemes were intended to support access to coal reserves and to restructure the coal sector in these countries. On the basis of the Regulation, the Commission also approved an aid scheme aimed at supporting the implementation of environmental rehabilitation projects in coal mines in Bulgaria that have ceased operating.99 The projects were targeted at cleaning up the previous coal mining sites and recultivating land for agricultural and forestry purposes. When assessing aid granted or envisaged, the Commission examined the purpose of the aid and the criteria to be fulfilled by the grantee in order to be able to apply for aid. Moreover, the Commission verified that the conditions laid down in previous Commission decisions, which have been taken on the basis of the ECSC Treaty, were respected. According to the Commission, the fact that the ECSC Treaty expired and Regulation 1407/2002 entered into force did not prejudice engagements of the past. In 2009/2010, in view of the forthcoming expiry of Regulation 1407/2002, the Commission approved a number of aid schemes and individual measures in relation to the coal sector including in Germany, Slovakia and Spain.100 A controversial decision adopted under the Regulation and subsequently appealed to the General Court concerned a Spanish measure to incentivise the production of electricity from domestic coal, which required Spanish electricity companies to purchase Spanish coal and to produce a fixed amount of electricity from this indigenous coal. In return for their compliance, the additional costs of implementation would be paid by the State. The power stations involved would benefit from preferential dispatch. Following a lengthy prenotification procedure, the Commission authorised the measure on 29 September 2010 until 31 December 2014. The Commission found that the obligations imposed on the electricity companies could be classified as

94 Commission Decision 2002/871/EC of 17 October 2002 establishing a joint framework for the communication of information needed for the application of Council Regulation (EC) No 1407/2002 on State aid to the coal industry [2002] OJ L300/42. 95 Secteur du charbon 2008–2015 (Case N 575/2007) [2008] OJ C137/2. 96 Aid for coal industry (2007–2010) (Case NN 3/2008) [2008] OJ C323/3. 97 Aides au charbon pour l’année 2008 (Case N 631/2007) [2009] OJ C42/7. 98 Aides à l’industrie du charbon pour l’année 2006 (Case NN 80/2006 and NN 81/2006) [2008] OJ C297/1; see also Gonzalez y Diez (Case C19/2003) Commission Decision 2004/340/EC [2004] OJ L119/26; State aid to the coal sector in Castilla Leon, 2000–2002 (Case C17/2003) Commission Decision 2005/140 [2005] OJ L48/30. 99 Coal sector (8 companies) (Case N261/2007) [2008] OJ C177/1. 100 German hard coal in 2009 (Case N563/2008) [2009] OJ C199/1; Baňa Dolina as (Case N347/2009) [2014] OJ C418/1; Coal sector in 2008–10 (Case NN20/2009) [2009] OJ C234/5.

Aid to Nuclear and Coal 225 services of general economic interest and were therefore compatible with the internal market. Applications for interim measures suspending the Commission decision were filed before the General Court.101 The General Court subsequently upheld the Commission decision in Case T-57/11 Castelnou.102 The appellant, Castelnou, had also argued that the Commission’s decision infringed Article 4(e) of Regulation 1407/2002 (which prohibits distortions of competition in the electricity market), as well as Article 6 of Regulation 1407/2002 (on degression of aid to the coal industry). The General Court noted that the principle of maintaining coal production capacity supported by State aid was affirmed by Regulation 1407/2002, and that Council Decision 2010/787, on State aid to facilitate the closure of uncompetitive coal mines, extended until 2018 the possibility for Member States to grant aid to cover, among other things, costs in connection with coal for the production of electricity. The General Court also confirmed the Commission’s reasoning that the revenue generated by the Spanish measure in favour of coal producers reduced the amount of direct aid authorised by Regulation 1407/2002. iii. Accession States The ECSC Treaty expired before the accession of the 10 new Member States in May 2004. Nevertheless, in accordance with the Europe Agreements (EAs) concluded prior to accession and prior to the expiry of the ECSC Treaty, the candidate countries were required to implement the acquis communautaire into national law. The EAs contained specific provisions in Protocol 2 and in the Annexes to the EAs, banning any aid not compatible with the exceptions laid down in the ECSC Treaty. Several candidate countries were however granted a five-year respite period from this requirement to allow them to grant restructuring aid, provided certain conditions were met and to allow for the continuation of VAT exemptions. Hence, State aid to coal undertakings in the new Member States remained partially exempted from the complete and immediate applicability of Articles 107 and 108 TFEU and Regulation 1407/2002. The Commission adopted a series of decisions approving aid to the coal sector in a number of the new Member States.103

101 See also P Romero, ‘The Coal Industry: Past or Future of Energy Production in Spain?’ (2011) 10 European State Aid Law Quarterly 565. 102 Case T-57/11 Castelnou Energía v Commission, EU:T:2014:1021. 103 See, eg, aid to the Slovakian coal sector: Commission, ‘Commission authorises Slovakia to grant €350.000 aid to mining company Hornonitrianske Bane Prievidza as’ (Press Release of 16 March 2005) IP/05/316; ‘The Commission authorises Slovakia’s investment plan for the coal industry for the years 2005 to 2010’ (Press Release of 25 January 2006) IP/06/76; and ‘Commission authorises Slovak state aid to the Hornonitrianske Bane Prievidza as mining company’ (Press Release of 13 September 2006) IP/06/1189; aid to the Czech coal industry: Commission, ‘European Commission authorises Czech Republic to grant €74 million aid to its coal industry’ (Press Release of 20 October 2005) IP/05/1312; aid to the Polish and Hungarian coal industry: Commission, ‘European Commission approves “coal package” authorising restructuring plans for the Polish, German and Hungarian coal industry until 2010’ (Press Release of 22 June 2005) Memo 05/217.

226 Leigh Hancher and Max Klasse C. The Current State Aid Regime and Practice i. The Closure of Uncompetitive Coal Mines: The Coal Decision of 2010 Regulation 1407/2002 is now replaced with Council Decision 2010/787 on State aid to facilitate the closure of uncompetitive coal mines—the Coal Decision.104 a. Background Regulation 1407/2002 foresaw an early evaluation by the Commission, the results of which are described in a Commission report in 2007.105 The report noted that, by the end of 2005, only eight of the then 25 Member States produced coal (Poland, Germany, Hungary, Great Britain, Spain, the Czech Republic, Slovakia and Greece (lignite)). France had closed its last mine in 2004. Aid to the sector in the Czech Republic and Slovakia comprises only aid to meet inherited liabilities and the UK provided investment aid only. While the Regulation was considered by the Commission to have resulted in a substantial reduction in the volume of aid to the coal industry,106 expiry of the Regulation without further transitional measures would have undermined the financial viability of a number of coal mines and forced their immediate closure. The prospect of a hard cut had sparked an extensive debate in both Parliament and Council, and between the EU institutions about the future of State aid in the sector.107 The Commission’s proposal108 foresaw a complete phasing out of all operating aid for coal mines by October 2014. In the words of the then Competition Commissioner Joaquín Almunia: The aim of the proposal is to ensure a definitive closure of uncompetitive mines by 1st October 2014. There should be no doubt about this. Companies need to be viable without subsidies. This is a question of fairness vis à vis competitors that operate without State aid. This is also in the interest of taxpayers and of government finances that are considerably constrained. The Commission will only allow operating aid to mining companies that have a closure plan and the subsidies should go increasingly towards supporting the social and environmental costs of doing so. Renewable, clean energy is the way to go, but we cannot ignore the dire regional economic and social consequences that would follow a sudden closure of the loss-making mines at this time of low or no growth and high unemployment.109

Coal producing Member States and other stakeholders criticised the proposed timing as being too short, both in terms of a healthy energy mix that is not overly

104 Council Decision 2010/787/EU of 10 December 2010 on State aid to facilitate the closure of uncompetitive coal mines (Coal Decision) [2012] OJ L336/24. 105 Commission, ‘Report on the Application of Council Regulation 1407/2002 on State Aid to the Coal Industry’ COM(2007) 253 final. 106 See Memo 10/348 of 20 July 2010. 107 See M Lang, ‘State Aid for the Coal Sector—Inevitable or Dispensable?’ (2012) 11 European State Aid Law Quarterly 113. 108 Commission, ‘Proposal for a Council Regulation on State aid to facilitate the closure of uncompetitive coal mines’ COM(2010) 372 final; see also Press Release IP/10/984 (n 78). 109 Press Release IP/10/984 (n 78).

Aid to Nuclear and Coal 227 dependent on imports and from the perspective of the regions concerned.110 Already in 2007, the then German government, on the basis of a compromise it had reached with stakeholders in Germany, had unilaterally decided to phase out State aid to coal mines (most of which are in the Ruhr region) by 2018 and it was not a coincidence that Germany pressed for that date in the Council.111 In view of strong political pressure from the Member States (in particular Germany) and the European Parliament, the Commission agreed to a compromise which culminated in the Coal Decision.112 The Coal Decision allows for the possibility to grant aid for the closure of uncompetitive coal mines until 31 December 2018, subject however to a gradual reduction of the aid in the years until 2018. Aid may be given to cover coal production if there is a closure plan whose deadline does not extend beyond 31 October 2018. Beyond that date, aid may be granted until 2027 to cover exceptional costs associated with the closure of mines (such as social welfare, rehabilitation of sites or removal of waste water). Any closure aid must decrease over time with a specified rate and accompanied by measures to mitigate the negative environmental impact of coal production. No investment aid to hard coal mining may be granted under the decision.113 The Council’s power to adopt the Coal Decision was unsuccessfully challenged in Case T-176/11 Carbunion v Council.114 The ECJ upheld the lower court’s ruling in Case C-99/14P of 11 December 2014.115 b. The Coal Decision in Brief The Coal Decision has been applied as of 1 January 2011 and it is due to expire on 31 December 2027. It aims to make any further operating aid to the sector conditional on a closure plan for lossmaking mines, and to direct State aid towards financing the social and environmental consequences of closures and to cover certain exceptional costs resulting from the closure.116 As its recitals make clear, the Union’s policy of encouraging renewable energy and a sustainable low-carbon economy does not justify indefinite support to uncompetitive coal mines. The Coal Decision marks the transition for the coal sector from the application of sector-specific rules to the application of the general State aid rules. The Coal Decision applies to high-, medium- and low-grade coal category A and B within the meaning of the international UNECE coal classification system (ie, hard coal). As with its predecessor, Regulation 1407/2002, it does not apply to lignite, or brown coal.

110

See for a comprehensive overview of the debate Lang (n 107) 115. The German ‘coal compromise’ as agreed by the German coalition government in 2007 and which culminated in the German Coal Financing Act. A revision clause, which could have led to a prolongation of State aid for coal in Germany was repealed in response to the Coal Decision. 112 Coal Decision (n 104). 113 Commission, ‘Commission Staff Working Paper Accompanying the Report from the Commission on Competition Policy 2010’ SEC(2011) 690 final. 114 Case T-176/11 Carbunión v Council, EU:T:2013:686. 115 Case C-99/14 P Carbunión v Council, EU:C:2014:2446. 116 P Staviczky and P Nicolaides, ‘State aid rules in the coal sector and linked energy sector under the Energy Community Treaty and European Law’ (2015) CEE Bankwatch Network 1, 10. 111

228 Leigh Hancher and Max Klasse In accordance with the Coal Decision, a mine receiving closure aid must be wound down by the end of 2018 at the latest but aid to cover exceptional costs resulting from the closure may be disbursed up until 2027. Hence, the decision allows operating aid to be granted as an exceptional tool for inefficient coal mines only if irreversible and indefinite closure has been decided and the aid is limited to the exceptional costs of the closure (Article 3(1)(a)–(c) of the Coal Decision). Closure aid is only available to mines that had been in operation on 31 December 2009. The annual amount of the closure aid must be reduced significantly in comparison to previous years (Article 3(1)(f) of the Coal Decision). All aid must be based on an agreed closure plan. If the coal production units to which aid is granted are not closed by the authorised closure date, the Member State shall recover all aid granted in respect of the whole period covered by the closure plan (Article 3(3) of the Coal Decision). The Decision also provides for the possibility of allowing for aid to cover exceptional costs resulting from activities related to the closures. This includes aid to mitigate social costs such as the costs of social welfare benefits or early retirement, costs incurred in safety or site rehabilitation for the production units subject to closure, as well as the pumping and cleaning of water from decommissioned mines. Such aid can be paid out after the closures until 2027 but must also be based on an agreed closure plan (Article 4 of the Coal Decision). c. Commission Practice Since the adoption of the Coal Decision, the Commission has dealt with a number of notifications concerning aid for the closure of coal mines in different Member States. In all these cases, the mines had received operating aid under the previous regime (Regulation 1407/2002). However, under the Coal Decision any additional aid is now subject to an agreed closure plan. Also, the aid is limited to the costs until the closure and the closure itself. A number of decisions were issued in the year following the entry into force of the Coal Decision, some of which solely concerned aid for exceptional costs (Slovenia117 and Poland),118 but the more significant aid volumes were in relation to the closure of uncompetitive coal mines relating to aid covering production costs.119 With the end of 2018 fast approaching, there has been another peak of decisions concerning closure aid lately, including in relation to mines, the restructuring of which had proven unsuccessful in the past. The examples below again prove the magnitude of the aid involved. In none of these cases, the qualification of the measure in question as State aid had been disputed by the Member State and the Commission has not had difficulties in applying the provisions of the Coal Decision. For instance, in May 2016, the Commission approved Spanish plans to grant €2.13 billion aid to support the operators of 26 coal mines that are due to be shut down until 2018.120 The Commission decision takes into consideration previous closure aid granted to the same beneficiaries in the period from 2011 to 2012, 117

Postponement of the closure of mine Trbovlje Hrastnik Ltd [2011] OJ C294/3. Coal plan for the period 2011–2015 (Case SA.33013) [2013] OJ C122/6. German Coal mine closure plan 2008–2018 (Case SA.24642) [2012] OJ C284/6 and National Hard Coal Company Petroşan (Case SA.33033) [2013] C23/3. 120 Aid to facilitate the closure of coal mines in Spain (Case SA.34332) [2016] OJ C471/1. 118 119

Aid to Nuclear and Coal 229 a measure which was complemented by a preferential dispatch mechanism involving financial compensation to coal-fired power plants burning indigenous coal but had been phased out in 2014. The revised closure plan foresees the possibility to receive additional aid with a view to mitigating the risk of disorderly closure. The Commission concluded that the measures were in line with the Coal Decision. The Commission found that the aid aims to ease the closure process by covering production losses of the mines until closure. It will also provide financial support to those workers who have lost or will lose their jobs due to the closures, by funding severance payments and social security benefits. Furthermore, it will finance the safety and remediation works necessary after the mine closures. The mines must be wound down by the end of 2018 at the latest and the Spanish authorities have given a commitment to recover any aid from mines that have not been closed by that date. In November 2016, the Commission cleared Polish funding of €1.79 billion to ensure the orderly closure of coal mines in Poland.121 Similar to the Spanish scheme, the new aid is partially an extension of schemes that had been approved by the Commission but had meanwhile expired.122 The Commission’s assessment found that, in line with the Coal Decision, the aid aims to ease the closure process by providing financial support totalling approximately €1.71 billion to workers who have lost, or will lose their jobs due to the closures. In particular, the State support will fund severance payments, compensatory pensions and social security benefits for these workers. Furthermore, it will be used to secure mine shafts and the decommissioning of mine infrastructures, repair damage to the environment caused by mining and recultivate land after the mine closures. The remainder of the aid will cover production losses of the mines until closure. Another example concerns the decision approving aid to alleviate social costs of closing down the Paskov mine in the Czech Republic which will allow the authorities to fund one-off severance payments as well as to provide special bonuses to workers exposed to occupational health risks.123 Closure of the mine had proven unsuccessful in the past. In November 2016, the Commission cleared approximately €99 million to close two uncompetitive coal mining units in Romania.124 According to the Commission’s assessment, the aid aims to ease the closure process by providing financial support totalling approximately €52 million for social security benefits of workers and to rehabilitate the mining sites and recultivate land after the mine closures. The remainder of the aid would cover production losses of the mines until closure. ii. Other Forms of Aid All other forms of aid to the sector are assessed in accordance with the general State aid rules and, since 1 July 2014, in accordance with the relevant guidelines where applicable.125 121

State Aid to Polish coal mining in the period 2015–2018 (Case SA.41161) [2017] OJ C51/1. See Coal plan for the period 2011–2015 (n 118). 123 Closure of the Paskov mine (Case SA.39570) [2015] OJ C309/1. 124 Commission, ‘Commission clears RON 447.8 million Romanian support for closing coal mines’ (Press Release of 24 November 2016) IP/16/3981. 125 See, for a comprehensive overview, Staviczky and Nicolaides (n 116) 12. 122

230 Leigh Hancher and Max Klasse a. Aid for the Conversion of Coal Power Plants In relation to the conversion of coal power plants to biomass, the Commission opened an in-depth investigation into a contract for differences that the UK government plans to conclude with the Drax coal fired power station for a conversion to biomass. On the basis of its analysis, the Commission concluded that the planned premium will not result in overcompensation.126 In its decision on the conversion of a coal-fired station at Lynemouth to run on wood pellets, the Commission approved a contract for difference having established that no risk of overcompensation would arise.127 On the basis of its analysis, the Commission concluded that the project’s contribution to the European renewable energy and carbon dioxide emissions reduction targets clearly outweighs any potential distortions of competition that could be triggered by the State support. b. Closure of Lignite-Fired Power Plants: Compensation for Damages or State Aid? As mentioned above, the Coal Decision does not apply to lignite which can generally be extracted without the need for subsidies. Putting external factors aside,128 lignite is a relatively cheap source of energy. Energy from lignite is a significant factor in the energy mix of some Member States, notably Germany, where lignitefired power plants play a significant role in the baseload supply. In 2013, Germany was the world’s largest lignite consumer, followed by China, Russia and the US.129 However, lignite also causes the highest levels of carbon dioxide emissions and it has become clear that Germany will not be able to meet its emissions targets if it does not significantly reduce energy production from lignite as part of its energy transition programme.130 It is quite clear though that lignite will continue to play some role in providing baseload power in Germany up to and after 2022, ie, by the time Germany has wound down its nuclear industry. Energy companies in Germany have responded differently to the mid- to long-term prospect of the phasing out of lignite mining and consumption for energy production. RWE has been considering construction of a new, highly efficient lignite-fired power plant in the ‘BoAplus’ project for some time now and has shown its commitment to the project by applying for

126 Drax 3rd Unit Biomass Conversion (Case SA.38760), OJ 2017 L217/1. See, Commission, ‘Commission authorises UK support to convert unit of Drax power plant from coal to biomass’ (Press Release of 19 December 2016) IP/16/4462. 127 Lynemouth Biomass Conversion (Case SA.38762), OJ 2017 L205/70. See Commission, ‘Commission authorises UK support to convert Lynemouth power station to biomass’ (Press Release of 1 December 2015) IP/15/6214. 128 According to a report commissioned by the European Commission, even including externalities, coal/lignite is the cheapest source of energy. See Ecofys (n 82). 129 About 90% of lignite is used to generate electricity and district heating in public and industrial power plants, close to where deposits are found. As a result, there is no international market for lignite. cf German Federal Ministry of Economic Affairs and Energy, at: www.bmwi.de/Redaktion/EN/Artikel/ Energy/coal.html. The remaining 10% is used in industrial processes. 130 See R Bajczuk, ‘The Uncertain Future of the Coal Energy Industry in Germany’ (2015) 188 OSW Commentary 4.

Aid to Nuclear and Coal 231 the necessary approvals in 2016. Vattenfall, by contrast, sold its lignite power plants and lignite assets in Eastern Germany to Czech energy incumbent EPH with a view to focusing on renewables and cutting its exposure to lignite.131 As mentioned above, unlike in relation to hard coal, State aid has not played a significant role in the lignite sector. A notable exception concerns closure aid for a number of lignite-fired power plants in Germany in 2016. In May 2016, the Commission approved €1.6 billion public financing for mothballing and subsequently closing eight lignite-fired power plants in Germany to be in line with State aid rules.132 Under the plans notified by Germany, the operators of these power plants would be compensated for foregone profits. The Commission found that the remuneration is primarily based on the foregone profits that the operators of the eight plants would have earned, had they continued operating in the electricity market for four more years until their final closure and decommissioning. In addition, during the mothballing phase, the companies will be remunerated for the extra costs of the mothballing and for producing electricity in emergency situations if other security reserve mechanisms do not provide sufficient backup energy (the so-called ‘security readiness’). While formally separate from and not part of the German capacity mechanisms cleared by the Commission in late 2016,133 the purpose of the mothballing has to be seen in the context of a set of measures to manage and ultimately overcome capacity shortages in parts of Germany. The case sparked an intense debate about what some considered a ‘golden handshake’ for the lignite industry in Germany.134 In particular, it was questioned whether the lignite plants were suitable to contribute in a meaningful way to the envisaged security of supply objectives in view of their slow response times to sudden increases in demand.135 Also, critics argued that the plants’ location would mean that they are ill-suited to provide backup energy in the regions where this was likely needed. From a State aid perspective, the case raised two interesting questions under Article 107(1) TFEU: first, Germany took the view that the advantage, if any, is not financed from State resources and not imputable to the State. Relying on the PreussenElektra case law,136 Germany argued that, as the measure is financed via increased network tariffs levied on the consumers, no State resources would be involved. The network tariffs would be collected by TSOs who subsequently transfer it to the beneficiaries without intervention from the State. The Commission rejected the

131 In May 2016, a State aid complaint was lodged in relation to Vattenfall’s sale of its lignite power plants and assets in Eastern Germany, albeit it seems that the Commission has not taken up the complaint. The complainant alleges that the sale at a negative purchase price entails elements of State aid because it was not on commercial terms in accordance with the MEO principle. The agreement includes a €1.7 billion premium to be paid by Vattenfall, a utility owned and controlled by the Swedish State, for its obligations of renaturalising former mines. See Power Engineering International, ‘Vattenfall lignite sale comes under EU State aid microscope’ (8 May 2016). 132 Closure of German lignite plants (Case SA.42536) [2016] OJ C258/1. 133 None of the lignite-fired plants in question will take part in the Network Reserve, see Commission, German Network Reserve (Case SA.42955) [2017] OJ C68/1. See also ch 14. 134 See eg, Frankfurter Allgemeine Zeitung, ‘Kabinett beschließt Aus für Braunkohle-Meiler’ (4 November 2015). 135 Even during the mothballing period, the activation of the plants will require a 10-day notice period. 136 Case C-379/98 PreussenElektra, EU:C:2001:160.

232 Leigh Hancher and Max Klasse argument. With reference to the parallel question in relation to the EEG surcharge, pending before the CJEU,137 the Commission found that even absent any direct impact on the State budget there was sufficient State control to bring the payments within the ambit of State resources. The Commission held that ‘in the present case it is indeed the German State that has established via legislative and regulatory provisions the remuneration for the lignite blocks and the obligation for the TSOs to disburse it’.138 Second, Germany argued that measure should be viewed as constituting a compensation for damages and therefore, in principle, as excluding the existence of a selective advantage for the companies involved. The Commission decision assesses the argument in quite some detail. Referring to its case practice it concluded that where a measure solely grants to the affected undertaking the compensation for damages incurred, this compensation does not provide it with an advantage and hence does not constitute State aid in the meaning of Article 107(1) TFEU. The Commission also considers that a legal framework is in place in German law that may entitle operators whose rights as owners are affected to a compensation and that it cannot be excluded that such right to compensation exists. However, for the measure at hand the Commission concludes that on the basis of the information provided by the German authorities in the present case it cannot be concluded with a sufficient degree of certainty that a right to compensation of an amount resulting from the formula adopted … for the operators exists.139

Ultimately, the Commission left the question open as to whether the measure provides the operators with an advantage and thus constitutes State aid pursuant to Article 107(1) TFEU because it considered that, even if State aid were involved, the measure is compatible with the internal market. The Commission went on to assess the measure under Article 107(3)(c) TFEU but not the EEAG as these do not contain provisions for aid to compensate for the closure of electricity generation plants. The Commission considered that with the mothballing and closure of the eight power plants, the German government aims to reduce carbon dioxide emissions by 1112.5 million tons per year as of 2020, when all eight plants will have stopped operating. While the costs for closing the plants will be borne by the operators themselves, Germany would only compensate the operators for their foregone profits. The Commission assessed this support in light of the significant contribution to Germany reaching its emissions target. It also came to the conclusion that the remuneration paid to the operators concerned would not confer an undue advantage over their competitors. c. SGEI In Case T-57/11 Castelnou,140 the General Court assessed whether the Commission had correctly considered that security of supply concerns on the Spanish electricity

137 See Case T-47/15 Germany v Commission, EU:T:2016:281, paras 81–128, currently on appeal in Case C-405/16 P. 138 Closure of German lignite plants (n 132) para 29. 139 ibid, paras 49, 50. 140 Castelnou Energía (n 102).

Aid to Nuclear and Coal 233 market justified an SGEI in the form of a preferential dispatch mechanism according to Article 11(4) of Directive 2003/54/EC. This mechanism provided that electricity produced by power plants using indigenous (ie, Spanish) coal must be bought in preference to electricity produced by power plants using imported coal or other fuels. The power plants benefiting from the measure received compensation related to their additional production costs. The General Court concluded that the Commission had not committed a manifest error when recognising the justified and proportionate nature of the measure. In addition, when assessing the measure, the Commission was not required to verify that the measure did not infringe EU provisions relating to the protection of the environment given that the aid measure did not pursue an environmental objective. The case confirmed the earlier position of the Commission that the use of indigenous primary energy resources for power generation may, under certain circumstances, be regarded as a genuine SGEI, where the measure is in line with the rules applicable to State aids in the form of public service compensation.141

141 Commission, ‘Commission Staff Working Document Accompanying the Report from the Commission on Competition Policy 2008’ SEC(2009)1004 final, para 163.

9 Projects of Common Interest and Energy Infrastructure Projects JULIA SACK, KRISTINA ZYCH AND KAI UWE PRITZSCHE*

I. INTRODUCTION

‘A

WELL-INTERCONNECTED ENERGY infrastructure is a pre-condition for an integrated, competitive and sustainable internal energy market in the European Union’,1 the European Commission stated when the second list of Projects of Common Interest (PCIs) was adopted in 2015. ‘It is also a prerequisite for a resilient Energy Union which provides EU consumers with secure, sustainable, competitive and affordable energy’.2 To establish this precondition, the existing trans-European energy transmission infrastructure needs to be upgraded and new infrastructure must be built. The Commission estimated that investment requirements in energy infrastructure with cross-border relevance up to 2020 will amount to €200 billion. The development of high-voltage electricity transmission systems, storage infrastructure and smart grids requires an expected amount of €140 billion. Gas pipelines, storage, LNG terminals and bidirectional gas infrastructure development are estimated to cost €70 billion, and €2.5 billion is expected to be needed for the development of carbon dioxide transport infrastructure.3 For the attraction of private investment in these fields, a transparent and aligned regulatory framework within the European Union (EU) that avoids lengthy and ineffective permit granting procedures is essential. Further, and though energy infrastructure should be financed generally by the market and through tariffs paid by users, projects which are not economically viable per se may require the provision of public funds to help leverage the investment needed. Therefore, in the context of the present treaties, an analysis of the extent to which such funding stems from public sources with European State aid rules seems to be advisable. * The authors would like to thank Christian Crohn and Hannah Rubin for their valuable help in preparing this chapter. 1 European Commission, ‘Commission Staff Working Document accompanying the Document Commission Delegated Regulation amending Regulation (EU) 347/2013 of the European Parliament and of the Council as regards the Union list of projects of common interest’ SWD(2015) 247, 2. 2 ibid. 3 Commission, ‘Commission Staff Working Document Impact assessment accompanying the document Proposal for a Regulation of the European Parliament and of the Council on guidelines for transEuropean energy infrastructure and repealing Decision 1364/2006/EC’ SEC(2011) 1233, 9.

236 Julia Sack, Kristina Zych and Kai Uwe Pritzsche As the European legislator had perceived these needs, the Regulation on Guidelines for trans-European energy infrastructure (TEN-E Regulation)4 was adopted in 2013. The TEN-E Regulation provides a regulatory framework for the timely development and interoperability of nine priority geographic corridors and three thematic priority areas of trans-European energy infrastructure. The implementation of these priorities is considered essential for the achievement of the EU’s energy and climate policy objectives, consisting of completing the internal market in energy, guaranteeing security of supply, in particular for electricity and gas, reducing greenhouse gas emissions, increasing the share of renewable energy in final energy consumption and achieving an increase in energy efficiency, whereby energy efficiency gains may contribute to reducing the need for the construction of new infrastructures. At the same time, the EU intends to prepare its infrastructure for further decarbonisation of its energy system in the longer term towards 2050.5 Besides the definition of the energy infrastructure priorities, the TEN-E Regulation sets the rules to identify PCIs, which are infrastructure projects necessary to implement these priorities and, thereby, contribute to well-interconnected energy networks in Europe. The first part of this chapter describes the regulatory regime applicable to energy PCIs, in particular the selection criteria for PCIs and the selection process under the TEN-E Regulation. Examples illustrate the diverse shape of PCIs with regard to parameters such as geographical scope, maturity and funding. Further, the specifics linked to the PCI status and the possible access to EU funding are presented. The second part of this chapter portrays the State aid regimes applicable to the various forms of State support to these key infrastructure projects. Two regimes of the Commission’s legal practice implementing the exemptions under Article 107(3) of the Treaty on the Functioning of the European Union (TFEU) are analysed with regard to their historic background and content as well as their application by the Commission. Finally, it is assessed to which extent PCIs are privileged under both regimes.

II. THE TEN-E REGIME FOR PCIS

The TEN-E Regulation provides for the regulatory framework for selecting PCIs (see II.A below) and a set of measures which facilitate their timely implementation (see II.B.i below). This includes rules for streamlining the permit granting, rules and guidance for the cross-border allocation of costs and risk-related incentives as well as rules determining the conditions for the eligibility of PCIs for Union financial assistance (see II.B.ii below).

A. Obtaining PCI Status According to the Commission, PCIs are key infrastructure projects, which will help Member States to physically integrate their energy markets, enable them to 4 Regulation (EU) 347/2013 of the European Parliament and of the Council of 17 April 2013 on guidelines for trans-European energy infrastructure (TEN-E Regulation) [2013] OJ L115/39. 5 ibid, recital 7.

Projects of Common Interest 237 diversify their energy sources and help bring an end to the energy isolation some of the Member States are facing.6 They are the primary European tool to accelerate the deployment of the infrastructure necessary for the completion of the European energy market and to ensure that the European Union meets its goals of affordable, secure and sustainable energy.7 If a project is recognised as a PCI, it benefits from the advantages that are conferred to it under the TEN-E Regulation and from opportunities for better financing. For the assessment of whether a project will be able to fulfil the criteria for a PCI that the Commission assigns to PCIs, the TEN-E Regulation defines selection criteria and a multi-stage selection process. i. Selection Criteria There are several general and specific selection criteria a project needs to meet to be designated as a PCI. The general precondition is that the project concerns the planning or construction of energy infrastructure pursuant to the exhaustive list of eligible infrastructure types in Annex II of the TEN-E Regulation (including, eg, transmission lines or pipelines, storage facilities, associated equipment and installations). In addition, the project must comply with three general criteria set out in Article 4(1) of the TEN-E Regulation, which are applicable to all energy infrastructure types. The first of these general criteria is that the project is necessary for at least one of the priority corridors or areas of trans-European energy infrastructure set out in Annex I of the TEN-E Regulation. These priorities cover different geographic regions or thematic areas in the field of electricity transmission and storage, gas transmission, storage and liquefied or compressed natural gas infrastructure, carbon dioxide transport and oil infrastructure (eg, the development of the North Sea offshore grid, north–south electricity interconnections in Central Eastern and South Eastern Europe, Southern Gas Corridor, oil supply connections in Central Eastern Europe). The second of the general criteria requires that the potential overall benefits of the project, assessed according to the respective specific criteria in Article 4(2) of the TEN-E Regulation, outweigh its costs, including in the long term. Finally, as a third criterion, the project must involve at least two Member States by directly crossing the border of two or more Member States or must be located on the territory of one Member State with a significant cross-border impact or must cross the border of at least one Member State and a European Economic Area country. Besides these general selection criteria, a project must meet the specific requirements that Article 4(2) of the TEN-E Regulation lays down for the five infrastructure type categories listed in Annex II respectively. For electricity, gas and oil projects, the criteria in Article 4 are assessed in accordance with indicators set out in Annex IV(2)–(5) of the TEN-E Regulation. Finally, Article 4(4) stipulates that the Groups assessing the projects shall give due consideration to certain defined circumstances of the project.

6 Commission, ‘Fact Sheet Projects of common interest in energy—questions and answers’ MEMO/15/6108. 7 Commission, ‘Commission unveils key energy infrastructure projects to integrate Europe’s energy markets and diversify sources’ (Press Release of 18 November 2015) IP/15/6107.

238 Julia Sack, Kristina Zych and Kai Uwe Pritzsche Table 1 below provides an overview of the selection criteria for PCIs as defined in the TEN-E Regulation: Table 1: Overview of the Selection Criteria for PCIs as Defined in the TEN-E Regulation Precondition Energy infrastructure for electricity, gas, oil, carbon dioxide pursuant to Annex II I. General Selection Criteria (Art 4 (1)) 1. Necessary for at least one of the 12 energy infrastructure priority corridors and areas (listed in Annex I). 2. Potential overall benefits of the project (such as market integration, sustainability, security of supply). 3. Any of the following: — —



Involvement of at least two Member States through directly crossing the border of two or more States. Project is located in one Member State with a significant cross-border impact as set out in Annex IV(1) (indicators differ for the respective electricity/gas/oil/carbon dioxide infrastructure). Project crosses the border of at least one Member State and a European Economic Area country.

II. Specific Selection Criteria (Art 4 (2)) 1. Electricity transmission and storage projects Significant contribution to at least one of the following: —

Market integration, inter alia through lifting the isolation of at least one Member State and reducing energy infrastructure bottlenecks; competition and system flexibility (indicator for market integration in Annex IV(2)(a). — Sustainability, inter alia through the integration of renewable energy into the grid and the transmission of renewable generation to major consumption centres and storage sites (indicator for transmission of renewable generation in Annex IV(2)(b). — Security of supply, inter alia through interoperability, appropriate connections and secure and reliable system operation (indicator for security of supply in Annex IV(2)(c). 2. Electricity smart grid projects Significant contributions to all of the following (indicators in Annex IV(4)): — — — — — —

Integration and involvement of network users with new technical requirements with regard to their electricity supply and demand. Efficiency and interoperability of electricity transmission and distribution in day-today network operation. Network security, system control and quality of supply. Optimised planning of future cost-efficient network investments. Market functioning and customer services. Involvement of users in the management of their energy usage. (continued)

Projects of Common Interest 239 Table 1: (Continued) 3. Gas projects Significant contribution to at least one of the following: —

Market integration, inter alia through lifting the isolation of at least one Member State and reducing energy infrastructure bottlenecks; interoperability and system flexibility (indicator for market integration in Annex IV(3)(a). — Competition, inter alia through diversification of supply sources, supplying counterparts and routes (indicator for competition in Annex IV(3)(b). — Security of supply, inter alia through appropriate connections and diversification of supply sources, supplying counterparts and routes (indicator for security of supply in Annex IV(3)(c). — Sustainability, inter alia through reducing emissions, supporting intermittent renewable generation and enhancing the deployment of renewable gas (indicator for sustainability in Annex IV(3)(d). 4. Oil transport projects Significant contribution to all of the following: — — —

Security of supply reducing single supply source or route dependency (indicator for security of supply in Annex IV(5)(a). Interoperability (indicator in Annex IV(5)(b). Efficient and sustainable use of resources through mitigation of environmental risks (indicator for efficiency and sustainability in Annex IV(5)(c).

5. Carbon dioxide transport projects Significant contribution to all of the following: — Avoidance of carbon dioxide emissions while maintaining security of energy supply. — Increasing the resilience and security of carbon dioxide transport. — Efficient use of resources by enabling the connection of multiple carbon dioxide sources and storage sites via common infrastructure and minimising environmental burden and risks. III. Circumstances for Due Consideration (Art 4 (4)) 1.

Urgency of the proposed project in order to meet the Union energy policy targets of market integration, inter alia through lifting the isolation of at least one Member State and competition, sustainability and security of supply.

2.

Number of Member States affected by each project, whilst ensuring equal opportunities for projects involving peripheral Member States.

3.

Contribution of each project to territorial cohesion.

4.

Complementarity with regard to other proposed projects.

Smart grids, in order to qualify as PCIs, additionally have to comply with the provisions under the Assessment Framework of the Smart Grids adopted by the Expert Group on Smart Grid Infrastructure Deployment (Expert Group 4).8 8 Smart Grid Task Force, ‘Definition of an Assessment Framework for Projects of Common Interest in the Field of Smart Grids under the EC “Proposal for a regulation of the European Parliament and of the Council on guidelines for trans-European energy infrastructure and repealing Decision No 1364/2006/ EC”’ (COM(2011) 658) (2012).

240 Julia Sack, Kristina Zych and Kai Uwe Pritzsche ii. Selection Process The final list of projects qualifying as PCIs (PCI or Union list) is the result of a thorough identification and evaluation process conducted, inter alia, by 12 Regional Groups (nine regional, three thematic) established under the TEN-E Regulation.9 Regional Groups for gas or electricity projects consist of the representatives of the Member States, national regulatory authorities (NRAs), transmission system operators as well as the Commission, the Agency for Cooperation of Energy Regulators (ACER) and ENTSO-E and ENTSO-G respectively.10 For oil and carbon dioxide transport projects each Group is composed of the representatives of the Member States and project promoters concerned by each of the relevant priorities under the TEN-E Regulation.11 During their evaluation, Regional Groups shall invite promoters of a project potentially eligible for selection as a PCI as well as further representatives, for example, of national administrations.12 Further, they shall consult the organisations representing relevant stakeholders—and, if deemed appropriate, the stakeholders directly—including distribution system operators, suppliers, consumers and organisations for environmental protection.13 Finally, the rules on the selection process stipulate that the Commission, ACER and the Regional Groups shall strive for consistency between the different Groups, and emphasises that the participation of NRAs and ACER may not jeopardise the fulfilment of the objectives under the TEN-E Regulation.14 The identification and evaluation process conducted by the Regional Groups is an essential part of the multi-stage selection procedure for PCIs under the TEN-E Regulation, which is illustrated below in Figure 1 (with estimated time intervals for the selection round in 2017):15

Figure 1: Overview of the Selection Procedure for PCIs under the TEN-E Regulation

9 European Commission, ‘Energy: Commission unveils list of 250 infrastructure projects that may qualify for €5,85 billion of funding’ (Press Release of 14 October 2013) IP/13/932. 10 TEN-E Regulation (n 4) Annex III, s 1(1). 11 ibid. 12 ibid, Annex III s 1(4). 13 ibid, Annex III s 1(5). 14 ibid, Annex III s 1(7). 15 Based on the European Commission’s timeline for the 2015 selection process, see: www.ecologic.eu/ sites/files/event/2014/01_viksne.pdf.

Projects of Common Interest 241 For a project to be included in the Union list, project promoters need to submit an application for their project within the deadlines of the respective selection rounds. There are separate calls for applications for electricity and gas on the one hand, and oil, carbon dioxide and smart grids on the other. Gas and electricity project applications are made via the online project portals of ENTSO-G and ENTSO-E, which will distribute the proposals to the respective Regional Groups in charge of the assessment. Applications for all other projects are addressed directly to the Regional Groups at Thematic Group Meetings for project promoters in Brussels. Pursuant to Annex III section 2(1) of the TEN-E Regulation, the applicant needs to submit the following documents: —

An assessment of the project’s contribution to the relevant priority corridor/ thematic area. — An analysis of the fulfilment of the relevant criteria defined in Article 4 TEN-E Regulation. — A societal cost–benefit analysis, in order to demonstrate the project’s economic viability and cost-effectiveness (for sufficiently mature projects). — Any other relevant information for the evaluation of the project. Further, as Annex III sections 2(3) and (4) of the TEN-E Regulation stipulates, proposed electricity and gas projects must be part of the latest available Ten-Year Network Development Plan (TYNDP) for electricity or gas, which are prepared on a biannual basis by ENTSO-E and ENTSO-G respectively. As there are no TYNDPs for carbon dioxide transport projects, they need to be presented as part of a plan, developed by at least two Member States, for the development of cross-border carbon dioxide transport and storage infrastructure, to be presented by the Member States concerned or entities designated by those Member States to the Commission.16 In case of proposals for electricity and gas projects, the NRA, and if necessary ACER, check the consistent application of the criteria/cost–benefit analysis methodology and evaluate their cross-border relevance.17 For proposed oil and carbon dioxide transport projects, the Commission evaluates the application of the criteria set out in Article 4 of the TEN-E Regulation. It thereby takes into account the potential for future extension to include additional Member States.18 The findings of these assessments are presented to the Regional Groups, which meet to examine and rank the proposed projects taking into account the assessment of the regulators or the Commission.19 The draft regional lists, which are the results of the Regional Groups’ evaluations are, in the case of electricity and gas projects, submitted to ACER six months before the adoption date of the Union list. ACER assesses the draft regional lists (and accompanying opinions)20 within three months of receipt and provides an 16

TEN-E Regulation (n 4) Annex III, s 2(6). ibid, Annex III s 2(7). 18 ibid, Annex III s 2(8). 19 ibid, Annex III s 2(11). 20 A Member State to whose territory a proposed project does not relate, but on which the proposed project may have a potential net positive impact of significant effect, may present an opinion to the Group specifying its concerns. See TEN-E Regulation (n 4) Annex III, s 2(9). 17

242 Julia Sack, Kristina Zych and Kai Uwe Pritzsche opinion on the regional lists, in particular on the consistent application of the criteria and the cost–benefit analysis across regions.21 Within one month of the date of receipt of ACER’s opinion, the decision-making body of each Group (composed of Member States and the Commission)22 adopts its final regional list, respecting the provisions set out in Article 3(3) of the TEN-E Regulation, based on the Group’s proposal and in view of the opinion of ACER and the assessment of the NRAs or the Commission.23 The 12 final regional lists (together with any opinions) are submitted to the Commission that adopts one Union-wide PCI list as an amendment to the TEN-E Regulation via the delegated act procedure.24 The Commission submits the Union list to the European Parliament and Council and, should neither of the two institutions oppose the list within two months, the Union list enters into force.25 The Union list is updated every two years following a new selection round.26 So far, there have been two selection rounds that resulted in, first, the Union list of 2013,27 and second, the Union list of 2015.28 The first Union list included 248 projects (132 electricity projects, 107 gas projects, seven oil projects and two smart grids projects); the second Union list 195 projects (108 electricity projects, 77 gas projects, seven oil projects and three smart grids). iii. Implementation and Delisting Once a project has been designated as a PCI, the project promoters need to draw up an implementation plan, including a timetable of feasibility and design studies, approval by the NRAs or by any other authority concerned, construction and commissioning and a permit-granting schedule.29 Further, by 31 March of each year following the year of inclusion of a PCI in the Union list, project promoters shall submit an annual progress report to the NRA and to ACER (or the Regional Group in case of oil and carbon capture storage projects), indicating the state of play, delays

21

ibid, Annex III, s 2(12). ibid, Art 3(1). 23 ibid, Annex III, s 2(13). 24 ibid, Art 3(4) and Annex III, s 2(13). If based on the regional lists received, and after having taken into account ACER’s opinion, the total number of proposed PCIs on the Union list exceeds a manageable number, the Commission may consider excluding in the Union list the projects that were ranked lowest by the Group concerned according to the ranking established pursuant to Art 4(4) of the TEN-E Regulation (ibid, Annex III, s 2(14)). 25 Commission, ‘Projects of common interest in energy—questions and answers’ (Press Release of 18 November 2015) MEMO/15/6108. 26 Commission, ‘Questions and Answers: projects of common interest in energy’ MEMO/13/880; TEN-E Regulation (n 4) Art 3(4). 27 Commission Delegated Regulation (EU) 1391/2013 of 14 October 2013 amending Regulation (EU) 347/2013 of the European Parliament and of the Council on guidelines for trans-European energy infrastructure as regards the Union list of projects of common interest [2013] OJ L349/28. 28 Commission Delegated Regulation (EU) 2016/89 of 18 November 2015 amending Regulation (EU) 347/2013 of the European Parliament and of the Council as regards the Union list of projects of common interest [2015] OJ L19/1. 29 TEN-E Regulation (n 4) Art 5(1). 22

Projects of Common Interest 243 and revised planning.30 ACER will submit a consolidated report to the Regional Groups.31 PCIs on the Union list that do not meet the eligibility criteria under the TEN-E Regulation any more or that are not resubmitted in the new PCI selection round will be delisted in the subsequent selection round. According to Article 5(9) TEN-E Regulation, projects which are no longer on the Union list shall lose all rights and obligations linked to the PCI status arising from the TEN-E Regulation. iv. Examples of PCIs In the following, three examples of PCIs are presented that contribute to different priorities under the TEN-E Regulation.

Figure 2:

a. Green-Me The name ‘Green-Me’ is the abbreviation of ‘Grid integration of Renewable Energy sources in the North Mediterranean’, a project that will contribute to the decarbonisation efforts of the EU. Green-Me aims at enhancing the integration of renewable energy sources in France and Italy by implementing automation, control and monitoring systems substations, including communication with the renewable generators and storage in primary substations, as well as new data exchange to allow for better cross-border interconnection management. It is part of the Priority Thematic Area ‘Smart Grids Deployment’ and is one of the three smart grid projects on the 2015 Union list (number 10.2).

30

ibid, Art 5(4) and 5(5). Commission, ‘Commission Staff Working Document Implementation of TEN-E, EEPR and PCI Projects accompanying the document Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions Progress towards completing the Internal Energy Market’ SWD(2014) 314, 10. 31

244 Julia Sack, Kristina Zych and Kai Uwe Pritzsche The project area of Green-Me covers a large territory between the North of Italy and the South of France, a consumption area with 700,000 customers consuming electricity of about 11 TWh/year. The electricity grid in its current state is not able to efficiently accommodate increasing RES as the respective primary substations are already close to their limit of maximum hosting capacity and reserve power flow problems occur.32 Green-Me will allow the active management of RES, especially photovoltaic (PV). Moreover, the project will implement several innovative functionalities such as an enhanced cross-border interconnection management, power system observability and controllability, and emergency action on generation and loads.33 The project promoters are a consortium of the two transmission system operators Terna (Italy) and RTE (France) and of the two distribution system operators ENEL Distribuzione (Italy) and ERDF (France). The construction is taking place between 2014 and 2019 and the commissioning date is scheduled for 2019. The overall cost amounts to about €72 million.34 b. Kriegers Flak Combined Grid Solution The Kriegers Flak Combined Grid Solution project is an offshore electricity interconnection utilising national grid connections to offshore wind farms and sub-sea cables for both the transmission of wind energy and energy trade. It is going to provide a 400 MW offshore interconnection between Bentwisch/Germany and Tolstrup Gaarde/Denmark via the offshore wind farms Kriegers Flak (Denmark, under construction) and Baltic 1 and 2 (Germany, operating). The two wind farms are located less than 30 km away from each other and will be connected by two sub-marine cables to establish the interconnector. With eastern Denmark and Germany being different synchronous areas, a frequency transformation is necessary35 and a double converter will be installed in Bentwisch. The Kriegers Flak project is part of the Priority Corridor ‘Baltic Energy Market Interconnection Plan in Electricity’ (number 4.1 on the 2015 Union list). It will improve the electricity interconnectivity in the Baltic Sea region and facilitate the integration of generated renewable energy in Germany and in eastern Denmark. The project promoters are the German transmission system operator 50Hertz Transmission GmbH and the Danish transmission system operator Energinet.dk. Construction is planned to take place between 2017 and 2018 and the commissioning date is scheduled for December 2018. The total cost of the project amounts to €900 million, mainly financed by the project promoters. The project has been granted €150 million of financial support from the European Energy Programme for Recovery (EEPR).36 32

www.download.terna.it/terna/0000/0607/37.PDF. L Schiavo et al, ‘Cost/benefit assessment for large-scale smart grids projects: The case of Project of Common Interest for smart grid “GREEN-ME”’ Paper 1658 (23rd International Conference on Electricity Distribution, Lyon, 15–18 June 2015). 34 Commission, ‘Smart Grid PCI GREEN ME for SGTG meeting on 08.09.2016’ (presentation), see: www.ec.europa.eu/energy/sites/ener/files/presentations.zip. 35 www.energinet.dk/EN/ANLAEG-OG-PROJEKTER/Anlaegsprojekter-el/Forbindelse-til-TysklandKriegers-Flak-CGS/Sider/Anlaegget.aspx. 36 Commission, ‘Offshore Wind. Baltic-Kriegers Flak: Combined Grid Solutions. Fact Sheet’ (2013). 33

Projects of Common Interest 245 c. BalticConnector The BalticConnector will consist of an 82 km bi-directional submarine gas transmission pipeline between Paldiski/Estonia and Inkoo/Finland, 22 km onshore gas transmission pipeline in Finland, 47 km onshore gas transmission pipeline in Estonia, gas metering stations, and compressor stations. It is part of the Priority Corridor ‘Baltic Energy Market Interconnection Plan in Gas’ and listed as number 8.1.1 on the 2015 Union list. It will contribute to a further integration of the Baltic States in the European energy market. The project promoters of the BalticConnector are the Estonian transmission system operator Elering AS and the Finnish State-owned company Baltic Connector OY.37

B. Consequences of PCI Status A project’s PCI status has (i) procedural and regulatory consequences and (ii), under certain circumstances, provides it with access to EU funding. i. Procedural and Regulatory Consequences Once a project has been designated as a PCI, this has certain procedural consequences in the permitting process (cf Article 7 et seq TEN-E Regulation)38 and for the regulatory treatment of the respective PCI (cf Article 11 et seq TEN-E Regulation). a. Priority Status Pursuant to Article 7, PCIs will be allocated the highest priority status available under national law in order to benefit from accelerated permit granting procedures. This priority status applies without prejudice to the exact location, routing or technology of the PCI. With regard to the environmental assessment procedures for PCIs, the Commission provided the Member States with guidance on how to define adequate legislative and non-legislative measures to streamline national environmental assessment procedures for PCIs and to ensure the coherent application of these procedures.39 The Member States have been required to take the identified measures.40 For all environmental impact assessments, PCIs will be considered as being of particular public interest.

37 A Vahtla, ‘Elering, Baltic Connector OY sign agreement to build gas interconnection’ News. err.ee (Estonian Public Broadcasting, 17 October 2016). See also www.ec.europa.eu/inea/en/connecting-europe-facility/cef-energy/projects-by-country/multi-country/8.1.1%E2%80%930004-fieew-m-2016. 38 For details see M Appel and A Burghardt, ‘The New Planning Regime for the Expansion of the German Onshore Electricity Grid—A Role Model for Europe?’ 2013 4(1) Renewable Energy Law and Policy Review 13, 26 ff. 39 Commission, ‘Streamlining environmental assessment procedures for energy infrastructure Projects of Common Interest (PCIs)’ (2013). 40 TEN-E Regulation (n 4) Art 7(2) and (3).

246 Julia Sack, Kristina Zych and Kai Uwe Pritzsche b. Organisation and Duration of the Permit Granting Process Article 8 of the TEN-E Regulation requires the Member States to designate one national competent authority which is responsible for facilitating and coordinating the permit granting process for PCIs (‘one-stop-shop’). This responsibility and the tasks related to it, however, may be delegated to, or carried out by, another authority in certain circumstances as long as there is one authority which is responsible for each PCI, which is the sole point of contact for the project promoter in a process leading to the comprehensive decision for the project and which coordinates the submission of all relevant documents and information.41 Member States can choose among three possible schemes to follow in the decision-making process (and may apply different schemes to onshore and offshore PCIs): the integrated, the coordinated and the collaborative scheme.42 Further, as many PCIs are of a cross-border nature, Article 8(5) requires the designated competent authorities to take all necessary steps for efficient and effective cooperation and coordination among the different countries. To ensure rapid treatment, Article 10 of the TEN-E Regulation sets an overall timetable of three years and six months for the permit granting process with an indicative period of two years for ‘pre-application procedures’—for example, the preparation of the necessary schedules, the concept for public participation and public consultation on PCI proposals—and one year and six months for the determination of applications for ‘permits’. The pre-application procedure starts with the written acknowledgement of the project notification by the competent authority. The second phase starts with the acceptance of the submitted application file and ends with the comprehensive decision. The total time limit of three years and six months may be extended by nine months on a case-by-case basis43 and does not include appeals or other judicial process.44 In practice, national laws can provide substantial hindrances to the quick and efficient planning and permitting of PCIs due to neighbours’ interventions, law suits etc. Therefore, due to a lack of efficient means of enforcement against national laws, the planning timetable may not always be achievable. This seems to be one of the significant weaknesses of the process. c. Transparency and Public Participation The European legislator expressly referred in the recitals of the TEN-E Regulation to the established standards for the participation of the public in environmental decision-making procedures. Nevertheless, additional measures were considered necessary to ensure the highest possible standards of transparency and public participation for all relevant issues in the permit granting process for PCIs:45 —

41 42 43 44 45

First, Article 9(1) obliges the Member States or their competent authorities to publish and update a (non-binding) manual of procedures for the permit granting ibid, Art 8(2). ibid, Art 8(3) and (4). ibid, Art 10(2). ibid, Art 10(6). ibid, recital 30.

Projects of Common Interest 247 process of PCIs, including at least the information specified in section 1 of the guidelines for transparency and public participation attached to the TEN-E Regulation as Annex VI (eg, the relevant law, name of the competent authority and a checklist for the submission of documents). — Moreover, Article 9(2) stipulates that all parties involved are required to follow the principles for public participation set out in section 3 of Annex VI.46 — In addition, project promoters need to draw up and submit a concept for public participation to the competent authority within three months of the start of the permit granting process to be approved by the authority47 and at least one public consultation shall be carried out before the final and complete application file is submitted. The results of the public participation are to be summarised in a report by the project promoters which is to be submitted together with the application file.48 — Finally, the rules on transparency require that the project promoters or competent authorities establish a website with all relevant information on a PCI which is not commercially sensitive.49 d. Regulatory Treatment The TEN-E Regulation also contains specific rules on the regulatory treatment of PCIs: Article 11 of the TEN-E Regulation stipulates how the methodologies for a harmonised energy system-wide cost–benefit analysis are set. Thereby, the provision takes into account that the cost–benefit analysis is of importance at several stages of the PCI selection and implementation process. For instance, and as noted before, projects having reached a sufficient degree of maturity are required to submit a cost– benefit analysis, which will be considered by the Regional Groups during the PCI selection process.50 Article 12 lays down the rules for a cross-border cost allocation which aim to facilitate PCIs implementation by envisaging decisions by NRAs or by ACER. The prerequisite for a cross-border cost allocation is that project promoters submit an investment request including a request for the cost allocation to all the NRAs concerned. After having received such a request, the NRAs are required to issue a coordinated decision on the allocation of investment costs to be borne by each party as well as their inclusion in tariffs to be paid by the national users. In the event that the NRAs are not able to reach an agreement within six months, the decision will be taken by ACER.51 Finally, Article 13 grants incentives where a 46 This requirement is without prejudice to any requirement under the Aarhus Convention (The United Nations Economic Commission for Europe (UNECE) Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters) and the Espoo Convention (Convention on Environmental Impact Assessment in a Transboundary Context) and relevant Union law. 47 TEN-E Regulation (n 4) Art 9(3). 48 For details see ibid, Art 9(4), (5) and (6). The minimum requirements applicable to the public consultation are specified in s 5 of Annex VI. 49 For details see ibid, Art 9(7). 50 ibid, Annex III, s 2(1). 51 In 2015, ACER issued its Recommendation No 05/2015 to help project promoters and NRAs in future cross-border cost allocation procedures and to facilitate a consistent EU-wide approach for CBCA cases by sharing good practices.

248 Julia Sack, Kristina Zych and Kai Uwe Pritzsche project promoter incurs higher risks for the development, construction, operation or maintenance of a PCI compared with the risks normally incurred by a comparable infrastructure project. ii. Access to EU Funding According to the TEN-E Regulation, a three-step logic applies to investments in PCIs: — —



First, the market should have the priority to invest. Second, if investments are not made by the market, the regulatory framework should be explored from the perspective of the PCI’s feasibility and, if necessary, it should be adjusted, and its correct application should be ensured. Third, where the first two steps are not sufficient to deliver the necessary investments in PCIs, Union financial assistance could be granted if the project fulfils eligibility criteria.52

In recent years, the EU has initiated a number of funding facilities with the aim of providing support to infrastructure projects which are not viable under the existing regulatory framework and market conditions, for example, (1) the Connecting Europe Facility (CEF), (2) the European Regional Development Fund (ERDF) and the EU Cohesion Policy Funds as well as (3) the European Fund for Strategic Investment (EFSI), which are described below. a. CEF Launched in December 2013, the CEF has complemented the reform of the TEN-E, TEN-T and TEN-Telecom53 regulatory regimes. The CEF Regulation54 provides a framework for financial support that aims at accelerating investment in the field of all three TEN and at leveraging funding from both the public and private sectors. The CEF budget for 2014 to 2020 is €30.4 billion (€5.35 billion for energy projects).55 Under the CEF, support is provided in two forms—grants and financial instruments.56 These two forms are subject to different application procedures and implementation plans and involve different stakeholders. PCI promoters are encouraged to explore using financial instruments before applying for a grant.

52

TEN-E Regulation (n 4) recital 42. Regulation (EU) No 283/2014 of the European Parliament and of the Council of 11 March 2014 on guidelines for trans-European networks in the area of telecommunications infrastructure and repealing Decision 1336/97/EC [2014] OJ L86/14. 54 Regulation (EU) No 1316/2013 of the European Parliament and of the Council of 11 December 2013 establishing the Connecting Europe Facility, amending Regulation (EU) No 913/2010 and repealing Regulations (EC) No 680/2007 and (EC) No 67/2010 (CEF Regulation) [2013] OJ L348/129. 55 Art 5 of the CEF Regulation allocates €3.85 billion to energy projects. However, after the CEF Regulation entered into force, the initial CEF budget was reduced to €30.4 billion in June 2015 because part of the budget was redirected to the EFSI and the amount designated for energy projects adjusted accordingly. See European Parliament, ‘Assessment of Connecting Europe Facility’ PE 572.677. 56 CEF Regulation (n 64) Art 15(4). 53

Projects of Common Interest 249 Financial instruments consist of debt instruments (enhanced loans, project bonds), equity instruments and credit enhancement measures (first loss cover, eg, subordinated mezzanine debt and other risk-sharing mechanisms with financial intermediaries, guarantees, letters of credit).57 Financial instruments are available outside any call for proposals. They are offered and managed by international financing institutions (entrusted entities), for example, the European Investment Bank. Pursuant to Article 14(2) of the CEF Regulation, only 10 per cent of the total CEF budget can be dedicated to co-funding projects through financial instruments. However, unlike grants, financial instruments can be combined with other parts of the CEF and other instruments, programmes and budget lines in the Union budget. PCIs to be supported by financial instruments shall be selected on the basis of maturity and shall seek sectoral diversification as well as geographical balance across the Member States. They shall represent European added value, respond to the objectives of the Europe 2020 Strategy, present a leverage effect with regard to Union support, ie aim at mobilising a global investment exceeding the size of the Union contribution according to the indicators defined in advance.58 Grants can be made for studies or construction works. They are allocated semiannually, following competitive calls for proposals. The maximum co-financing rates for projects in the energy sector are listed in Article 10(3) of the CEF Regulation. As a matter of principle, they should not exceed 50 per cent of the eligible costs of studies and/or works (with a maximum of 75 per cent under special circumstances). The Innovation and Networks Executive Agency (INEA) is responsible for the grant management under the CEF of all three trans-European networks. So far, grants amounting to about €2.21 billion have been provided since the CEF was launched. Table 2: Overview of Supported PCIs and Amounts of Financial Assistance Call for proposals (round/year)

PCIs supported (grants for works/studies)

Total amount of financial assistance €

II/2016

18

444 million

I/2016

9

263 million

II/2015

15

550 million

I/2015

20

150 million

2014

34

647 million

There are a number of criteria a project needs to fulfil to be eligible for a grant. One of the criteria is that the project is included in the current PCI list at the time of the application for the grant. Moreover, Article 7 of the CEF Regulation refers to Article 14 TEN-E Regulation, which contains the eligibility criteria for energy PCIs that apply for CEF support. According to Article 14(1) of the TEN-E Regulation,

57 58

ibid, Annex I, Part III. ibid, Art 15(1).

250 Julia Sack, Kristina Zych and Kai Uwe Pritzsche all PCIs falling under the categories set out in Annex II(1) (electricity), (2) (gas) and (4) (carbon dioxide) of the TEN-E Regulation are eligible for grants for studies and financial instruments. This means oil infrastructure PCIs are not eligible for grants for studies. According to Article 14(2) of the TEN-E Regulation, PCIs falling under the categories set out in Annex II(1)(a)–(d) and (2) of the TEN-E Regulation, except for hydro-pumped electricity storage, are also eligible for EU financial assistance in the form of grants for works if the following three criteria cumulatively apply to the PCI applicant: a. b. c.

The cost–benefit analysis shows significant positive externalities (security of supply, solidarity or innovation). The project has received a cross-border cost-allocation decision. The PCI is commercially not viable according to the business plan and other assessments carried out.

As electricity smart grids and carbon dioxide PCI cannot receive a cross-border costallocation decision, Article 14(4) of the TEN-E Regulation stipulates that those PCIs are eligible for grants for works if the project promoters can clearly demonstrate the significant positive externalities generated by the project and a lack of commercial viability. This means oil infrastructure PCIs are also not eligible for grants for works. Additional award criteria are established by the Commission taking into account certain general orientations defined in Annex I Part V of the CEF Regulation, for example, the maturity of the action in the project development, the soundness of implementation plan or the stimulating effect of the CEF financial assistance.59 b. European Regional Development Fund (ERDF) and Cohesion Fund (CF) The European Structural and Investment Funds (ESIF), including the ERDF and the CF, make up one of the largest items of the EU budget. Whereas the ESIF mainly target projects at the national and regional levels, cross-territorial activities may be supported as well. Thus, PCIs may be included in the investment priorities of the ERDF and CF. Partnership Agreements between the Commission and individual Member States set out the national authorities’ plans on how to use the ESIF funding in the period from 2014 to 2020. The Partnership Agreements cover all support from the ESIF in the Member State concerned and link each country’s strategic goals and investment priorities to the overall aims of the Europe 2020 strategy for smart, sustainable and inclusive growth. Project promoters need to submit their applications for funding to the national or regional authority managing the relevant programme. The applications for funding must comply with the investment priorities, eligibility criteria and application procedure of the programmes in the applicant’s region and country and the general criteria laid down in the fund specific frameworks.60 59

ibid, Art 17(5). Regulation (EU) No 1301/2013 of the European Parliament and of the Council of 17 December 2013 on the European Regional Development Fund and on specific provisions concerning the investment for growth and jobs goal and repealing Regulation (EC) No 1080/2006 (ERDF Regulation) [2013] OJ L347/289 and Regulation (EU) No 1300/2013 of the European Parliament and of the Council of 17 December 2013 on the Cohesion Fund and repealing Council Reg (EC) No 1084/2006 (CF Regulation) [2013] OJ L347/281. 60

Projects of Common Interest 251 In the field of energy, both the ERDF and CF focus on energy efficiency and renewable energy, high-efficiency cogeneration, smart energy distribution grids and integrated low-carbon and sustainable energy action plans for urban areas. In addition, the ERDF supports research and innovation in low-carbon technologies. The ERDF aims to strengthen economic and social cohesion in the EU by correcting imbalances between its regions. The ERDF budget for 2014–20 is €276.8 billion. The majority of its resources are devoted to R&D (€39.9 billion), low-carbon economy (€30.1 billion) and transport and energy infrastructure (€25.6 billion).61 Similar to the ERDF, the CF aims at reducing economic and social disparities and to promote sustainable development throughout the EU. The CF is only accessible for Member States whose Gross National Income per inhabitant is less than 90 per cent of the EU average (eg, Bulgaria, Croatia, Portugal or Slovenia).62 The CF fund budget for 2014–20 is €75.5 billion.63 Energy infrastructure supported by the CF essentially comes down to infrastructure for the distribution and use of renewable energy as well as the development and implementation of smart distribution systems. c. European Fund for Strategic Investment (EFSI) The EFSI is a joint initiative of the Commission and the EIB Group intended to complement other EU actions taken to reduce the investment gaps in the EU by acting as a guarantee fund and stimulus for new investments in various sectors, not only in the energy sector. It resulted from the 315 billion Euro Investment Plan for Europe, the so called ‘Juncker Plan’. The EFSI is intended to support strategic investments such as, but not limited to, PCIs which aim at completing the internal market in the transport, telecommunications and energy infrastructure sectors. According to Article 3 of the EFSI Regulation,64 the purpose of the fund is to support a. b.

investments and increased access to financing for entities having up to 3000 employees, with a particular focus on SMEs and small mid-cap companies. An independent investment committee decides which projects will be backed by the EU guarantee.

The EFSI was established for an initial period of three years from 2015 to 2017; however on 14 September 2016 the Commission declared its intention to extend the fund until 2020. In the new proposal, the Commission plans to increase the EU guarantee from currently €16 billion to €26 billion and the EIB contribution from currently €5 billion to €7.5 billion. 61

www.cohesiondata.ec.europa.eu/funds/erdf. Regulation (EU) No 1303/2013 of the European Parliament and of the Council of 17 December 2013 laying down common provisions on the European Regional Development Fund, the European Social Fund, the Cohesion Fund, the European Agricultural Fund for Rural Development and the European Maritime and Fisheries Fund and laying down general provisions on the European Regional Development Fund, the European Social Fund, the Cohesion Fund and the European Maritime and Fisheries Fund and repealing Council Regulation (EC) No 1083/2006 (Common Provisions Regulation) [2013] OJ L347/320, recital 77. 63 www.cohesiondata.ec.europa.eu/funds/cf. 64 Regulation (EU) No 2015/1017 of the European Parliament and of the Council of 25 June 2015 on the European Fund for Strategic Investments, the European Investment Advisory Hub and the European Investment Project Portal and amending Regulations (EU) No 1291/2013 and (EU) No 1316/2013—the European Fund for Strategic Investments [2015] OJ L 169. 62

252 Julia Sack, Kristina Zych and Kai Uwe Pritzsche As described above, most of the EU funding initiatives and programmes are available as grants and financial instruments. However, the EFSI operates differently as it concentrates on financial instruments and first loss pieces rather than grants. The Commission hopes that with a first loss piece, such as guarantees or subordinated debt, the funds can be reused rather than paid out once. d. Development Bank Loans Besides the above described EU funds, PCIs may apply for loans from European banks like the EIB or the European Bank for Reconstruction and Development (EBRD). In the period from 2007 to 2012, the EIB supported about €19 billion of investment for the development and modernisation of the European electricity network infrastructure and €10.4 billion for the development and modernisation of the European natural gas networks.65 It has, for example, also granted a loan of €800 million to the NordLink PCI in December 2016.66 The EBRD invested more than €8.6 billion in 172 energy sector projects in the period from January 2006 to October 2013.67 In 2012, the EBRD provided a €75 million loan for the construction of an LNG terminal in Swinoujście/Poland.68

III. STATE AID COMPATIBILITY OF TRANS-EUROPEAN ENERGY INFRASTRUCTURE PROJECTS

There may be strategically important projects that are not economically viable under the existing market conditions and that rely on the leverage of public funds, not only on a European but on a national level. With the TEN-E Regulation but also, in general, with its aim of an integrated European energy market69 the EU aims at promoting the interconnection and interoperability of national networks as well as access to the thereby established TENs within the framework of a system of open and competitive markets (Article 170(2) TFEU). State intervention supporting PCIs, which are necessary to establish the TEN-E, may nevertheless not come at the price of a distortion of competition and negative effects on trade between Member States. This second part will thus illustrate how this is handled by the EU State aid regime.

A. Types of State Aid to Energy Infrastructure and Legal Framework Article 107(1) TFEU provides that any support measure 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

65 66 67 68 69

www.eib.europa.eu/attachments/strategies/eib_energy_lending_criteria_en.pdf. www.eib.org/projects/pipelines/pipeline/20150911. www.ebrd.com/what-we-do/sectors-and-topics/ebrd-energy-strategy.html. www.ebrd.com/news/2012/ebrd-is-helping-poland-open-up-to-liquefied-gas.html. See, eg, above (n 1).

Projects of Common Interest 253 goods and affects trade between Member States constitutes State aid which is, in principle, incompatible with the internal market. State aid can be awarded as an individual aid or as an aid scheme and can take a myriad of shapes, inter alia, grants, loans, preferential interest rates, favourable taxation or tax exemptions, guarantees, adjustment of feed-in tariffs, exemption from regulatory regimes. The Commission has an exclusive competence to determine the legality of State aid measures. It distinguishes between investment and operating aid to energy infrastructure projects of which in general, for competition reasons, only the first is considered acceptable while the latter is subject to a more severe review. In the absence of applicable exemptions, State aid measures are subject to the Member State’s notification obligation pursuant to Article 108(3) TFEU and Member States must not put the aid into effect before the Commission has approved their compatibility with the internal market. Before turning to the compatibility assessment of State aid to PCIs and energy infrastructure, it is worth recalling the various EU aids presented earlier and contemplate their relevance in the context of the Commission’s State aid regime. The cornerstone of EU State aid law, Article 107(1) TFEU, stipulates that State aid ipso jure can only be any aid granted by a Member State or through State resources. One might argue over the legal nature of the EU, however the EU is not a State itself. Pursuant to Article 311 TFEU (implemented by Council Decision 2014/335/EU), the budget of the EU is composed of contributions from the Member States. Yet, this does not make the EU budget a pool of State resources in terms of Article 107(1) TFEU. From the moment the State contributions feed into the EU budget, they become the EU’s own resources (Article 311 TFEU). Consequently, the State aid rules of Article 107 et seq TFEU do not apply directly to EU aids.70 Notwithstanding, certain EU aids from EU funds that are not centrally managed by the Commission with the support of its executive agencies are worth a closer look in this regard. In fact, the majority of EU funds are administered in shared or indirect management, so that decisions about final beneficiaries and the actual transfers of money to them will not be taken by the Commission but by the Member States or other delegated EU bodies.71 When there is shared management of Union funds involving the Member States and the Commission, the national authorities are directly responsible for managing their share of the EU fund, with reporting obligations to the Commission. About 80 per cent of the EU’s funding is managed by the Member States themselves (national governments, regional managing authorities).72 Such Union aid is subject to State aid control. Among the EU funds presented under

70 See Regulation (EU) No 651/2014 of 17 June 2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty (General Block Exemption Regulation) [2014] OJ L187/1, recital 26; Commission, ‘Communication from the Commission— Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG) [2014] OJ C200/1; Commission, ‘Communication from the Commission—Criteria for the analysis of the compatibility with the internal market of State aid to promote the execution of important projects of common European interest’ (IPCEI Communication) [2014] OJ C188/4, para 20. 71 Regulation (EU, Euratom) No 966/2012 of the European Parliament and of the Council of 25 October 2012 on the financial rules applicable to the general budget of the Union and repealing Council Regulation (EC, Euratom) No 1605/2002 (Financial Regulation) [2012] OJ L298/1, Arts 59, 60. 72 www.ec.europa.eu/info/how-eu-funding-works/types-funding/management-eu-funding_en.

254 Julia Sack, Kristina Zych and Kai Uwe Pritzsche section II.B.ii, the ESIF is an example of shared fund management.73 Consequently, ERDF and CF resources spent in support of PCIs are considered as State resources and therefore are subject to State aid control.74 The following two sections will present two State aid regimes applicable to State support for priority TEN infrastructure projects. As a starting point, one has to recall that there is no absolute ban on State aid under the TFEU. Article 107(2) and (3) TFEU contain important exemptions from the general prohibition of State aid laid down in Article 107(1) TFEU. Article 107(3) TFEU confers discretion to the Commission to declare State support, that was found to constitute aid, compatible with the internal market. However, the Commission will only resort to the provisions of Article 107 et seq TFEU if the State support does not fall under the thresholds of the de minimis Regulation75 or the General Block Exemption Regulation.76 Moreover, the Commission may specify its aid policy, for example through communications, frameworks or guidelines as long as these rules do not depart from the legal provisions under primary EU law.77 Unlike delegated Regulations that can be passed by the Commission pursuant to Article 290 TFEU, those measures of self-regulation are considered generalised legal practice or also ‘soft law’ that will be de facto legally binding since its publicity and application sets a self-imposed limit on the Commission’s exercise of discretion.78 In practice, such soft law will also block the direct application and ad hoc assessment of the aid under the more general provisions of Article 107 TFEU that it specifies. With regard to energy infrastructure, the Commission adopted the ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG)79 and the Commission’s Communication 2014/C 188/0280 on important projects of common European interest (IPCEI Communication). Both will be discussed below.

B. Regime 1: Article 107 (3)(c) TFEU and EEAG 2014–2020 Reading Article 107(3)(c) TFEU, one would not instinctively subsume support to PCIs under ‘aid to facilitate the development of certain economic activities or of

73

Common Provisions Regulation (n 62) Art 4(7). See, eg, Upgrading of the Port of Patras (Case SA.38048) [2014] OJ C280/1; and National Broadband (Case N201/2006) [2006] OJ C296/33. 75 Regulation (EU) 1407/2013 of 18 December 2013 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid [2013] OJ L352/1. 76 Regulation (EU) 651/2014 of 17 June 2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty (General Block Exemption Regulation) [2014] OJ L187/1. 77 Case C-313/90 CIRFS and Others v Commission, EU:C:1993:111; Case C-288/96 Germany v Commission, EU:C:2000:537; Case C-310/99 Italy v Commission, EU:C:2002:143; Case T-319/11 ABN Amro Group v Commission, EU:T:2014:186. 78 See Case T-380/94 AIUFASS and AKT v Commission, EU:T:1996:195, para 57; Case T-214/95 Vlaamse Gewest v Commission, EU:T:1998:77, para 89. 79 EEAG (n 70). 80 IPCEI Communication (n 70). 74

Projects of Common Interest 255 certain economic areas’. The addition that this aid may ‘not adversely affect trading conditions to an extent contrary to the common interest’ does not link the common interest to the economic activity or sector either. However, even before the adoption of the Communication on EEAG, the Commission had assessed priority TEN-E81 (and TEN-T)82 PCIs under Article 107(3) (c) TFEU. Both having been passed in 2014, the EEAG (in July) and the TEN-E Regulation (in April) are clearly dovetailed with regard to their contribution to reaching the 2020 energy goals. While the latter lays down the framework for the identification and (regulatory) support of trans-European energy infrastructure, the former provides the corresponding State aid regime in case Member States join the Union and the market participants in establishing the necessary infrastructure. The EEAG are not the first guidelines the Commission ever passed in the field of environmental protection. For the first time, in 1994, it passed the ‘Community Guidelines on State aid for environmental protection’.83 The 1994 Guidelines (EAG) were revised in 200184 and 2008,85 but none of these versions dealt with the construction of energy infrastructure. With the adoption of the EEAG, which is planned to remain in force from 1 July 2014 to 31 December 2020, the Commission laid down common assessment principles for aid to environmental protection and, for the first time, energy infrastructure projects, which will be declared compatible with the internal market pursuant to Article 107(3)(c) TFEU if they comply with the EEAG provisions. Aid to energy infrastructure however is only one among 14 different State measures that are covered by the EEAG,86 so that the Guidelines have a much broader scope than just targeting PCIs and other non-priority energy infrastructure. In Chapter 3, the Guidelines stipulate a regime of general assessment principles and compatibility provisions that apply to all these types of State aid. The specific compatibility criteria set out in Chapter 4 of the EEAG provide additional and more specific provisions derived from the general assessment principles for 10 out of these 14 State aid types. In the following, the conditions will be presented that national aids to energy infrastructures have to cumulatively comply with ((i)–(viii)). Information on other State aid types with relevance for PCIs (especially aid to renewable sources of energy) will selectively be included at the relevant issues of the Commission’s compatibility assessment.

81 See, eg, Aid to PGNiG for underground gas storage in Poland (Case N660/2009) [2010] OJ C213/9; Construction of interconnection and cross-border power line between Poland and Lithuania (Case N542/2010) [2011] C79/1; Construction of a LNG Terminal in Swinoujsciu (Case SA.31953) [2011] OJ C361/1; 1st upgrade LNG Terminal at Revithoussa (Case SA.35165) [2013] OJ C117/1. 82 See eg Flughafen Kassel-Calden (Case NN14/2007) [2009] OJ C97/5; Upgrading of the Port of Patras (n 74); Ferry port Sassnitz—Breakwater (Case SA.45849) [2016] OJ C369/1. 83 Commission, ‘Community guidelines on State aid for environmental protection’ [1994] OJ C72/3. 84 Commission, ‘Community guidelines on State aid for environmental protection’ [2001] OJ C37/3. 85 Commission, ‘Community guidelines on State aid for environmental protection’ (EEG 2008) [2008] OJ C82/1. 86 EEAG (n 70) para 18(k).

256 Julia Sack, Kristina Zych and Kai Uwe Pritzsche i. Notification Thresholds The EEAG sets out aid thresholds above which aid to energy projects must be notified to the Commission. When projects either are selected through a competitive bidding process or the planned aid stays below the notification thresholds, no notification to the Commission is necessary under the EEAG and such State support will be deemed compatible with the internal market. Accordingly, investment aid for energy infrastructure exceeding €50 million for one undertaking for each investment project and investment aid for Carbon Capture and Storage exceeding €50 million for each investment project will be subject to notification and State aid scrutiny under the EEAG.87 Operating aid granted to renewable electricity installations has to be notified for projects where the resulting renewable electricity generation capacity per site exceeds 250 megawatts.88 ii. Objective of Common Interest As already mentioned, Article 107(3)(c) TFEU provides an exemption from the general ban on State aid depending on whether the aid in question is not contrary to the common interest. Member States intending to grant environmental or energy aid have to define precisely the objective pursued by the aid and how the measure will contribute towards this objective.89 Aid granted without a specific purpose distorts competition without being necessary in the common interest. That is why operating aid is, in principle, considered incompatible with the internal market90 and the EEAG stipulates that aid to energy infrastructure should, in principle, be investment aid, including its modernisation and upgrade.91 The general objective of common interest applicable to all aid measures under the EEAG92 is to increase the level of environmental protection (for environmental aid) and to ensure an affordable, sustainable, secure and well-functioning internal energy market (for aid in the energy sector). For aid to energy infrastructure, the Commission presumes that such infrastructure is a precondition for a functioning internal energy market and thus contributes to an objective of common interest.93 It is not only the PCI label which implies that the project’s contribution to the common interest will be the Union’s internal energy market. Energy infrastructure projects that do not carry this label, whose common objective was previously derived from Articles 170(2), 194(1)(d) TFEU,94 also now benefit from the presumption of paragraph 202 EEAG.95 87

ibid, para 20(e). ibid, para 20(b). 89 ibid, para 31. 90 Case C-301/87 France v Commission, EU:C:1990:67, para 50; P Nicolaides, M Kekelekis and P Buyskes, State Aid Policy in the European Community: A Guide for Practitioners (The Hague, Kluwer Law International, 2005) 40. 91 EEAG (n 70) para 201. 92 ibid, para 30. 93 ibid, para 202. 94 See, eg, Electricity cable between mainland Finland and Aland Commission Decision (Case SA.33823) [2012] OJ C382/4. 95 The reasoning is identical for PCIs (see Individual Aid Measure for Gas Infrastructure in Poland Commission Decision SA.39050 [2015]) and non-priority energy infrastructure (see SA.36290) with regard to the common objective based on para 202 EEAG (n 70). 88

Projects of Common Interest 257 For aid to renewable energy projects, such aid will be deemed as contributing to an objective of common interest if the project helps to meet the targets set by the EU as part of its EU 2020 and 2030 energy strategies.96 iii. Need for State Intervention (Market Failure) As a further requirement, the planned aid must be necessary for the achievement of the objective of common interest. It must be granted in a situation where aid can bring about a material improvement that the market cannot deliver itself, for example, by remedying a clearly defined market failure.97 The EEAG cite four main reasons for market failures occurring in the field of environmental protection and energy infrastructure.98 With regard to energy infrastructure investments, the EEAG identify coordination failures and positive externalities as the main market failures.99 The identification of a market failure is however not sufficient to prove the necessity of the State aid in question. State intervention must furthermore be restricted to the residual market failure, that is to say, the market failure that remains unaddressed by other policies and measures already in place such as sector regulation, pricing mechanisms and carbon taxes. Moreover, the case for the necessity of the aid is weaker if it counteracts such other policies targeted at the same market failure.100 In the case of individual aids, the Commission carries out a case-by-case assessment considering the following factors: a.

To what extent a market failure leads to a sub-optimal provision of the necessary infrastructure, to what extent the infrastructure is open to third party access and subject to tariff regulation and to what extent the project contributes to the Union’s security of energy supply.101

b. c.

TEN-E PCIs benefit from a presumption in favour of a residual market failure that State aid support can help overcome.102 This presumption extends to aid for energy from renewable sources,103 but does not apply to oil infrastructure projects.104 iv. Appropriateness In the next step of its compatibility assessment, the Commission examines whether the proposed aid measure, provided it is suitable for its purpose, represents an appropriate instrument to achieve the specific objective of common interest. To that end, the Commission considers to what extent the same objective is achievable through other less distortive policy instruments or other less distortive types

96 97 98 99 100 101 102 103 104

EEAG (n 70) paras 107–08. ibid, para 34. ibid, para 35. ibid, para 203. ibid, para 36. ibid, para 207. ibid, para 206. ibid, para 115. ibid, para 208.

258 Julia Sack, Kristina Zych and Kai Uwe Pritzsche of aid instruments.105 From the pool of energy aid measures described above in section III.A, Member States should choose the form that is likely to generate the least distortions of trade and competition (eg, repayable advances are preferable to direct grants, tax credits are preferable compared with tax reductions).106 The Commission considers that, in principle, tariffs are the appropriate primary means to fund energy infrastructure. However, PCIs also benefit from a presumption at this stage of the compatibility assessment because for them it is assumed that market failures cannot be remedied fully through tariff regulation.107 For support to non-priority infrastructure, Member States have to demonstrate its appropriateness, whereas operating aid, being a more distortive measure than investment aid, is subject to a more severe review.108 As for PCIs, the appropriateness for aid to energy from renewable sources is also presumed, provided that all other conditions are met.109 v. Incentive Effect Furthermore, the State aid must have an incentive effect, which occurs if the aid induces the beneficiary to change its behaviour, ie, leading the beneficiary to a behaviour he would not show without the aid. The aid must not merely subsidise the costs of a project that would anyhow incur and must not compensate for the normal business risk of an economic activity.110 The incentive effect is, in principle, identified through a counterfactual scenario analysis, which essentially amounts to checking the profitability of the project in the absence of the aid.111 It is incumbent upon the national authorities to conduct a credibility check of the counterfactual scenario when assessing the aid application.112 The incentive effect of aid for energy infrastructure will be assessed on the basis of the general conditions set out in section 3.2.4 of the EEAG.113 The required incentive effect is negated by the Commission where work on the project had already started before the beneficiary submitted the aid application to the national authorities.114 However, in some cases one may have to analyse closely whether the project already started and the one for which the application is made really are one and the same project. If a competitive bidding process is carried out, most of the conditions115 are disregarded and the incentive effect is assumed.116

105 106 107 108 109 110 111 112 113 114 115 116

ibid, para 40. ibid, para 45. ibid, para 209. ibid, para 47. ibid, para 116. ibid, para 49. ibid, para 60. ibid, paras 52, 58. ibid, para 210. ibid, para 50. ibid, precisely those in paras 50–51. ibid, para 52.

Projects of Common Interest 259 vi. Proportionality The proportionality of an aid measure is given when its amount is limited to the minimum needed to incentivise the additional investment or activity needed to achieve the objective of common interest.117 This step in the Commission’s assessment assures that no overcompensation is provided that could result in undue distortive effects but that the State aid is limited to what is necessary for allowing the project to achieve a reasonable rate of return. Prior to the adoption of the EEAG, the Commission had not shown a consistent approach with regard to the maximum admissible amount of aid that can be considered as proportional. However, in order to ensure predictability and a level playing field, Annex I of the EEAG now contains a list of maximum aid intensities, ie, the maximum amount admissible for aid expressed as a percentage of the project’s eligible costs. The table below (Table 3) highlights the aid intensities for investment aid of those State support measures among the types of State aid covered by the EEAG that have relevance for PCIs. Table 3: Aid Intensity of Investment Aid Small118 enterprise

Medium-sized119 enterprise

Large120 enterprise

Aid for environmental studies

70%

60%

50%

Aid for renewable energies or cogeneration installations

65%, 100% if bidding process

55%, 100% if bidding process

45%, 100% if bidding process

Aid for energy infrastructure

100%

100%

100%

Aid for CCS

100%

100%

100%

Aid type

With regard to aid schemes and individual aid granted on the basis of such schemes, the maximum aid intensities effectively function as thresholds for the assessment of their proportionality. For individual aid, however, these maximum aid intensities are merely used as a cap; compliance with them is not sufficient to ensure proportionality. The proportionality of individually notifiable aid rather has to be assessed individually, based on the net extra costs of the aided investment (being the difference between the economic benefits and costs of the aided project and those of the counterfactual scenario), taking into account all relevant costs and benefits over the lifetime of the supported project.121 For operating aid granted by way of a competitive bidding process pursuant to the conditions applying from 1 January 2017, the proportionality of individual aid is presumed to be met if the general conditions are fulfilled.122

117

ibid, para 69. Defined in Commission Recommendation of 6 May 2003 concerning the definition of micro, small and medium-sized enterprises [2003] OJ L124/36, Annex, Art 2(2). 119 ibid, Annex, Art 2 (1). 120 EEAG (n 70) para 19, sub-para 18. 121 ibid, para 84. 122 ibid, paras 87, 126. 118

260 Julia Sack, Kristina Zych and Kai Uwe Pritzsche Aid to infrastructure can be granted up to an aid intensity of 100 per cent of the eligible costs.123 This shows the priority that the Commission confers on energy infrastructure projects in the common interest. If any of the projects supported through State aid is to be awarded additional EU financing, the relevant eligible costs will be decreased accordingly, so that the total amount of public (State and Union) funding granted in relation to the same eligible costs will not exceed the maximum acceptable funding rate under the EEAG. vii. Avoidance of Undue Distortion of Competition Having successfully jumped all previous hurdles, the State aid measure will have to pass this last step of the compatibility assessment where the Commission examines whether the distortions of competition and the effect on trade between Member States caused by the aid are limited in such a way that the overall balance remains positive.124 The proportionality criterion established in the previous step of the compatibility assessment contributes in principle to softening negative impacts and thus positively influences the overall balance.125 When assessing aid to energy infrastructure, the Commission will consider that support to infrastructure subject to internal market regulation does not have undue distortive effects, because the existing requirements under this legislation are aimed at strengthening competition.126 Since PCIs usually are subject to internal market legislation, Member States will more easily make the case for a positive balance of aid to such projects. Infrastructure partially or wholly exempted from, or not subject to, internal energy market legislation and underground gas storage facilities will be assessed on a case-by-case basis in view of the degree of third-party access to the aided infrastructure, access to alternative infrastructure and the market share of the beneficiary.127 Additional conditions are that the appropriate, proportionate and not unduly distortive State aid measure has to comply with transparency obligations and the principle of non-violation of other provisions of EU law. With regard to the former, as of 1 July 2016, Member States have to publish certain information on a comprehensive State aid website.128 With regard to the latter, a State aid measure, or the conditions attached to it, that entail a non-severable violation of Union law cannot be declared compatible with the internal market. In the field of energy infrastructure, any levy that has the aim of financing a State aid measure needs to comply, in particular, with the provisions of Articles 30 and 110 TFEU.129

123 124 125 126 127 128 129

ibid, para 212. ibid, para 88. ibid, para 98. ibid, para 214. ibid, para 215. ibid, paras 104, 106. ibid, para 29.

Projects of Common Interest 261 viii. Energy Infrastructure within the Scope of Application The EEAG use the same definition of energy infrastructure as Annex II of the TEN-E Regulation.130 This is another sign of the synchronisation of both regimes and the fit of Article 107(3)(c) TFEU, as specified by the EEAG, for the State aid assessment of TEN-E PCIs. Consequently, aid to transmission, storage and distribution and associated infrastructure for electricity, gas, oil and carbon dioxide is covered, but aid to nuclear power plants is not. According to paragraph 13, the EEAG also apply to State aid granted to projects in sectors that are subject to specific Union rules on State aid themselves (transport, coal, agriculture, forestry, fisheries and aquaculture) provided that such projects realise environmental protection or energy objectives and that the sector-specific State aid rules do not provide an exclusive assessment framework for them. All in all, under the EEAG regime adopted to specify the exemption from Article 107(1) pursuant to Article 107(3)(c) TFEU, priority projects as well as PCIs benefit from certain presumptions with regard to the objective of common interest, the need for State intervention, incentive effect and appropriateness. Hence, they are more readily declared compatible with the internal market by the Commission than other energy or environmental protection projects.

C. Regime 2: Article 107(3)(b) TFEU and IPCEI Communication Another State aid regime potentially relevant for PCI projects is provided by the exemption of important projects of common European interest (IPCEI) under Article 107(3)(b) TFEU and the Commission’s IPCEI Communication specifying the Commission’s practice laid down therein. Article 107(3)(b) TFEU provides that ‘aid to promote the execution of an important project of common European interest … may be compatible with the internal market’. The Commission’s IPCEI Communication provides a cross-sector guidance. Its scope includes all transnational projects of strategic significance for the EU and for the achievement of the Europe 2020 objectives in all sectors of economic activity, including environmental protection and energy.131 IPCEIs are large-scale projects that bring together knowledge, expertise, financial resources and actors from public and private sectors throughout the Union, so as to overcome important market or systemic failures and societal challenges which could not otherwise be addressed.132 Despite its broader scope of application, the Communication applies predominantly to TEN-T PCIs, for it is principally aimed at priority European transport infrastructure.133

130

ibid, para 19, sub-para. 31. IPCEI Communication (n 70) paras 4, 9, 23. ibid, para 3. 133 L Sandberg, ‘Infrastructure and State Aid in Light of the IPCEI Communication’ (2016) 17 ERA Forum 57. 131 132

262 Julia Sack, Kristina Zych and Kai Uwe Pritzsche Both the EEAG and the IPCEI Communication were adopted in July 2014 and apply until 31 December 2020. Thus, from a temporal perspective, the Commission was well aware of their potentially overlapping scopes of application. Since the EEAG specifically target, for the first time, energy infrastructure and explicitly refer to TEN-E PCIs, the argument that the IPCEI Communication fills the lacuna that would otherwise exist for guidance on State aid supporting TEN-T PCIs,134 is convincing. Together with the Broadband Guidelines135 that apply to State aid in support of TEN-Telecom PCIs, there is now not only a regulatory framework for all three Trans-European Networks and the PCIs identified therein, but also a framework for the corresponding State aid regimes for each TEN. TEN-T, not TEN-E, PCIs are explicitly mentioned and privileged under the IPCEI Communication in so far as for them (and for interconnected research infrastructures), the transEuropean importance is assumed because they are part of a physically connected cross-border network or are essential to enhance cross-border traffic management or interoperability.136 Prior to the adoption of the IPCEI Communication, provisions on public financing of IPCEIs were scattered in the R&D&I Framework137 and in the EAG 2008,138 which gave a rather fragmented guidance on the application of Article 107(3)(b) TFEU to IPCEIs. The EAG 2008 stipulated that IPCEIs that were of an environmental priority could be considered compatible with the common market if they contributed in a concrete, exemplary and identifiable manner to the Community interest in the field of environmental protection, such as by being of great importance for the environmental strategy of the European Union.139 The EEAG 2014 do not include such a provision. It could be concluded that State aid to environmental protection projects is primarily to be assessed under Article 107(3)(c) TFEU and the EEAG. Since the EEAG equally prioritise environmental protection and energy infrastructure, this argument could be applied to energy infrastructure as well. However, it should be noted that the IPCEI Communication is not an exclusive regime and leaves open the possibility that aid to promote the execution of IPCEIs may also be found compatible with the internal market on the basis of, notably, Article 107(3)(c) TFEU and its implementing rules in the EEAG.140 Thus, in the absence of an established decision-making practice, it cannot be excluded that TEN-E PCI projects will also be assessed by the Commission under the IPCEI Communication. With regard to projects of the Eureka Initiative (see below, section III.C.ii), the Commission has applied Article 107(3)(b) and (c)

134

ibid. Commission, ‘Guidelines for the application of state aid rules in relation to the rapid deployment of broadband networks’ [2013] OJ C25/1. 136 IPCEI Communication (n 70) para 16. 137 Commission, ‘Community framework for state aid for research and development and innovation’ [2006] OJ C323/1. 138 EEG 2008 (n 85). 139 ibid, para 147. 140 IPCEI Communication (n 70) para 8. 135

Projects of Common Interest 263 TFEU simultaneously. The Commission usually assessed such projects under the (c) alternative but deviated from that practice in the presence of a project also fulfilling the IPCEI criteria and felt the need to double check the measure under the regime normally applied to projects labelled Eureka for the sake of coherence.141 There is no reason why this may not also be applied in some cases to energy projects. Because of its potential application to TEN-E PCI projects, it is worth looking into the particulars of the IPCEI Communication in the following section. i. Criteria Defining an IPCEI Unlike PCIs, IPCEIs are not selected and put on a continuously evolving list. Their compliance with the eligibility criteria under the IPCEI Communication is a necessary transitional phase in order to come to the compatibility assessment of State aid to such IPCEI projects. From an early stage, and confirmed by the European Court of Justice, the Commission considered that only projects forming part of a transnational European programme supported jointly by a number of Member States or arising from a concerted action by a number of Member States to combat a common threat (such as environmental pollution) can constitute IPCEIs.142 Later on, the Commission emphasised in greater detail the four cumulative eligibility criteria that have to be met in order to assume the presence of an IPCEI in its decision on German airlines.143 The IPCEI Communication picks up those criteria and further elaborates on them. First, undertakings have to qualify as a project pursuant to paragraphs 12 and 13 of the IPCEI Communication. To that end, they must be either single projects or integrated projects. A single project has to prove it has a clearly defined objective, list of participants, implementation and funding schedules. An integrated project is a group of single projects inserted in a common structure, road map or programme aiming at the same objective and based on a coherent systemic approach whereas the individual components of the integrated project may relate to separate levels of the supply chain but must be complementary and necessary for the achievement of the important European objective. Second, the single or integrated project has to meet the eligibility criteria set out in section 3 of the IPCEI Communication. The table below (Table 4) lists these general cumulative criteria and their general positive indicators as well as the specific criteria that the project must fulfil in order to be considered an IPCEI.

141 Recherche & Développement lié au regime MEDEA+ (T 207) Commission Decision (Case N478/2003) [2005] OJ C131/13, 3.3.2. and R&D lié au régime MEDEA+ (T 404) Commission Decision (Case N8/2003) [2003] OJ C301/6, 3.2 (both decisions deal with State aid to R&D in microelectronics under the Medea+ Eureka Programme). 142 Joined Cases C-62/87 and C-72/87 Exécutif regional wallon v Commission, EU:C:1988:132. 143 Commission Decision 96/369/EC of 13 March 1996 concerning fiscal aid given to German airlines in the form of a depreciation facility [1996] OJ L146/42.

264 Julia Sack, Kristina Zych and Kai Uwe Pritzsche Table 4: Criteria and Positive Indicators of an IPCEI I. General Cumulative Criteria, sections 3.2.1, 3.3 1. Contribution to one or more Union objectives and significant impact on competitiveness of the EU, sustainable growth, addressing societal challenges or value creation across the Union. 2. Representation of an important contribution to the Union´s objectives through being of major importance for the strategies listed in paragraph 15 (among others, the Energy Strategy 2020). 3. All of the following: — —

Involvement, in principle, of more than one Member State and benefits must extend to a wide part of the Union (not only the financing Member States). Benefits shall have spill-over effects (benefits must not be limited to one sector).

4. Co-financing by the beneficiary. 5. Respect of the principle of phasing out of environmental harmful subsidies. 6. Importance of the project (section 3.3): In order to qualify as an IPCEI, a project must be important quantitatively or qualitatively. It should either be particularly large in size or scope and/or imply a very considerable level of technological or financial risk. II. General Positive Indicators, section 3.2.2 1. Possibility for all Member States to participate. 2. Involvement of the Commission or any legal body the Commission delegated the respective powers to, through any of or more than one of the following elements of the project: — — —

Design Selection Governance structure

3. Project involves important collaborative interactions. 4. Co-financing by a Union fund. III. Specific Criteria, section 3.2.3 [1. R&D&I projects, paragraph 21] [2. Industrial deployment projects, paragraph 22] 3. Environmental, energy and transport projects (paragraph 23): Great importance for/significant contribution to either of the following: — —

Environmental, energy, including security of energy supply, or transport strategy of the Union. Internal market, including, but not limited to those specific sectors.

IPCEIs should not be confused with PCIs. PCIs are, in themselves, a category of priority energy infrastructure projects with their own regulatory regime and financial support (eg, under the CEF) whereas the IPCEI status is rather awarded on a case-by-case basis and a necessary precondition for subjecting those projects under the State aid regime of Article 107(3)(b) TFEU and the IPCEI Communication.

Projects of Common Interest 265 With regard to the criteria listed above, PCIs are particularly well positioned to prove compliance thereto, since most of them are very similar to the PCI eligibility criteria. Their objective, in most cases, being the furthering of the Energy Union by completing the internal energy market contributes to increase the Union-wide competitiveness of several sectors that are relying on a stable and secure energy supply. They hence foster the goals of the Energy Strategy 2020, explicitly included in paragraph 15. PCIs usually involve more than one Member State and are assumed to create positive externalities in the shape of spillover effects for other regions or multiple sectors. PCIs are also eligible for co-financing from several Union funds; however the project promoters (beneficiaries) must also provide financing. ii. Examples of IPCEIs It is obvious from the eligibility criteria that there is a high bar for a project to be considered an IPCEI. As a result, there is only a fraction of the amount of PCI examples available to illustrate the character of such important projects of common European interest. One early example is the Channel Tunnel High Speed Rail Link (CTRL) connecting London to the Channel Tunnel that links the UK and France. The Commission Decisions acknowledged that this transport infrastructure is a concrete, specific and clearly defined project of common European interest that is deemed important, quantitatively as well as qualitatively. As a result, it approved several State aid measures granted to the CTRL under (now) Article 107(3)(b) TFEU.144 Another ‘typical example’ given by the Commission for what actually can be considered an IPCEI is the European Spallation Source linear proton accelerator (ESS) in Sweden, which is a project involving 17 Member States and the EU.145 State support to the Eureka European Network however is, in the majority of cases, not assessed under Article 107(3)(b) but (c) TFEU, which indicates the Commission’s reluctance to assume the Initiative is an IPCEI in itself. The Eureka Initiative is an intergovernmental initiative of 17 countries launched in 1985, which aims to enhance European competitiveness by supporting businesses, research centres and universities which carry out pan-European projects to develop innovative products, processes and services. It provides project partners with rapid access to knowledge, skills and expertise across Europe and facilitates access to national public and private funding schemes. In 1988, the Commission received 18 notifications for State measures in support of the Eureka Initiative. Among those Eureka schemes or projects, it considered only the Netherlands aid for the development of a common standard for high definition colour television (HDTV) to support an IPCEI.146 More recently,

144 Channel Tunnel Rail Link Commission Decision (Case N706/2001) [2002] OJ C130/4; Channel Tunnel Rail Link (CTRL IV) Commission Decision (Case N523/2002) [2002] OJ C262/3; Channel Tunnel Rail Link—Hedging for section 1 securitisation Commission Decision (Case N687/2002) [2003] OJ C216/2; EWSI Channel Tunnel Freight Support Funding Commission Decision (Case N159/2005) [2005] OJ C314/2. 145 Commission, ‘State aid: Commission facilitates support for important projects of common European interest’ MEMO/14/423. 146 European Commission, 18th Report on Competition Policy 1988 (Luxembourg, The Commission, 1989) para 172.

266 Julia Sack, Kristina Zych and Kai Uwe Pritzsche other Eureka projects have been classified as IPCEIs and assessed under (b) as well, for example (in the field of microelectronics) a project aiming at the development of a 65 nanometric complementary metal-oxide semiconductor.147 Under the IPCEI Communication, the Commission so far seems to have assessed (and approved) State aid to only two IPCEI projects in 2015. These projects are at the same time both TEN-T PCIs and key components in the main north–south route between central Europe and the Nordic countries: the Øresund Link and the Fehmarn Belt. In case of the Øresund Link, Denmark and Sweden granted funding through State guarantees and tax advantages that were assessed by the Commission and found compatible with Article 107(3)(b) TFEU and (by analogy) the IPCEI Communication.148 The Fehmarn Belt is supported by the Danish State with State guarantees and State loans. This State support was also found compatible with the regime of Article 107(3)(b) TFEU and the IPCEI Communication by the Commission.149 Both the Øresund Link and Fehmarn Belt are among the 30 top TEN-T priority projects (mentioned in section II.A.iii above). The other 28 projects include highspeed and freight railway axes, motorway axes, multimodal axes, waterway axes, inland waterways and strategic airports. It is likely that other projects from that list would fulfil the IPCEI eligibility criteria so that State aid to such projects could be assessed under the IPCEI Communication. As mentioned above, Project 21 on that list (Motorways of the Sea) includes 18 single TEN-T undertakings that have links to energy or environmental protection and might therefore, in principle, fall under the EEAG and/or the IPCEI Regime. iii. State Aid Assessment Once it has been determined with the criteria outlined above that the beneficiary in question is an IPCEI, it must be assessed whether State support granted to that project is compatible with the internal market. The Commission will apply compatibility assessment principles outlined in the IPCEI Communication to all notified aid projects,150 even where the projects were notified prior to the publication of the IPCEI Communication.151 On the whole, the compatibility assessment under both regimes basically follows the same steps. The objective of common interest and the presence of a market failure (or other important systemic failures) will be presumed based on the project’s compliance with the IPCEI eligibility criteria.152

147

MEDEA+ (T 207) (n 141). State Aid granted to Øresundsbro Konsortiet Commission Decisions (Case SA.36662, Case SA.36558, Case SA.38371) [2014] OJ C437/1. 149 Financing of the Fehmarn Belt Fixed Link Project Commission Decision (Case SA.39078) [2015] OJ C325/1. 150 However, for non-notified State aid, the European Commission will apply the IPCEI Communication if the aid was granted after its entry into force pursuant to para 52 IPCEI Communication and by analogy if the aid was granted prior to its entry into force (State Aid granted to Øresundsbro Konsortiet, see n 148). 151 IPCEI Communication (n 70) para 51. 152 ibid, para 27. 148

Projects of Common Interest 267 With regard to the appropriateness, the IPCEI Communication stipulates that repayable aid instruments will generally be considered as a positive indicator and gives the examples that for a lack of access to finance liquidity support, such as a loan or guarantees, and for a boost of credibility through risk-sharing, a repayable advance would be the appropriate instrument of choice to overcome the market or systemic failure.153 Where beneficiaries are selected through a competitive, transparent and non-discriminatory tender, this will be considered as a positive indicator of the appropriateness.154 Aid to IPCEIs must have an incentive effect. Like aid to PCIs under the EEAG, it must not subsidise the costs that a project would incur anyway or that merely compensate for the normal business risk of the economic activity. The aid application must precede the start of the works, which is any commitment that makes the investment irreversible, with the exception of the purchase of land and preparatory works, such as obtaining permits and conducting preliminary feasibility studies.155 To assess the incentive effect, the Commission will consider the counterfactual scenario.156 Where the aid beneficiary faces a clear choice between carrying out either an aided project or an alternative project without aid, the Commission will compare the expected net present values of the investment in the aided project and the counterfactual project including considerations of the probabilities of the different business scenarios.157 In the absence of an alternative project, the Commission will assess whether the aid amount exceeds the minimum necessary for the aided project to be sufficiently profitable in comparison to the normal rates of return required by the same beneficiary in other investment projects of a similar kind or returns commonly observed in the industry concerned.158 An insufficient level of profitability serves as an indicator for the incentive effect.159 As to the proportionality of the aid, the maximum aid level will also be determined with regard to the identified funding gap in relation to the eligible costs. If justified by the funding gap analysis, public support may cover up to 100 per cent of the funding gap on the basis of a large set of eligible costs laid down in the Annex to the IPCEI Communication.160 Eligible costs include the costs of feasibility studies; of instruments, equipment, other materials, supplies and similar products; of the acquisition (or construction) of buildings, infrastructure and land; for obtaining, validating and defending patents and other intangible assets as well as personnel and administrative costs directly incurred for research, development and innovation (R&D) activities. Operating aid is only considered legitimate if it is related to a project of first industrial deployment, as long as the industrial deployment follows on from and itself contains a very important R&D component which constitutes an integral and necessary element for the successful implementation of the project. Unlike the EEAG, the IPCEI Communication does not list maximum aid intensities. 153 154 155 156 157 158 159 160

ibid, para 36. ibid, para 39. ibid, para 28. ibid, para 29. ibid, para 32. ibid, para 30. ibid, para 33(b). ibid, para 31.

268 Julia Sack, Kristina Zych and Kai Uwe Pritzsche The maximum aid intensity of a State support measure may not exceed the funding gap ratio, which will be calculated as the funding gap divided by the net present value of the eligible investment costs.161 The Commission will consider projects more favourably that include a significant co-funding by the beneficiaries or by independent private investors.162 When balancing the negative effects of the aid measure against the dimension of its contribution to the objective of common European interest, the Commission will check the avoidance of undue distortions. In assessing the negative effects, it will focus its analysis on the foreseeable impact the aid may have on competition between undertakings in the product markets concerned (including up- or downstream markets) and on the risk of overcapacity.163 Projects involving the construction of an infrastructure must ensure open and non-discriminatory access to the infrastructure and non-discriminatory pricing.164 The Commission not only checks the effects of the aid on the product markets concerned, but also considers the risk of a subsidy race between Member States that may arise in particular with respect to the choice of a location.165 It also does not limit its considerations to distortion of intra-Union trade and can acknowledge the competitive advantage enjoyed by a third-country competitor. The Commission may take account of the fact that, directly or indirectly, competitors located outside the Union have received (in the last three years) or are going to receive, aid of an equivalent intensity for similar projects.166 Finally, the IPCEI Communication also sets out transparency obligations that Member States have to comply with from 1 July 2016 onwards.167 These obligations can be waived with respect to individual aid awards below €500,000.168

IV. CONCLUSION

Trans-European energy projects are considered crucial for the achievement of the EU’s energy and climate policy objectives and in particular for completing the internal energy market. Therefore, in 2013, the Regulation on Guidelines for trans-European energy infrastructure (TEN-E Regulation) was adopted in order to provide for the necessary support for such energy projects. PCIs, if they qualify under the TEN-E Regulation, are subject to a special regime for the planning and permitting processes and may also qualify for financial support. EU State aid rules under Article 107 TFEU in principle apply to such energy projects as to any other project. They prohibit aid granted by EU Member States which distorts competition by favouring market participants if it affects trade between

161 162 163 164 165 166 167 168

Financing of the Fehmarn Belt Fixed Link Project (n 149) para 107. IPCEI Communication (n 70) para 38. ibid, para 42. ibid, para 43. ibid, para 44. ibid, para 34. ibid, para 45. ibid, para 46.

Projects of Common Interest 269 Member States. However, importantly for cross-border energy projects, exceptions may apply to: —



Projects of common interest (PCI) which facilitate the development of an integrated European energy market by integrating across borders EU Member States’ energy markets (see Article 107(3)(c) TFEU). Important projects of common European interest (IPCEI) which are transnational projects of strategic significance for the EU and for the achievement of Europe 2020 objectives in all sectors of economic activity, including environmental protection and energy (see Article 107(3)(b) TFEU).

In order to bring the support for PCIs and IPCEIs in line, among other things under the TEN-E Regulation, in 2014 the Commission bound itself in its assessment practice by the EEAG and by the IPCEI Communication. Energy projects, and in particular PCIs, are mostly considered and assessed under the EEAG. PCIs, if they qualify under the Regulation on Guidelines for Trans-European energy infrastructure (TEN-E Regulation), receive support for the planning and permitting processes and may also qualify for financial support. PCIs and IPCEIs may also receive EU funding under programmes like the Connecting Europe Facility, the European Regional Development Fund and the EU Cohesion Policy Funds. However, State aid rules do not apply to this EU funding if it is also managed only by the EU as Article 107 TFEU only applies to aid granted by EU Member States. PCIs benefit from certain presumptions with regard to the criteria of the aid’s compatibility with the internal market, among those criteria, and in particular the project meeting the objective of the common EU interest, assessing the need for State intervention, the necessary incentive effect of the aid to the beneficiary and the aid’s proportionality and appropriateness as well as a lack of undue distortion of competition. Hence, PCIs are more readily declared compatible with the internal market by the Commission than other energy or environmental protection projects which do not meet the criteria for PCIs.

10 Services of General Economic Interest and EU State Aid Control in Energy Markets ADRIEN DE HAUTECLOCQUE, FRANCESCO MARIA SALERNO AND SIMINA SUCIU

I. INTRODUCTION

S

ERVICES OF GENERAL Economic Interest (SGEIs) have been a pillar in the architecture of the EU from its very inception and, in some respects, are at the core of the social market economy model that is associated with the EU. The declarations in the Lisbon Treaty about the importance of SGEIs are therefore nothing more than the latest step in a journey that began with the foundation of the EU. Yet, as Mario Monti put it in 2010 ‘[s]ince the nineties, the place of public services within the single market has been a persistent irritant in the European public debate’.1 This tension stems from two main sources. First, SGEIs are common in telecommunications, energy, transport and postal services, all sectors in which the EU has, since the 1990s, sought to create a liberalised market governed by the rules of competition, hence the difficult balance between free market and State responsibility for delivering what are regarded as essential services. The second source of tension is between Member States and the EU. Statements about SGEIs being the precinct of Member States, with EU institutions—the European Commission and Courts—having the limited role of checking for a ‘manifest error’, abound in decisions and judgments. However, given the extent to which the EU has regulated these sectors, more often than not Member States and EU institutions have clashed over the boundaries of what is and what is not an SGEI, and whether Member States have indeed committed a ‘manifest error’. In an effort to improve clarity, in 2012 the European Commission adopted a new SGEI package in the framework of State aid control (replacing the Monti–Kroes 2005 package) which includes a number of instruments: —

A new Communication that clarifies key State aid concepts as regards SGEIs.2

1 M Monti, A New Strategy for the Single Market. At the Service of Europe’s Economy and Society. Report to the President of the European Commission Jose Manuel Barroso (9 May 2010) 73. 2 ‘Communication from the Commission on the application of the European Union State aid rules to compensation granted for the provision of services of general economic interest’ [2012] OJ C8/4 (Communication).

272 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu —

A new de minimis Regulation, pursuant to which public service compensation under €500,000 over three years is deemed to be free of aid within the meaning of Article 107 TFEU because of its lack of effect on the internal market.3 — A revised Decision that simplifies the conditions under which public service compensation is considered to be compatible with the internal market pursuant to Article 106(2) TFEU and escapes the notification obligation under Article 108(3) TFEU.4 — A revised Framework that outlines the conditions under which large compensation amounts, falling outside the scope of the Decision, must be notified to the Commission for thorough assessment and could be found compatible with the internal market.5 State aid for SGEIs not covered by the Decision can be declared compatible under Article 106(2) TFEU. However, the SGEI package provides only limited guidance on the notion of an SGEI, confining itself to the usual deferential statement to the effect that ‘[i]n the absence of specific Union rules defining the scope for the existence of an SGEI, Member States have a wide margin of discretion in defining a given service as an SGEI and in granting compensation to the service provider’.6 Behind the surface of these declarations, it is submitted that the EU courts and the Commission have laid the groundwork for defining the boundaries of what is an acceptable definition of an SGEI from an EU perspective, thereby reducing the ‘wide discretion’ allegedly enjoyed by Member States. The aim of this chapter is to analyse court cases and Commission acts (decisions as well as the package described above) to distil this trend. The energy sector offers an excellent observation point to address this question. At the same time, because SGEIs cover a wide variety of activities and this chapter focuses on energy, the findings do not purport to cover all the many facets of this complex debate. This chapter is structured as follows: section II discusses the emerging body of case law and Commission practice from which one draws a first approximation to an EU-wide notion of SGEIs. Section III takes the findings of section II on the notion of an SGEI and links them to the case law and Commission practice on State aid, as Member States often justify aid as a form of compensation for the discharge of an SGEI. Section IV discusses SGEI compensation as a basis for compatible aid. Section V provides concluding remarks, drawing on the findings of the previous sections.

3 Commission Regulation (EU) 360/2012 of 25 April 2012 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid granted to undertakings providing services of general economic interest [2012] OJ L114/8. 4 Commission Decision (EU) 2012/21 of 20 December 2011 on the application of Article 106(2) of the Treaty on the Functioning of the European Union to State aid in the form of public service compensation granted to certain undertakings entrusted with the operation of services of general economic interest [2012] OJ L7/3. 5 ‘Communication from the Commission on the European Union framework for State aid in the form of public service compensation’ [2012] OJ C 8/15 (2012 Framework). 6 ‘Communication from the Commission of 11 January 2012 on the application of the European Union State aid rules to compensation granted for the provision of services of general economic interest’ [2012] OJ C 8/4, para 46.

Services of General Economic Interest 273 II. THE NOTION OF SGEIS

The existence of a genuine SGEI is (i) the first requirement of the ‘Altmark’ case law (see below at section III.B),7 and (ii) the first step in the assessment of compatibility of aid in light of Article 106(2) TFEU. But the question of the existence of a genuine SGEI can also be looked at without any aid being involved, for example, when the question arises purely in the context of a regulatory obligation, such as the obligation to supply electricity to final customers at affordable prices. Thus, this section looks at the case law and Commission practice on the existence of a genuine SGEI as a stand-alone question, while the next section will add the State aid dimension. As the question of the existence of a genuine SGEI cuts across sectors, where necessary, reference is made to case law and/or decisions pertaining to sectors other than energy.

A. The Existence of a Market Failure as a Pre-condition for Designating an Activity as an SGEI A good starting point for sketching the notion of an SGEI in EU law is the General Court’s stance in Case T-79/10, COLT,8 where it stated that, ‘l’appréciation de l’existence d’une défaillance du marché constitue un préalable à la qualification d’une activité de SIEG et ainsi à la constatation de l’absence d’aide d’État’.9 This matches the position in the SGEI Communication, where the Commission stated that it would not be appropriate to attach specific public service obligations to an activity which is already provided or can be provided satisfactorily and under conditions, such as price, objective quality characteristics, continuity and access to the service, consistent with the public interest, as defined by the State, by undertakings operating under normal market conditions.10

In identifying which market failures justify State intervention in the energy sector, Hinkley Point C sheds further light on the issue. In its Opening Decision,11 the Commission doubted that the aid addressed a market failure related to electricity generation or to nuclear energy. In assessing whether the measure addressed market failures related to electricity generation, the Commission seemed not to have sufficient information to understand how the measure interplayed with other already existent EU or UK policies, whether there were other less distortive technologies at stake and how the measure addressed potential failures

7

Case C-280/00 Altmark Trans and Regierungspräsidium Magdeburg, EU:C:2003:415 (Altmark). Case T-79/10 France v Commission, EU:T:2013:463. Note that this judgment has not been published in the Court’s report and has not been appealed. 9 para 154. In English, ‘the assessment of the existence of a market failure is a prerequisite for classifying an activity as a SGEI and thus for finding that there is no State aid’. 10 para 49. 11 Support to Hinkley Point C New Nuclear Power Station (Case SA.34947) Commission Decision [2014] OJ C69/60 (Opening Decision). 8

274 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu (such as carbon emission and pollution standards). Additionally, because potential market failures related to ‘electricity generation [were] not sufficient to justify state intervention to support nuclear generation’,12 the Commission went on to assess potential market failures specific to nuclear generation. Although the investment and long duration that characterise nuclear power plants are ‘unparalleled’, the Commission was not convinced that tail risks and catastrophic risks qualified as market failures. In the attempt to distance nuclear energy technology from other types of electricity generating technology, doubts were expressed in relation to the underlying costs pertaining to radioactive material and the potential for nuclear accidents. However, the Commission considered in its Final Decision13 that the project could not be built without public support because of the ‘unprecedented nature and scale’. More specifically, the lack of market-based financial investments to hedge against risks related to high capital costs, lengthy construction times and long periods of operation to recover the investment, constituted a market failure.14 Interestingly, the Commission considered that the risk of political hold-up that an investor might incur did not merit its own characterisation as a market failure but it could be qualified as a factor making investment in nuclear more difficult. The Commission’s Final Decision has been appealed before the General Court by Greenpeace Energy and Austria. Greenpeace Energy claimed that (i) the Commission wrongly found the existence of a market failure resulting from the alleged impossibility of financing the nuclear power station on the financial markets, and (ii) it neglected other nuclear power stations, including those using the same technology, that managed without comparable State aid. However, the General Court dismissed the order in its entirety as inadmissible, as the applicants were not individually concerned by the contested decision, without going into detail about the two claims referred to above. In particular, the Court held that while the applicants and Hinkley Point C were competitors (because electricity produced from renewable sources, through cogeneration or from other primary sources, is part of the same relevant market),15 the effects of the measure on the applicants’ competitive situation could not be differentiated from the competitive situation of other operators affected by the aid measure.16 Similarly, Austria argued that the market was wrongly defined and an incorrect assumption of market failure was applied, and highlighted the incorrect qualification

12

ibid, para 269. Support to Hinkley Point C New Nuclear Power Station (Case SA. 34947) [2014] OJ L109/44 (Final Decision). 14 ibid, paras 382 ff. 15 para 64. The nature of electricity and the difficulty in determining the energy source once electricity has been put in the grids/distribution network, the lack of reference in the Electricity Directive 2009/72 on the potential sub-division of the electricity EU market into sub-markets based on production method, the priority access for renewable energy installation, which ensures that the renewable source is integrated in the EU market, were considered of relevance in reaching this conclusion. 16 Case T-382/15 Greenpeace Energy and Others v Commission, EU:T:2016:589. This matter is pending before the Court of Justice in Case C-640/16 P, Greenpeace Energy v Commission. 13

Services of General Economic Interest 275 of a nuclear station as a ‘new technology’.17 This appeal is pending and it will be interesting to see whether the General Court reverses the Commission’s finding on market failure in nuclear power (which is indeed a mature technology). The existence of a market failure may thus be considered a constituent element of the EU notion of SGEIs, at least for the Commission. By the same token, the absence of a market failure limits the material scope of the public service mission that a Member State may devise; if the activity can be provided by the market, then it cannot be an SGEI.

B. Entrustment and the Compulsory Character Although somewhat useful, the market failure préalable is still wanting to filter between ‘normal’ economic activities and SGEIs, because a market failure is also the justification for providing aid for any economic activity that a Member State seeks to promote. And this aid can be cleared under Article 107(3) TFEU, not Article 106(2) TFEU. A quid pluris factor to distinguish an SGEI from any other economic activity may appeal more to the lawyer than to the economist: the compulsory character. Once a Member State is confronted with a market failure in connection with an activity that it deems essential for its citizens, the next step is normally to use public law powers to identify one or more undertakings that can fill the gap left by market forces. This implies that a genuine SGEI is compulsory in nature for the undertakings that must deliver the service. The compulsory nature sets an SGEI apart from a ‘normal’ economic activity for which a Member State may provide aid ex Article 107(3) TFEU. While, by providing aid, the Member State seeks to modify the incentive structure of undertakings, but accepts that undertakings remain free to take the aid or not, in the case of SGEIs the Member State acts to ensure that the service is delivered. For this reason, an SGEI mission must be enshrined in an act of entrustment, which is a public law act, whereby the obligated party or parties, their obligations, and the sanctions for non-compliance must be clearly designated. In practice, this means that a blanket designation of a sector as being of ‘general economic interest’ is not enough to claim a valid SGEI designation, which entails a clear demarcation of the obligations within the sector that a Member State considers to be an SGEI and the obligated party or parties within the sector. The Hungarian PPA case is relevant for this point. When dismissing an argument about the existence of an SGEI in connection with electricity generation, the Commission found that the interested parties failed to submit any document of the Hungarian State which would clearly define a SGEI and entrust a specific generator (or several generators) with providing that precise service. The PPAs themselves are similar from this aspect and they fix the obligations of the parties but do not define a specific public service obligation. The fact that all the ten power plants under PPAs have to reserve their capacities to MVM [the single buyer], does not in itself mean that they are specifically entrusted with a public service obligation.

17

Case T-356/15 Austria v Commission, pending.

276 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu Again, such approach could lead to conclude that the whole sector of power generation fulfils a SGEI—which would clearly be in breach of the spirit Community legislation and practice means to give to such notion.18

A failure to draw up an act of entrustment and/or to identify clearly the obligated party or parties for delivering the SGEI in question constitutes a manifest error by a Member State. The act of entrustment need not be contained in a single act, but can be composed of, for example, an act laying out the requirements in general terms, and then single implementing acts identifying the obligated parties and their obligations. This can be the case, for instance, when the SGEI provider is chosen through tenders. However, it must be clear that the obligations stem from a public law act, not from a private law contractual framework, which the parties remain free to enter into or not, and for which the only rules applicable in case of non-compliance are those of agreements between private parties under civil law (termination and damages). For instance, a public law body may decide to buy products or services and put forward a tender process to select providers. This is not enough to automatically turn these contracts into SGEI acts of entrustment. Otherwise, any contract in which a public law body is a part would be an SGEI act of entrustment. Thus, besides the activity falling within the material scope of what can be validly designated as an SGEI, the quid pluris factor is the compulsory nature; the designation of the service as an SGEI must take place in the framework of public law and the act must specify sanctions for non-compliance which go beyond the means available to private parties in a private law contract. For instance, the public body could have the power to compel the obligated parties into performing the service through sanctions or public orders. The energy sector has probably one of the clearest applications of these principles. In Hinkley Point C, the UK sought to argue that ensuring the investment in new generation nuclear capacity to be delivered within a specific timeframe was an SGEI. In dismissing this argument, the Commission primarily relied on the analysis of the contract with the public authorities. In particular, the Commission found that despite the special nature of the project, the contractual provisions are typical contractual obligations that any contractual parties would try to include in a similar deal, rather than a public service obligation imposed by the UK authorities. NNBG [NNB Generation Company Limited] is actually not obliged to build the nuclear plant, nor is it obliged to build it by a certain date. The UK authorities cannot enforce any obligation in this respect; they can only terminate the contract.

The emphasis on the remedial action available (termination) clearly illustrates the difference—in the Commission’s mind—between a ‘simple’ private contract and an SGEI.

18 Commission Decision of 25 September 2007 on State aid awarded by Poland as part of Power Purchase Agreements and the State aid which Poland is planning to award concerning compensation for the voluntary termination of Power Purchase Agreement [2009] OJ L83/1.

Services of General Economic Interest 277 Moreover, the Commission found that there is no obligation imposed on HPC [Hinkley Point C—the power plant] to produce electricity, to produce a certain amount of electricity or to make that electricity available on the market. Indeed, under the CfD, HPC will have high incentives to produce as much electricity as possible to increase its gains, but it is not obliged to do so … The Commission considers that these conditions cannot be viewed as public service obligations or as demonstrating that NNBG is be entrusted with a SGEI.

This reasoning encapsulates the difference between incentivising a ‘normal’ economic activity through aid, and submitting it to the rigors of SGEI designation. The Commission has made a similar argument in the decision to open an investigation on a French capacity payments case.19 In the Final Decision, the Commission reiterated its arguments, but did not have to take a final position because the French authorities did not make any new comments.20

C. Security of Energy Supply and the SGEI Scope Court cases and Commission decisions show that in the energy sector SGEI designation almost invariably comes with a link to security of supply. ‘Security of supply’ constitutes both a pillar of the EU energy policy (Article 194 TFEU)21 and an objective of common interest which may justify derogation from the rules of free movement.22 These cases bring to life the courts’ holding that a Member State needs to show what is distinct about the activity that makes it worthy of an SGEI designation.23 In the energy sector, this distinctiveness is often linked to security of supply, but a Member State has an obligation to define precisely in what way the activity that it wishes to designate as an SGEI contributes to the goal of security of supply. Selected cases illustrative of this approach are presented below. i. Electricity Generation from a Specific Fuel and Security of Supply Electricity generation should not, in principle, be amenable to SGEI designation because it is a liberalised activity which can be provided by the market. There have been, however, a few exceptions to this principle based on security of supply considerations. What these cases have in common is that the security of supply argument

19

Mécanisme de capacité en France (Case SA.39621) [2015] OJ C46/35. Mécanisme de capacité en France (Case SA.39621) [2016] paras 206–08. 21 ‘In the context of the establishment and functioning of the internal market and with regard to the need to preserve and improve the environment, Union policy on energy shall aim, in a spirit of solidarity between Member States, to: (a) ensure the functioning of the energy market; (b) ensure security of energy supply in the Union’. 22 See, eg, Case 72/83 Campus Oil Limited and Others v Minister for Industry and Energy and others, EU:C:1984:256, paras 32–51; Case C-503/99 Commission v Belgium, EU:C:2002:328, paras 46–55. 23 See, eg, Case T-289/03 BUPA v Commission, EU:T:2008:29, para 178, quoting Case C-172/80 Gerhard Züchner v Bayerische Vereinsbank AG, EU:C:1981:178, para 7 and case C-7/82 GVL v Commission, EU:C:1983:52, paras 31–32. 20

278 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu was predicated on certain specific qualities of the fuel used to generate electricity. An analysis of these cases shows in what way security of supply was considered and attempts to draw more general lessons. In 2010, the Commission accepted that production of electricity from an indigenous primary energy resource—in this case coal—was an SGEI.24 In Castelnou,25 the court case that followed the Commission decision in Spanish Coal, the General Court held that, when a Member State invokes a risk of security of supply as a justification for an SGEI, the standard of proof that it must meet is one of a ‘sufficiently plausible’ threat.26 The Court then engaged in a detailed review of the elements provided by Spain—and the Commission’s assessment thereof—to substantiate ‘the existence of specific risks jeopardising security of electricity supply’,27 thus showing that ‘security of supply’ is not an automatic passe-partout for any measure to withstand EU law scrutiny. The Commission followed the Court’s exacting stance on the burden of proof when invoking the security of supply argument for an SGEI in the 2011 first Delimara case, where it found that it did not have sufficient evidence to determine whether €15.5 million in financing from the European Regional Development Fund to support the modification of two boilers at Delimara power station contributed to enhancing security of supply or whether the imposed Public Service Obligation (PSO) was defined without manifest error.28 More specifically, the Commission concluded that Malta’s arguments were not substantiated and no evidence was put forward in relation to, inter alia, how long and to what extent the capacity ensured by the boilers was needed in order to maintain security of supply.29 In 2017, Malta was able to win State aid approval for a new measure, where it was able to show Malta’s specific handicaps (‘the Maltese energy market cannot adequately deliver the necessary level of service due to its particularly small and isolated character’). The precedent on generation from indigenous fuel sources as an SGEI raises interesting questions as regards the possibility of designating generation from RES as an SGEI. After all, RES are indigenous and their contribution to security of supply is arguably even more important than that of other sources which are subject to depletion like coal.

24 Spain Public service compensation linked to a preferential dispatch mechanism for indigenous coal power plants (Case N178/2010) [2010] OJ C312/6. This exception to the general principle of generation being a market activity is foreseen in EU regulation (see Art 11(4) of Directive 2003/54/EC and the identical provision in the Third Electricity Directive—‘[a] Member State may, for reasons of security of supply, direct that priority be given to the dispatch of generating installations using indigenous primary energy fuel sources, to an extent not exceeding in any calendar year 15% of the overall primary energy necessary to produce the electricity consumed in the Member State concerned’). In the Slovenian tariff case (Case C 7/2005 Slovenian Electricity tariffs [2007] OJ L219/9) the Commission had already made an application of this provision. 25 Case T-57/11 Castelnou Energía v Commission, EU:T:2014:1021 (Castelnou). 26 ibid, para 136. 27 ibid, paras 137 et seq. 28 Case C 32/10 Malta, Environmental project for Delimara power station [2011] OJ 52/3, para 97 (Delimara power station). This case was assessed under the 2005 SGEI Framework. The notification was later withdrawn. 29 ibid, paras 87, 90–01, 96.

Services of General Economic Interest 279 While no Member States so far have invoked SGEI qualification for RES, if ever one Member State were to do so, such an attempt would probably fail for the following reasons: —

When an activity is open to competition, State measures distorting competition on public interest grounds must be proportionate in the sense that the distortion should be the least possible to achieve the goal at hand. — If a Member State detects a market failure in connection with the generation of electricity from RES, the least distortive measure is to provide incentives for generators to overcome the market failure. — Only if this measure failed, should a Member State consider imposing an SGEI, which is, by definition, a more restrictive measure because of its obligatory character, compared with the mere provision of an incentive. Finally, the insertion of a new Article on energy in the TFEU, sanctioning Member States’ freedom to choose their energy mix, adds further complexity to the issue,30 because, if a Member State were to argue that its energy policy is to have electricity from a certain source, and assuming market forces did not deliver that goal, it would probably be difficult for the Commission to object, even if the chosen type of fuel were to be considered a mature technology. As regards nuclear, in Hinkley Point C, the UK government put forward that the investment in new nuclear could be designated as an SGEI, because it contributed towards the UK’s security of supply of low-carbon electricity on a long-term basis, [it was supposed] to be delivered and operated within a specific timeframe and its operation within the framework of the CfD [was] directed towards achieving a general or public interest … [N]ew baseload capacity and in particular nuclear projects [were] not provided by undertakings operating under normal market conditions in a timescale sufficient to meet the UK’s general interest objectives.31

In its Opening Decision, the Commission seemed reluctant to find there was an SGEI, as the market could have provided nuclear energy (‘baseload electricity at technical standards consistent with the TSO specifications’)32 without the proposed measures. When reaching the conclusion that investment in new generation nuclear capacity to be delivered within a specific time frame did not constitute a genuine SGEI, the Commission seemed to suggest that compatibility under Article 107(3)(c) TFEU was the preferred approach: The Commission considered that a CfD is an appropriate means of granting State aid for electricity generation that was approved as compatible with the internal market in accordance with Article 107(3)(c) TFEU. Therefore, there would be no reason for the Commission to distance itself for the assessment performed therein and consider that support to electricity production by way of a CfD could be subject to an SGEI (emphasis added).33

30 See L Hancher and F Salerno, ‘Energy Policy After Lisbon’ in A Biondi, P Eeckhout and S Ripley (eds), EU Law After Lisbon (Oxford, Oxford University Press, 2012). 31 Final Decision, paras 177–78, 272–73. 32 Opening Decision, para 202. For a more detailed discussion on baseload electricity versus reserve capacity, see paras 102 ff. 33 Final Decision, para 307. In the same vein, see ibid, paras 308–09.

280 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu While the Commission did not question the characterisation of security of supply as a common objective in its Opening Decision, it initially doubted its application in this case due to mismatching factual information around the predicted shortfall in demand and the time when the plant would have been available, and it did not know whether there were other technologies that addressed the need for new capacity. However, without going into detail, the Commission found in its Final Decision that ‘the measure contribute[d] to long-term security of supply’ due to capacity forecast and the plant’s role once operating.34 ii. Specific Type of Electricity Generation as a Source of Security of Supply In 2003, the Commission accepted that ‘bringing to the Irish electricity grid new electricity reserve generation capacity’ could be validly considered an SGEI.35 According to the Commission, [o]ne may indeed wonder whether investors are prepared to invest in peaking capacity to cover the very highest periods of demand or incidents where a large proportion of other generation is not available. It is arguable that such investment might not occur because such events are infrequent and their occurrence is unpredictable. Accordingly, there may be a case for governments to provide further measures, in addition to market mechanisms, to ensure adequate capacity is available.36

This early finding could be important in today’s markets where the massive deployment of renewables calls for electricity generators to provide a specific service; an availability to act as a back-up. The Guidelines include a section on capacity mechanisms,37 which are measures geared towards security of supply, thus suggesting that such measures qualify under Article 107(3) TFEU, but not as SGEIs under Article 106(2). In the recent Commission decision opening an in-depth investigation into the French capacity remuneration system, in response to an argument that the capacity payments were the remuneration for an SGEI, the Commission did not take a position, opting instead to attack the lack of a compulsory character.38 The Final Decision, however, does not elaborate any further on the potential SGEI aspects.39 By contrast, in the Staff Working Document on Generation Adequacy in the internal electricity market—guidance on public intervention, accompanying the Communication delivering the internal electricity market and making the most of

34

ibid, paras 373–74. Ireland public service obligation in respect of new electricity generation capacity for security of supply (Case N475/2003) [2004] OJ C34. In 2004, Ireland adopted a temporary measure for the provision of the reserve capacity. This temporary solution was the imposition on the TSO of a public service obligation consisting mainly of the procurement of temporary generation units. The Commission found that the Altmark case law was not applicable because the fourth condition was not met. However, it authorised the measure as compatible aid based on Art 106(2) TFEU. See Ireland public service obligation—Electricity supply board (ESB) (Case N143/2004) [2005] OJ C242/5. 36 Ireland public service obligation in respect of new electricity generation capacity (n 35) para 60. 37 See ch 6 regarding capacity mechanisms and auctions. 38 Mécanisme de capacité en France (n 19) para 132. 39 ibid, paras 206–08. 35

Services of General Economic Interest 281 public intervention,40 the Commission recognised that ‘public intervention to promote generation adequacy may entail public service obligations imposed on generators, suppliers and/or TSOs’.41 It thus remains open whether a capacity remuneration system properly designed as regards its compulsory character could be validly designated as an SGEI in spite of the Commission position in the Guidelines.42 iii. Energy Infrastructure and Gas Security of Supply: Diversification The most important case where an energy infrastructure project won SGEI designation is the 2013 Lithuania LNG terminal case.43 Lithuania is dependent on a single gas supplier (Gazprom). The Commission agreed with Lithuania that the obligation to operate the LNG terminal under non-discriminatory conditions at regulated tariffs was necessary in order to ensure ‘security of supply through diversification of gas sources’ (emphasis added).44 Qualification of the operation of the terminal as an SGEI was criticised because the LNG market was open, internationally competitive, commercial and without market failure. The Commission stated that [w]hile [it] does not deny that the LNG market in general is an open and internationally competitive market, it also notes that Lithuania has so far no access to that market. Rather, it is fully dependent on one single supply source, which raises security of supply issues.45

It is difficult to gauge whether this case has precedential value, due to its specificities. However, it is interesting that security of supply in the gas context was interpreted as the diversification of gas supply.

D. EU Regulation and SGEI Designation The SGEI Communication states that ‘[w]here specific Union rules exist, the Member States’ discretion [in defining an SGEI] is further bound by those rules’.46 For instance, in Germany v Commission,47 the Court struck down an SGEI designation because it was incompatible with the ‘polluter pays principles’, which is the overarching principle in the field of environmental regulation. For the energy sector, the EU legislator’s solution to the conundrum of respecting Member States’ prerogatives in SGEI in sectors harmonised through EU legislation

40 Commission Staff Working Document of 05 November 2013, available at: ec.europa.eu/energy/ sites/ener/files/documents/com_2013_public_intervention_swd01_en.pdf. 41 ibid, 3. 42 See L Hancher, A de Hauteclocque and M Sadowska (eds), Capacity Mechanisms in the EU Energy Market: Law, Policy, and Economics (Oxford, Oxford University Press, 2015). 43 Klaipedos Naltfa LNG Terminal (Case SA.36740) [2016] OJ C161/2, para 134 (LNG Terminal). 44 ibid, para 202. 45 ibid, paras 206–11. 46 ibid, para 46. 47 Case T-295/12 Germany v Commission, EU:T:2014:675.

282 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu has been to enshrine public service objectives as a justification for national regulation in the EU legislation directly. Thus, while Article 3(1) of Directive 2009/72/ EC and Directive 2009/73/EC provides that the sector is open to competition, both Directives state that [h]aving full regard to the relevant provisions of the Treaty, in particular Article 86 thereof [now 106 TFEU], Member States may impose on undertakings operating in the electricity [gas] sector, in the general economic interest, public service obligations which may relate to security, including security of supply, regularity, quality and price of supplies and environmental protection, including energy efficiency, energy from renewable sources and climate protection. Such obligations shall be clearly defined, transparent, non-discriminatory, verifiable and shall guarantee equality of access for EU gas companies to national consumers.48

Article 3(2) of the Gas Directive and its link with Article 106(2) TFEU has been the subject of the Federutility case (on gas retail supply),49 followed by Enel Produzione (on electricity generation)50 and by the recent ANODE case (again on gas retail supply).51 These cases have added another layer of EU law on top of a what should allegedly be the exclusive precinct of Member States because, under the aegis of compliance with the proportionality principle, the Court of Justice distilled the requirements that Member States have to meet when they seek to exercise their SGEI-designation/ regulation powers in the energy sector: —

First, the Court of Justice held that regulation must be temporary, emphasising the need for periodic re-examinations at close intervals of the need for intervention. — Second, ‘the method of intervention used must not go beyond what is necessary to achieve the objective which is being pursued in the general economic interest’. — Third, the Court of Justice held that the principle of proportionality entails an examination of the scope of the measure ratione personae. Although the Directive did not exclude that undertakings may also benefit from public service obligations, the referring court must verify whether the measure takes into account, and in light of its objectives, the differences between undertakings and domestic consumers and between undertakings of different sizes. — Fourth, an SGEI on an electricity generator entailing a cost entitled the generator to a ‘fair remuneration’ (Enel Produzione, paragraph 69). The Court’s Federutility test stands out as a vademecum for good regulation overall. At the same time, this test imposes a limit on a Member States’ ability to impose public service obligations, because such obligations can be struck down if they run contrary to this test.

48 The provisions in the Third Gas Directive are identical (see Art 3(2)). The same provisions were set out in the Second Electricity and Gas Directives. 49 Case C-265/08 Federutility and Others v Autorità per l’energia elettrica e il gas, EU:C:2010:205. 50 Case C-242/10 Enel Produzione SpA v Autorita per l’energia elettrica e il gas, EU:C:2011:861. 51 Case C-121/15 Association nationale des opérateurs détaillants en énergie (ANODE) v Premier ministre and Others, EU:C:2016:637.

Services of General Economic Interest 283 III. SGEI AND THE NOTION OF AID

A. Before Altmark As has already been mentioned in different parts of this book, a finding of State aid pursuant to Article 107(1) TFEU requires the existence of an economic advantage. Accordingly, the main question when it comes to qualifying public funding granted to finance PSOs as aid, pursuant to the Treaty, has long been to define whether this financing was to be considered as granting an economic advantage to the beneficiary. In response to this question, a first approach in the case law is the ‘net’ or ‘compensatory’ approach, according to which these funds are mere compensation for the costs incurred. In this sense, there is no advantage when costs are covered. Under this strict condition, public funds do not alter competitive conditions in the single market. Public financing should thus create a level playing field. However, in case funds would come in excess of net costs, aid would be found and, in addition, could not be considered compatible with the internal market under Article 106(2) TFEU. This approach was followed until the mid-1990s and the decisions by the General Court and the ECJ in the FFSA cases.52 The second—so-called ‘gross’—approach, in essence, used to consider that any funds granted to support the performance of a PSO constitute an economic advantage and should therefore be considered as State aid under Article 107(1) TFEU. However, in contrast with the former approach, the measure under review could be considered compatible with the internal market under Article 106(2) TFEU. In SIC v Commission, the General Court stated, concerning State funding for public television in Portugal, that the fact that a financial advantage is granted to an undertaking by the public authorities in order to offset the cost of public service obligations which that undertaking is claimed to have assumed has no bearing on the classification of that measure as aid within the meaning of Article [107](1) of the Treaty, although that aspect may be taken into account when considering whether the aid in question is compatible with the common market under Article [106](2) of the Treaty.53

This was justified by the fact that ‘the Treaty does not distinguish between measures of State intervention by reference to their causes or aims but defines them in relation to their effects’ and therefore, the reason support is granted is irrelevant as to its effects on competition. The ECJ, in a much discussed judgment, came back to the compensatory approach in the Ferring case.54 It was thus high time to clarify the situation, which the ECJ did in a ‘grand chamber’ judgment of 2001 in the Altmark case.

52 Case T-106/95 FFSA and Others v Commission, EU:T:1997:23; Order of Case C-174/97 P FFSA and Others v Commission, EU:C:1998:130. 53 Case T-46/97 SIC v Commission [2000], EU:T:2000:123, para 84. 54 Case C-53/00 Ferring, EU:C:2001:627.

284 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu B. The Altmark Case In essence, Altmark is coherent with the compensatory approach as it clearly states: [W]here a State measure must be regarded as compensation for the services provided by the recipient undertakings in order to discharge public service obligations, so that those undertakings do not enjoy a real financial advantage and the measure thus does not have the effect of putting them in a more favourable competitive position than the undertakings competing with them, such a measure is not caught by Article [107](1) of the Treaty.55

However, the Altmark ruling reflects on the critics by defining four cumulative conditions to be fulfilled to ensure clearance under Article 107(1) TFEU, the interpretation of which proves to be difficult as some of the terms are fairly abstract. First, there must be a clearly defined and ‘entrusted’ obligation to carry out an SGEI, the definition of entrustment being sometimes problematic. The company which benefits from the compensation must therefore be effectively vested with a PSO to discharge and this obligation must be clearly defined, in particular in the national legislation and public contracts. It is not necessary that the obligations are enacted in a law or regulation, it is sufficient that there is an ‘act of a public authority’ which may be the grant of a concession or the approval by the State of a particular set of arrangements or contracts. Meeting this first condition raises the set of issues discussed in section II on the definition of an SGEI. Second, the parameters on the basis of which the compensation is calculated must be established in advance in an objective and transparent manner. The existence of an objective and transparent compensation mechanism, set out in advance (no specific formula being imposed), aims to avoid the conferral of an economic advantage which would favour the beneficiary over its competitors. However, ‘objective’ does not necessarily mean ‘efficient’, which casts doubt on what the Court intended to say in Altmark. Third, the compensation should not go beyond what is necessary to cover the costs incurred in discharging the PSO, taking into account possible benefits and a reasonable profit. Compliance with this condition is deemed necessary to ensure that the recipient undertaking is not given any advantage which distorts or threatens to distort competition by strengthening that undertaking’s competitive position. Of course, avoidance of overcompensation, defined as no compensation in excess of net costs and reasonable profit, is not easy to do in practice. Reasonable profit is approached as a rate of return on the capital required by a typical company for such a service, which implies a benchmark (for instance a return on capital employed that equals the interbank rate plus 1 per cent seems reasonable in the absence of commercial risk).56 In addition, Member States may provide incentive criteria for quality of service and productive efficiency.57

55 56 57

Altmark, para 87. Communication (n 2) para 61. ibid.

Services of General Economic Interest 285 Fourth, where the undertaking which is to discharge PSOs in a specific case is not chosen pursuant to a public procurement procedure which allows for the selection of the tenderer capable of providing those services at the least cost to the community, the level of compensation needed must be determined on the basis of an analysis of the costs which a typical, well-run undertaking would have incurred in discharging those obligations, taking into account the relevant receipts and a reasonable profit for discharging the obligations. In other words, the awarding authority must choose the provider by either a public procurement procedure that allows for the provision of the service at the least cost to the community, or by other means, provided the compensation does not exceed the amount required by a typical, well-run undertaking. In this context, an open tender is obviously best and awarding authorities would be well advised to follow EU procurement law procedures, even when non-mandatory. Negotiated procedures without publication of a contract notice seem possible, but only in exceptional cases. As stated in the SGEI communication, ‘negotiated procedures with prior publications confers wide discretion upon the adjudicating authority and may restrict the participation of interested operator’.58 Finally, we may notice that for a typical, well-run undertaking, making profit is not enough; the awarding authority must use objective criteria, such as analytical ratios.

C. Application of the Altmark Case Law in Energy: The Old and the New There is a difference between cases concerning partially liberalised markets, essentially around the early to mid-2000s, and the cases concerning fully liberalised markets a decade later. i. Altmark in Partially-Opened Markets In the Irish CADA scheme case of 2003,59 which concerned a CRM involving capacity payments to builders of new capacity, the Commission accepted that this scheme contributed to ensure the performance of an SGEI concerning security of supply in Ireland. It was however already emphasised that demand-side measures and interconnection should be developed before such measures were implemented, even though in this particular case the geographical situation had to be taken into account. This sort of reasoning would be at the core of the first wave of Commission decisions regarding the new generation of CRMs a decade later. It is worth mentioning that qualifying generators were selected on the basis of a competitive process and that the details defining all indispensable parameters were established in the tender documents. The Commission therefore concluded that all four of the Altmark conditions had been met and ruled that Article 107(1) TFEU did not apply.

58 59

ibid, para 66. Ireland Public Service Obligation—Electricity Supply Board (Case N475/2003) [2004] OJ C34/8.

286 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu Back in its decision of July 14, 2004, Public Service Obligation—Electricity Supply Board,60 again concerning Ireland, the Commission considered the applicability of the Altmark case law to a scheme involving a public service compensation mechanism financed by a levy collected from end consumers. The purpose of the compensation mechanism was to finance the provision of short-term reserve capacity by the Electricity Supply Board (ESB), the public electricity supplier. The Commission concluded that the PSO measure did not comply with all the criteria of the Altmark judgment.61 In particular, the Commission focused on the fulfilment of the fourth criteria. It noticed that ESB had not been selected in a public procurement procedure and that the Irish authorities had not provided information on the comparative costs of ESB and a typical well-run undertaking. The Commission could therefore not conclude that the PSO would be discharged at the required level of cost-efficiency. In 2007 and 2008, the Commission, in its decisions on the Hungarian and Polish PPAs,62 again concluded that the test devised in Altmark had not been met. Most of the analysis of the Commission concerned the fact that the PPAs in question did not contribute to the fulfilment of an SGEI. However, the Commission also analysed the other criteria. In particular, the Commission considered that in the absence of a clear definition of the SGEI to be provided, specifically one making a clear distinction between the service to be rendered and the power plants’ normal business operations, it was impossible to establish parameters for compensation and to determine whether the compensation exceeds the amount necessary to cover related costs. Further, the existence of certain parameters for establishing the PPA prices was not deemed to be equivalent to the existence of precise parameters for calculating compensation for SGEIs, since the price was not equal to the compensation. Furthermore, the fact that the price covers only the costs of generating electricity, plus a margin for profit, does not mean that it does not include any excess compensation, since many of the costs of generating electricity may be the normal costs covered by any electricity generator, as opposed to the surplus costs associated with SGEIs. This type of reasoning is also relevant for the new generation of CRMs, in particular for the market-wide mechanisms where there is a general risk of compensating market operators who could produce electricity or build new capacity without public support. The Commission also noted that various criteria were applied in the tender procedure, not just the price, and that the inclusion of criteria related to wider policy objectives made it impossible to conclude automatically that the level of compensation was correct. Furthermore, neither the Polish or Hungarian authorities nor the interested parties provided an analysis of the costs of the generators in question to support the contention that they tally with the costs incurred by a typical undertaking. 60

ibid. Nevertheless, the Commission found the measure compatible with the common market under the derogation of Art 106(2) TFEU arguing that both the necessity and the proportionality tests were met. 62 [2009] OJ L83/1 and L225/53. See judgment in case T-80/06 Budapesti Erőmű v Commission [2012] and Case T-182/09, EU:T:2012:65. For an excellent analysis of these cases, see L Hancher, ‘Long-Term Contracts and State Aid: A New Application of the EU State Aid Regime or a Special Case?’ in JM Glachant et al, Competition, Contracts and Electricity Markets: A New Perspective (London, Edward Elgar Publishing, 2011). 61

Services of General Economic Interest 287 Last, we can mention the decision of April 2007 concerning the Trbovllje power plant.63 The Commission found that the four Altmark conditions were met. With respect to the fourth condition, the authorities had demonstrated that the choice of Trbovllje was the one that led to providing the SGEI at the least cost to the community, even though a tender procedure was not used. Only two plants could have fulfilled the SGEI in question and the other plant, Sostanj, would have had to burn lignite and this would have required expensive upgrading. A public tender would not have delivered a cheaper solution as the upgrading of existing plants would have required high investment costs and any bidder would have demanded compensation including a reasonable profit by way of return on invested capital. ii. Altmark in Liberalised Markets Three cases are particularly noticeable here: the Castelnou energía case,64 and the two French CRM cases.65 a. Castelnou Energia Case (Commission Decision) In this case, Spain notified a scheme implementing a preferential dispatch mechanism with a view to grant State aid for the provision of electricity under the 15 per cent rule set by the Electricity Directive. The scheme, foreseen by Royal Decree 134/2010, involved granting preferential dispatch and compensation to 10 electricity producers holding an indigenous coal procurement agreement and using domestic hard coal up to 15 per cent of the domestic demand. Based on this mechanism, which was set for a definitive period of four years, every day, the outcome of the clearing of the day-ahead electricity market was modified to the extent necessary to ensure that the above-mentioned coal-fired power plants could place on that market predefined volumes of electricity generated out of indigenous coal. This meant that those producers were granted stable demand for their product at a preferential price. Spain argued that the measure enhanced security of supply by creating backup for electricity generated from renewable energy sources that were highly variable due to weather conditions. Spain also mentioned that the Spanish electricity system was not well connected to other major European systems, so security of supply was needed. The capacity payment mechanism was financed through a levy imposed by national law on electricity retail suppliers as well as direct consumers and was managed by an entity entrusted by the Spanish TSO REE. REE had no margin of discretion on these amounts. When there was a shortfall in the system, the missing amount was provided by the CNE. Payments to the generators were also made by order of the TSO under a very complex mechanism.

63 State aid scheme implemented by Slovenia in the framework of its legislation on qualified energy producers (Case C 7/2005) [2007] OJ L219/9. 64 Spain Public service compensation linked to a preferential dispatch mechanism for indigenous coal power plants (n 24). 65 In Hinkley Point, since the first Altmark criterion was not met (and all criteria are cumulative), the Commission did not consider that the investment itself could qualify as an SGEI, and as such, it was also unable to fulfil the requirements under Art 106(2) TFEU.

288 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu The list of undertakings concerned, the parameters for compensation and the methodology to assess them were precisely defined in Royal Decree 134/2010. Costs taken into account covered variable costs (fuel, acquisition of emission allowance, variable operation and maintenance costs and costs associated with coal losses at the level of stock), contribution to fixed costs (a fraction of the fixed operation and maintenance costs and of the depreciation of investment costs that can be attributed to production in the framework of the SGEI) and return on the capital assigned to the SGEI (equal to the regulated rate of return applied to the regulated electricity production activity in the Spanish insular and extra-peninsular system, which is 5.5 per cent post tax, thus lower than the weighted average cost of capital (WACC) of the Spanish electricity sector). Certain parameters that influenced unit production costs, such as fuel specific consumption, were based on historical values. The national regulator CNE would monitor ex post the actual costs incurred by the companies concerned. Compensation would then be adjusted upwards or downwards. The companies operating in the scheme would remain covered by the ETS. Interestingly, the Commission considered the application of the Altmark criteria to the present case whereas Spain had only argued that the scheme should be deemed compatible pursuant to Article 106(2) TFEU. It found that the fourth Altmark criteria was not met and hence that the compensation scheme entailed an economic advantage pursuant to Article 107(1) TFEU. In particular, an open tender was not used and Spain had not provided any comprehensive analysis of the costs which a typical owner of such a plant, well run and adequately equipped, would have incurred. In the foreseen scheme, the national regulator had to monitor the actual cost incurred by the companies concerned so that the actual compensation could be calculated depending on the actual costs and revenues, but not on the costs of a typical, well-run and well-equipped power generating undertaking. The Commission noted that indigenous coal producers also obtained an economic advantage as they would be able to sell more coal thanks to the priority dispatch mechanism. b. The Two French CRM Cases In the case concerning the market-wide French CRM,66 the French authorities argued that capacity remuneration coming from the certificates traded did not constitute an advantage, but only a compensation for availability. The compensation would necessarily be proportionate as the price of certificates will be fixed freely in the certificates market, without public intervention. The Commission, in its Opening Decision adopted on the basis of Article 108(2) TFEU, did not follow that logic. The Commission noted that buyers of certificates have an obligation to procure certificates and are thus not enjoying an advantage. Similarly, generators would have the obligation to be certified. However, this obligation does not exist for providers of demand response, which entails that certain participants in the capacity mechanism will participate on a voluntary basis. Also, existing producers have a choice as to which volume of capacity they certify. Certification is thus a choice and not an obligation, hence not a ‘public service’ obligation. 66

Mécanisme de capacité en France (n 19).

Services of General Economic Interest 289 In addition, the Commission highlighted that capacity providers will obtain monies that they would not have obtained without the market specifically organised by the French authorities. The French capacity mechanism would therefore constitute an advantage for capacity providers and the first Altmark criteria would not be met. As noted, this analysis was not modified in the Final Decision of 8 November 2016. In the Opening Decision concerning the French CRM Brittany gas-fired power plant case,67 there is also a very interesting (preliminary) discussion concerning the application of the Altmark criteria. In this case, the French authorities considered that the primary objective of the measure was to secure electricity supply in Brittany, which would be under threat from low generation capacity in that region, network constraints, increased consumption and high thermal sensitivity. They considered that this constituted an SGEI and that the support offered to the successful tenderer constituted compensation for the discharge of PSOs which met all the conditions of the Altmark case. The project that was finally selected was to be managed by Direct Energie and Siemens, who would benefit from a 20-year PPA including a fixed yearly premium and a tariff equal to 95 per cent of the hourly price recorded on the EPEX spot market. The Commission had doubts that the first, third and fourth Altmark criteria were met. First, the Commission questioned whether there was an actual SGEI. According to the Commission, the French authorities did not provide enough evidence that there had been a security of supply problem in the past, even when temperatures were extremely low, as in the winter of 2012. Then the Commission considered that the production capacities necessary to ensure security of supply in Brittany could have been procured under normal conditions if the French authorities had allowed market forces to play freely. In particular, the fact that the French authorities would have defined a single price zone for the entire French territory would have hampered relevant price signals for localisation, which was not contested by French authorities. The Commission also noted that the notified measure is not technology neutral as it discriminates against technologies other than combined-cycle gas turbine (CCGT), which, again, was not contested by the French authorities. As a consequence, the first Altmark criteria was not met and Article 106(2) TFEU could not be invoked. Second, the Commission addressed the question as to whether the notified measure was adequate and proportionate. The Commission noted that the French authorities had not provided precise data as to the need for new capacities in Brittany and that, if 450MW were truly needed, that number could have been reached by a combination of demand response and smaller generation capacities. On top of that, the tender could exacerbate the alleged security of supply problem in the long term, as (a) private investors could delay investment in the hope of benefiting from additional tenders, (b) national regulation tends to create a missing money problem and (c) the new CCGT could accelerate the exit of older plants. As to the compensation parameters, the Commission considered that they were transparent and objective. However, the third Altmark condition was not met, as

67

France—Call for additional capacity in Brittany (Case SA.40454) [2016] OJ C 46/69.

290 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu the notified measure did not include a clawback mechanism, despite the difficulty of estimating trends in electricity prices over a 20-year period. The Commission was consequently unable to rule out the possibility of the measure resulting in overcompensation. Last, the Commission made a thorough analysis of the selection process and considered that, even though the Direct Energie/Siemens offer was the cheapest, the selection criteria, in particular regarding localisation and technology, had not allowed for fair competition. The Commission therefore doubted whether the notified measure actually allowed for the selection of the tenderer capable of providing the requested services at the ‘least cost to the community’.

IV. SGEI AND COMPATIBILITY WITH THE INTERNAL MARKET

A. Article 106(2) TFEU and its Application in Energy Article 106(2) TFEU states that undertakings entrusted with the operation of services of general economic interest or having the character of a revenue-producing monopoly shall be subject to the rules contained in the Treaties, in particular to the rules on competition, in so far as the application of such rules does not obstruct the performance, in law or in fact, of the particular tasks assigned to them. The development of trade must not be affected to such an extent as would be contrary to the interests of the Union.

In order to be able to rely on Article106(2) TFEU as grounds for compatibility, the recipient of an aid granted as compensation for the discharge of a PSO must meet three conditions: —

The undertaking must have been given a specific PSO to discharge, defined as such by the Member State concerned. — The task must have been granted by the State pursuant to a measure of the public authorities, ie, a public act or ‘entrustment’. — The (Necessity and) Proportionality test (SGEI versus Union interest) must be met.

The Altmark conditions and the Article 106(2) TFEU conditions bear similarities and it can legitimately be wondered whether it is possible to fulfil the first and second Article 106(2) TFEU conditions if the first Altmark condition is not fulfilled, and whether it is possible to fulfil the third Article 106(2) TFEU condition if the third Altmark condition is not fulfilled. In in this regard, the General Court stated in paragraph 160 of the BUPA judgment68 that the two sets of criteria overlap to a large extent. A remaining question would concern the link between the fourth Altmark condition and the third Article 106(2) TFEU condition, even though it is fairly clear that

68

Case T-289/03, BUPA and Others v Commission, EU:T:2008:29.

Services of General Economic Interest 291 the purpose of Article 106(2) TFEU is not to define if an SGEI is provided under competitive market conditions and at the least possible cost. As regards the proportionality test, according to paragraph 23 of the SGEI Communication proportionality implies that the means used to fulfil the SGEI shall not create unnecessary distortions of trade. Specifically, it has to be ensured that any restrictions to the rules of the Treaty do not exceed what is necessary to guarantee effective fulfilment of the SGEI.

In other words, the public financing can be considered compatible only in so far as the application of State aid rules would prevent the aid beneficiary from effectively performing the PSO, taking into account the fact that the derogation should not affect trade ‘to such an extent as would be contrary to the interests of the Union’. In general, the Court has not attached much importance to the last sentence of Article 106(2) TFEU. The application of Article 106(2) TFEU has arisen in a number of cases, but only few of them recently. A recent illustration can be found in the judgment of the General Court in Castelnou Energía.69 After a lengthy analysis of the existence of an SGEI in that particular case, the General Court went on to discuss whether the support scheme for indigenous coal under review was manifestly inappropriate given the objective pursued. In this regard, it recalled the existence of the 15 per cent rule set by the Electricity Directive and the fact that the Court had held, in Ålands Vindkraft,70 that Member States could design support schemes limited to producers established in their territory. It also acknowledged that the aid creates distortions, but not manifestly inappropriate in this case due to the 15 per cent threshold. It also noted that coal power plants can act as a back-up to RES, even if less flexible than gas-fired power plants and that switching to import coal would be expensive and take time. The General Court therefore found that that the Commission did not err when it found that it was not manifestly inappropriate given the objective pursued.

B. SGEI Framework The compatibility conditions under the 2012 Framework have changed significantly in comparison with the 2005 package. The following subsections discuss some of the major novelties and highlight the limited number of cases assessed by the Commission in the energy sector under the new framework.71

69 Castelnou (n 25). The Commission decision is discussed above regarding the application of the Altmark conditions. 70 Case C-573/12 Alands Vindkraft AB v Energimyndigheten, EU:C:2014:2037. 71 Without going into detail about conditions which have already been set out in this chapter, such as the existence of a genuine SGEI. In this regard, see section II of this chapter.

292 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu i. Entrustment Act The condition related to the entrustment act did not change substantially under the 2012 Framework. Irrespective of the act’s form,72 it should include some mandatory elements which are not always assessed individually by the Commission in its decisions (most likely because some of these elements constitute conditions for compatibility on their own, and are thus assessed in any event). These are the content and duration of the PSO, the undertaking and territory concerned, the nature of the rights, the compensation mechanism and details related to its calculation, monitoring and review, and measures to ensure the avoidance of overcompensation.73 ii. Duration of the Entrustment Period The entrustment period duration is a new condition compared with 2005. Without providing a specific term, the 2012 Framework states that the entrustment period should not exceed the period for depreciation of the most significant assets required to provide the SGEI.74 However, by correlating Article 2(2) of the Decision with paragraph 17 of the 2012 Framework,75 the Commission considers that a duration exceeding 10 years, not justified by objective criteria, would involve incompatible aid. Through this mechanism, long entrustment periods that could lead to competition distortion are avoided. The 10-year duration threshold is not set in stone and the Commission seems to be amenable to accepting longer periods. For instance, a period of 15 years was accepted as ‘appropriate in view of the lifetime of [a peat-fired generation plant]’ in Sale of state assets.76 Also, without making a distinction between physical and financial depreciation, the Commission found that a lengthy period of 55 years was objectively justified because of the duration of depreciation and amortisation of the pipeline connecting the LNG terminal to the gas network. As some commentators argued, the Commission seemed to be focusing on the natural life of the pipeline rather than the time necessary to recoup the cost of investment in assets.77 Compared with the postal sector, in which the Commission lists the main assets, their amortisation time and their relationship with relevant SGEIs,78 in LNG terminal,

72 The entrustment act can take various shapes: legislation, statutory documents, resolutions of the government, decisions, codes, schemes and award decisions, letter, agreements. 73 2012 Framework, para 16. 74 ibid, para 17. 75 This provision is in relation to the Decision’s scope of application rather than the compatibility conditions. See N Pesaresi et al ‘The New State Aid Rules for Services of General Economic Interest (SGEI): The Commission Decision and Framework of 20 December 2011’, available at: ec.europa.eu/competition/publications/cpn/2012_1_11_en.pdf, 12. 76 Ireland, Public service obligations imposed on the Electricity Supply Board with respect to the generation of electricity out of peat (Case N6/A/2001) [2002] OJ C 77, and Ireland, Sale of State Assets (Case SA.37030) [2015] OJ C63/3, para 31 (Sale of State Assets). 77 P Nicolaides, ‘PART II: Combining Infrastructure Aid with SGEI Aid’, available at: stateaidhub.eu/ blogs/stateaiduncovered/post/6567. 78 Belgium, State compensations to Bpost for the delivery of public services over 2013–2015 (Case SA.31006) [2013] OJ C279/2, table 2.

Services of General Economic Interest 293 the duration is tied exclusively to the depreciation period of the pipeline, characterised as ‘one of the most important assets required to provide the SGEI (as no gas could be supplied through the LNG terminal without the pipeline connection)’.79 However, if gas would not even reach the pipeline without the conversion from a liquid to gas state, should the gasification plant not also be qualified as an important asset? The Commission’s focus on a single component rather than the whole LNG terminal, which is a complex project by its very nature, is questionable.80 iii. Compliance with Directive 2006/111/EC Although compliance with Directive 2006/111/EC on the transparency of financial relations between Member States and public undertakings as well as on financial transparency within certain undertakings has been added as a new condition in the 2012 Framework, it is one of the least important conditions for compatibility.81 Its purpose (more transparency) is already achieved to a certain extent via the separation of accounts obligation for activities falling in and outside of the SGEI scope.82 Interestingly, the threshold for fulfilling this condition differs from one case to another. For instance, in NII, the Greek State confirmed that PPC complied with the conditions related to the transparency and verifiability of cost data for SGEI, and provided detailed explanations regarding the separation of accounts for each activity.83 A similarly detailed justification was provided in POL, where the UK confirmed that POL’s public funds would be identified in its statutory accounts and the Department for Business, Innovation & Skills’ annual report.84 At the other end of the spectrum, in LNG terminal, the Commission appears to have been satisfied with much less, accepting Lithuania’s reference to the national implementing legislation of Directive 2006/111/EC and the mere promise/confirmation that the terminal operator would abide by the rules.85 iv. Compliance with EU Procurement Rules The Almunia SGEI package reinforced the link between the Treaty (State aid) and secondary EU law (public procurement rules). As of 2012, the existence of a competitive tender is not merely linked to the fourth Altmark criterion, but also a compatibility condition.86 At first glance, the addition of this requirement makes the 79

para 221. Nicolaides (n 77). 81 Published in [2006] OJ L 318/17 and 2012 Framework, para 18. 82 The financial and organisational structure is established in separate accounts in a way that the costs and revenues associated with different activities and full details of the methods by which costs and revenues are assigned or allocated to different activities are clearly set out (Arts 2(d) and 1(2) of the Directive). 83 Alleged illegal aid for discharging public service obligations in the non-interconnected islands (Case SA.32060) [2015] OJ C203/1, paras 27–33, 159 (NII). 84 paras 94–98. 85 paras 227–28. 86 2012 Framework, para 19. For the relationship between State aid and public procurement rules and the impact of the violation of the latter on the former, see D Grespan, ‘Services of General Economic Interest’ in W Mederer et al (eds), EU Competition Law. Volume IV. State Aid. Book Two (Deventer, Claeys & Casteels Publishing, 2008) 1206–08. 80

294 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu 2012 Framework seem stricter. However, its precise language actually gives some flexibility as to how this condition can be applied.87 For instance, instead of undergoing a tender procedure, the Member State may merely promise to use a tender in the future88 or may choose to trade this compatibility condition with commitments that remove exclusive or special rights. In practice, this compatibility condition has been stretched even further, as the Commission did not shy away from using the exemptions in the EU procurement package and case law. Generally, entrustment should be preceded by a certain procedure, such as an open or restricted procedure, negotiated procedure with prior call for competition.89 a. Essential Security Interests One exemption from such a procedure, and more generally from procurement rules, is the essential security interest grounds, which reads that this Directive shall not apply to contracts and design contests … to the extent that the protection of the essential security interests of a Member State cannot be guaranteed by less intrusive measures, for instance by imposing requirements aimed at protecting the confidential nature of information which the contracting entity makes available in a contract award procedure as provided for in this Directive [or] to the extent that the application of this Directive would oblige a Member State to supply information the disclosure of which it considers contrary to the essential interests of its security [or if] the procurement and performance of the contract or design contest are declared to be secret or must be accompanied by special security measures in accordance with the laws, regulations or administrative provisions in force in a Member State … provided that the Member State has determined that the essential interests concerned cannot be guaranteed by less intrusive measures.90

In LNG terminal, the Commission concluded that Lithuania met the ‘essential interest’ exemption and thus a public procurement selection procedure was unnecessary. Irrespective of the importance of the LNG terminal to Lithuania’s independence from a single-gas supplier, it is really questionable whether entrustment of the SGEI to the terminal operator was really required to ensure essential interests. Put differently, if the SGEI had been assigned to a company not controlled by the Lithuanian State how would have this impacted Lithuania’s essential interests?91 Interestingly, the Commission did not conduct the assessment of this alternative scenario, but merely justified the exemption based on the requirement that the entrusted company

87 See JL Buendia, ‘A Turn of the Screw’, available at: www.chillingcompetition.com/2012/03/14/ a-turn-of-the-screw-jl-buendia-on-sgeis/. 88 A mere declaration of Ireland to conduct the sale of the plants in accordance with the principles of openness, transparency, objectivity, fairness and non-discrimination was considered to be sufficient, see Sales of Assets, para 35. By the same token, see case Compensation to Post Office Limited for costs incurred to provide SGEIs 2015–2018 (Case SA.38788) [2015] OJ C188/4, para 103 (POL). 89 Directive 2014/25/EU of the European Parliament and of the Council of 6 February 2014 on procurement by entities operating in the water, energy, transport and postal services sectors and repealing Directive 2004/17/EC [2014] OJ L94/243, Article 44(2) (Directive 25/2014). 90 Directive 25/2014, Art 24(2)(3). 91 For an interesting discussion on this topic, see Nicolaides (n 77).

Services of General Economic Interest 295 did not have or could not have developed any corporate and economic ties with the single supplier: In order to develop access to alternative gas supply sources, the project can only fruitfully be implemented by an entity that is independent both on the corporate but also on the economic level from the single supplier. However, if the project manager (operator) would be selected in a transparent competitive procedure under Directive 2004/18/EC or the Treaty, there would be a risk that it would have (at the time of the tender) or later develop ties to the single supplier that would allow the latter to influence its market behaviour in a way that could negatively affect the fulfilment of its SGEI mission. In fact, currently the single gas supplier has corporate and/or economic bonds with a vast number of undertakings in Lithuania, in particular companies active in the gas sector and Lithuania has provided examples of energy related projects that had to be cancelled following various interventions at different levels of the single gas supplier (emphasis added).92

b. The ‘In-House’ Concept Besides the essential security interest grounds, the Commission excused the application of EU procurement rules to certain measures based on the ‘in-house provision of services’ principle established by the Teckal case,93 which essentially states that if an economic operator is controlled by a public authority and conducts the essential part of its activities towards this controlling authority, it will be qualified as an in-house entity in relation to this controlling authority.94 This concept was applied in Skelleftea airport, where Skelleftea Municipality ‘fully owned and controlled Skelleftea airport and the airport activity [was] its sole economic activity’.95 c. Negotiated Procedure A middle ground between fulfilling the compatibility condition and the exemptions from EU procurement rules is the negotiated procedure without prior call for competition.96 The Commission has generally viewed this procedure as applicable in situations where the company to whom the SGEI was entrusted was in a unique position to provide the SGEI because there was no other company that met the requirements for providing the service or the specificities of the market.97

92

LNG Terminal (n 43) para 232. Case C-107/98, Teckal Srl v Comune di Viano and Azienda Gas-Acqua Consorziale (AGAC) di Reggio Emilia, EU:C:1999:562, para 50. For an indepth view of this principle and the Teckal conditions, see ‘Commission Staff Working Paper concerning the application of EU public procurement law to relations between contracting authorities (“public–public cooperation”)’ SEC(2011) 1169 final. 94 The Commission considers that there is no proper definition of the term, and the interpretation should be imported from public procurement case law. For more details, see Pesaresi et al (n 75) 12; and Communication (n 2) paras 13, 37. 95 Sweden Skelleftea Airport entrustment of a service of general economic interest (Case SA.38757) [2016] OJ C406/2, paras 13, 104–05 (Skelleftea Airport). 96 Directive 25/2014, Art 50. 97 POL, paras 101–02; NII, para 106. However, on the contrary, the Commission is of the opinion that the negotiated procedure without publication is not sufficient to meet the fourth Altmark criterion (see Pesaresi et al (n 75) 12). 93

296 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu v. Absence of Discrimination Lack of discrimination is a new requirement under the 2012 Framework and entails the obligation to calculate the compensation granted to several undertakings subject to the same SGEI based on the same method. This condition occurred as, in practice, authorities used to compensate undertakings based on different methods, and the Commission received several complaints in the health insurance sector.98 This condition is less relevant than the others, as in most of the Commission’s decisions enacted under the 2012 Framework, the SGEI is entrusted only to one undertaking.99 vi. Compensation Under the 2012 Framework, the principle remained the same as in the 2005 package: compensation should not exceed what is necessary to cover the net cost for discharging the PSO plus a reasonable profit. However, there are a few novelties described below that are worth mentioning. First, the 2012 Framework provides for a net avoided cost methodology, where the cost is calculated as the difference between the net cost for the undertaking of operating the SGEI and the net cost for the same undertaking without the SGEI entrustment,100 as an alternative to the cost allocation methodology. As explained by the Commission, the advantage of shifting to a different methodology is that it would reflect a ‘better estimate of the economic burden of the [PSO], because it takes account of the decisions which would be made in the absence of such an obligation’.101 The Commission considers that when the identity and activities of the future PSO provider are not known or where support has already been granted for a long duration based on cost allocation, the net avoided cost methodology is not feasible.102 It is not clear to what extent other methodologies, besides cost allocation, could be used. Second, unless Member States justify that it is not feasible or appropriate, they must introduce incentives for the efficient provision of SGEIs. In designing the type of compensation, Member States have the option to choose between an upfront payment (where the fixed payment will include expected efficiency gains) and a payment made depending on the extent to which efficiency targets have been met. Efficiencies must be based on objective and measurable factors set out a priori, in the entrustment act, and should be subject to an ex post assessment conducted by a third party (such as the NRA). The Commission’s practice varies in relation to the provision of efficiency incentives. For instance, while there are no details on how efficiencies were constructed in Sale of State assets,103 the LNG terminal case includes various efficiencies. 98

Pesaresi et al (n 75) 13. For an exceptional case, see Health Insurance Risk Equalization Scheme 2013 (Case SA.34515) [2013] OJ C204/2, paras 104–05. 100 2012 Framework, para 25. 101 Pesaresi et al (n 75) 18–19. 102 Sale of State Assets, para 38. 103 ibid, para 40. 99

Services of General Economic Interest 297 These include upper limits for the price of the regasification service and other services in the natural gas sector, costs that could be considered for the payment of the LNG supplement, adjustment to inflation ‘only by half of the inflation coefficient and not more than 3%’.104 The lack of efficiency incentives can be temporarily justified by particular situations, including a period of time required for the implementation of national legislation where the efficiencies are provided.105 In contrast to Altmark, there is no obligation for the provider of the PSO to be efficient. However, by introducing efficiency incentives in the compensation, the Commission hopes that providers become more efficient in the long term. Should such efficiency incentives not be provided, and the net cost compensated ex post and in full, then the reasonable profit should not exceed the swap rate plus 100 basic points. This benchmark may however not always be appropriate. As reflected in LNG terminal, when the project is not totally risk-free or the operator is not compensated for the total costs because the compensation is paid ex post and it does not cover the whole variable operating costs, the swap rate plus 100 basic points benchmark is not appropriate.106 The Commission has compared instead the terminal’s internal rate of return over a duration of 55 years with the WACC for the sector and the regulated WACC established every five years. Also, compensation now goes beyond incurred costs: for upfront payments, it is based on expected costs and, for payments made if efficiency targets were met, there is a combination of expected and incurred costs. Third, in relation to the upfront fixed compensations that incorporate expected efficiency gains, the 2012 Framework provides for an ex ante multi-annual approach for checking overcompensation. Like this, the Commission ensures that no overcompensation (which constitutes incompatible State aid, not necessary for the operation of SGEI) is paid throughout the whole duration of entrustment. Ex ante cost estimates can be ex post rectified based on actual data.107 vii. Substantive Competitive Assessment Test Under exceptional circumstances, although the conditions provided under the 2012 Framework are met, the aid can be particularly distortive, ie, the aid could affect trade to such an extent as it could be contrary to the interest of the Union. Such circumstances can include when the aid denies undertakings in important sectors of the economy the possibility to achieve the scale operations necessary to operate efficiently, … the entrustment has a duration which cannot be justified by objective criteria or bundles a series of tasks, … the SGEI has been entrusted without a competitive selection procedure in a non-reserved market, … the entrustment is connected with special and exclusive rights which seriously restrict 104

paras 253–59. In NII, implementation of the NII code coupled with the fact that Greece was benefiting from a derogation decision granted by the Commission based on the Third Electricity Directive because of the specificities of the Greek electricity market (eg, substantial problems related to the operation of conventional power plants within the systems) paras 78–84, 186. 106 para 249. 107 Sale of State Assets, para 42. 105

298 Adrien de Hauteclocque, Francesco Maria Salerno and Simina Suciu competition, … the aid allows the beneficiary to finance the creation or use of an infrastructure that is not replicable and enables it to foreclose the market where the SGEI is provided or related relevant markets or if the entrustment hinders the effective implementation or enforcement of Union legislation aimed at safeguarding the proper function of the internal market.108

Until now, this test has not yet been conducted in the energy-related cases assessed by the Commission under the 2012 Framework.

V. CONCLUDING REMARKS

The foregoing shows that there are three elements in the Commission’s and Court’s decision-making that may lead towards more control over SGEI at EU level: —

The need to show a market failure as a préalable for defining an SGEI, a trend recently reiterated by the Court in the SNCM judgment.109 — The rigorous conditions of the Altmark case law, which means that, more often than not, compensation for SGEI needs to be approved as State aid by the Commission. — The adoption of the SGEI packages, which means that, to win State aid approval, Member States must conform to the mould provided in the EU law instruments. As a result, as regards State aid, in practice, Member States may find that in fact their margin of discretion in defining and entrusting SGEIs is not so wide. The decisions in Hinkley and the French CRM cases are illustrative: having started by arguing for an SGEI, these Member States later acquiesced to have their measures analysed—and approved—under Article 107 TFEU. The trend towards more EU control over SGEIs in the energy sector is also visible in the recent proposal for a Directive on Common Rules for the Internal Market in Electricity.110 On the one hand, the first sentence of Recital 15 reiterates the usual formula about Member States’ wide discretion. On the other hand, the following sentence of Recital 15 impinges on this discretion by ruling out the use of regulated retail prices. Accordingly, under ‘Public Service Obligation’, the proposal provides that ‘public service obligations which concern the price setting for the supply of electricity shall comply with the requirements set out in Article 5’, which in turn sets out a time limit for any such measure and a set of additional requirements. Whether the proposal will remain intact following the legislative process remains to be seen. However, the trend towards greater control over SGEI at the EU level seems well established. 108

NII, para 200. Case T-454/13, Société nationale maritime Corse Méditerranée (SNCM) v Commission, EU:T:2017:134. 110 Commission, ‘Proposal for a Directive of the European Parliament and of the council on common rules for the internal market in electricity (recast) COM(2016) 864 final, available at: eur-lex.europa.eu/ legal-content/EN/TXT/?uri=COM:2016:0864:FIN. 109

11 State Aid Recovery and the Energy Sector1 JACQUES DERENNE2

I. INTRODUCTION

A

BOUT 10 YEARS ago, on 25 October 2007, the European Commission (Commission) adopted a communication on the implementation of Commission decisions ordering Member States to recover unlawful and incompatible State aid in order to improve the application of recovery decisions (the Recovery Notice).3 The Recovery Notice also clarifies the role of national courts in aid recovery cases, which may not necessarily require an action from the Commission. Taking into account the lessons learned from a study carried out in 2005–06,4 the Commission’s objective was to remedy the excessive slowness of the recovery of aid by Member States. Under European Union law, the recovery of aid is the responsibility of the Member States under the close scrutiny of both the Commission and national courts. Since the adoption of Regulation No 659/1999 (repealed by Council Regulation No 2015/1589),5 the Commission is obliged to recover from the Member

1 This chapter is an updated version of a previous article in French from the author: J Derenne, ‘La récupération des aides illégales et incompatibles: Un tour d’horizon sélectif de la jurisprudence européenne depuis la communication “récupération” de 2007 de la Commission européenne’ (2012) 1 Concurrences 73. 2 The assistance of Caroline Ruiz Palmer, Ciara Barbu-O’Connor, and Ira Valsamaki at Sheppard, Mullin, Richter & Hampton LLP, Brussels, is gratefully acknowledged. 3 Commission Notice, ‘Towards an effective implementation of Commission decisions ordering Member States to recover unlawful and incompatible State aid’ [2007] OJ C272/05, 4. 4 European Commission, Study on the Enforcement of State Aid Law at National Level (March, 2006), coordinated by T Jestaedt, J Derenne and T Ottervanger (part I: ‘Application of EC State aid rules by national courts’ and part II: ‘Recovery of unlawful State aid: enforcement of negative Commission decisions by the Member States’. See DG COMP website (section ‘Studies and Reports’): ec.europa. eu/competition/state_aid/studies_reports/study_part_1.pdf; and ec.europa.eu/competition/state_aid/studies_reports/study_part_2.pdf. 5 Council Regulation (EU) 2015/1589, laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union [2015] OJ L248/9. Regulation 2015/1589 is a consolidated version of Regulation 659/1999, as amended in 2013.

302 Jacques Derenne State concerned any unlawful aid6 and aid declared incompatible with the internal market. The Member States are responsible for the enforcement of the Commission’s decisions and, in the event of a negative decision, the effective and immediate recovery of the unlawful and incompatible aid. Being aware that the lack of clarity in some of its aid decisions is one of the reasons for the slow recovery process, the Commission undertook, in its Recovery Notice, to provide as many guidelines to the Member States as possible, such as the identification of the beneficiary, the amount of aid to be recovered and the period within which the recovery must be executed. The Commission also reminded the Member States of their role in the recovery process and of the actions they could take to accelerate the implementation of its decisions. Within the Member States, national courts are a key player in the process of recovering aid. Once seized by a Member State following a decision of the Commission or by a third party affected by the unlawful grant of the aid, national courts have very broad obligations in the recovery process of that aid.7 Unlike the Commission, the national courts have no competence to assess the compatibility of the aid with the internal market. They must confine themselves to assessing the unlawfulness of the aid. When an aid is found to be unlawful, national courts have full powers (and duties) to order the recovery and cessation, as well as any other necessary measures to restore competition distorted by the unlawful grant. National courts are also called upon to intervene after a negative decision of the Commission, in support of the Member State concerned if it faces difficulties in complying with its obligation to recover the unlawful and incompatible aid. In its settled case law, the Court of Justice of the European Union (CJEU) has thus established the distinct, but complementary, roles of the Commission on the one hand, and of the national courts on the other, in the area of State aid: with the Commission lies the substantive assessment; with the national court lies the protection of the subjective rights of third parties by ensuring the regularity of the obligation to notify which allows the Commission to fully exercise control of compatibility with the internal market.

6 Aid is ‘unlawful’ (the term ‘illegal’ is not appropriate) when it has not been notified or it has been put into force without the compatibility assessment of the Commission (only the Commission can decide on the compatibility of aid with the internal market), or when a compatible aid is misused. The EU case law has ruled that the Commission cannot limit itself to the unlawfulness of the aid to require its recovery (Case 301/87 France v Commission, EU:C:1990:67): the Commission must establish the incompatibility of the aid with the internal market in order to order its recovery. The compatibility of the aid (the substance of the aid) must therefore be carefully distinguished from its ‘legality’ or lawfulness (formal regularity of aid). Unlike the Commission, the action of national courts is limited to unlawful aid (to that extent, national courts have more powers than the Commission). 7 See, among others, E Righini, ‘Godot Is Here: Recovery as an Effective State Aid Remedy’ in Le droit des aides d’État dans la CE Liber Amicorum Fransisco Santaolalla Gadea (The Hague, Kluwer Law International, 2008); and J-P Keppenne and K Gross, ‘Quelques considérations sur le rôle du juge national dans le contrôle des aides d’État’ in Le droit des aides d’État dans la CE Liber Amicorum Fransisco Santaolalla Gadea (The Hague, Kluwer Law International, 2008). See also J Derenne and C Kaczmarek, ‘La récupération des aides illégales: le rôle du juge national dans le “private enforcement” du droit des aides d’État’ (‘Recovery of unlawful State aid: the role of national courts in the private enforcement of State aid rules’) (2009) 10 ERA Forum 251.

State Aid Recovery and the Energy Sector 303 The various levels of intervention of national courts with regard to State aid was also the subject of a specific communication from the Commission in 2009 (the Enforcement Notice),8 following, in particular, an update of the aforementioned study of 2006. This update also contributed, among other things, to the publication of new pages on DG Competition’s website, where complainants, lawyers but also national courts can be informed of the (proper) precedents in the application of State aid law at national level.9 Accordingly, the recovery of an unlawful and incompatible aid may arise in different situations with different actors, depending on whether the aid is: —

Unlawful, whereby the national courts alone are empowered (obliged) to intervene according to Article 108(3) TFEU and the Commission must make an assessment of the merits. — Unlawful but declared compatible by the Commission: in this case, national courts are not obliged by EU law to order the recovery of the aid itself, but must nonetheless require the beneficiary to pay the ‘unlawful interest’ corresponding to the anticipated benefit of the aid unlawfully granted, prior to the decision of the Commission. — Unlawful but declared incompatible by the Commission: the Commission is required to adopt a negative decision ordering the Member State concerned to recover the aid at stake and the national courts are obliged to assist the Member State in the recovery. The purpose of this chapter is to review the application of these principles on the basis of selected cases from the Commission decisional practice and EU case law since the Recovery Notice. The review of the cases, selected by themes (section III), will be preceded by a reminder of the main principles developed previously and on which the Recovery Notice is based (section II).

II. GENERAL PRINCIPLES REGARDING THE RECOVERY OF AID

The Recovery Notice, which was the first of the Commission on the subject since the CJEU confirmed the principle of the recovery of unlawful and incompatible aid in 1973, constitutes a remarkable synthesis of the principles which relate to the recovery of an aid. These principles are summarised below, giving ample space to the extracts of judgments of the CJEU and of the General Court of the European Union (GCEU), often cited in their original text. 8

Commission notice on the enforcement of State aid law by national courts [2009] OJ C85/1. It should be noted that over the last 50 years, it seems that there are more ‘State aid’ cases than ‘antitrust’ cases that have been handled by national judges. Reference is made, in particular, to the book published as part of the 2009, update of the 2006, State aid study: J. Derenne, A Müller-Rappard and C Kaczmarek (eds), Enforcement of EU State Aid Rules at National Level: 2010—Reports from the 27 Member States (Berlin, Lexxion, 2010) 363. See also, State Aid Thesaurus of Hogan Lovells, available at: www.concurrences.com; and P Nemitz (ed), The Effective Application of EU State Aid Procedures: The Role of National Law and Practice, vol 29 (The Hague, Kluwer Law International 2009) 425, published after the FIDE Congress of 2006. 9

304 Jacques Derenne A. The Principle of Restitution and Obligations of the Commission and the National Courts The Commission is competent, when it has found that aid is incompatible with the common market, to decide that the State concerned must abolish or alter it; … to be of practical effect, this abolition or modification may include an obligation to require repayment of aid granted in breach of the Treaty.10

Since 1999, Article 16(1) of Regulation 2015/1589 (formerly Article 14(1) of Regulation 659/1999) requires the Commission to order the recovery of unlawful and incompatible aid, subject to compliance with the general principles of European law. The national courts must also, in principle, order the reimbursement of an unlawful aid: Any other interpretation would encourage the Member States to disregard the prohibition laid down in Article [108](3). Thus, if national courts could only order suspension of any new payment, aid already granted would subsist until the Commission’s final decision finding the aid incompatible with the common market and ordering its repayment. Having regard to the importance for the proper functioning of the common market of compliance with the procedure for prior review of planned State aid, national courts must in principle allow an application for repayment of aid paid in breach of Article [108](3) of the Treaty. However, … there may be exceptional circumstances in which it would be inappropriate to order repayment of the aid. … a national court requested to order the repayment of aid must grant that application if it finds that the aid was not notified to the Commission, unless by reason of exceptional circumstances repayment is inappropriate.11

B. The Purpose of Recovery The recovery of aid consists in ‘eliminating the advantage’ gained by the beneficiary in order to restore the competitive situation: [T]he obligation on a State to abolish aid regarded by the Commission as being incompatible with the common market has as its purpose to re-establish the previously existing situation … That objective is attained once the aid in question, increased where appropriate by default interest, has been repaid by the recipient … By repaying the aid, the recipient forfeits the advantage which it had enjoyed over its competitors on the market, and the situation prior to payment of the aid is restored.12

Recovery is not a ‘sanction’: its objective is to ensure that its recipient forfeits the advantage which it had enjoyed over its competitors on the market and to restore the situation existing prior to the payment of the aid …

10

Case 70/72 Commission v Germany, EU:C:1973:87, para 13. Case C-39/94 Syndicat français de l’Express international (SFEI) and others v La Poste and others, EU:C:1996:285, paras 68–71. 12 Case C-348/93 Commission v Italian Republic, EU:C:1995:95, paras 26–27. See also Case C-350/93 Commission v Italian Republic, EU:C:1995:96. 11

State Aid Recovery and the Energy Sector 305 Thus, even if the recovery of unlawful aid is implemented long after the aid in question was granted, it cannot constitute a penalty not provided for by Community law … In other words, decisions of the Commission ordering the recovery of State aid are measures designed to restore the previously existing situation and are not penal in character.13

Recovery is to be understood to include interest: [T]he restoration of the situation as it was prior to the payment of the illegal aid presupposes that all of the financial advantages resulting from the aid which adversely affect competition in the common market have been eliminated. It follows that a decision of the Commission ordering the recovery of illegal aid … may also require interest to be recovered on the sums granted in order to eliminate any financial advantages incidental to such aid.

The interest is equivalent to the financial advantage arising from the availability of the funds in question, free of charge, over a given period. Consequently, such interest may not start to run before the date on which the recipient of the aid actually had those funds at its disposal.14

These principles are set out in detail in Article 16(2) of Regulation No 2015/1589 and in Articles 9–11 of Regulation No 2015/2282.15 However, the Commission does not necessarily have to fix the exact amount of the aid to be repaid: [N]o provision of [Union] law requires the Commission, when ordering the recovery of aid declared incompatible with the common market, to fix the exact amount of the aid to be recovered. It is sufficient for the Commission’s decision to include information enabling the recipient to workout itself, without overmuch difficulty, that amount.16

C. Compliance with the Principles of Equivalence and Effectiveness in Recovery Procedures In the absence of Community rules governing the matter, … it is for the domestic legal system of each Member State to designate the courts and tribunals having jurisdiction and to lay down the detailed procedural rules governing actions for safeguarding rights which individuals derive from the direct effects of Community law, provided that such rules are not less favourable than those governing similar domestic actions (the principle of equivalence) and that they do not render practically impossible or excessively difficult the exercise of rights conferred by EU law (the principle of effectiveness).17

13 Case T-366/00 Scott v Commission, EU:T:2003:113, para 94, confirmed on appeal: Case C-276/03 P Scott v Commission, EU:C:2005:590 and Case C-290/07 P Commission v Scott, EU:C:2010:480. Also, Case T-366/00 Scott v Commission, EU:T:2007:99. 14 Case T-459/93 Siemens v Commission, EU:T:1995:100, paras 97–101, confirmed on appeal: Case C-278/95 P Siemens v Commission, EU:C:1997:240. 15 Commission Regulation (EU) 2015/2282 of 27 November 2015 amending Regulation (EC) No 794/2004 as regards the notification forms and information sheets (the Implementing Regulation). 16 Case C-441/06 Commission v France, EU:C:2007:616, para 29. Also see Case T-473/12 Aer Lingus Ltd v Commission, EU:T:2015:78. 17 Case C-526/04 Laboratoires Boiron, EU:C:2006:528, para 51.

306 Jacques Derenne The Commission may suspend the payment of an aid to a company with a pending recovery decision (so-called ‘Deggendorf ’ principle): when the Commission considers the compatibility of a State aid with the internal market, it must take all the relevant factors into account, including, where relevant, the circumstances already considered in a prior decision and the obligations that a previous decision may have imposed on a Member State. The Commission has the power to take into consideration, firstly, any accumulated effect of the old aid and the new aid and, secondly, the fact that the [old] aid declared unlawful had not been repaid.18 Finally, recovery may take place from the purchaser in the case of assets being misappropriated. For instance, the transfer of all the assets of the company Olympic Airways, free of all debt, to the new company Olympic Airlines appeared to have been structured in such a way as to make it impossible, under national law, to recover the debts of the former company Olympic Airways from the new company Olympic Airlines. The operation created an obstacle to the effective implementation of Decision 2003/372 and to the recovery of the aid, which was challenged by the Commission and condemned by the CJEU.19

D. Obstacles to Recovery and Defences for the Member State: ‘Absolute Impossibility’ of Implementing a Decision A Member State that is prosecuted for failing to execute a negative decision of the Commission may not resort to any defence mechanism unless there is an absolute impossibility to recover the aid: [T]he only defence available to a Member State in opposing an infringement action by the Commission … is to plead that it was absolutely impossible for it to implement the decision in question; … a Member State which, in giving effect to a Commission decision on State aid, encounters unforeseen and unforeseeable difficulties, whether of a political, legal or practical kind, or becomes aware of consequences overlooked by the Commission, must submit those problems to the Commission for consideration, together with proposals for suitable amendments to the decision in question. In such cases, the Commission and the Member State concerned must work together in good faith with a view to overcoming the difficulties whilst fully observing the provisions of the [TFEU] and in particular those on aid.20

The Member State (its administration and its courts) must disregard any national law that would prevent the recovery of an aid: The application of national procedures should not … impede the restoration of effective competition by preventing the immediate and effective execution of the Commission’s decision. To achieve this result, Member States should take all necessary measures ensuring the effectiveness of that decision. By providing for the suspensory effect of actions brought against demands for payment issued for the recovery of aid granted, the procedure laid down by [national] law … in the present case cannot be considered to allow the ‘immediate 18 19 20

Joined Cases T-244/93 and T-486/93 TWD Deggendorf v Commission, EU:T:1995:160, para 56. Case C-415/03 Commission v Greece, EU:C:2005:287, paras 33–34. Case C-441/06 Commission v France, EU:C:2007:616, paras 27–28.

State Aid Recovery and the Energy Sector 307 and effective’ … Thus, by failing to have regard to the objectives pursued by the Community rules on State aid, that national procedure has prevented the immediate restoration of the previously existing situation and prolonged the unfair competitive advantage resulting from the aid at issue. [This procedure] … should therefore have been left unapplied … the suspensory effect of actions brought before national courts cannot be considered essential for ensuring effective judicial protection in the light of Community law.21

Finally, the principle of Union institutional balance takes precedence over the principle of res judicata. A national judicial decision which has become res judicata cannot prevent the recovery of State aid granted in breach of Union law.22 Obstacles deriving from national law must be side-lined if necessary to ensure conformity with European law. Any failure to comply will constitute a breach of the Member State’s international obligations (under Articles 258 TFEU and 260 TFEU), but may also hold the Member State liable towards any person harmed by the breach of his obligations under the TFEU.23

III. ILLUSTRATION OF THE PRINCIPLES IN THE CASE LAW

These principles relating to the recovery of unlawful and incompatible aid are illustrated hereafter by recent European cases, in accordance with the following nomenclature: —



The subject matter and the terms of the refund (annulment of acts implementing unlawful aid measures, rules on limitation periods, identification of beneficiaries, determination of amounts to be recovered and forms of recovery). The defences available to the Member State, against the principle of effectiveness (absolute impossibility of implementing a negative decision, protection of legitimate expectations, delays in implementation and correlative sanctions by the Member State).

A. Recovery: A Logical Consequence of the Unlawfulness of the Aid In the Residex Capital judgment of 8 December 2011, the CJEU was asked to answer a preliminary question posed by the Dutch Supreme Court. This concerned the issue of whether Article 108(3) TFEU dictates that, where the unlawful aid measure was implemented by granting the lender a guarantee that enabled the borrower to obtain a loan which would not have been available under normal market conditions, the national courts are bound—or at any rate authorised—to cancel the guarantee, even if that does not result in the cancellation of the loan.

21

Case C-232/05 Commission v France (Scott), EU:C:2006:651, paras 50–60. Case C-119/05 Lucchini, EU:C:2007:434. See below. 23 Joined Cases C-6/90 and C-9/90 Andrea Francovich and Danila Bonifaci ea v Italy, EU:C:1991:428; Joined Cases C-46/93 and C-48/93 Brasserie du Pêcheur SA c/ Bundesrepublik Deutschland and The Queen v Secretary of State for Transport, ex parte: Factortame Ltd ea, EU:C:1996:79; Case C-224/01 Köbler, EU:C:2003:513 and Case C-173/03 Traghetti del Mediterraneo, EU:C:2006:391. 22

308 Jacques Derenne The CJEU ruled that the national court may, in the absence of less onerous procedural measures, declare the cancellation of the guarantee if it takes the view that cancellation may lead to or facilitate the restoration of the competitive situation which existed before that guarantee was provided. When underlying the applicable rules, the CJEU refers to the Enforcement Notice.24 The CJEU of course confirms the national court’s full jurisdiction to draw all the consequences of the illegality found, including the validity of the acts involving the enforcement of the guarantee in question. As to whether the law of the Union requires the national courts to annul a guarantee granted in those circumstances, the CJEU reaffirms the fundamental principle according to which ‘the logical consequence of a finding that aid is unlawful is to remove it by means of recovery in order to restore the previous situation’.25 The main objective pursued in recovering unlawfully paid State aid is ‘to eliminate the distortion of competition caused by the competitive advantage which such aid affords. By repaying the aid, the beneficiary forfeits the advantage which it had over its competitors on the market, and the situation prior to payment of the aid is restored’.26 On the question of the beneficiaries’ identity (to which reference is made below), the CJEU considers in essence that, irrespective of the beneficiary of the aid, Union law does not require the annulment of a particular act (in this case the guarantee in question). However, courts must ensure that the measures which they take with regard to the validity of the above-mentioned acts make it possible to restore the competitive situation existing prior to the payment of the aid in question. As a result, the cancellation of the guarantee is not necessarily the most effective measure. However, the annulment of the underlying contract, ‘liable to lead to the mutual restitution of the services performed by the parties or the disappearance of an advantage for the future, may be better able to achieve the objective of restoring the competitive situation which existed before the aid was granted’,27 since it will enable recovery of the overall amount of the aid in question, which was not limited to the guarantee but to the credit which had been granted by virtue of that guarantee. With regard to the recovery of the aid, which is the more or less automatic consequence where that aid is held to be unlawful and incompatible, subject to the sole condition that an order for recovery is not contrary to a general principle of EU law, the GCEU considers, in its judgment Regione autonoma delle Sardegna of 20 September 2011, that the Commission has no discretion in this respect. Accordingly, ‘where negative decisions are taken in cases of unlawful aid, the Commission shall decide that the Member State concerned shall take all necessary measures to recover the aid from the beneficiary’.28

24

Case C-275/10 Residex Capital IV CV v Gemeente Rotterdam, EU:C:2011:814, paras 25–26 and 29. ibid, para 33. See also, Case C-403/10 P Mediaset v Commission, EU:C:2011:533, para 122. 26 ibid, para 34, and reference see Case C-350/93 Commission v Italian Republic, EU:C:1995:96, para 22. 27 ibid, para 47. 28 Cases T-394/08, T-408/08, T-453/08 and T-454/08 Regione autonoma della Sardegna and others v Commission, EU:T:2011:493, para 152 (appeal case C-630/11 P, EU:C:2013:387). 25

State Aid Recovery and the Energy Sector 309 In practice, there is a certain reluctance from national courts to make full use of their powers, in compliance with their EU law obligations, in particular when it comes to ordering the recovery of unlawful aid.29

B. The Limitation Period The obligation to recover has only two limits: first, where such recovery would be contrary to a general principle of EU law (the principles of the protection of legitimate expectations or legal certainty are most often invoked but they are interpreted in a very restrictive manner by the case law—see below); and second, where the limitation period provided for in Article 17 of Council Regulation 2015/158930 applies. In Département du Loiret, the GCEU considered that the expression in Article 17 of Regulation 2015/1589 that ‘any action taken by the Commission … shall interrupt the limitation period’ concerned, inter alia, for simple requests for information by the Commission to the Member State concerned. The decision to initiate the formal investigation procedure or an injunction procedure is therefore not the only act interrupting the limitation period. The starting point of the limitation period (the day on which the aid was unlawfully granted) has also been the subject of case law. In its Cipriani judgment of 2008, the GCEU held that the limitation period did not begin until the date on which the unlawful aid was paid out, not from the date on which the aid measure was adopted. Thus, ‘in the case of an aid scheme introduced more than ten years before the first interruption of the limitation period, the unlawful aid incompatible with the common market granted under that scheme during the last ten years is therefore subject to recovery’.31 This question was raised in the case of France Telecom (business tax), which gave rise to a judgment on appeal by the CJEU on 8 December 2011. In that case, the Commission had decided that the limitation period was to run from the day that the

29 An example is a judgment of the President of the Bruges Commercial Court who ordered the suspension of the implementation of the unlawful aid to the Minque of Ostend until the end of the formal investigation procedure. And all of this, without having any regard to the consequences of the unlawfulness of the aid in question, in particular by ordering the recovery of the aid. (Pres Trib Com Bruges, of 12 February 2009, Zeebrugse Visveiling v Exploitatie Vismijn Oostende NV/Pakhuizen Oostende NV, 00886/08). See J Derenne and M Smeets, ‘Chronique “Aides d’État” —Application des règles sur les aides d’État dans des affaires concernant la Belgique (juridictions belges et européennes, Commission européenne)—Années 2009 et 2010’ Tijdschrift voor Belgische Mededinging (2011–12) Revue de la Concurrence Belge—Tijdschrift voor Belgische Mededinging RCB-TBM 100. 30 ‘1. The powers of the Commission to recover aid shall be subject to a limitation period of 10 years. 2. The limitation period shall begin on the day on which the unlawful aid is awarded to the beneficiary either as individual aid or as aid under an aid scheme. Any action taken by the Commission or by a Member State, acting at the request of the Commission, with regard to the unlawful aid shall interrupt the limitation period. Each interruption shall start time running afresh. The limitation period shall be suspended for as long as the decision of the Commission is the subject of proceedings pending before the Court of Justice of the European Union. 3. Any aid with regard to which the limitation period has expired shall be deemed to be existing aid’. 31 Joined Cases T-254/00, T-270/00 and T-277/00 Hotel Cipriani v Commission, EU:T:2008:537, para 364.

310 Jacques Derenne actual granting of the aid to France Télécom took place (annually) and not from the date that the tax scheme was adopted.32 In its judgment of 8 December 2011, the CJEU confirmed the GCEU’s interpretation by holding that ‘the determination of the date on which aid was granted may vary depending on the nature of the aid in question’. In the present case, France Telecom’s tax system was a multi-annual scheme, entailing payments or advantages granted on a periodic basis. In this case, the date on which an act forming the legal basis of the aid is adopted and the date on which the undertakings concerned will actually be granted the aid may be a considerable period of time. In such a case, for the purpose of calculating the limitation period, the aid must be regarded as not having been awarded to the beneficiary until the date on which it was in fact received by the beneficiary.33 The CJEU therefore confirmed the reasoning of the GCEU whereby, in view of the annual nature of business tax, the aid at issue could not be regarded as having been awarded before the year 1994, because that was the year in which the binding legal acts were adopted which made it possible, for the first time, to establish the existence of a tax differential.34

Finally, the GCEU held in its judgment Gibraltar, that Article 15 of Regulation 659/199 (now Article 17 of Regulation 2015/1589), ‘does not in any way express a general principle whereby new aid is transformed into existing aid but merely precludes recovery of aid established more than 10 years before the Commission first intervened.’35

C. Identification of the Beneficiary and Recovery from a Third-Party Purchaser i. Identification of the Beneficiary The question of the identification of beneficiaries (of those who actually benefited from the aid) was mentioned above in the Residex case. While the Dutch Supreme Court’s question did not relate to this issue, the CJEU invited the national court to examine it, since it was closely connected to the question

32 Commission Decision of 2 August 2004 on the State aid implemented by France for France Télécom [2005] OJ L269/30, para 49: ‘in the case of aid schemes, the limitation period for recovery starts to run, not on the day on which the scheme was introduced, but on the day on which the aid was actually granted to the recipient … Consequently, the limitation period in respect of the aid granted to FT under this specific tax scheme starts to run, not on the day on which Law No 90-568 was adopted, but on the day on which the aid was actually granted to FT, that is to say annually at the time when the business tax was due’; confirmed by the GCEU in Joined Cases T-427/04 and T-17/05 France v Commission, EU:T:2009:474, paras 202–03 and 322–23, confirmed by the CJEU on appeal in Case C-81/10 P, EU:C:2011:811. 33 Case C-81/10 P, France Télécom v Commission, EU:C:2011:811, para 82. 34 ibid, para 86. 35 Joined Cases T-195/01 and T-207/01 Government of Gibraltar v Commission, EU:T:2002:111, para 130.

State Aid Recovery and the Energy Sector 311 raised by the latter, ie, whether the annulment of the guarantee was effective for the recovery of the aid. The CJEU notes that in order to carry out this repayment, it is essential that the national courts identify the beneficiary or, as the case may be, the beneficiaries of the aid. In the case where aid is granted in the form of a guarantee, the beneficiaries of that aid may be either the borrower or the lender or, in certain cases, both of them together.36

In the present case, the lender Residex would also have been able to derive an economic advantage from the guarantee at stake since the borrower was in a financial situation such that he would not have been able to obtain a loan on the capital markets and that Residex was able to grant a loan at a rate that was preferential in comparison with that in force on the market only on the basis of the guarantee obtained. The CJEU entrusts the referring court with the task of examining those questions with a view to ‘the recovery of the total amount of the aid in question’ (see paragraph 43 of the judgment cited above).37 The issue was also examined in the case of DHL/Leipzig-Halle which concerned commitments made by the public operator of the Leipzig-Halle airport to guarantee the continuous operation of a specific runway that would allow at least 90 per cent of the air transport to carried out by or for DHL at any time from this runway. DHL contested that those commitments had been ‘made available’ and that it had benefited from them, these commitments were thus to be viewed as null and void ab initio (nullity of a contract for infringement of Article 108(3) TFEU). In its judgment of 7 October 2010, the GCEU rejected the applicant’s argument and held that even assuming that the commitments at issue were considered null and void under German law, ‘that would not affect the recovery obligation or the lawfulness of the Decision’.38 According to the GCEU, admitting the applicants’ argument would make it possible for Member States whose domestic law treats measures granting unlawful aid as null and void to grant undertakings the benefit of such aid, while precluding its recovery should the Commission find the aid to be incompatible and order it to be recovered.39

ii. The Notion of Economic Continuity The general rule is that once the Commission establishes the incompatibility of an aid, it decides that the Member State concerned takes all necessary measures to recover the aid from the beneficiary (Article 16 of Regulation 2015/1589). For the recovery to be effective, it is possible the aid be recovered from undertakings or persons other than the initial beneficiary when there are signs of ‘circumvention’ of

36

Case C-275/10 Residex Capital, EU:C:2011:814, para 37. ibid, para 43. Case T-452/08 DHL Aviation SA/NV and DHL Hub Leipzig GmbH v Commission, EU:T:2010:427, para 34. 39 ibid, para 43. 37 38

312 Jacques Derenne the recovery obligation. This may be the case when the original beneficiary sells the entirety or part of its assets that benefited from the aid. In this scenario, the obligation to repay the aid may lie with the acquiring undertaking on the basis of the principle of the economic continuity. The economic continuity is established on the basis of certain conditions, namely: the purpose of the transfer, the transfer price, the identity of the shareholders or owners of the acquiring undertaking and of the original undertaking, the moment at which the transfer was carried out and the economic logic of the transaction.40 These conditions must be objective and the burden of proof lies with the Commission.41 With regard to the purpose of the transfer, it must be determined whether the sale would allow the acquirer to continue the subsidised activities and thus perpetuate the disruption of competition.42 For example, the sale of active elements without the passive is an indication of economic continuity.43 The Commission may also examine the moment of the transfer, which ‘is also objective and does not imply the existence of an intention to evade. It must be understood as meaning that the time of the transfer is capable of constituting, according to the circumstances of the case, evidence of evasion’.44 The Commission will examine whether the transfer took place before the opening of the proceedings, during the assessment of the aid or after the adoption of the recovery decision. With regard to the economic logic of the transaction, the Commission examines whether the buyer will utilise the actives in the same way/manner as the previous owner or whether the former will establish a different economic strategy. Finally, the transfer price must reflect the price on the market. According to the CJEU in Banks: Since those undertakings bought the companies in question under non-discriminatory competitive conditions and, by definition, at the market price, that is to say at the highest price which a private investor acting under normal competitive conditions was ready to pay for those companies in the situation they were in, in particular after having enjoyed State aid, the aid element was assessed at the market price and included in the purchase price. In such circumstances, the undertakings to which the tenders were granted cannot be regarded as having benefited from an advantage in relation to other market operators.45

It is worth mentioning here a couple of particular recent cases. In the Val Saint-Lambert case,46 the Commission concluded that the incompatible aid at hand should be recovered from VSL. However, VSL had declared bankruptcy, so the national authorities asked the Commission to specifically confirm the absence of economic continuity between VSL and a company that wished to acquire the

40

See, eg, Joined Cases C-328/99 and C-399/00 Italy v Commission, EU:C:2003:252, para 78. Case T-415/05 Greece v Commission, EU:T:2010:386, para 146 and Joined Cases C-328/99 and C-399/00, para 85. 42 Commission Decision of 15 October 2014, SA.33797—Slovakia, para 149. 43 Case C-415/03 Commission v Greece, EU:C:2005:287, para 33. 44 Case C-415/05 Commission v Greece, EU:T:2010:386, para 146. 45 Case C-390/98 Banks, EU:C:2001:456, para 77. 46 Commission Decision (EU) 2015/1825 of 31 July 2014 on non-notified State aid SA.34791 (2013/C) [2015] OJ L269/47; SA.38810 (purchase of certain assets following the bankruptcy of VSL). 41

State Aid Recovery and the Energy Sector 313 former in the context of the liquidation. In those circumstances, the Commission issued a separate decision in which it decided that there was indeed no economic continuity between VSL and the potential buyer. This was the first sui generis decision on economic continuity, which provided a third party with clarification on the risks of a takeover operation. In the Sernam case, the Commission adopted a decision, following the opening of the formal investigation, declaring that aid granted to Sernam should be recovered from the said company. Following this negative decision, the national authorities requested the Commission to confirm the absence of continuity between Sernam and the companies that had submitted offers in the context of the former’s legal redress. The Commission indeed established the absence of economic continuity through such a sui generis decision without opening a new formal investigation.47

D. Recovery from a Third-Party Purchaser The sale of shares in a company which is the beneficiary of unlawful aid by a shareholder to a third party does not affect the requirement for recovery. The transfer of the advantage to be recovered may result from the acquisition by a third party at a price below the market price of the assets or shares of the original beneficiary of the aid. This third party can sometimes be an entity created to circumvent the obligation of repayment and with which there is economic continuity (in particular where the transfer does not have any economic logic other than to deprive of concrete effect the recovery decision).48 In the Olympic Airways case, following the CJEU’s judgment of 2005,49 which identified the actual beneficiaries of the aid referred to in a Commission decision of 11 December 2002, the application of those principles arose as a result of the privatisation of Olympic Airways, now Olympic Airways Services (OA), and of the creation of Olympic Airlines (NOA). The question of the economic continuity between the two entities was at the heart of the debate. In a decision of 14 September 2005, challenged before the GCEU by the Hellenic Republic, the Commission found that various aids were granted to OA and NOA 47 Case T242/12 Société nationale des chemins de fer v Commission, EU:T:2015:1003 [2016]. These decisions and the subsequent litigation are further described in the AG Opinion in the appeal against the recovery order in Case 27/16 P of 20 July 2017. 48 Case C-277/00 Germany v Commission, EU:C:2004:238, paras 75 ff; Joined Cases C-328/99 and C-399/00 Italy and SIM 2 Multimedia v Commission, EU:C:2003:252, para 83; Case C-390/98 Banks, EU:C:2001:456, para 77. See Recovery Notice, paras 32–35. 49 Case C-415/03 Commission v Greece, EU:C:2005:287. The CJEU found a failure in the execution of the decision of 11 December 2002 and it found also a transfer made by Greece of the most profitable assets of OA, free of all debts, to NOA, belonging to the State and enjoying special protection vis-a-vis creditors (by way of derogation of the provisions of common law and the obligations of commercial law). According to the CJEU, that legal construction made impossible, under national law, the recovery of the aid granted and hindered the effective implementation of the decision of 11 December 2002. In Case T-68/03 (Olympiaki Aeroporia Ypiresies v Commission, EU:T:2007:253) the decision was partially annulled in the part that tolerates the continuation of the non-payment of, on the one hand, the airport’s royalties by OA to the Athens International Airport (AIA) and, on the other, the value added tax by OA on fuel and spare parts.

314 Jacques Derenne and thus ordered their repayment.50 The Commission concluded, on the basis of the analysis of the CJEU’s judgment of 2005 regarding the economic continuity between OA and NOA, to the extension of the recovery obligation to NOA with regard, in particular, to aid at issue granted to OA prior to the splitting. It also referred to this element in its compatibility assessment, since the new aid to NOA could not be declared compatible as long as the aid prior to the split had not been recovered, in particular from NOA, which was the successor to OA for the recovery of the aid. In its judgment of 13 September 2010, notwithstanding the absence of clear indications in the operative part of the decision (the Commission did not expressly identify NOA as the effective recipient of part of the aid granted to OA), the GCEU considered that the Commission duly imposed a duty upon the Greek State to recover the aid, which had been granted to OA prior to the split, from OA or NOA. As regards the apportionment of the recovery obligation between recipients of aid, the Commission is not required to state to what extent each recipient undertaking has benefited from the amount of the aid in question. It is for the Member State concerned to determine the amount which must be repaid by each of those undertakings in its recovery of the aid.51

The GCEU concluded that there was no manifest error of assessment by the Commission in its finding of economic continuity between OA and NOA for the recovery of the aid granted to OA before the splitting.52 In its view, the Commission could transpose the CJEU’s reasoning in that regard even though at the time of its judgment in May 2005, the transfer in question had already been carried out.53 The GCEU added that ‘the criteria laid down in the case law for identifying the effective recipient of aid are objective’ (such as the absence of any payment in consideration of the transferred assets, or of a price consistent with market conditions, or the objective fact that the effect of the transfer is to evade the obligation to repay the aid at issue). ‘The presence of an intentional element is necessary to find that the obligation to repay aid has been evaded by the transfer of assets’. ‘[T]he Court cannot accept the applicants’ argument that the restructuring of the Olympic Airways Group and the transfer of the flight operations to NOA were required by the economic logic of a more effective recovery of the aid granted to OA through the

50 By a judgment of 14 February 2008 (Case C-419/06 Commission v Greece, EU:C:2008:89), the CJEU found a breach of that decision. 51 Joined Cases T-415/05, T-416/05 and T-423/05 Hellenic Republic, Olympiakes Aerogrammes AE, and Olympiaki Aeroporia Ypiresies AE v Commission, EU:T:2010:386, para 126. 52 The GCEU notes, in particular, the transfer timing of the assets from OA to NOA, in the flight operations sector, which was made after the decision of 2002, but prior to the initiation of the formal investigation procedure leading to the adoption of the contested decision of 14 September 2005. 53 Joined Cases T-415/05, T-416/05 and T-423/05 Hellenic Republic, Olympiakes Aerogrammes AE, and Olympiaki Aeroporia Ypiresies AE v Commission, paras 138 and 139. An injunction to provide information was sent to the Hellenic Republic on 8 September 2003 and the Hellenic Republic and OA could not ignore, when the NOA was created, that the measures in favour of OA prior to the split could be investigated by the Commission and these measures were in line with certain previous aid. Due to the similarity of the context, the analysis of the judgment of 12 May 2005 can also be applied to the aid prior to the split in question.

State Aid Recovery and the Energy Sector 315 privatisation of NOA’. ‘The purpose of the obligation to recover aid is to restore the competitive situation in the economic sector concerned’.54 i. Determination of the Amount to be Recovered The determination of the amount of aid to be recovered has also been the subject of litigation before the EU courts. However, the Commission stresses in its Recovery Notice that, when it has the necessary data at its disposal, it will (…) endeavour to quantify the precise amount of aid to be recovered. It is clear, though, that the Commission cannot and is legally not required to fix the exact amount to be recovered. It is sufficient for the Commission’s decision to include information enabling the Member State to determine the amount, without too much difficulty.55

This general principle was reaffirmed by the CJEU in the Mediaset case. In this case, an infringement of the principle of legal certainty was alleged against a judgment of the GCEU that dismissed an action brought against a decision which did not allow for an adequate recovery methodology to be established. The CJEU held, inter alia, that the Commission can leave it to the national authorities to calculate the exact amounts to be repaid.56 Furthermore, the GCEU’s judgment of 24 March 2011 in the case of Freistaat Sachsen also provides an illustration of the risk of the Commission indicating the amount when it can, even in cases where it declared unlawful aid to be compatible. In the present case, the Commission declared compatible a €350 million aid for the construction of a new runway and other airport infrastructure at the Leipzig-Halle airport financed by public funding. However, the Commission also stated in its decision that some of the costs involved were part of the exercise of public tasks and could not be regarded as State aid. However, the Commission considered that the entire capital contribution constituted State aid but it was not necessary to make a definitive assessment as to the costs involved since the aid was declared compatible.

54 ibid, paras 146 and 148–50. The GCEU also notes the special circumstance that ‘the restructuring of OA and the formation of NOA were not short-term transactions, intended to facilitate the privatisation. The transfer to NOA of the flight operations sector of the Olympic Airways Group was effected by legislation, derogating from the general law, and the entire capital of that new company was immediately vested in the Hellenic Republic. In those circumstances, in the absence of payment of any consideration, by a new acquirer, as long as the privatisation of the airline had not been concluded, there was no need to determine whether the amount of the aid granted to OA prior to the hiving-off could be regarded as included in a purchase price consistent with market conditions’. 55 See para 37 of the Recovery Notice; see Case C-102/87 France v Commission, EU:C:1988:391, para 33 and Case C-310/99 Italy v Commission, EU:C:2002:143, para 91. 56 Case C-403/10 P Mediaset v Commission, EU:C:2011:533, para 127; Case C-69/13 Mediaset SpA v Ministero dello Sviluppo economico, ECLI:EU:C:2014:71: in this reference for a preliminary ruling, the CJEU affirmed that the national court is bound by the decision of the Commission finding an aid unlawful and incompatible and ordering the recovery. Nevertheless, the national court is not bound by the institution’s later statements of position, but must take them into consideration by virtue of the principle of sincere cooperation. In that regard, the calculations made by the national court to quantify the amounts to be repaid may, on the basis of all the factors of which it has been made aware, result in an amount equal to zero.

316 Jacques Derenne After recalling the principle that there is no obligation to indicate the exact amount to be recovered in the event of illegal and incompatible aid, the GCEU criticises the Commission’s position: [W]ithout it being necessary to consider whether such a principle also applies where the Commission declares aid compatible with the common market, it must be considered that, where it decides to state, in the operative part of a decision, an amount of State aid within the meaning of Article 107, paragraph 1 [TFEU], the Commission must indicate the correct amount.57

The GCEU went on to suggest that the correctness of the amount of unlawful aid stated by the Commission in the operative part of a final decision finding that the aid is compatible with the internal market is all the more important as it is likely to affect the amount of interest that the recipient can be required to pay. National courts before which proceedings might be brought are required to order the recipient of the aid to pay interest in respect of the period of unlawfulness and the amount of that interest depends, inter alia, on the amount of the State aid as such. The interest in question must be calculated on the basis of the total amount of the State aid and not merely on the amount of the unlawful aid regarded as compatible with the internal market. The amount of the State aid appeared to be incorrect having regard to the reasons stated since the amounts falling within the public policy remit do not constitute State aid and must therefore be deducted from the total amount of the capital contribution, which has been classified as State aid. As the GCEU suggests, ‘the relevant question … is not to determine the amount of the aid compatible with the [internal] market but to determine the amount of the aid as such’.58 In the above-mentioned France Telecom case, France Télécom also alleged an infringement of the principle of legal certainty against the GCEU’s judgment. According to France Télécom, the GCEU was wrong in its ruling that the Commission should be justified in ordering the recovery of aid on the basis of approximations provided by the French authorities (estimate of a possible range of tax aid). According to the GCEU, the principle of legal security is not ignored by the mere fact that the amount of the aid must be specified for purposes of its restitution.59 No provision of European Union law requires the Commission, when ordering the recovery of aid declared incompatible with the common market, to fix the exact amount of the aid to be recovered. It is sufficient for the Commission’s decision to include information enabling the recipient to calculate the amount itself, without much difficulty. With French authorities having been unable to calculate exactly the amount of the advantage from which France Télécom had benefited under the regime at issue, the Commission was entitled to rely on the data provided and the

57

Case T-443/08 Freistaat Sachsen, EU:T:2011:117, para 228. ibid, para 232. 59 The Commission estimated the amount of aid between €798 million and €1.14 billion, which is the range within which the final amount was to be established. In that context, the GCEU held that the contested decision contained the appropriate information enabling the determination of that amount, without undue difficulty. 58

State Aid Recovery and the Energy Sector 317 GCEU correctly concluded that the contested decision did not constitute a breach of the principle of legal certainty.60 Under these circumstances, the GCEU was therefore entitled to find that the contested decision is not vitiated by unlawfulness because it simply refers to an indicative range as regards the amount of aid to be recovered.

IV. FORMS OF RECOVERY

Since there are no EU provisions on the procedure for recovery of wrongly paid amounts, unlawful aid must, in principle, be recovered in accordance with the relevant procedural provisions of national law. The CJEU has accordingly confirmed that aid may be repaid otherwise than by a payment in cash, in this case, under the compensation mechanism. In a judgment where the CJEU ordered Greece to pay a daily penalty payment and a lump sum on account of insufficient recovery of aid incompatible with the common market, the CJEU held that ‘so long as it is provided for under the national legal system as a mechanism for extinguishing debts, a set-off operation can constitute an appropriate means by which State aid may be recovered’.61 As regards the implementation of a decision requiring the recovery of unlawful aid, whereby a Member State recovers that aid by means other than a cash payment, it must provide the Commission with all the information enabling it to establish that the means chosen constitute an adapted implementation of the decision.62 As regards compensation, the debt in question, in particular, must be payable. Other types of reimbursement, particularly in the application of the principles of tax law, are also possible under the aforementioned principles. They are often the subject of confidential exchanges between Member States and the Commission.

A. Absolute Impossibility of Implementation What defence can a Member State attempt to invoke against a decision ordering the repayment of unlawful and incompatible aid, when it cannot challenge the validity of that decision? EU case law is extremely severe on this issue, showing a rigidity that may not be compatible with the reality of legal, economic and political circumstances. However, it must be borne in mind that this assessment takes place in proceedings for failure to fulfil obligations resulting from a decision of the Commission. And, most of the time, the facts are there: the unlawful and incompatible aid has not been recovered, either not within the prescribed period, or not in full, or sometimes not at all.

60

Case C-81/10 P France Télécom v Commission, EU:C:2011:811, paras 102–04 Case C-369/07 Commission v Hellenic Republic, EU:C:2009:428, para 68. 62 Case-209/00 Commission v Germany, EU:C:2002:74; Case C-369/07 Commission v Hellenic Republic, EU:C:2009:428, paras 40–41. 61

318 Jacques Derenne The only admissible defence recognised by EU case law is the absolute impossibility of giving effect to the decision in question. And the case law has made it hardly possible for Member States to rely upon such a defence. In order to avoid a finding of failure to fulfil obligations, the national authorities should especially adopt the following measures strictly: — — — —



Inform the Commission of the legal, political or practical difficulties involved in implementing the decision (obligation to state reasons).63 Propose to the Commission an alternative arrangement for implementing the decision which could have enabled those difficulties to be overcome. Draw up a list of companies subject to recovery of the aid and to carry out actual recovery. Where an undertaking is subject to collective proceedings, the re-establishment of the previous situation and the elimination of the distortion of competition resulting from the unlawfully paid aid may, in principle, be achieved by registration of the liability relating to the repayment of the aid in question in the schedule of liabilities. In the case of divested businesses, carry out any concrete steps to examine the situation of each of these businesses (eg, check the price of the business transfer), possibly by focusing on the most important transfers of assets, after approval from the Commission.

Two cases involving France may be mentioned here to illustrate these items. On 13 November 2008, the CJEU ruled that France failed to execute a Commission decision of 16 December 2003. This decision ordered France to abolish an aid scheme granting tax exemptions for the takeover of companies in difficulties and to recover the illegal aid already granted from the beneficiaries.64 The debates which took place before the CJEU, in the context of an action brought against a failure to fulfil its obligations under Article 260 TFEU, focused in particular on the recovery of aid from beneficiaries who had not ceased their activities and those who had ceased their activities. The CJEU cites the conditions laid down in (then) Article 14(3) of Regulation No 659/1999: [R]ecovery shall 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. To this effect and in the event of a procedure before national courts, the Member States concerned shall take all necessary steps which are available in their respective legal systems, including provisional measures, without prejudice to Community law.65 63 See, eg, C-219/13 Ferraci v Commission, EU:C:2014:2207: if recovery is impossible, there is an obligation to establish sincere cooperation. The Member State must submit the reasons why it may not recover and the Commission must carefully examine those reasons. 64 Case C-214/07 Commission v France, EU:C:2008:619. See J-Y Chérot, ‘Récupération: La Court de justice rend un nouvel arrêt illustrant les règles relatives á la récupération des aides d’État (France)’ (2009) 1 Concurrences 146. With regard to certain national judicial decisions on the aid scheme, see J Derenne, ‘La récupération des aides illégals et incompatibles: Un tour d’horizon sélectif de la jurisprudence européenne depuis la communication “récupération” de 2007 de la Commission européenne’ (2012) 1 Concurrences 73. See also Case C-37/14 Commission v France, EU:C:2015:90. 65 Case C-214/07 Commission v France, EU:C:2008:619, para 43.

State Aid Recovery and the Energy Sector 319 The CJEU then reiterates that the only defence available to a Member State in opposing an application by the Commission under Article 108(2) TFEU for a declaration that it has failed to fulfil its obligations is to plead that it was absolutely impossible for it to implement the decision properly and, in the event of difficulties, the Commission and the Member State must, pursuant to the rule imposing on the Member States and the EU institutions reciprocal duties of genuine cooperation, work together in good faith with a view to overcoming those difficulties. Regarding the recovery of the aid from recipients which have not ceased their activities, the CJEU states that the aid relating to the 1991 to 1993 financial years is not at issue in the present proceedings. In respect of that aid, the Commission, prior to commencing the action, had already accepted the absolute impossibility of recovery, a view which it expressly confirmed in its application. On the other hand, the CJEU rejects the plea for all the beneficiaries who have not ceased their activity in respect to the fiscal years after 1994. The difficulties alleged by the French authorities (eg, the recipients’ identity, the calculation of the amount of the aid to be recovered and the choice and implementation of recovery procedures) are ‘internal difficulties caused by the national authorities’ own acts or omissions’.66 Moreover, the French authorities have not drawn up an exhaustive list of companies in receipt of aid and have not failed to implement the decision within the prescribed period. With regard to the recovery of the aid from recipients who have not ceased their activities, the Commission had not, either in its exchanges prior to the present proceedings or in its application, complained that the defendant Member State had failed to implement the decision with regard to the companies which simply disappeared without finding a buyer. In the case of the recipients who have ceased their activity and transferred their assets, the CJEU points out, that it is for the national authorities to check whether the financial conditions of the transfer complied with market conditions. To that end, the national authorities may take into consideration, in particular, ‘the form of the transfer, for example, public tendering, deemed to ensure that a sale takes place under market conditions’ or ‘any expert’s report prepared at the time of the transfer’. In the case of a privately negotiated transfer of assets, the recovery of the aid from the transferee cannot be dependent on an express reference in the legal instrument of a transfer of that aid. Recovery can take place where the transferee should have been aware of the existence of the aid and of a review procedure instituted by the Commission.67

In the present case, the CJEU concludes that France ‘may not, in order to avoid a declaration that it has failed to fulfil its obligations, simply make general and abstract statements without referring to specific individual cases, analysed in the light of all the steps actually taken to implement the decision’. ‘To allow the contrary would amount to precluding already a priori any implementation with regard to the whole category of undertakings which have ceased their activity, whereas, in their

66 67

ibid, para 50. ibid, paras 59, 60 and 62.

320 Jacques Derenne case, an absolute impossibility of implementation may be accepted, where relevant, only on the basis of their individual circumstances’.68 However, France claimed, in its exchanges with the Commission, an absolute impossibility of implementation vis-a-vis 204 undertakings which have ceased their activity but provided no evidence that it has taken any concrete steps to examine the situation of each of them and to determine whether or not it necessitated recovery pursuant to the criteria set out above. It [did] not provide evidence even of having taken advantage of the Commission’s acceptance, … of a review restricted only to the most significant asset transfers.69

In those circumstances, the CJEU concluded that there was no absolute impossibility of implementation. In a case of 2011 where France had failed to implement the decision of the Commission concerning certain aid to fish farmers and fishermen, the CJEU took a similar approach and rejected all the arguments put forward by France.70 The CJEU reiterates the above-mentioned principles and, with regard to time limits, notes that the fact that the time limits have been postponed to reply to requests for information may not be interpreted as granting an additional period to recover the aid in question pursuant to the decision. The CJEU then notes that, in their exchanges with the Commission, the French authorities did not allege any absolute impossibility of implementing the decision but merely ‘explained various justifications for the delay in the recovery procedure and the need to analyse the situation of each of the undertakings concerned. Since those difficulties constitute internal difficulties caused by the national authorities’ own acts or omissions …, the arguments based on those difficulties cannot be upheld’.71 Moreover, the CJEU stresses that the difficulty in identifying the beneficiaries of the aid in question, due to the need to verify the individual situation of each undertaking concerned, is not such as to justify the failure to implement a decision.

B. Protection of Legitimate Expectations: Exceptional Circumstances In the same way that the argument based on the absolute impossibility of executing a negative decision, those based on the alleged breach of the principle of the protection of legitimate expectations or exceptional circumstances also seem to be bound to perpetuate failure before the European Courts. In view of the mandatory nature of the review of State aid by the Commission, undertakings to which aid has been granted may not, in principle, entertain a legitimate expectation that the aid is lawful unless it has been granted in compliance with the procedure laid down

68

ibid, paras 63–64. ibid, para 65. 70 Case C-549/09 Commission v France, EU:C:2011:672. See also Case C-305/09 Commission v Italy, EU:C:2011:274; Case C-331/09 Commission v Poland, EU:C:2011:250, paras 70–73; Case C-507/08 Commission v Slovakia, EU:C:2010:802, paras 42–64. 71 Case C-549/09 Commission v France, EU:C:2011:672 paras 37–38 69

State Aid Recovery and the Energy Sector 321 in Article [108 TFEU] and a diligent business operator should normally be able to determine whether that procedure has been followed. In particular, where aid … is unlawful …, the recipient of the aid cannot have at that time a legitimate expectation that its grant is lawful.72

In the case of France Télécom (business tax), the tax regime in question had not been notified to the Commission. The CJEU noted that ‘neither the purported complexity of the tax regime at issue nor the periodic nature of the aid measure can release the Member State from its obligation to notify or give rise to any legitimate expectation on the part of the company receiving the aid.73 In ThyssenKrupp, the GCEU conducted a more detailed analysis of the circumstances put forward by a beneficiary that claimed that the principle of the protection of legitimate expectations had been breached. The GCEU provides a genuine theoretical account of how the Commission must take account of the possible legitimate expectations of a beneficiary when it adopts a decision ordering the recovery of unlawful aid. The GCEU stresses the changes brought about by the adoption of the procedural regulation (now Regulation No 2015/1589): ‘The adoption of Regulation No 659/1999 created a new situation as regards the recovery of incompatible aid. All legal consequences should be drawn’. The GCEU then finds that the Commission did verify whether there were exceptional circumstances in this case and rightly concluded that there were no such circumstances. C. The Principle of Effectiveness The intervention of a national court is sometimes a cause of delay in the implementation of a negative decision by the Commission, whether or not it is challenged before the GCEU. An action brought by the beneficiary may be brought before the national court against recovery operations under national law itself taken in execution of the Commission’s negative decision. This may result in the Commission’s action against the Member State concerned for obstacles raised by its national courts to the aid recovery.74 The CJEU has delivered a number of judgments finding failure to implement the Commission’s negative decisions following delays in the implementation, in particular due to domestic court proceedings. 72 Case C-81/10 P France Télécom v Commission, EU:C:2011:811, para 59. See also Case C-403/10 P Mediaset v Commission, EU:C:2011:533, paras 171–77; and Joined Cases C-465/09 P and C-470/09 P Territorio Histórico de Vizcaya—Diputación Foral de Vizcaya, Territorio Histórico de Álava— Diputación Foral de Álava, and Territorio Histórico de Guipúzcoa—Diputación Foral de Guipúzcoa v Commission, EU:C:2011:372, paras 148–66 ; Cases T-233/11 and T-262/11 Greece v European Commission and Ellinikos Chrysos v Commission, ECLI:EU:T:2015:948. 73 Case C-81/10P, France Telecom v Commission, EU:C:2011:811, para 62. 74 See the Opinion of the Advocate General Geelhoed prior to Case C-129/00 Commission v Italy, EU:C:2003:656. For an example of where the Commission finally withdrew its request due to the execution of the Member State, see Case C-187/06 Commission v Belgium, EU:C:2008:527 [2006] OJ C154/10. About the implementation of the Commission decision of 24 April 2002 (aid to group Beaulieu—Ter Lembeek [2002] OJ L296/30, confirmed by Case T-217/02 Ter Lembeek International v Commission, EU:T:2006:361; confirmed on appeal, Case C-28/07 P, EU:C:2007:764).

322 Jacques Derenne i. Failure to Use All Judicial Means Available to the State In a judgment condemning the Slovak Republic on 22 December 2010,75 the CJEU heard arguments concerning the scope of the principle of res judicata. A beneficiary of unlawful and incompatible aid (debt forgiveness in the context of a concordat) had been summoned by the Slovak tax authority to repay the aid. An action was brought against him before Slovak courts which had dismissed the claim. A court of appeal confirmed the dismissal on the grounds that it was not possible to revise the decision of the concordat once it had become final. The Slovak authorities had not filed an extraordinary remedy against the court of appeal’s decision. The CJEU reiterates that ‘a Member State which, pursuant to a Commission decision, is obliged to recover unlawful aid is free to choose the means of fulfilling that obligation, provided that the measures chosen do not adversely affect the scope and effectiveness of European Union law’.76 The CJEU goes on to observe that ‘the measures taken by the competent Slovak authorities did not lead to the recovery of the wrongful aid and normal conditions of competition were therefore not restored’.77 The judgment of 14 July 2004 of the court with jurisdiction approving the agreement in the arrangement with creditors had acquired the force of res judicata in 2004 and therefore predates the Commission’s decision to recover the contested aid. The CJEU stated that the situation at issue here is distinguishable from that in Lucchini, relied on by the Commission, where the CJEU held that European Union law precludes the application of a provision of national law which seeks to lay down the principle of res judicata in so far as the application of that provision prevents the recovery of State aid granted in breach of European Union law which has been found to be incompatible with the common market in a decision of the Commission which has become final (see, to that effect, paragraph 63 of Lucchini).78

The CJEU then underlines the importance, both in the European Union legal order and in the national legal orders, of the principle of res judicata: European Union law does not in all circumstances require a national court to disapply domestic rules of procedure conferring the force of res judicata on a judgment, even if to do so would make it possible to remedy an infringement of European Union law by the judgment in question.79

But, without going into any further discussion, the CJEU goes on to say that the Slovak authorities under national law had resources which, if diligently used, could have ensured that the Slovak Republic was able to recover the aid at issue. The Slovak government had not provided any precise information on the circumstances in which it used the resources which were available to it.

75 76 77 78 79

Case C-507/08 Commission v Slovakia, EU:C:2010:802. ibid, para 51. ibid, para 54. ibid, para 56. Case C-507/08 Commission v Slovakia, EU:C:2010:802, para 60.

State Aid Recovery and the Energy Sector 323 The CJEU held ‘that the information provided by the Slovak Republic is insufficient to allow the conclusion that it took, within the prescribed period, all the measures which it could have employed in order to obtain the repayment of the aid at issue’.80 Thus, by failing to take within the prescribed period all the measures necessary to recover the aid from the beneficiary, the Slovak Republic failed to fulfil its obligations. The Lucchini and Klausner judgments,81 in 2007 and 2015 respectively, show how different the interpretation is of the interaction between the principle of the primacy of EU law and the principle of res judicata in a State aid context on the one hand, and in a general EU law infringement (including antitrust) on the other. While, in general, res judicata should not be set aside when EU law is infringed (there is no obligation to remedy the violation of EU law if this violation is covered by the principle of res judicata),82 in a State aid context, res judicata should be set aside in order to remedy the violation of EU law, which concerns in this case the fundamental balance of institutional powers between the Commission (the exclusive competence for the aid compatibility assessment) and the national courts. When the exclusive powers of the Commission are at stake, the res judicata principle should be set aside.83 ii. Decisions of National Courts Delaying the Recovery of the Aid In two cases involving Italy, the CJEU revisits the more typical situations of delays in execution due to judicial decisions quashing requests for recovery on grounds of national law not related to the legality of the negative decision of the Commission. In the first judgment of May 2011, the CJEU referred to its judgment in Scott in which the ‘annulment of a national measure implementing a Commission decision ordering recovery of unlawful aid, which impedes the immediate and effective implementation of that decision, is irreconcilable with the requirements arising from Article 14(3) of Regulation No 659/1999’.84 In the present case, although the annulment of the national injunction to pay was reformed on appeal, this caused a delay in the implementation of the negative decision which led the CJEU to establish that the Member State had failed to fulfil its obligations.

80

ibid, para 64. Case C-119/05 judgment of 18 July 2007 Lucchini, EU:C:2007:434 and Case C-505/14 judgment of 11 November 2015 Klausner Holz, EU:C:2015:742. 82 Case C-234/16 judgment of 16 March 2006 Kapferer, EU:C:2006:178. But, in Olimpiclub, the CJEU ruled that, in substance, a new decision cannot perpetuate the definitive illegal decision covered by the principle of res judicata. In this case, the latter principle should be set aside in order to avoid a new infringement of EU law (Case C-2/08 judgment of 3 September 2009 Fallimento Olimpiclub, EU:C:2009:506). 83 On all these issues, see J Derenne, ‘L’autorité de chose jugée à l’ épreuve du droit de l’Union européenne—Du principe d’effectivité en général et des règles spécifiques en matière d’aides d’État en particulier’ in V Giacobbo and C Verdure (eds), Contentieux du droit de la concurrence de l’Union européenne: questions d’actualité et perspectives (Brussels, Larcier, 2017). See also G SkovgaarD Ølykke, ‘State Aid as a Defence for Public Authorities?’ (2016) 2 European State Aid and Law Quarterly 286. 84 Case C-305/09 Commission v Italy, EU:C:2011:274, para 47 referring to para 30 of Scott case below. 81

324 Jacques Derenne In a second judgment of October 2011, the CJEU found that a national court had suspended the execution of numerous national recovery orders and had stayed the proceedings on the grounds that the Commission’s negative decision had been challenged before the GCEU. In doing so, the national court misused Article 278 TFEU by giving a de facto suspensive effect to the action brought before the GCEU against that decision, although the GCEU did not order a suspension under Article 278 TFEU. The Italian State could not rely on such national decisions to justify its failure to comply with the negative decision, considering the fact that the provisional measures taken by the national court had clearly failed to comply with the strict conditions set by the case law of the European courts which, if met, would allow a national court to suspend a national act implementing a European act, the legality of which would be called into question.85 iii. Possible Annulment of a National Recovery Order Provided it does not Result in the Return of the Amount of the Unlawful Aid In Scott, which was referred to the CJEU for a preliminary ruling, the CJEU delivered a ruling which struck a balance between the general principle of EU law of effective judicial protection (which guarantees control by national courts of the procedural legality of a national measure ordering the recovery of unlawful aid) and the principle of effectiveness under Article 14(3) of Regulation No 659/1999, which requires an immediate and effective recovery of unlawful aid. In the case of Scott—Kimberly Clark,86 the Administrative Court of Appeal of Nantes had been asked to rule on the regularity of assessments issued for the recovery of unlawful aid. The assessments did not refer either to the scope of the powers conferred by the mayor on the deputy who signed the assessments, or to his surname and first name. Finding that the assessments in question did not comply with the formal requirements laid down in Article 4 of Law No 2000-321 on the rights of citizens in their dealings with administrative authorities, and that that infringement was liable to lead to the annulment of those assessments, the referring court had doubts as to the compatibility of the annulment of the assessments on the grounds of procedural defect with Article 14(3) of Regulation No 659/1999. In those circumstances, the Administrative Court of Appeal of Nantes decided to stay the proceedings and to refer the following question to the CJEU for a preliminary ruling, given that France had already been condemned in 2006 by the CJEU for failure to comply with the Commission’s negative decision.

85 Joined cases C-143/88 and C-92/89 Zuckerfabrik, EU:C:1991:65, and Case C-465/93 Atlanta, EU:C:1995:369. 86 Case C-232/05 Commission v France, EU:C:2006:651 (automatic suspensive effect of the recourse against the revenue securities issued by local authorities for the recovery of the unlawful aid in question). This judgment was partially cancelled by the GCEU in 2007 (Case T-366/00 Scott SA v Commission, EU:T:2007:99; annulment of the decision for its most important part relating to the preferential price of the land, leaving only a preferential tariff for the recovery charge to be recovered) but this judgment was cancelled by the CJEU after the appeal of the Commission (Case C-290/07 P Commission v Scott, EU:C:2010:480).

State Aid Recovery and the Energy Sector 325 The CJEU’s judgment is balanced.87 The CJEU states that ‘the annulment of an assessment may not be criticised per se’. However, such annulment might, in principle, confer an advantage on an aid recipient who has been successful in legal proceedings in the form of the right to claim, under national law, that sums corresponding to the previously-reimbursed aid be paid out to him once again; that potential consequence must therefore be examined in the light of the obligations laid down in Article 14(3) of Regulation No 659/1999. It follows from the very wording of the question referred that the competent authority which issues the assessments in question has the power to rectify the procedural defect vitiating those assessments, which permits the inference that the annulment of the assessments in question will not necessarily lead to the sums which the companies concerned paid pursuant to the assessments being paid once again to the companies.88

iv. Double Infringement: Pecuniary Sanctions Finally, it should be noted that the CJEU may condemn Member States to pecuniary sanctions under Article 260 TFEU for failure to fulfil obligations to recover unlawful and incompatible aid (failure to comply with a CJEU judgment ordering the Member State to recover aid).89 Article 260 TFEU is therefore the Commission’s ultimate weapon vis-a-vis the Member States and it has used it on several occasions, in particular in State aid cases, over the past few years.

87

Case C-210/09 Scott SA et Kimberly Clark SAS v Ville d’Orléans, EU:C:2010:294 paras 26–27. Case C-496/09 Commission v Italy, EU:C:2011:740 (a pecuniary sanction of an amount calculated by multiplying the basic amount of €30 million by the percentage of the unlawful aid that has not yet been recovered, at the end of the period concerned, compared with the total amount not yet recovered on the date of delivery of the present judgment, for every six months of delay in implementing the necessary measures to comply from the judgment of 1 April 2004). 89 Case C-496/09 Commission v Italy, EU:C:2011:74; see also Case C-367/14 Commission v Italy, EU:C:2015:611; Case C-184/11 Commission v Spain, EU:C:2014:33; Case C-610/10 Commission v Spain, EU:C:2012:781. 88

12 Arbitration JOHANNES KOEPP, ALEJANDRO ESCOBAR, LAURIE FREY AND ERNESTO FELIZ

I. INTRODUCTION

C

ERTAIN CURRENT EUROPEAN Union (EU) Member States, prior to joining the EU, entered into investment treaties with certain other Member States. Notably, formerly Communist Eastern and Central European States entered into investment agreements with EU Member States to reassure potential foreign investors at a time when investors’ confidence in the stability and rule of law in these countries was not as high as it is today. These States eventually acceded to the EU, but many of these pre-accession intra-EU bilateral investment treaties (BITs) remain in force today. Attitudes to these treaties have changed drastically in recent years. Today, there is a growing tension between EU law and investment treaty arbitration. Specifically, the European Commission (EC) today takes the view that the accession of new Member States has made intra-EU BITs between old and new Member States superfluous. All Member States are subject to the same legal rules under EU law, and all investors in these Member States benefit from the same protections, for example, against discriminatory treatment or disproportionate interference with property rights. The EC considers intra-EU BITs not only superfluous but also illegal under EU law. It has initiated infringement proceedings against certain EU Member States under Article 108(2) of the Treaty on the Functioning of the European Union (TFEU) and it has also intervened in annulment proceedings, which related to an arbitral award based on an intra-EU BIT, to advance its view that intra-EU BITs violate EU law. Among other things, the EC claims that these BITs confer certain rights to some but not to all EU investors, on a bilateral basis, leading to discriminatory treatment based on nationality, which is incompatible with EU law. This arguably discriminatory treatment has both a procedural and a substantive dimension. Procedurally, BITs grant foreign investors of one Contracting State the right to sue the other Contracting State in international arbitral proceedings for violation of the substantive rights granted by these treaties. This right does not extend to investors from other States that are not party to a BIT, regardless of whether they are investors from EU Member States or not. Substantively, tensions between EU law and BITs exist because arbitral tribunals have held that the scope of the rights and protections granted to investors under BITs are not identical to those provided

328 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz by EU law to investors from other Member States. This tension becomes particularly apparent with respect to the EU law rules on State aid. Under EU law, any advantage, economic or otherwise, granted by a Member State on a selective basis to a particular company or business, that has the effect of distorting the competition of the internal market, is defined as illegal State aid.1 Although EU State aid law offers some protections for an investor’s legitimate expectations, the substantive protections granted by BITs go beyond the scope of the protections under EU mandatory State aid law. This is not a theoretical problem, as has been illustrated by the widely publicised Micula matter. In the Micula case, an investment arbitration based upon an intra-EU BIT produced a result that the EC considered plainly incompatible with EU law, as compliance with the arbitral award would, in the view of the EC, constitute illegal State aid. The issues are made more complex by the fact that the EU itself has, together with each of its Member States, entered into the Energy Charter Treaty (ECT) which is a multilateral investment treaty. The analysis that follows discusses the different investment protections under investment agreements and under EU law (in particular, the different conceptions of the principle of protection of legitimate expectation under each regime) and examines how these differences result in a seemingly irreconcilable clash between investment arbitration and EU State aid law. Following this introduction, section II explains the principle of legitimate expectations in EU State aid law. Section III explains the fair and equitable treatment (‘FET’) standards and the principle of legitimate expectations under international investment agreements (IIAs or IIA for singular). Section IV contains a summary of prominent cases in which arbitral tribunals applied investment protection standards to measures taken in the pursuance of compliance with EU State aid rules. Section V explains the practical implications of these different standards of protection as it relates to the enforcement of arbitral awards inside and outside of the EU. Finally, the conclusion attempts to explore the reasons for the diverging standards applied by EU institutions, on the one hand, and investment arbitration tribunals, on the other.

II. PROTECTION OF LEGITIMATE EXPECTATIONS IN EU LAW STATE AID PROCEEDINGS

To understand the tension between EU State aid law and investment treaty protection, it is first necessary to compare the standard by which the hopes and expectations of recipients of State aid are protected within the EU legal order against the standard applied under IIAs. Seen as ‘the corollary of the principle of

1 Art 107(1) ‘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, in so far as it affects trade between Member States, be incompatible with the internal market in Consolidated Version of the Treaty of the Functioning on the European Union’ [2012] OJ C316/01.

Arbitration

329

legal certainty’,2 the principle of the protection of legitimate expectations is well established under EU law. It has been described as a ‘superior rule of law’ for the protection of individuals,3 ‘one of the fundamental principles of the Community’,4 and as a ‘general principle’5 of EU law against which the legality of the acts of the Community institutions are to be reviewed, including in proceedings aimed at the recovery of State aid. Article 16 of Council Regulation (EU) 2015/1589 of 13 July 2015 providing detailed rules for the application of Article 108 of the TFEU makes this requirement express. It obliges the EC not to demand recovery of State aid if to do so would be contrary to a ‘general principle of Union law’. While ‘general principles of Union law’ include the principle of protection of legitimate expectations, in practice, the EU courts narrowly construe the principle of protection of legitimate expectations.

A. Legitimate Expectations as a Defence to Recovery Demands in Administrative and Judicial Proceedings Relating to State Aid A recipient of illegal State aid may assert its legitimate expectations as a defence to a demand for recovery of such aid6 in both administrative and judicial proceedings: —

First, in proceedings before the EC dealing with a request for the recovery of State aid.7 — Second, before the EU courts in support of a claim for annulment of the EC’s decision requiring recovery.8

2 Joined Cases C-182/03 and C-217/03 Kingdom of Belgium and Forum 187 v Commission, Opinion of AG Leger, EU:C:2006:89, para 366. 3 Case C-74/74 CNTA v Commission, EU:C:1975:59, para 44; Case C-152/88 Sofrimport SARL v Commission EU:C:1990:259, para 26. 4 Case C-104/97 P Atlanta v European Community, EU:C:1999:498, para 52; Case C-17/03 Vereniging voor Energie, Milieu en Water v Directeur van de Dienst uitvoering en toezicht energie, EU:C:2005:362, para 73. 5 Joined Cases C-104/89 and C-37/90 Mulder and Others v Council and Commission, EU:C:2000:38, para 15; Case C-403/99 Italian Republic v Commission, EU:C:2001:507, para 35. 6 The principle of protection of legitimate expectations cannot be relied upon by the Member States. To do so would deprive Arts 107 and 108 TFEU of practical force, ‘since national authorities would thus be able to rely on their own unlawful conduct in order to render decisions taken by the Commission under those provisions of the Treaty ineffectual’ in Joined Cases T-116/01 and T-118/01 P & O European Ferries (Vizcaya) v Commission, EU:T:2003:217, para 202. 7 There are a number of decisions in which the Commission has refrained from demanding the recovery of illegal aid in order to secure the protection of legitimate expectations. See generally Industrie Ottiche Riunite (ORI) Commission Decision 92/329/EEC [1992] OJ L183/30, para VIII; Pari Mutuel Urbain (PMU) Commission Decision 93/625/EEC [1993] OJ L300/15; Temporary Withdrawal of Vessels Commission Decision 95/195/EC [1995] L126/32, para VI; Acciaierie di Bolzano Commission Decision 96/617/ECSC [1996] OJ L274/30; Large Firms in Crisis (Case C68/1999) Commission Decision 2001/212/EC [2001] OJ L79/29; Régime fiscal des centres de coordination (Case C15/200) Commission Decision 2003/755/EC [2003] OJ L282/25, para VI.4. 8 Case T-62/08 ThyssenKrupp Acciai Speciali Terni SpA v European Commission, EU:T:2010:268, paras 272–76.

330 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz — —

Third, in administrative proceedings before the Member State national authorities dealing with the requests for the recovery of State aid. Finally, before Member State national courts as a defence in State aid recovery cases, particularly in circumstances where the EC has examined the general characteristics of a State aid scheme without conducting an analysis of the aid granted in individual cases.9

To successfully invoke the principle of legitimate expectations in any of these proceedings, the recipient of State aid must be able to satisfy three conditions: (i) there must be a reasonable basis for the legitimate expectation; (ii) the adverse change must not have been foreseeable to a prudent and diligent operator; and (iii) the legitimate expectation of the beneficiary must outweigh the public policy interests that are at stake.

B. Basis for Legitimate Expectations under EU Law The beneficiary of legitimate expectations must show that there was an act or conduct of the EU administration capable of creating legitimate expectations. This includes specific statements or assurances by the EU institution in question, as well as conduct that creates confusion in the mind of a recipient of State aid acting in good faith.10 For example, the principle of protection of legitimate expectations precludes the recovery of State Aid if the beneficiary has complied with aid conditions laid down by the EC in an authorising decision.11 A person not acting in good faith (eg, a person who has manifestly violated the applicable law), in contrast, cannot successfully invoke the protection of legitimate expectations.12 While silence or inaction normally cannot preclude the recovery of State aid, this may be different when the EC fails to act over a very protracted period of time. For example, in RSV, the European Court of Justice (ECJ) held that where, without sufficient reason, it took the EC 26 months to declare that a given State aid measure was incompatible with the common market and order its cancellation, that delay

9 Case T-394/08 Regione autonoma della Sardegna v Commission, EU:T:2011:493, para 91: ‘Furthermore, the specific circumstances of the individual recipients of an aid scheme can be assessed only at the stage of recovery of the aid by the Member State concerned’. 10 Joined Cases C-182/03 and C-217/03 Kingdom of Belgium, Opinion of AG Leger (n 2) para 366; AP Rossi, ‘Recovery of Unlawful and Incompatible Aid’ in A. Santa Maria (ed), International Competition Law Series: Competition & State Aid, 2nd. edn. (The Hague, Kluwer Law International, 2015) para 4.03[A]. 11 Joined Cases T-227/99 and T-134/00 Kvaerner Warnow Werft GmbH v Commission, EU:T:2002:54, para 92. 12 Case T-394/08 Regione autonoma della Sardegna v Commission (n 9) para 261: ‘In accordance with settled case-law, any person on whose part an institution has given rise to justified hopes has the right to rely on the protection of legitimate expectations. However, the principle of the protection of legitimate expectations may not be relied upon by a person who has committed a manifest infringement of the rules in force (Case C-96/89 Commission v Netherlands [1991] ECR I-2461, para 30; Joined Cases C-65/02 P and C-73/02 P ThyssenKrupp v Commission [2005] ECR I-6773, para 41; and Case T-217/01 Forum des migrants v Commission [2003] ECR II-1563, para 76)’.

Arbitration

331

could give rise to a legitimate expectation on the part of the beneficiary such as to prevent the EC from ordering the national authorities to demand the refund of the aid.13 On the other hand, a failure to act on the part of the Commission is ‘irrelevant when an aid scheme has not been notified to [the Commission]’.14 Not even a positive decision by the EC on a duly notified State aid measure is sufficient to create a legitimate expectation that the aid has been lawfully granted. As long as such a positive decision can be challenged by third parties (ie, the beneficiary’s competitors) in the EU courts, a recipient of State aid cannot be said to have any legitimate expectations that the measure is lawful.15 Consequently, prior to the expiry of the relevant time periods for challenging the act in question, as well as the conclusion of any proceedings that might have been initiated, the recipient of aid uses that aid at its own peril. This case law significantly limits the efficacy of the legitimate expectations doctrine in EU State aid cases. In the view of the European courts, the narrow scope of protection of State aid beneficiaries is necessary to promote public interest in enforcing EU competition rules. Moreover, to protect the recipient’s hopes more efficiently could otherwise frustrate the beneficiary’s competitors’ right to have a State aid measure reviewed by an independent judicial body.16 The incentive to bring an action against a positive decision by the EC would be severely diminished if there was only little prospect of the unlawful State aid measure being revoked.

C. Lack of Foreseeability and the Diligent Business Man Second, the recipient of State aid can only successfully invoke legitimate expectations if the measure in question that contradicts the recipient’s alleged expectations was not reasonably foreseeable.17 Expectations are only ‘legitimate’ and worthy of protection where the recipient of the State aid could ‘reasonably rely on the maintenance or the stability of the situation thus created, in the same way as a “prudent and circumspect” trader’.18 This is not the case where the recipient could have ascertained that the aid had not been duly notified and approved, even if the Member State has mistakenly indicated that the aid is covered by a Commission approval.19 The beneficiary’s duty to ascertain whether the procedure in Article 108(3) TFEU has been duly followed also applies to small businesses.20

13 Case C-223/85 Rijn-Schelde-Verolme (RSV) Machinefabrieken en Scheepswerven NV v Commission, EU:C:1987:502. 14 ibid. 15 ibid. 16 Joined Cases T-116/01 and T-118/01 P & O European Ferries (Vizcaya) (n 6), para 219. 17 Case C-265/85 Van den Bergh en Jurgens BV v Commission, EU:C:1987:121, para 44: ‘On the other hand, if a prudent and discriminating trader could have foreseen the adoption of a community measure likely to affect his interests, he cannot plead [the principle of protection of legitimate interests] if the measure is adopted’. 18 Joined Cases C-182/03 and C-217/03 Kingdom of Belgium, Opinion of AG Leger (n 2), para 369. 19 Case T-109/01 Fleuren Compost BV v Commission, EU:T:2004:4, paras 143–44. 20 ibid, para 140.

332 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz Thus, under normal circumstances, only the EU administration’s conduct, but not the acts of the State granting the aid, can give rise to a legitimate expectation that the aid is lawful.21 It is only in exceptional circumstances that a beneficiary can successfully plead legitimate circumstances despite the fact that it has not verified that the Article 108(3) TFEU procedure has been followed. Such cases may include atypical instances in which it is unclear whether a given measure constitutes State aid at all.22

D. Legitimate Expectations Must Outweigh Public Policy Interest Finally, in balancing the interests at stake, the public policy interest in recovering the unlawful State aid must not override the aid beneficiary’s interest in the maintenance of a situation which it was entitled to regard as stable.23 In the field of State aid, the relevant public interest consists of preventing the operation of the market from being distorted by State aid injurious to competition, a fact which, in accordance with settled case-law, requires unlawful aid to be repaid in order to re-establish the previously existing situation … That public interest thus encompasses, in particular, the protection of competitors who, themselves, have a clear interest in being able to challenge Commission measures which adversely affect them.24

In the relatively rare cases where the beneficiary is deemed to have a legitimate expectation, this balancing act prevents the recovery of the unlawful State aid, while its termination or modification remains possible.25 Thus, as this brief overview illustrates, in balancing the different interests involved, the European courts attach more significance to public policy interests (in enforcing the EU competition rules to prevent the distortion of the relevant markets by State aid), as well as the interests of affected third parties (ie, the State aid beneficiary’s competitors), than to the hope of the beneficiary that its situation would not be

21

Rossi (n 10), para 4.03[A]. Case C-5/89 Commission v Federal Republic of Germany, Opinion of AG Darmon, EU:C:1990:187, para 26: ‘However, both the principle of the protection of legitimate expectations itself and the jurisdiction of the national courts to determine such matters must be preserved, and allowance must therefore be made for cases in which the fundamental rights of an undertaking, although it has not verified whether the aid had been notified, are such that it should none the less be accorded the benefit of the protection of legitimate expectations. In such cases, the national court must be able to assess the conduct of the recipient undertaking in concreto, if necessary after having referred a preliminary question to this Court. The doubts with which some undertakings may be assailed, when faced with ‘atypical’ forms of aid, as to whether notification is necessary should not be made light of. But the concrete nature of the assessment to be carried out by the national court must be contrasted with the abstract concept of legitimate expectations on which Germany relies in support of its refusal to implement the Community decision ordering the recovery of aid in question. The existence of legitimate expectations is not presumed, it must be proved’. 23 Joined Cases C-182/03 and C-217/03 Kingdom of Belgium, Opinion of AG Leger (n 2), para 366. 24 Joined Cases T-116/01 and T-118/01 P & O European Ferries (Vizcaya) (n 6), para 208. 25 D Grespan, A Pelin and L Rossi, ‘Recovery of Unlawful and Incompatible State Aid’ in N Pesaresi and K Van de Casteele (eds), EU Competition Law, Vol IV: State Aid, 2nd edn (Leuven, Claeys & Casteels, 2016), para 5.453. 22

Arbitration

333

subject to an adverse change. This narrow conception leaves only a very limited scope of application for the legitimate expectations principle in EU law State aid recovery proceedings. By comparison, the concept of legitimate expectations under IIAs is more extensive.

III. LEGITIMATE EXPECTATIONS AS KEY CRITERION UNDER THE FET STANDARD PURSUANT TO IIAS

Under most modern IIAs, host States must uphold foreign investors’ legitimate expectations. The tribunal in Thunderbird v Mexico explained: [T]he concept of ‘legitimate expectations’ relates, within the context of the NAFTA framework, to a situation where a Contracting Party’s conduct creates reasonable and justifiable expectations on the part of an investor (or investment) to act in reliance on said conduct, such that a failure by the NAFTA Party to honour those expectations could cause the investor (or investment) to suffer damages.26

This statement in the context of NAFTA is relevant also for other IIAs. Legitimate expectations derive from the host State’s commitments to investors. They can be created by the host State’s conduct or specific legal undertakings given in individual contracts. Stabilisation clauses in investment contracts provide good examples. Stabilisation clauses freeze the regulatory framework applicable to investments. Through such clauses, host States commit not to alter (or not to apply fully) this regulatory framework.27 In turn, stabilisation clauses create a legitimate expectation under IIAs that the host State will abide by these commitments.28 IIAs protect investors’ legitimate expectations through the notion of fair and equitable treatment (FET). Modern bilateral investment treaties,29 multilateral 26 International Thunderbird Gaming Corp v United Mexican States, UNCITRAL (NAFTA) Award (26 January 2006), para 147. 27 Stabilisation clauses take many forms. But see, eg, the Oman Model Exploration and Production Sharing Agreement of 2002, Art 26.2: ‘In the event that any provisions contained in the laws, rules or regulations of the Government are inconsistent with the terms and conditions of this Agreement, the Government shall take such action as is necessary to relieve the Company from any loss, liability or jeopardy as a result of such inconsistency and shall also take such action as may be necessary to maintain the economic equilibrium existing between the Parties on the Effective Date of this Agreement’. Reprinted in R Doak Bishop, JR Crawford and WM Reisman, Foreign Investment Disputes: Cases, Materials and Commentary, 2nd edn (The Hague, Kluwer Law International, 2014) 259. 28 See, eg, Parkerings Compagniet AS v Republic of Lithuania, ICSID Case No ARB/05/8, Award (11 September 2007), para 332: ‘It is each State’s undeniable right and privilege to exercise its sovereign legislative power. A State has the right to enact, modify or cancel a law at its own discretion. Save for the existence of an agreement, in the form of a stabilisation clause or otherwise, there is nothing objectionable about the amendment brought to the regulatory framework existing at the time an investor made its investment. As a matter of fact, any businessman or investor knows that laws will evolve over time. What is prohibited however is for a State to act unfairly, unreasonably or inequitably in the exercise of its legislative power’ (emphasis added). 29 There are numerous examples. See, eg, the Agreement between the Government of the United Kingdom of Great Britain and Northern Ireland and the Government of the Republic of Mozambique for the Promotion and Protection of Investments (opened for signature 18 March 2004, entered into force 27 February 2007) 2490 UNTS 221, Art 2(2): ‘Investments of Nationals or Companies of each Contracting Party shall at all times be accorded fair and equitable treatment and shall enjoy full protection and security in the Territory of the other Contracting Party’.

334 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz investment treaties30 and free trade agreements31 routinely contain this standard of protection, obliging host States to treat foreign investments fairly and equitably. The ECT provides for the FET of investments protected under that treaty.32 The FET standard is broad. It obliges States to treat foreign investments transparently and without coercion, arbitrariness, discrimination, due process or bad faith. According to some tribunals, legitimate expectations are the ‘dominant element’ of the FET standard.33 Recent attempts to define the FET standard in major free trade agreements expressly protect legitimate expectations as part of that standard.34 International arbitration tribunals have identified at least four factors to give content to the legitimate expectations protected under IIAs. In the words of the tribunal in LG&E v Argentina: It can be said that the investor’s fair expectations have the following characteristics: they are based on the conditions offered by the host State at the time of the investment; they may not be established unilaterally by one of the parties; they must exist and be enforceable by law; in the event of infringement by the host State, a duty to compensate the investor for damages arises except for those caused in the event of state of necessity; however, the investor’s fair expectations cannot fail to consider parameters such as business risk or industry’s regular patterns (emphasis added).35

First, the investor must prove that the host State has made a ‘specific’ commitment to the investor that gives rise to legitimate expectations. The investor must prove that the State has made a representation or promise, which ‘must be precise as to its content and clear as to its form’.36 Investment arbitration tribunals have held that legitimate expectations arise ‘as a result’ of such commitments.37

30 ASEAN Comprehensive Investment Agreement (opened for signature 26 February 2009, entered into force 29 March 2012) Art 11(1): ‘Each Member State shall accord to covered investments of investors of any other Member State, fair and equitable treatment and full protection and security’. 31 See, eg, North American Free Trade Agreement (opened for signature 17 December 1992, entered into force 1 January 1994) Art 1105(1): ‘Each Party shall accord to investments of investors of another Party treatment in accordance with international law, including fair and equitable treatment and full protection and security’. 32 Energy Charter Treaty (opened for signature 17 December 1994, entered into force 16 April 1998) 2080 UNTS 95, Art 10(1): ‘Each Contracting Party shall, in accordance with the provisions of this Treaty, encourage and create stable, equitable, favourable and transparent conditions for Investors of other Contracting Parties to make Investments in its Area. Such conditions shall include a commitment to accord at all times to Investments of Investors of other Contracting Parties fair and equitable treatment’. 33 Saluka Investments BV v Czech Republic, PCA (UNCITRAL) Partial Award (17 March 2006), para 302. 34 EU–Canada Comprehensive Economic and Trade Agreement (pending ratification): trade.ec.europa. eu/doclib/docs/2014/september/tradoc_152806.pdf (CETA) Art 8(10)(4): ‘When applying the above fair and equitable treatment obligation, a Tribunal may take into account whether a Party made a specific representation to an investor to induce a covered investment, that created a legitimate expectation, and upon which the investor relied in deciding to make or maintain the covered investment, but that the Party subsequently frustrated’. 35 LG&E Energy Corp, LG&E Capital Corp and LG&E International Inc v Argentine Republic, ICSID Case No ARB/02/1, Decision on Liability (3 October 2006), para 130. 36 Crystallex International Corporation v Bolivarian Republic of Venezuela, ICSID Case No ARB(AF)/ 11/2, Award (4 April 2016), para 547. 37 ECE Projektmanagement International GmbH and Kommanditgesellschaft Panta Achtundsechzigste GmbH & Co v Czech Republic, PCA Case No 2010-5 (UNCITRAL) Award (19 September 2013), para 4.762.

Arbitration

335

The host State’s commitment may be express (eg, ‘an explicit promise or guaranty from the host-State’)38 or may be implied from the circumstances (eg, the host State’s conduct).39 In the absence of such commitments, an investor’s claims for legitimate expectations may fail. In Ulysseas v Ecuador, an investor in electricity barges claimed a legitimate expectation that the fuel credits available at the time it made its investment would remain in place throughout the lifetime of the investment. The tribunal dismissed this claim. It found that these credits formed part of an emergency regime subject to modification. It therefore found that the investor had no legitimate expectation ‘in the absence of a specific commitment in that regard by the State’.40 Second, and building on the above, legitimate expectations cannot be established unilaterally.41 Legitimate expectations ‘cannot exclusively be determined by foreign investors’ subjective motivations and considerations’.42 Thus, in order to prove the existence and scope of the host State’s commitment, the investor must point to objective elements (eg, a document or contract).43 Consequently, tribunals may enquire whether the investor’s claimed expectations are reasonable in light of the host State’s representations.44 Third, the investor must show that it based its decision to make an investment on the legitimate expectations it claims to have relied upon.45 In this regard, the investor must normally show that the host State made its assurances ‘prior to or at the time of the making of the investment’.46 The investor’s FET claim routinely fails if it is based on the post-investment conduct of the host State. In Duke v Ecuador, for example, the investor based some of its expectations on mediation and arbitration agreements concluded after the investment. The tribunal dismissed this claim,

38 Parkerings Compagniet AS, ICSID Case No ARB/05/8, Award (n 28), para 331: ‘The expectation is legitimate if the investor received an explicit promise or guaranty from the host-State, or if implicitly, the host-State made assurances or representation that the investor took into account in making the investment. Finally, in the situation where the host-State made no assurance or representation, the circumstances surrounding the conclusion of the agreement are decisive to determine if the expectation of the investor was legitimate. In order to determine the legitimate expectation of an investor, it is also necessary to analyse the conduct of the State at the time of the investment’ (footnotes omitted; emphasis added). 39 See ibid. 40 Ulysseas, Inc v Republic of Ecuador, PCA Case No 2009-19 (UNCITRAL) Final Award (12 June 2012), para 254. 41 LG&E Energy Corp, ICSID Case No ARB/02/1, Decision on Liability (n 35), para 130. 42 Saluka Investments BV, PCA (UNCITRAL) Partial Award, para 304. 43 See, eg, Franck Charles Arif v Republic of Moldova, ICSID Case No ARB/11/23, Award (8 April 2013), para 535: ‘a claim based on legitimate expectations must proceed from the exact identification of the origin of the expectation alleged, so that its scope can be formulated with precision’. 44 Charanne BV and Construction Investments SÀRL v Kingdom of Spain, SCC Arb No 062/2012, Final Award (21 January 2016), para 495. 45 See, eg, Duke Energy Electroquil Partners and Electroquil SA v Republic of Ecuador, ICSID Case No ARB/04/19, Award (18 August 2008), para 365: ‘the legitimate expectations which are protected are those on which the foreign party relied when deciding to invest’. 46 ECE Projektmanagement International GmbH and Kommanditgesellschaft Panta Achtundsechzigste GmbH & Co, PCA Case No 2010-5 (UNCITRAL), Award (n 37), para 4.762. There is extensive support for this notion. See, eg, Duke Energy Electroquil Partners and Electroquil SA, ICSID Case No ARB/04/19, Award (n 45), para 365.

336 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz as ‘[t]he Med-Arb Agreements were concluded more than two years later and can thus in no event give rise to expectations protected under the fair and equitable treatment standard’.47 Fourth, the investor must take into account all relevant circumstances surrounding its legitimate expectations. IIAs do not exempt the investor from all business or political risk. In particular, the investor must become aware of any relevant host State legislation, regulation, or policies as part of its due diligence when making an investment.48 Recently, the vagueness of standards protected under IIAs has attracted scrutiny and criticism. The crux of defining the FET standard is to strike the right balance between States’ right to regulate their economies, on the one hand, and an investor’s legitimate expectations, on the other. A broad and undefined standard allows (and arguably obliges) arbitrators to use their own criteria in defining the FET standard. There is a perception that broad and undefined standards included in traditional IIAs and arbitrator discretion hinder States’ right to regulate. Because of this, recent IIAs have sought to define more precisely (and thus to circumscribe) both the FET standard49 and legitimate expectations.50

IV. INVESTMENT TREATY CASES DEALING WITH STATE AID

As shown in sections II and III, EU State aid law and IIAs take different approaches to the protection of an investor’s legitimate expectation. How do these differences play out in practice when IIAs and the State aid rules coincide? This section focuses on three cases in which arbitral tribunals applied investment protection standards to measures taken in pursuance of EU State aid rules. The first two cases were brought simultaneously against Hungary by two power generators (Electrabel and AES) whose power purchasing agreements (PPAs or PPA for singular) were affected by Hungary’s accession to the EU. These two cases are discussed in order of relevance, as the AES case ultimately turned on a different issue.51 Their common feature is that their respective outcomes each in their own way avoided open contradiction between investment treaty protections and EU State aid rules. The third case, Micula v Romania, concerned the repeal of free zone benefits in anticipation of

47 Duke Energy Electroquil Partners and Electroquil SA, ICSID Case No ARB/04/19, Award (n 45), para 365. 48 Frontier Petroleum Ltd v Czech Republic, PCA (UNCITRAL) Final Award (12 November 2010), para 287: ‘a foreign investor has to make its business decisions and shape its expectations on the basis of the law and the factual situation prevailing in the country as it stands at the time of the investment’. 49 See, eg, CETA, Art 8.10(2); Free Trade Agreement between the European Union and the Socialist Republic of Vietnam (pending ratification): www.trade.ec.europa.eu/doclib/press/index.cfm?id=1437 (EU–Vietnam FTA) ch 8, s 2, Art 14; Trans-Pacific Partnership (pending ratification), available at: www.ustr.gov/trade-agreements/free-trade-agreements/trans-pacific-partnership/tpp-full-text, Art 9.6. See also US Model Bilateral Investment Treaty (2012), Art 5. 50 See, eg, CETA, Art 8.10(4); EU–Vietnam FTA, ch 8, s 2, Art 14(5). 51 The award in a third case against Hungary, EDF International v Hungary, UNCITRAL Award (3 December 2014), is not public.

Arbitration

337

EU accession, and its outcome has, by contrast, led to an outright clash between investment treaty arbitration and EU rules on State aid.

A. Electrabel v Hungary Electrabel, a Belgian electricity company brought a claim against Hungary for terminating a PPA, signed prior to Hungary’s accession to the EU. After Hungary became a Member State, the EC, which has the exclusive mandate to determine whether a State measure distorts the common market, found that Hungary had not notified the EC about the PPA and that the PPA constituted illegal State aid.52 The EC equally held that the PPAs about which Hungary did notify the EC were also illegal State aid. The EC ordered Hungary to ‘end the State aid contained in the PPAs’ and recover the aid received by the beneficiaries since the effective date of Hungary’s accession to the EU (ie, since 1 May 2004).53 Once Hungary began to implement the EC’s decision, Electrabel commenced arbitration proceedings under the ECT against Hungary claiming a breach of fair and equitable treatment and regulatory expropriation of its investment, among other things. The tribunal allowed the EC to submit a brief as a non-disputing party pursuant to Article 37(2) of the ICSID Arbitration Rules.54 The EC put forward four arguments on why the tribunal should refuse jurisdiction. First, the EC submitted that the claim should have been brought against the EU because State aid falls within the exclusive mandate of the EU. The tribunal stated that it agreed with the EC’s analysis only if and to the extent that the relevant dispute engaged the legal responsibility of the EU under the ECT for a decision of the EC. The tribunal held, however, that this was not the issue before it. Electrabel did not bring a claim against the EU or the EC or under EU law, or against a Community measure,55 and did not dispute the EC’s decision. This was a claim brought under the ECT for measures taken by Hungary and, as such, Hungary was the right respondent.56 Second, the EC submitted that Hungary, by applying the EC’s decision to revoke the PPA and recover the aid, was not breaching the ECT, but merely complying with its EU obligations under the State aid regime. There is a presumption under the ECT, it argued, that compliance with EU law on State aid cannot be a breach of the ECT, if EU law offers substantive and procedural guarantees equivalent to the protections contained in the ECT (as, according to the EC, it does).57 The tribunal did not accept

52 Electrabel SA (Belgium) v Republic of Hungary, ICSID Case No ARB/07/19, Decision on Jurisdiction, Applicable Law and Liability (30 November 2012), para 4.97. 53 ibid. 54 ibid, para 4.89. 55 ibid, para 5.33. 56 ibid, para 4.171. 57 ibid, para 5.13–5.14.

338 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz the EC’s jurisdictional arguments nor did it adopt its terms when the tribunal turned to the merits of the dispute. The issue on the merits was whether Hungary’s view that it was required to terminate the claimant’s PPA to comply with State aid law was consistent with Hungary’s ECT obligations, given that termination of the PPAs was not expressly provided for in the EC’s decision on State aid. The tribunal’s ruling on this question was broadly consistent with the EC’s desired outcome (although not with the EC’s view as to who should have the final word on the issue). The tribunal held that Hungary’s implementation of the EC decision regarding State aid in connection with the PPAs was not irrational or arbitrary and therefore not contrary to the ECT duty to afford fair and equitable treatment. Third, the EC submitted that the ECT must be interpreted in such a way so that it does not conflict with EU law as there is a general international law principle of harmonious interpretation.58 The tribunal held this was an ideal objective but that international law did not prescribe a particular method or conflict rule for this purpose. It concluded that there was ‘no material inconsistency between the ECT and EU law’.59 Fourth, the EC submitted that EU Member States are obliged to comply with the EC’s decisions on State aid and if they do not, the EC can bring proceedings against the State in front of the ECJ.60 If the tribunal were to issue an award substituting compensation for State aid which has been declared unlawful under EU law, such an award would be unenforceable in any Member State court as it would be contrary to EU law.61 The tribunal addressed this argument in its discussion concerning the applicable law. The tribunal noted that it operated under international law, with no seat within the EU and its award was enforceable both within and outside the EU. It concluded that none of these factors affected its jurisdiction.62 The tribunal left for a further phase of the proceedings the issue of the compensation method applied by Hungary for Electrabel’s stranded costs following cancellation of the PPA. That method resulted in Hungary foregoing the recovery of the State aid it had paid for Electrabel’s PPA, paying no additional amounts of compensation. The tribunal applied the tests of arbitrariness and protection of legitimate expectations, as invoked by the claimant. The tribunal first concluded that Hungary’s method of compensation was reasonable and non-arbitrary, given that under EU rules it was required to undertake a balancing of relevant considerations in use of its discretion. Such a balancing exercise would be negated if the claimant was entitled to the maximum degree of possible compensation. Hungary’s determination of compensation demonstrated a reasonable connection to a rational policy.63 The tribunal also held that Electrabel had no reasonable grounds (and, in particular, had received no representations) to expect that Hungary would apply the maximum

58 59 60 61 62 63

ibid, para 5.15. ibid, para 4.196. ibid, para 5.16. ibid, paras 4.110, 5.16. ibid, para 4.199. ibid, paras 179 and 209–21.

Arbitration

339

possible level of compensation under EU rules, and that therefore Hungary did not violate the claimant’s legitimate expectations.64 As a result, the tribunal dismissed Electrabel’s claims in their entirety. Electrabel did not seek the annulment of the award under the ICSID Convention.

B. AES v Hungary In this second case involving a power generator whose PPA was affected by EU State aid rules (in this case applied by Hungary as a prospective EU Member State), the claimants also alleged, among other things, regulatory expropriation and a breach of fair and equitable treatment against Hungary under the ECT. The PPA that Hungary had signed with an AES subsidiary prior to becoming an EU Member State was being examined by the EC on the grounds that it could constitute illegal State aid. In anticipation of a possible decision by the EC to terminate the PPA and recover the aid, Hungary reintroduced administrative pricing in the form of a price cap to avoid having to terminate the PPA.65 The EC again intervened as a non-disputing party in the proceeding. The disputing parties and the EC discussed the issue of the compatibility of investor protection under the ECT with the obligations of Member States under the EU State aid regime. The tribunal held that if Hungary wanted to address the eventual concerns of the EC regarding illegal State aid, the reintroduction of administrative pricing would have been a legitimate exercise of public policy. However, because the EC had not issued a decision on State aid, Hungary had no legal obligation to comply with any particular EU decision, even if Hungary believed the EC would issue such a decision in due course.66 During the hearing, it became clear that when the government issued the price cap regulations, the Hungarian State aid agency had not been consulted and the price cap was not related to State aid at all.67 As a result of the tribunal, it was decided that Hungary’s decision to reintroduce an administrative price cap was not motivated by pressure from the EC.68 Co-arbitrator Brigitte Stern disagreed with the majority. She expressed the view that, even before the EC’s decision on State aid was announced, it was made clear to Hungary that the PPA raised serious State aid concerns for the EC and would have to be terminated or at the very least renegotiated. In arbitrator Stern’s view, the evidence ‘is overwhelming[ly clear that the] decision to reintroduce maximum administrative prices was a rational, non-arbitrary response to a complex set of legitimate policy concerns’.69

64

ibid, para 155. AES Summit Generation Limited and AES-Tisza Erőmű Kft v Hungary, ICSID Case No ARB 07/22, Award (23 September 2010), para 9.2.13. 66 ibid, para 10.3.16. 67 ibid, para 10.3.17. 68 ibid, para 10.3.18. 69 ibid, para 10.3.19. 65

340 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz Hungary prevailed on the merits for other reasons. The tribunal found that Hungary had reintroduced price caps to control the ‘luxury profits’ accrued by power generators, that this was a legitimate public policy objective, and that the price cap measures were reasonable and non-discriminatory.

C. Micula v Romania The third case was the case that brought the tension between EU State aid law and investment treaty law to the fore. In 1993, Romania entered into an agreement laying out the procedures for its accession to the EU, including the steps required to harmonise matters relating to State aid with the requirements of EU law. With a view towards its EU accession, Romania enacted a series of regional aid measures around 1998 to attract foreign investment to ‘disfavoured regions’. The Miculas, Swedish nationals, were among those investors who took advantage of the incentive programmes, which involved Romania’s provision of assurances that were to be valid until 2009. However, after formal accession negotiations were initiated, the EU began to raise questions as to the compatibility of Romania’s existing State aid measures with EU law, and Romania repealed the regional aid measures in 2004. In 2005, the Miculas commenced an ICSID Convention arbitration against Romania and alleged that Romania’s repeal of the regional aid measures was a breach of its obligations under the Sweden–Romania BIT. In 2013, the arbitral tribunal issued an award in favour of the Miculas, despite Romania having argued, with support from the EC, that the relevant BIT had to be interpreted consistently with EU law, including the State aid rules. The tribunal allowed the EC to make an amicus submission as to the applicable law of the dispute. The EC submitted that the tribunal should interpret the BIT taking into account the BIT’s European context and origin.70 The EC also argued that Article 30(3) of the Vienna Convention on the Law of Treaties (VCLT) required the primacy of EU rules on State aid over any incompatible provisions of the BIT.71 The tribunal disagreed with the EC’s position, making the following three points: —

First, the BIT is the primary source of the tribunal’s jurisdiction,72 and there are no references in it to EU accession.73 Furthermore, there were no references to the BIT in the Accession Treaty that Romania signed in 2005, so the tribunal cannot assume that the parties intended Romania’s EU membership to modify its obligations under the BIT.74

70 Ioan Micula, Viorel Micula, SC European Food SA, SC Starmill SRL and SC Multipack SRL v Romania, ICSID Case No ARB/05/20, Award (11 December 2013), para 317. 71 ibid, para 317. Art 30 of the VCLT concerns the application of successive treaties concerning the same subject matter. Art 30(3) of the VCLT provides as follows: ‘When all the parties to the earlier treaty are parties also to the later treaty but the earlier treaty is not terminated or suspended in operation under Art 59, the earlier treaty applies only to the extent that its provisions are compatible with those of the later treaty’. 72 ibid, para 318. 73 ibid, para 321. 74 ibid.

Arbitration —



341

Second, Article 31 of the VCLT provides that a treaty should be interpreted in good faith in accordance with the ordinary meaning of the words and in light of the treaty’s object and purpose, including taking into account any preamble or annexes.75 The tribunal proposed to follow this rule of interpretation. Third, the tribunal would interpret all the treaties at issue having due regard to all the other applicable treaties assuming that parties entered into each of those treaties fully aware of their legal obligations under all other treaties.76 The treaties at issue included the Europe Treaty of 1995 on which Romania placed particular reliance.

The EC made further submissions arguing that any award issued by the tribunal ordering compensation for the investors would be unenforceable as it would constitute illegal State aid under EU law. The tribunal refused to consider these arguments, saying it would not speculate about the future actions of different actors and stating that the enforceability of the award is not an appropriate consideration for a tribunal.77 The tribunal was divided on the merits of the dispute. By majority, the tribunal held that the claimants had not shown that Romania was bound to keep in place and enforce the free zone benefits for the entire stipulated 10-year period under Romanian law. Romania therefore did not breach the BIT’s so-called umbrella clause.78 By majority, the tribunal also found that Romania had violated the claimants’ legitimate expectations with respect to the availability of the free zone benefits. The tribunal held that, on the basis of the pattern of State conduct over several years, it was objectively reasonable for the claimants to rely on Romania’s stipulation that it would maintain the free zone benefits for the full 10-year period, despite finding that the claimants had not proven a legal obligation to do so under Romanian law.79 As a result, the tribunal held that Romania had breached its duty of FET towards the claimants. The tribunal also addressed additional arguments made by the Parties. It held that Romania had not acted in bad faith. The tribunal found, however, that Romania had not been forthcoming to the affected investors regarding the status and potential consequences of its preparations for EU accession. The tribunal therefore held that ‘the manner in which Romania carried out’ the termination of the free zone benefits ‘was not sufficiently transparent to meet the fair and equitable treatment standard’.80 The tribunal also held that Romania acted unreasonably by maintaining the claimants’ obligations under the free zone regime while abolishing the free zone benefits (but otherwise did not act unreasonably or arbitrarily in repealing the free zone benefits).

75 76 77 78 79 80

ibid, para 322. ibid, para 326. ibid, para 340. ibid, paras 418–19. ibid, paras 690–717 and 726. ibid, para 864.

342 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz The tribunal issued an award for damages for increased costs and the lost profits occasioned by the revocation of these incentives, notwithstanding the fact that from an EU law perspective, the claimants’ hopes that Romania would maintain the free zone benefits for the full 10-year period were not worthy of protection. When considering the same facts, the EC emphasised that under its conception of legitimate expectations, the claimants could not expect that the State aid was lawful as long as the national authority had not authorised that aid: Absent such an authorisation from the Competition Council and in line with the caselaw of the CJEU, which applied by virtue of the Article 64 of the 1995 Europe Agreement, Article 1 of the Implementing Rules to Decision No 4/2000 of the EU-Romanian Association Council and the acquis communautaire, SC European Food SA could never have entertained a legitimate expectation that the incentives granted under EGO 24 constituted compatible State aid, regardless of the subsequent actions of the Romanian Government after Decision No 244/2000 was adopted.81

Because the EC concluded that the claimants’ expectations were not worthy of protection, it also considered that compliance with the award would constitute a reinstatement of illegal State aid, violative of EU law. It was that conflict that subsequently gave rise to extremely difficult conflicts during the enforcement stage. These conflicts are still unresolved, as will be shown in the following section.

V. THE ENFORCEMENT OF INVESTMENT ARBITRATION AWARDS IN CONFLICT WITH EU LAW

As a practical matter, an arbitral award against a Member State will have little value to an investor unless the investor can successfully enforce it. For an investor, therefore, a primary question will be whether the investor can enforce that award in a jurisdiction where the Member State has assets, if they obtain an award in an arbitration under an IIA. The interaction between international investment arbitration and EU law complicates that question, particularly where the award is based on a measure that may constitute illegal State aid under Article 107 TFEU. The Micula case shows that the tension between international investment arbitration and EU State aid law emerges most clearly at the enforcement stage. If a tribunal issues an award granting an investor compensation for the discontinuation of a Member State measure where such measure would have constituted illegal State aid under Article 107 TFEU, payment of such compensation may be seen (as it is viewed by the EC) as a reinstatement of the illegal State aid. Under such circumstances, will a court enforce the award if the payment of compensation under the award will result in the reinstitution of the illegal State aid by the Member State? The answer will likely depend on whether the enforcing court is located in an EU Member State (courts outside the EU appear far more likely to enforce such an award).

81 Arbitral award Micula v Romania (Case SA.38517) Commission Decision 2015/1470 [2015] OJ L232/43 (Micula Commission Decision), para 159.

Arbitration

343

This section examines the likelihood of the enforcement of an award linked to illegal State aid in courts inside and outside the EU. As a starting point, subsection V.A briefly reviews the two principal instruments under which the enforcement of such an award against a Member State may be sought, the ICSID Convention and New York Convention. Subsection V.B discusses the difficulties of enforcing such an award in a Member State. Finally, subsection V.C evaluates the significantly better chances of enforcement in non-EU courts.

A. Recognition and Enforcement Under the ICSID and New York Conventions In the IIA context, an investor seeking to enforce an award against an EU Member State will likely rely on one of two instruments: the ICSID Convention or the New York Convention. This subsection provides a summary of the recognition and enforcement provisions of each of these instruments. ICSID awards are binding and cannot be subject to appeal or review before a court.82 Once an award becomes final, the ICSID Convention provides for the automatic recognition and enforcement of the award in any State party to the ICSID Convention. Article 54(1) requires State parties to recognise the award as binding and enforce monetary obligations imposed by the award ‘as if it were a final judgment of a court in that State’. The ICSID Convention offers no grounds for national courts to refuse recognition and enforcement of an award. ICSID awards are thus extremely favourable to claimants, as the national court should have no option but to recognise and enforce the award (at least from a treaty perspective).83 As for non-ICSID awards, the New York Convention will likely be the relevant instrument for enforcement. The New York Convention facilitates the enforcement of foreign arbitral awards by requiring each of its 153 State parties to recognise and enforce an award made in the territory of another State.84 However, unlike the ICSID Convention, the New York Convention provides on whether a national court may refuse to enforce a foreign award, although these grounds are limited. Under Article V of the New York Convention, a national court may refuse to enforce an award only if either (i) the party against whom the award is invoked is able to prove one of several enumerated grounds85 or (ii) the court finds that the subject matter of the award is not capable of arbitration under the law of the country where the court is located or recognition or enforcement would violate the public policy of that country.86 The general view is that these grounds should be ‘construed narrowly’,

82 ICSID Convention, Art 53(1) (ICSID Convention awards ‘shall not be subject to any appeal or to any other remedy except those provided for in this Convention’). 83 See L Reed and J Paulson et al, Guide to ICSID Arbitration (Netherlands, Kluwer Law International, 2010) 179–82. 84 New York Convention, Art III. 85 These grounds are the lack of a valid arbitration agreement (Art V(1)(a)); the violation of due process (Art V(1)(b)); an excess of the tribunal’s authority (Art V(1)(c)); irregularity in the tribunal’s composition or in the arbitral procedure (Art V(1)(d)); and if the award has not yet become binding (Art V(1)(e)), or the award has been set aside or suspended (Art V(1)(e)). 86 New York Convention, Art V(2).

344 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz and with respect to the public policy exception, courts should hesitate to find a public policy violation except ‘in extreme cases only’.87 Both Conventions take an extremely pro-enforcement approach. With this background in mind, we now turn to the question whether EU State aid concerns may result in deviation from this approach.

B. Enforcement Inside the European Union For an investor holding an award against a Member State, the most attractive forum for enforcement may be a jurisdiction within the EU. Unfortunately for such an investor, after the EC’s decision in Micula, an EU Member State court may be reluctant to enforce an award if the compensation could be considered incompatible new aid under the State aid rules, in particular where the award was rendered in a non-ICSID Convention arbitration. In this subsection, we examine, first, the general rule that enforcement of an award requiring a Member State to pay compensation pursuant to an IIA does not constitute the provision of illegal State aid and, second, the exception to this rule in cases where the compensation has been awarded as a remedy for the Member State’s discontinuation of illegal State aid. i. Generally, Enforcement of an Arbitral Award Granting Compensation Pursuant to an IIA does not Constitute Illegal State Aid Generally, payment of compensation by an EU Member State pursuant to an award rendered in an investment arbitration is unlikely to fall foul of the EU State aid rules, and therefore an investor should encounter no difficulties in this regard when seeking to enforce such an award in Member State courts. Although neither the ECJ nor the EC has explicitly expressed a view on this issue, the general position may be extrapolated from decisions holding that payment of compensation under a Member State’s domestic law will not constitute illegal State aid. According to the ECJ and the EC, an economic advantage granted by a Member State must be voluntary for such an advantage to amount to illegal State aid within the meaning of Article 107 TFEU.88 A series of ECJ and EC decisions have made clear that, where a Member State’s domestic law requires it to pay compensation if the Member State infringes on an otherwise legally protected right or interest of a private party, such payment is not voluntary and therefore does not constitute

87 A van den Berg, The New York Convention of 1958: An Overview (International Council for Commercial Arbitration, available at: www.arbitration-icca.org/media/0/12125884227980/new_york_ convention_of_1958_overview.pdf, 13. 88 C Tietje and C Wakernagel ‘Enforcement of Intra-EU ICSID Awards: Multilevel Governance, Investment Tribunals and the Lost Opportunity of the Micula Arbitration’ (2015) 16 The Journal of World Investment & Trade 201, 220–21.

Arbitration

345

illegal State aid.89 In particular, in the Asteris case, the ECJ indicated that State aid ‘is fundamentally different in its legal nature from damages which the competent national authorities may be ordered to pay to individuals in compensation for the damage they have caused to those individuals’.90 In Akzo Nobel, the EC further explained that compensation ‘usually’ does not amount to State aid, even if paid in the absence of a court order requiring payment, so long as (i) it does no more than compensate the injured party for the damages caused by government action and (ii) the right to compensation arises under a general rule of national law.91 Relying on that same rationale, it seems reasonable to consider the payment of compensation pursuant to an award under an IIA as comparably involuntary, as it arises out of a binding international law obligation to pay compensation for the government’s actions in accordance with a substantive standard of protection.92 It could also be argued that the enforcement of an award under an IIA cannot be illegal State aid because it is not imputable to the Member State. However, this line of reasoning is less satisfactory. Under the doctrine of imputability, a measure must be attributable to the Member State for it to be State aid under Article 107 TFEU. For example, in Deutsche Bahn AG v Commission, the Court of First Instance (now known as the General Court) of the Court of Justice of the European Union (CJEU) concluded that a measure was not imputable to the Member State because the measure was required for the implementation of mandatory EU law.93 It could be argued, at least with respect to ICSID awards, that Articles 53 and 54 of the ICSID Convention leave the Member State with no autonomy as to whether to enforce the award. Therefore, as with the implementation of EU law, the implementation of such provisions cannot be attributable to the State.94 However, it has been suggested that the same reasoning does not translate quite as easily to awards enforced under the New York Convention.95 This is because of the public policy exception to enforcement under the New York Convention. Article V(2)(b) provides a safety valve for courts to refuse to enforce an award where it would be contrary to public policy, and according to the Eco Swiss ruling,96

89 ibid; Case C-61/79 Amministrazione delle finanze dello Stato v Denkavit italiana [1980] ECR 1205 (repayment for improper tax levies); Joined Cases C-106/87-120/87 Asteris v Greece & EEC [1988] ECR 5515, 5539 et seq (payment of damages); Commission Decision 2008/408/EC on the State Aid C 36/A/06 (ex NN 38/06) [2008] OJ L144/37 (compensation for expropriation); Commission Decision 304/2005 of Authorisation for State aid pursuant to Articles 87 and 88 of the EC Treaty—Akzo Nobel [2005] OJ C81/48 (compensation for expropriation). 90 Asteris v Greece (n 89), paras 22–23. 91 Letter from the European Commission to Netherlands re Aid in favour of Akzo-Nobel in order to minimise chlorine transports (Brussels, 16 June 2004) para 18: www.ec.europa.eu/competition/state_aid/ cases/138128/138128_460621_53_2.pdf. 92 Tietje and Wakernagel (n 88) 221; P Ortolani, ‘Intra-EU Arbitral Awards vis-à-vis Art 107 TFEU: State Aid Law as a Limit to Compliance’ (2015) 6 Journal of International Dispute Settlement 118, 122–23. 93 Case T-351/02 Deutsche Bahn AG v Commission, EU:T:2006:104. 94 Tietje and Wakernagel (n 88) 221–22. 95 Ortolani (n 92) 124. 96 Case C-126/97 Eco Swiss China Time Ltd v Benetton International NV, EU:C:1999:269.

346 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz an EU Member State’s public policy must incorporate the core rules of EU competition law.97 Therefore, a violation of the key EU competition law provision contained in Article 107 TFEU would likely constitute a breach of EU public policy and accordingly provide a ground for denying enforcement under Article V(2)(b). At least one commentator has observed that this has the potential to create circular reasoning.98 The public policy exception in the New York Convention would not apply because, due to the applicability of the New York Convention, there is no imputability to the Member State.99 The New York Convention therefore provides an unsatisfactory source of non-imputability: ‘If we accept that the New York Convention applies because of non-imputability, the non-imputability cannot derive from the applicability of the New York Convention’.100 In any event, as the same commentator concludes, non-imputability can be derived from the fact that the compensatory obligation, when an arbitral tribunal has found it to exist in its award, arises out of an IIA and is non-voluntary.101 ii. An Award Requiring a Member State to Pay Compensation under an IIA is Unlikely to be Enforced where it could be seen as Resulting in the Indirect Reinstitution of Illegal State Aid Although the foregoing discussion shows that arbitral awards under IIAs generally should be enforced, there is an exception to the general rule whereby the obligation to pay compensation under the award arises from the Member State’s discontinuation of a measure that would itself have constituted illegal State aid under Article 107 TFEU. At least in the case of a non-ICSID Convention arbitration award, it seems uncertain whether any EU Member State court would enforce such an award, as the EC has made it clear that payment of compensation under such an award will constitute impermissible new State aid in violation of Article 107 TFEU. The saga surrounding the Micula case illustrates the type of scenario where concerns about illegal State aid make a Member State court unlikely to enforce an award under an IIA. After the tribunal issued the award, the situation became even more complex. Romania informed the EC in February 2014 that it had partially implemented the award by offsetting certain taxes owed by a Micula-owned company against the amount of the award. In response, the EC issued a suspension injunction in May 2014 ordering Romania not to comply with the arbitral award until the EC had made a final decision on the compatibility of the compensation awarded with EU State aid law. At the time, the EC took the view that the payment of the compensation awarded to the Miculas would effectively amount to new State aid because it would have the effect of reinstituting the discontinued regional aid measures.

97 98 99 100 101

Ortolani (n 92) 124. ibid. ibid. ibid. ibid.

Arbitration

347

After conducting an investigation, the EC issued a final decision on 30 March 2015. In its decision, the EC confirmed its conclusion that payment of compensation under the award would constitute illegal State aid in violation of Article 107 TFEU and ordered Romania not to make further payment pursuant to the award and to recover any portion of the award already paid to the Miculas. In its decision, the EC distinguished the facts of Micula from the cases discussed above, which hold that damages do not amount to State aid, on the grounds that the amounts the Miculas were seeking to recover as damages were in effect those amounts that, had they been received under the regional aid scheme, would have amounted to illegal State aid. It stated: [T]he purpose of the Award is to compensate the applicants for the incentives which Romania had promised under [the incentive programme] but had been required to abolish by the Union to complete the negotiation process for its accession to the Union. Thus, in contrast to the Asteris case, the reason the applicants claim compensation in this case is because they were denied the incentives Romania promised to grant them in violations of its obligations … not to grant unlawful State aid … [A]n award of damages equal to the sum of the amounts of aid that were envisaged to be granted would constitute an indirect grant of State aid found to be illegal and incompatible with the internal market.102

It further explained its view: In the present case, the compensation awarded to the claimants by the Tribunal refers to the losses directly linked to the revocation of the EGO 24 incentives and are aimed at placing the beneficiary in the position in which it would ‘in all probability’ have found itself in had the EGO 24 incentives had not been revoked. In effect, the implementation of the award re-establishes the situation in which the claimants would have in all likelihood, found themselves if EGO 24 had never been repealed by Romania. As the advantages granted under EGO 24 were connected to the recurrent costs of the claimants and were not linked to an initial investment, those advantages constituted operating aid. Therefore, placing the beneficiary in the position in which it would have been if the EGO 24 incentives had not been revoked and thus compensating the losses linked to this revocation constitutes operating aid.103

In its Micula decision, the EC further reasoned that, under the Asteris line of cases discussed above, damages must be based on a general rule of compensation in order to fall outside the EU State aid rules. However, the EC considers intraEU IIAs, such as the Sweden–Romania BIT relied on by the Micula claimants, as incompatible with the EU treaties and therefore invalid.104 As a result, according 102

Micula Commission Decision, para 103. ibid, para 146. 104 Micula Commission Decision, para 102. In a case involving a BIT between Slovakia and the Netherlands, the German Federal Court of Justice (the Bundesgerichtshof or BGH) has recently asked the ECJ to issue a preliminary ruling as to whether intra-EU BITS are incompatible with EU law on 3 March 2016: www.juris.bundesgerichtshof.de/cgibin/rechtsprechung/document.py?Gericht=bgh& Art=pm&Datum=2016&Sort=3&nr=74606&pos=1&anz=82. The case is now pending before the ECJ as C-284/16. Request for a preliminary ruling from the Bundesgerichtshof (Germany) lodged on 23 May 2016—Slovak Republic v Achmea BV. See also Eureko BV v Slovak Republic, UNCITRAL arbitration, PCA Case No 2008-13, Award on Jurisdiction, Arbitrability and Suspension (26 October 2010); Award upheld by the Frankfurt Higher Regional Court, 10 May 2012, 26 SchH 11/10: www.italaw.com/ documents/26schh01110.pdf. 103

348 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz to the EC, compensation awarded pursuant to the provisions of an intra-EU IIAs is not awarded based on a valid general rule of compensation that would make the payment of such compensation involuntary and exclude it from the application of Article 107 TFEU. Although some commentators have been vehemently critical of the EC’s position that compensation for breach of an IIA in a case like Micula constitutes illegal State aid,105 an EU Member State court is likely to accept the EC’s view as to EU law, unless and until the EU courts rule otherwise. In the Micula case, an action has been lodged with the General Court seeking annulment of the EC’s decision.106 However, until the EU courts render a decision, the EC’s view of EU law, as propounded in Micula, is likely to have authoritative import, which means that courts would be requiring a breach of EU law if they required payment pursuant to an award ordering compensation under an intra-EU IIA for the discontinuation of State aid measures. Further complicating matters, in the interim while the EC considered the award, Romania filed an application for annulment of the award on the basis of Article 52 of the ICSID Convention in April 2014. The EC received permission to intervene in the annulment proceeding, but the ad hoc ICSID committee refused to annul the award in its decision issued in February 2016. As discussed before, Articles 53 and 54 of the ICSID Convention theoretically oblige ICSID convention parties to treat the award as binding and automatically enforceable. This creates a conflict of obligations for EU Member State courts, which must decide whether to follow the EC’s decision under EU law or uphold their obligations under the ICSID Convention. For a Member State court faced with the question of whether to follow the commands of the EC or the ICSID Convention, the question is whether EU law should prevail. In its Micula decision, the EC has suggested that EU law does prevail. The prevalence of EU law over the ICSID Convention can be justified through multiple rationales, and an EU Member State court would rarely (or never) enter an order requiring the enforcement of an award contrary to EU law. Under one approach, if one views the EU treaties as a source of international law, then one could argue that Articles 53 and 54 of the ICSID Convention do not apply to the extent their provisions are incompatible with EU law, as the Lisbon Treaty requiring Member State compliance with EU law is the latter as between itself and the ICSID Convention and treaty under Articles 30(2) and 30(4) of the VCLT.107 Moreover, in the Kadi cases, the ECJ effectively ruled that acts mandated by 105 C Söderlund, ‘Does compensation for a treat breach qualify as state aid?’ (Global Arbitration Review, 13 May 2015). 106 Action brought on 28 November 2015. Case T-704/15 Micula ea v Commission [2016] OJ C68/30. The annulment application was lodged on 28 November 2015. The legal arguments advanced in the application include the following: (i) the IIA relied on by the Miculas in bringing their claim against Romania establish ‘a system of general liability that is equally applicable to any investor’, and therefore there is no selectivity for purposes of the EU State aid rules; (ii) there is no imputability because Romania has no discretion as to enforcing the award; (iii) the EU encouraged the conclusion of BITs and created a legitimate expectation that their enforcement would not be blocked by the EU State aid rules; and (iv) the underlying revoked measures had never been subject to an independent State aid incompatibility determination. 107 Tietje and Wakernagel (n 88) 224–28.

Arbitration

349

international law (in that case, UN Security Council Resolutions imposing targeted sanctions on individuals and entities associated with terrorism) could be challenged to the extent that they are contrary to the core founding principles of EU law.108 With the ECJ’s rulings in Kadi and the EC’s decision in Micula in the background, an investor holding an ICSID award should not expect an EU Member State court to enforce the award if the compensation could be viewed as a violation of the EU State aid rules. In one example of the not surprising reluctance of EU Member State courts to fall afoul of the EC’s view as to EU law, after the EC issued its Micula decision, a Belgian court reversed its earlier order allowing the Micula claimants to recover compensation through the garnishment of funds payable to a Romanian State-owned entity.109 While the Belgian court acknowledged the res judicata effect of the arbitral award, it held that the EC’s decision that payment of compensation would constitute illegal State aid in violation of Article 107 TFEU would render its enforcement illegal. A recent English court decision dealing with the Micula award took a similar approach to reconciling the Member State’s seemingly divergent obligations.110 The English court considered whether it was required to recognise (or ‘register’) and enforce the Micula award, despite the EC’s position. In order to reconcile the seemingly conflicting EU and ICSID regimes, the Court focused on the distinction between the registration of an award, on the one hand, and the enforcement of the award, on the other. The operative act that the EC had found would violate EU law was the payment of the amounts awarded, and only enforcement of the award, not its mere registration, would require Romania to make such prohibited payment. The Arbitration Act 1966, which implemented the ICSID Convention into English law, required only that the English court recognise and enforce an ICSID award as if it were a final judgment of an English court, and thus such an award (once registered) could still be denied enforcement on the same grounds as an English court judgment. One such example given by the Court for denying enforcement was English sovereign immunity law, which would equally prevent the enforcement of a court judgment or arbitral award. The Court concluded that it would not enforce an English court judgment if to do so would, in contravention of the principle of sincere cooperation enshrined in Article 4(3) of the TFEU, require it to make a decision that would be contrary to a decision of the EC.111 Accordingly, the Court could refuse to enforce an arbitral award on the same grounds, without going against the requirements of the Arbitration Act 1966. The Court summarised the crux of its reasoning: In the present case, a judgment of the High Court is subject to the EU rules as to State aid … [T]he court must refrain from taking a decision which conflicts with a decision of

108 Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission, EU:C:2008:461; See also the more recent decision C-584/10 P European Commission and Others v Yassin Abdullah Kadi, EU:C:2013:518. 109 Tribunal de première instance francophone de Bruxelles, S Civile, Jugement, 25 January 2016, 15/7241/A and 15/7242/A. 110 Micula v Romania [2017] EWHC 31 (Comm). 111 ibid, para 69 (citing Air Canada v Emerald Supplies [2015] EWCA Civ 1024, para 70).

350 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz the Commission. Whilst this case law may be consonant with what is seen as public policy, it is not based on it. As the Court of Appeal put it in the passage from Emerald Supplies cited above, ‘The general principle of legal certainty, which underpins the duty of sincere cooperation, requires Member States to avoid making decisions that could conflict with a decision contemplated by the Commission’. This applies to the court itself: ‘national courts must, in particular, refrain from taking decisions which conflict with a decision of the Commission’ (Deutsche Lufthansa, ibid, para 41). This court cannot therefore proceed to enforce the judgment consequent on registration of the Award in circumstances in which the Commission has prohibited Romania from making any payment under the Award to the claimants because in doing so, the court would, in effect, be acting unlawfully. This does not (in the court’s view) create a conflict with the international obligations of the UK as contained in the 1966 Arbitration Act implementing the ICSID Convention in UK law, because a purely domestic judgment would be subject to the same limitation.112

For those reasons, the Court did not reverse its registration of the Micula award but granted a stay of its enforcement pending the anticipated decision of the ECJ. Other EU Member State courts are likely to look for similar ways to justify adherence to the EC’s position as consistent with international law obligations. However, because the English court appears to have focused on the duty of cooperation as a bar to its acting contrary to an EC decision (or at least a contemplated EC decision), a question may remain as to how the court would decide in a case where there may be an EU-law problem with enforcing an award but there is no prospect of an EC decision. In the case of non-ICSID awards, it will likely be far less complicated for EU Member State courts to refuse to enforce awards under the New York Convention. Under that convention, the Article V(2)(b) public policy exception is available to courts, and after the EcoSwiss case, it is clear that the national courts of Member States must take into account EU public policy, which includes the EU principles on State aid.113 Therefore, if enforcement of an award would entail a violation of EU State aid rules, a Member State court could arguably deny recognition and enforcement on the basis of Article V(2)(b). This makes it extremely unlikely that such an award would be enforced within the EU. Some commentators have suggested that the public policy exception in Article V(2)(b) of the New York Convention should not be invoked as a ‘blanket rule’ covering all situations where a revoked measure might be considered incompatible with EU State aid rules.114 On the basis that, among other things, the public policy exception should be given a restrictive and pro-enforcement interpretation, the argument would be that a court should only rely on Article V(2)(b) to refuse to enforce an arbitral award where the EC has found a violation of Article 107 TFEU and issued a decision under Article 108 TFEU requiring recovery of the illegal State aid.115

112 113 114 115

Micula v Romania (n 110), paras 131–32. Case C-126/97 Eco Swiss China Time (n 96). Ortolani (n 92) 129. ibid, 129–31.

Arbitration

351

Under this approach, a court would not refuse to enforce an award where, for example, the measure had been revoked of the Member State’s own volition without any decision of the EC having been issued. However, the likelihood that a Member State would follow this course seems doubtful, particularly after the Micula case in which the EC had never formally found the revoked measures to be illegal State aid prior to the arbitration and only issued a decision finding the measures to be State aid after the fact as part of its consideration as to whether payment of compensation under the award would be illegal State aid. Moreover, the EC’s willingness to intervene in a number of recent arbitrations involving measures voluntarily revoked by Member States make it exceedingly unlikely that an EU Member State court must decide whether to enforce an award without the EC’s input on whether the measures involved constitute illegal State aid. Therefore, it appears that, if the measures on which the award is based are incompatible with EU State aid rules, the claimant will face an uphill battle when enforcing the award in Member State courts.

C. Enforcement Outside the European Union In courts outside the EU, an investor’s chances of successfully enforcing an award that is potentially incompatible with EU State aid law appear to be substantially better. This is seemingly the case for both ICSID and non-ICSID awards. As stated before, Articles 53 and 54 of the ICSID Convention require State parties to recognise and enforce ICSID awards. When these States are not bound by EU treaties, there should be no concern in non-EU national courts about treaty conflicts that might be invoked to prevent the application of the ICSID Convention provisions. Significantly, given the United Kingdom’s anticipated departure from the EU, English courts could soon be among those not bound by EU law, and therefore more likely to enforce an award even where there are EU law concerns. As one example, the US District Court for the Southern District of New York has confirmed the arbitral award in the Micula case, converting it into an enforceable judgment.116 In recognising the award, the District Court rejected arguments that it

116 In April 2014, one of the Micula brothers commenced an ex parte action seeking to confirm the arbitral award in the US District Court for the District of Columbia. The Court denied the petition without prejudice in April 2015 on the grounds that ICSID Convention awards may not be converted into judgments by way of summary ex parte proceedings—Micula v Gov’t of Romania, 104 F Supp 3d 42 (DDC 2015). Proceedings also were initiated in the District Court for the Southern District of New York, and that Court took a different view as to whether an ICSID award could be confirmed in summary ex parte proceedings. That Court converted the award into a judgment in summary ex parte proceedings in April 2015. In August 2015, the District Court permitted the EC to file an amicus curiae brief in support of Romania’s motion to vacate and/or stay the April 2015 judgment but then denied that motion— Micula v Gov’t of Romania, Case No 15 Misc 107 (Part I), 2015 WL 4643180 (SDNY 5 August 2015). Romania’s subsequent motion for reconsideration was also denied—Micula v Gov’t of Romania, Case No 15 Misc 107 (LGS), 2015 WL 5257013 (SDNY 9 September 2015). Romania has appealed the judgment to the Second Circuit Court of Appeals, and the EC has submitted an amicus curiae brief in support of the appeal.

352 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz should defer to the EU’s position in a ‘wholly internal matter’ and emphasised that there was no basis for reviewing an award outside the ICSID Convention process. In particular, in rejecting arguments that international comity required it to defer to the EC, it stated: A party with an ICSID award can convert it into a judgment in any Member State. As a party to the ICSID Convention, the United States has a compelling interest in fulfilling its obligation under Article 54 to recognise and enforce ICSID awards regardless of the actions of another state. To do otherwise would undermine the ICSID Convention’s expansive spirit on which many American investors rely when they seek to confirm awards in the national courts of the Convention’s other Member States.117

In the case of non-ICSID awards, non-EU national courts also seem fairly likely to enforce awards, even where such awards might involve compensation for the revocation of measures that breached the EU State aid rules. In particular, non-EU national courts appear reluctant to find that a violation of EU State aid rules would constitute a violation of their own jurisdiction’s public policy sufficient to refuse enforcement. For example, in its October 2015 decision 4A-34/2015, the Federal Supreme Court of Switzerland considered an application to set aside an award rendered against Hungary under the ECT.118 The award provided compensation relating to certain power purchase agreements (PPAs) that Hungary had terminated after the EC had concluded that the PPAs were incompatible with EU State aid rules. Hungary argued, inter alia, that the award should be set aside under Article 190(2) (e) of the Swiss Private International Law Act (PILA), which permits Swiss courts to set aside arbitral awards that are ‘incompatible with public policy’. Relying on the EC’s decision in Micula, Hungary asserted that the award obliged it to breach its obligations under international law (in particular, under the TFEU provisions relating to State aid) and therefore contravened Swiss public policy as a violation of the principle of pacta sunt servanda. The Swiss Supreme Court reiterated that an award is incompatible with public policy within the meaning of Article 190(2)(e) PILA ‘when it disregards the essential and broadly recognised values which, according to prevailing concepts in Switzerland, should constitute the basis of any legal order’.119 Although it decided the issue on other grounds, the Swiss Supreme Court concluded that it was not necessarily the case that ‘an award ordering a party to compensate its opponent fairly would be incompatible’120 with this definition of public policy, ‘even if it would contradict a standard of supranational law’.121 Although it left the question open, the Court thus suggested that concerns about the possible breach of

117

Micula v Gov’t of Romania, Case No 15 Misc 107 (Part I) (n 116) 7. Republic A v B International, 4A 34/2015, Federal Supreme Court of Switzerland, 6 October 2015: www.swissarbitrationdecisions.com/sites/default/files/6%20octobre%202015%204A%2034%20 2015.pdf. 119 ibid, para 5.1. 120 ibid, para 5.3.1. 121 ibid. 118

Arbitration

353

EU State aid rules under the TFEU may not sufficiently rise to the level of public policy sufficient to set aside an award of compensation pursuant to the obligations and standards in an IIA.

VI. CONCLUSION

The clash between investment treaty arbitration, on the one hand, and EU State aid rules, on the other, appears to arise out of the two regimes’ differing conceptions of the principle of legitimate expectations. Formally, the principal elements constituting legitimate expectations are quite similar under both international investment arbitration and EU law standards. Under both legal systems, legitimate expectations require (i) a commitment or assurance by the State or the EU institutions; (ii) reliance by the recipient or the investor, or a lack of foreseeability of adverse change; and (iii) a balance of interests tilting in the recipient’s or investor’s favour. However, in practice, the interpretation and application of the concept by the EU courts and the EC is narrower than those adopted by investment arbitration tribunals. The EU law concept imposes more constraints on the doctrine. For example, it imposes a duty on the recipient or investor to investigate that the Member State has complied with the notification procedure; it does not recognise legitimate expectations based on assurances given by a Member State in violation of EU procedure; and it does not acknowledge legitimate expectations as long as an EC decision approving State aid can be challenged (or is being challenged). By contrast, the concept of legitimate expectations under IIAs tends to be broader. The host State’s failure to comply with the applicable procedure is not held against the investor/recipient of State aid, but against the host State itself. The Micula case illustrates this divergence with particular clarity. While the Investment Arbitration Tribunal ruled that it was ‘objectively reasonable’ for the claimants to rely on Romania’s stipulation that it would maintain the free zone benefits for the full 10-year period, despite finding that the claimants had not proven a legal obligation to do so under Romanian law,122 the EC held that ‘the claimants cannot justifiably claim a legitimate expectation as to the validity and continued existence of that scheme until 1 April 2009’.123 Why is there such divergence in the scope of protections? The answer may be gleaned from the very different raisons d’etre of the two legal orders: EU institutions consider legitimate expectations in the context of a legal order designed to promote competition within the Single Market, while arbitration tribunals acting under IIAs

122 Ioan Micula, Viorel Micula, SC European Food SA, SC Starmill SRL and SC Multipack SRL ICSID Case No ARB/05/20, Award (n 70), paras 690–717 and 726. 123 Micula Commission Decision, para 158.

354 Johannes Koepp, Alejandro Escobar, Laurie Frey and Ernesto Feliz are acting within the framework of an order designed to resolve bilateral investorState disputes and ensure investor protection.124 Beyond this public policy interest that the EC and EU courts are mandated to protect, an additional reason for their restrictive interpretation of the concept of legitimate expectation is their duty to protect the interests of the aid beneficiary’s competitors as well. A broader interpretation of legitimate expectations in the area of State aid would inevitably risk harm to the competitors’ interests. The beneficiary’s competitors would have little incentive to bring actions against EC decisions approving State aid if the EU courts were not prepared to order recovery of that aid save in exceptional circumstances. Thus, the narrow conception of legitimate expectations ensures the control and implementation of the State aid rules. Investment arbitration tribunals, in contrast, have no mandate to ensure the effective implementation of State aid rules. Nor are they tasked with taking account of third-party interests. Unlike in the recovery proceedings before the EU and national courts where the beneficiary’s competitors may participate as parties or third parties, those competitors normally have no standing to participate in the arbitration proceedings. The primary mission of an investment arbitration tribunal is to resolve a specific dispute between the investor and the host State. In this regard, it is also relevant that the decision-makers are different in each legal regime: government entities with public agendas in the EU context, and partyappointed tribunals under IIAs. This difference in identity of the decision-makers impacts not only the motivations for the differing interpretations of legitimate expectations under the two regimes, but also how the tension between the two regimes is dealt with when cases like Micula arise. The EC would no doubt resolve such clashes from the outset in favour of the EU understanding of legitimate expectations, by arguing that IIA tribunals have no jurisdiction to rule over EU law as between an EU Member State and an EU investor. By contrast, IIA tribunals have found themselves competent to decide such questions of EU law, and to date there has been no judicial pronouncement to the contrary. The divergence in EU State aid law and investment arbitration creates serious practical problems for investors who, like those in Micula, hold a potentially unenforceable arbitration award. The predicament in Micula serves as a warning both to investment arbitration tribunals as well as to potential claimants, as it illustrates that it will be almost impossible to enforce an arbitral award against an EU Member State over the EC’s objection, at least within the EU where it is likely to be most desirable from an asset availability perspective. However, on an optimistic note, few investors are likely to find themselves in this predicament going forward. It seems unlikely that dilemmas like those posed by Micula will arise with any frequency, for several reasons.

124 C Saavedra Pinto, ‘The “Narrow” Meaning of the Legitimate Expectations Principle in State Aid Law Versus the Foreign Investor’s Legitimate Expectations’ (2016) 15(2) European State Aid Law Quarterly 270.

Arbitration

355

First, as discussed above, there has been a general backlash against a perceived overly investor-friendly interpretation of the FET standard under IIAs which hinders a State’s right to regulate. More modern IIAs purport to limit the FET concept and specifically clarify that the concept cannot be invoked against the discontinuation of State aid. For example, the EC proposed the following language for inclusion in the Transatlantic Trade and Investment Partnership under discussion between the EU and United States: For greater certainty, nothing in this Section shall be construed as preventing a Party from discontinuing the granting of a subsidy and/or requesting its reimbursement, or as requiring that Party to compensate the investor therefor, where such action has been ordered by one of its competent authorities listed in Annex III.125

Second, the EU has revealed plans for a multilateral investment court to replace the system of privately appointed arbitrators deciding investor–state disputes. The proposal would include a roster of publicly appointed judges in place of arbitrators and an appellate mechanism. If implemented, it is likely that the interpretation of the FET standard would take greater account of public policy interests, such as the promotion of competition within the Single Market. Third, investment arbitration tribunals increasingly permit third-party intervention in cases affecting those third parties’ interests. As a result, investment arbitration tribunals may consider a fuller picture, including third-party interests, in reaching decisions. Taken together, these factors may lead investment arbitration tribunals to adopt a narrower concept of legitimate expectations in the context of State aid. At the same time, the widespread criticism of EU institutions’ narrow interpretation of legitimate expectations may make an increased convergence of the two legal orders in this respect more likely.

125 Commission Draft Text Transatlantic Trade and Investment Partnership (14 July 2016) Ch II Investment, Art 2(3), available at: www.trade.ec.europa.eu/doclib/docs/2015/september/tradoc_153807.pdf.

13 Common EU Law Principles of Private Enforcement of State Aid GEORG M BERRISCH AND BRIAN R BYRNE*

I. INTRODUCTION

T

HE COURT OF Justice of the European Union (CJEU) stated as early as 1964 that the notification and standstill obligation imposed by Article 108(3) of the Treaty on the Functioning of the European Union (TFEU) has direct effect in the Member States.1 Private parties, therefore, can enforce the notification and standstill obligation by bringing proceedings before national courts. For example, if a company considers that its competitor is receiving, or has received, unlawful State aid (ie, aid that has not been notified to the Commission), its options are not limited to filing a complaint with the Commission. It may also, or as an alternative, initiate an action before a national court alleging that the Member State has violated Article 108(3) TFEU. There are a number of potential benefits associated with such private enforcement actions: they can offer effective remedies to claimants, which in turn create a deterrent effect, arguably leading to more robust State aid enforcement overall;2 they can also resolve infringements of State aid rules at national level without necessarily drawing on the Commission’s (limited) resources. Notwithstanding the apparent benefits, private enforcement actions of the kind described above have not yet attained prevalence throughout the EU. Thus far, the majority of State aid litigation before national courts, which continues to increase in absolute terms,3 is not concerned with ‘enforcement’ in the sense of seeking a * The authors wish to thank Sohra Askaryar and Buki Owa for their assistance with the research carried out for this chapter. 1 Case 6/64 Costa v ENEL, EU:C:1964:66, 596. See also: Case C-120/73 Lorenz GmbH v Germany and Others, EU:C:1973:152, para 8; Case C-354/90 Fédération nationale du commerce extérieur des produits alimentaires (FNCE) and Others v France, EU:C:1991:440, para 11; Case C-284/12 Deutsche Lufthansa, EU:C:2013:755, para 29; Case C-672/13 OTP Bank, EU:C:2015:185, para 76; Case C-368/04 Transalpine Ölleitung in Österreich, EU:C:2006:644, para 41; Joined Cases C-78/08 to C-80/08 Paint Graphos and Others, EU:C:2011:550, para 25; Joined Cases C-393/04 and C-41/05 Air Liquide Industries Belgium, EU:C:2006:403, para 41. 2 Commission Notice on the enforcement of State aid law by national courts (Enforcement Notice) [2009] OJ C85/1, para 5. 3 See the 2006 ‘Study on the Enforcement of State Aid Law at National Level’ (2006 Enforcement Study) coordinated by T Jestaedt, J Derenne and TR Ottervanger (Luxembourg, Office for Official Publications of the European Communities, 2006) and the 2009 update of the 2006 study (2009 Update) coordinated

358 Georg M Berrisch and Brian R Byrne level competitive playing field. Rather, many of the cases involve beneficiaries of aid seeking to resist a State aid recovery order4 or companies hoping to avoid a particular tax liability, which they claim is discriminatory.5 There are several reasons, all of which are explored further in this chapter, that might explain why true private enforcement actions are not ubiquitous. First, the likelihood of receiving financial compensation in the form of damages is low. Second, claimants must overcome procedural hurdles, such as establishing legal standing and satisfying applicable burdens of proof, all of which can vary from one Member State to the next. Third, claimants often have difficulty obtaining information that is key to their case. Fourth, the length of national court proceedings can be off-putting for claimants, especially considering the possibility that the national court will make a preliminary reference to the CJEU and/or seek assistance from the Commission. Finally, national judges sometimes lack the necessary expertise or experience to navigate the complexities of EU State aid jurisprudence. This chapter addresses the challenges associated with private enforcement of the State aid regime. We describe the respective roles of the Commission and the national courts in enforcing the Treaty provisions on State aid; the parties to national court cases; the types of remedy that might be available; and the obligations of national courts in cases of concurrent proceedings before the Commission or the EU Courts. As noted above, many State aid cases before national courts concern the enforcement of Commission decisions ordering the Member State to recover unlawful and incompatible aid. Such cases arise when a beneficiary challenges a recovery decision by a national authority before a national court or when the national authority must resort to litigation in order to recover the aid. These cases are outside the scope of this chapter, but are discussed in chapter eleven.

II. THE ROLE OF THE COMMISSION AND THE NATIONAL COURTS IN THE ENFORCEMENT OF ARTICLES 107 AND 108 TFEU

The CJEU has described the Commission and the national courts as fulfilling ‘complementary’ and ‘separate’ roles as regards their supervision of Member States’ compliance with Articles 107 and 108 TFEU.6 The roles are considered by J Derenne, C Kaczmarek and J Clovin (Brussels, Lovells, 2009). Both are also available at: ec.europa. eu/competition/state_aid/studies_reports/studies_reports.html [2006] and ec.europa.eu/competition/ state_aid/studies_reports/enforcement_study_2009.pdf [2009]. 4 See, eg, Case C-148/04 Unicredito Italiano, EU:C:2005:774; Joined Cases C-183/02 P and C-187/02 P Demesa and Territorio Histórico de Álava v Commission, EU:C:2004:701; Joined Cases C-182/03 and C-217/03 Belgium and Forum 187 v Commission, EU:C:2006:416. See also: Enforcement Notice (n 2) para 4. 5 See, eg, Case C-143/99 Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke, EU:C:2001:598, which was a reference by the Verfassungsgerichtshof Austria (Constitutional Court Austria) for a preliminary ruling in proceedings concerning the rejection of applications from undertakings who applied for energy tax rebates. See also: 2006 Enforcement Study (n 3) 68 and Enforcement Notice (n 2) para 4. 6 Joined Cases C-352/14 and C-353/14 Iglesias Gutiérrez and Elisabet Rion Bea v Bankia SA and Others, EU:C:2015:691, para 26; Transalpine Ölleitung in Österreich (n 1) para 37; Joined Cases C-261/01 and C-262/01 van Calster and Cleeren, EU:C:2003:571, para 74; Case C-39/94 SFEI and Others, EU:C:1996:285, para 41.

Private Enforcement of State Aid 359 ‘complementary’ because national courts help contribute to the Commission’s effective enforcement of EU State aid law in two important ways: first, by helping to safeguard the rights of parties affected by an unlawful implementation of an aid measure;7 and second, by playing an important role in the enforcement of recovery decisions adopted by the Commission.8 The roles are ‘separate’ in the sense that their respective competences are not aligned: national courts are not competent to assess the compatibility of an aid measure, as this falls within the exclusive competence of the Commission, subject to judicial review by the EU Courts.9 As regards the complementary nature of the relationship, the Commission has published two notices that help clarify the respective roles of the Commission and the national courts concerning State aid enforcement. The ‘Notice on the enforcement of State aid law by national courts’ (Enforcement Notice),10 provides guidance on a broad range of issues likely to arise in State aid litigation before national courts. The ‘Notice towards an effective implementation of Commission decisions ordering Member States to recover unlawful and incompatible State aid’ (Recovery Notice),11 focuses specifically on the recovery of aid.12 Several elements of both the Enforcement Notice and the Recovery Notice have since been codified by Council Regulation (EU) 2017/1589 laying down detailed rules for the application of Article 108 TFEU (Procedural Regulation).13

A. The Commission The Commission has several responsibilities with respect to State aid enforcement. Its main role is to examine whether a proposed measure constitutes ‘aid’ and, if so, to determine whether that aid is compatible with the common market.14 Such determinations may be made either in the context of notified aid, or non-notified ‘unlawful’ aid, following a formal investigation.15 As noted, deciding on whether

7 Case C-574/14 PGE, EU:C:2016:686, para 31; Deutsche Lufthansa (n 1) para 28; Transalpine Ölleitung in Österreich (n 1) para 38; van Calster and Cleeren (n 6) para 75. See also: Enforcement Notice (n 2) para 21(a). 8 Enforcement Notice (n 2) para 21(b). 9 Case C-590/14 P DEI and Commission v Alouminion tis Ellados, EU:C:2016:797, para 96; PGE (n 7) para 32; Iglesias Gutiérrez and Elisabet Rion Bea (n 6) para 26; OTP Bank (n 1) para 37; Deutsche Lufthansa (n 1) para 28; Case C-6/12 P Oy, EU:C:2013:525, paras 38–39; Case C-275/10 Residex Capital IV, EU:C:2011:814, para 27; SFEI and Others (n 6) para 42; FNCE and Others (n 1) para 14. 10 Enforcement Notice (n 2). 11 Notice from the Commission Towards an effective implementation of Commission decisions ordering Member States to recover unlawful and incompatible State aid (Recovery Notice) [2007] OJ C272/4. 12 The Recovery Notice is primarily a guidance tool for the Member State itself, as opposed to the national courts; however, because recovery often necessitates litigation, the Recovery Notice also addresses the interplay between national courts and the Commission. 13 Council Regulation (EU) 2015/1589 of 13 July 2015 laying down detailed rules for the application of Art 108 of the Treaty on the Functioning of the European Union (codification) [2015] OJ L248/9 (Procedural Regulation), eg, Art 29 regarding the cooperation between the Commission and national courts; and Art 16 regarding the recovery of aid. 14 Enforcement Notice (n 2) para 20. 15 Procedural Regulation (n 13) Arts 4 and 9 regarding ‘notified aid’, Arts 1(f), 12, 15 regarding ‘unlawful aid’.

360 Georg M Berrisch and Brian R Byrne aid is compatible with the common market is the exclusive competence of the Commission.16 Further, as discussed below in relation to the principle of sincere cooperation,17 the Commission is tasked with assisting the national courts when they are confronted with litigation founded on EU State aid law. Specifically, the Commission is obliged, upon request by the national courts, to transmit relevant information in its possession and/or issue a written opinion concerning the application of the State aid rules.18 The Commission also has the possibility of submitting written observations to the national court and, if the national court permits, to submit oral observations.19

B. The National Courts National courts have two principal roles with respect to the enforcement of EU State aid law. First, they are obliged to safeguard the rights of individuals that might be affected by the unlawful implementation of aid.20 Second, they may be called upon to enforce recovery decisions adopted by the Commission. As noted above, the present chapter deals only with the former task. As noted earlier, the notification and standstill obligation gives rise to directly effective rights for ‘individuals faced with a possible breach … of the prohibition laid down by Article [108(3) TFEU]’.21 According to the CJEU, ‘national courts do no more than preserve [those rights] until the final decision of the Commission’.22 Consequently, even if the Commission is examining the same purported aid measure that is the subject of the national court proceedings, the national court is not entitled to stay proceedings pending the outcome of the Commission’s compatibility assessment. As the CJEU has neatly summarised, ‘[u]ltimately, the first obligation of the national court is to make a ruling, whether positive or negative’.23 In order to safeguard or ‘uphold’24 these individual rights, the national courts must assess whether the disputed measure can be classified as ‘aid’ within the

16 DEI and Commission (n 9) para 96; PGE (n 7) para 32; Iglesias Gutiérrez and Elisabet Rion Bea (n 6) para 26; OTP Bank (n 1) para 37; Deutsche Lufthansa (n 1) para 28; P Oy (n 9) paras 38–39; Residex Capital IV (n 9) para 27; SFEI and Others (n 6) para 42; FNCE and Others (n 1) para 14. 17 This will be discussed in greater detail in section II.C.iii below. 18 Procedural Regulation (n 13) Art 29(1); Enforcement Notice (n 2) paras 77–96. See also: SFEI and Others (n 6) para 50; Case C-2/88-Imm, Zwartveld and Others, EU:C:1990:440, paras 17–18. 19 Procedural Regulation (n 13) Art 29(2). 20 PGE (n 7) para 31; Deutsche Lufthansa (n 1) para 28; Transalpine Ölleitung in Österreich (n 1) para 38; van Calster and Cleeren (n 6) para 75. 21 P Oy (n 9) para 39; Case C-199/06 CELF and ministre de la Culture and de la Communication (CELF I), EU:C:2008:79, para 38; See also: P Oy (n 9) para 39; and FNCE and Others (n 1) para 14. 22 FNCE and Others (n 1) para 14. See also: Enforcement Notice (n 2) para 46. 23 Case C-1/09 CELF and ministre de la Culture and de la Communication (CELF II), EU:C:2010:136, para 39. This will be discussed in greater detail in section IV below. 24 See Joined Cases C-393/04 and C-41/05 Air Liquide Industries Belgium (n 1) para 41; Case C-345/02 Pearle and Others, EU:C:2004:448, para 31; van Calster and Cleeren (n 6) paras 53 and 64; Case C-17/91 Lornoy and Others, EU:C:1992:514, para 30.

Private Enforcement of State Aid 361 meaning of Article 107(1) TFEU.25 Absent such an assessment, it is impossible for the national court to conclude whether there may have been a violation of Article 108(3) TFEU (as the notification and standstill obligation only applies to ‘aid’ measures).26 Thus, the national courts and the Commission have overlapping powers to interpret the notion of State aid.27 This overlap poses specific challenges in cases of concurrent proceedings, as discussed further in section IV below. Indeed, in certain cases, depending on the stage of the Commission’s proceedings, the national court’s hands might be tied as regards its assessment of whether the measure at issue constitutes ‘aid’. Determining the existence of State aid is often a complex exercise, even more so for national courts that might lack the necessary experience or familiarity with EU State aid rules. To help alleviate this burden, national courts have the option of requesting an ‘opinion’ from the Commission which ‘may, in principle, cover all economic, factual, or legal matters which arise in the context of the national proceedings’.28 Such opinions can address, for example, whether a measure qualifies as State aid; whether a measure might meet a requirement under the Block Exemption Regulation29 (such that Article 108(3) TFEU is not applicable); or whether a measure has already been notified and approved.30 National courts may also request that the Commission transmit relevant information in its possession, including factual data and economic analysis.31 The mechanisms for national courts to request a Commission opinion, and/or information in the Commission’s possession, are discussed further below in relation to the principle of sincere cooperation. As an alternative to seeking guidance from the Commission, national courts may avail of the procedure under Article 267 TFEU when confronted with a question concerning the interpretation of EU State aid law and seek a preliminary ruling from the CJEU.32 The ruling to be made by the national court ‘goes beyond that of a judge ruling on an application for interim relief ’.33 Rather, if the national court determines that

25 Deutsche Lufthansa (n 1) para 35; P Oy (n 9) para 38; Transalpine Ölleitung in Österreich (n 1) para 39; Pearle (n 24) para 31; SFEI and Others (n 6) paras 49 and 53; FNCE and Others (n 1) para 10; Case C-78/76 Steinike & Weinlig v Germany, EU:C:1977:52, para 14. 26 Procedural Regulation (n 13) Art 29(1). Transalpine Ölleitung in Österreich (n 1); SFEI and Others (n 6) paras 49 and 53; Case C-189/91 Kirsammer- v Sidal, EU:C:1993:907, para 14; FNCE and Others (n 1) para 10; Steinike & Weinlig (n 25) para 14. 27 Enforcement Notice (n 2) para 10, with reference to: Transalpine Ölleitung in Österreich (n 1) para 39; SFEI and Others (n 6) para 49; FNCE and Others (n 1) para 10; and Steinike & Weinlig (n 25) para 14. 28 Enforcement Notice (n 2) para 90. 29 Commission Regulation (EU) No 651/2014 of 17 June 2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty (Block Exemption Regulation) [2014] OJ L187/1. 30 Enforcement Notice (n 2) para 91(c). 31 Procedural Regulation (n 13) Art 29(1); Enforcement Notice (n 2) paras 82–88. 32 Because the assessment of the compatibility of aid with the common market falls within the exclusive competence of the Commission, national courts may not, when referring questions to the CJEU, seek guidance on whether the disputed measure is compatible with the common market. Case C-237/04 Enirisorse, EU:C:2006:197, para 23; Case C-297/01 Sicilcassa and Others, EU:C:2003:416, para 47. 33 SFEI and Others (n 6) para 67.

362 Georg M Berrisch and Brian R Byrne the Member State has violated Article 108(3) TFEU, it must ‘draw the necessary consequences’, in accordance with its national law, with regard to both the validity of the acts giving effect to the aid and the recovery of the aid granted in violation of Article 108(3) TFEU.34 In principle, the national court must order termination of the measure at issue and repayment of the aid that was granted in violation of Article 108(3) TFEU, in accordance with its domestic procedural law.35 If, however, the Commission subsequently determines that the measure constitutes compatible aid, the national authority that granted the aid can ‘re-grant’ it, notwithstanding that the national court ordered recovery. The ability of, and indeed the obligation on, national courts to order the recovery of aid, merely because it was granted in violation of Article 108(3) TFEU, is a point of distinction between the respective powers of the national courts and the Commission. The CJEU has held that, unlike the national courts, the Commission cannot order the recovery of aid solely on the ground that it was not notified in accordance with Article 108(3) TFEU.36 Instead, the Commission must first carry out a full compatibility assessment.37 While the Procedural Regulation provides the Commission with the possibility of ordering the provisional recovery of unlawful aid, pending the outcome of the Commission’s compatibility assessment, the legal threshold is extremely high. The Commission must have ‘no doubts about the aid character of the measure’; there must be an ‘urgency to act’; and ‘a serious risk of substantial and irreparable damage to a competitor’.38 In practice, it is very rare for the Commission to order the provisional recovery of aid.

C. General Principles Governing the Respective Roles of the Commission and National Courts in the Enforcement of Articles 107 and 108 TFEU Four general EU law principles are of particular importance in the context of the private enforcement of State aid law: the principle of procedural autonomy; the principle of equivalence; the principle of effectiveness; and the principle of sincere cooperation. Each of these principles imposes obligations on national courts and/ or impacts on the nature of the relationship between the national courts and the

34 Case C-71/04 Xunta de Galicia, EU:C:2005:493, para 50. See also: CELF I (n 21) para 41; Transalpine Ölleitung in Österreich (n 1) para 47; van Calster and Cleeren (n 6) para 64; SFEI and Others (n 6) para 40; Joined Cases C-144/91 and C-145/91 Demoor and Others v Belgian State, EU:C:1992:518, para 26; FNCE and Others (n 1) para 12. 35 Deutsche Lufthansa (n 1) para 30; Residex Capital IV (n 9) para 29; CELF I (n 21) para 41; Joined Cases C-393/04 and C-41/05 Air Liquide Industries Belgium (n 1) para 42; Xunta de Galicia (n 34) para 49; van Calster and Cleeren (n 6) para 64; SFEI and Others (n 6) paras 40 and 68; FNCE and Others (n 1) para 12. 36 SFEI and Others (n 6) para 43; FNCE and Others (n 1) para 12; Case C-301/87 France v Commission (Boussac), EU:C:1990:67, paras 9–24; and Case C-142/87 Belgium v Commission (Tubemeuse), EU:C:1990:125, para 20. 37 Enforcement Notice (n 2) para 25, with reference to: CELF I (n 21) para 38; FNCE and Others (n 1) para 14; Boussac (n 36) paras 17–23; and Tubemeuse (n 36) paras 15–19. 38 Procedural Regulation (n 13) Art 13(2).

Private Enforcement of State Aid 363 Commission. As such, the application of these principles can be key to a private claimant’s chances of success in national court proceedings. i. Procedural Autonomy As noted above, the primary role of the national courts with respect to State aid enforcement is to safeguard the directly effective individual rights associated with the notification and standstill obligation imposed by Article 108(3) TFEU. In carrying out this role, national courts enjoy the principle of procedural autonomy, which provides that in the absence of [EU] legislation, it is for the internal legal order of each Member State to designate the competent courts and lay down the detailed procedural rules for legal proceedings intended fully to safeguard the rights which individuals derive from [EU] law (emphasis added).39

Therefore, the options available to a private claimant, when seeking to initiate an action alleging a violation of the notification and standstill obligation in Article 108(3) TFEU, will depend on the domestic procedural laws of the Member State in question. These laws will not only determine the court(s) entitled to hear the case, but will, in principle, also determine critical procedural issues that may be fundamental to a claimant’s chances of success. Such issues include the assessment of whether the claimant has legal standing to bring the case and the applicable standard of proof that the claimant must satisfy. Naturally, the principle of procedural autonomy has the potential to cause disharmony and inconsistencies across the EU with respect to private litigation based on EU State aid law. Furthermore, this principle gives rise to a number of practical challenges, especially for the national court itself. On the one hand, the national court is obliged to apply substantive provisions of EU law to determine, for example, whether the disputed measure constitutes ‘aid’ within the meaning of Article 107(1) TFEU. On the other hand, the national court is bound by its own domestic procedural rules which may critically affect the outcome of the case. These practical challenges are discussed further in section III, in particular certain issues that have arisen with respect to legal standing. ii. Principles of Equivalence and Effectiveness The procedural autonomy afforded to national courts is not unfettered. Two important principles of EU law curtail the independence of the national courts: namely the principle of equivalence, and the principle of effectiveness.

39 Case C-224/01 Köbler, EU:C:2003:513, para 46. See also: Case C-505/14 Klausner Holz Niedersachsen, EU:C:2015:742, para 40; Transalpine Ölleitung in Österreich (n 1) para 45; Joined Cases C-392/04 and C-422/04 i-21 Germany, EU:C:2006:586, para 57; Case C-526/04 Laboratoires Boiron, EU:C:2006:528, para 51; Case C-382/99 Netherlands v Commission, EU:C:2002:368, para 90; and Case C-33/76 Rewe, EU:C:1976:188, para 5.

364 Georg M Berrisch and Brian R Byrne The principle of equivalence provides that national procedural rules governing legal actions for safeguarding rights which individuals derive from EU law must not be less favourable than those governing individual rights which originate in the Member State’s domestic law.40 The principle of effectiveness provides that national procedural rules must not render it ‘practically impossible’ or ‘excessively difficult’ to exercise rights conferred by EU law.41 This principle has immense practical significance for private claimants relying on an alleged breach of Article 108(3) TFEU. For instance, with respect to the applicable burden of proof, the CJEU has held that if it would be impossible or excessively difficult for a claimant to produce the evidence necessary to show that a particular measure constitutes State aid, the national court has an obligation ‘to use all procedures available to it under national law, including that of ordering the necessary measures of inquiry, in particular the production by one of the parties or a third party of a particular document’.42 Considering that one of the key challenges facing private claimants is the information asymmetry that exists between the claimant, on the one hand, and the Member State and/or beneficiary on the other, the obligation imposed on national courts by virtue of the principle of effectiveness may be critical to a private claimant’s ability to secure supporting evidence.43 Another example of the practical implications of the principle of effectiveness concerns actions for damages (discussed further in section III.B.iv below). National law cannot exclude the possibility that a Member State, found to have infringed Article 108(3) TFEU, can be held liable for the claimant’s loss of profit; in the event that national law contains such an exclusion, the national court would need to leave that provision unapplied.44 The principle of effectiveness is also capable of trumping well-known legal principles that are common to many Member States. For instance, in the Klausner Holz Niedersachsen case,45 which concerned the principle of res judicata, a German court had ruled that certain contracts between a German regional authority (‘Land’) and a private company, Klausner, for the supply of wood were enforceable. Under the contracts, the Land had committed itself to sell fixed amounts of wood for a period of seven years to Klausner at predetermined prices, and to refrain from making other sales at prices below those agreed in the contracts. Following the judgment

40 Transalpine Ölleitung in Österreich (n 1) para 45; i-21 Germany (n 39) para 57; Laboratoires Boiron (n 39) para 51; Netherlands (n 39) para 90; Rewe (n 39) para 5. 41 Transalpine Ölleitung in Österreich (n 1) para 45; i-21 Germany (n 39) para 57; Laboratoires Boiron (n 39) para 51; Netherlands (n 39) para 90. See also: CELF II (n 23) para 31; Case C-232/05 Commission v France (Scott), EU:C:2006:651, para 49; Case C-174/02 Streekgewest, EU:C:2005:10, para 18; and Rewe (n 39) para 5. 42 Laboratoires Boiron (n 39) paras 55 and 57. 43 Enforcement Notice (n 2) para 76; ibid, paras 55 and 57. 44 Enforcement Notice (n 2) para 49(a) with reference to: Joined Cases C-46/93 and C-48/93 Brasserie du pêcheur and Factortame, EU:C:1996:79, paras 87 and 90. See also: Case T-239/94 Association des Aciéries Européennes Indépendantes (EISA) v Commission, EU:T:1997:158, para 109; Joined Cases C-295/04 to C-298/04 Manfredi, EU:C:2006:461, para 93. 45 Klausner Holz (n 39).

Private Enforcement of State Aid 365 declaring that the contracts were enforceable, Klausner brought another action against the Land demanding performance, and payment of damages and the supply of information on prices applied in the sector. In an attempt to resist these claims, the Land argued that, despite the prior judgment declaring the contracts enforceable, the contracts could not be executed because it would involve a grant of aid in violation of Article 108(3) TFEU. The German court referred a question to the CJEU asking whether the national law principle of res judicata precluded the Land’s Article 108(3) argument (as the German court had already ruled that the contract was enforceable). The CJEU noted that the issue of State aid had not been argued in the case that gave rise to the binding national court judgment and stated that: [A] national rule which prevents the national court from drawing all the consequences of a breach of the third sentence of Article 108(3) TFEU because of a decision of a national court, which is res judicata, given in a dispute which does not have the same subject-matter and which did not concern the State aid characteristics of the contracts at issue must be regarded as being incompatible with the principle of effectiveness. A significant obstacle to the effective application of EU law and, in particular, a principle as fundamental as that of the control of State aid cannot be justified either by the principle of res judicata or by the principle of legal certainty.46

While the principle of effectiveness is indeed a valuable tool from the perspective of a private claimant, convincing a national court to set aside the rules of domestic law will not necessarily be easy. The CJEU has emphasised the need for the national court to undertake a case-by-case analysis, and its guidance in this regard is ripe with nuances: [E]ach case … must be analysed by reference to the role of [the national] provision in the procedure, its progress and its special features, viewed as a whole, before the various national instances. In that context, it is necessary to take into consideration, where relevant, the principles which lie at the basis of the national legal system, such as the protection of the rights of the defence, the principle of legal certainty and the proper conduct of the proceedings.47

iii. Principle of Sincere Cooperation As noted above, in the sphere of State aid enforcement, the responsibility of the national courts overlaps with that of the Commission insofar as both are competent to assess the existence of ‘aid’ within the meaning of Article 107(1) TFEU. For that reason, among others, the principle of loyal/sincere cooperation is of particular importance, especially in cases of concurrent proceedings (discussed further in section IV below). The principle of sincere cooperation is enshrined in Article 4(3) of the Treaty on European Union (TEU) which provides: Pursuant to the principle of sincere cooperation, the Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks which flow from the Treaties.

46 47

ibid, para 45. Case C-426/05 Tele2 Telecommunication, EU:C:2008:103, para 55.

366 Georg M Berrisch and Brian R Byrne The Member States shall take any appropriate measure, general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting from the acts of the institutions of the Union. The Member States shall facilitate the achievement of the Union’s tasks and refrain from any measure which could jeopardise the attainment of the Union’s objectives (emphasis added).

The principle of sincere cooperation is reciprocal in nature.48 This is clear both from the wording of Article 4(3) TEU and from the case law of the CJEU. Thus, it is not only Member States that are obliged to ‘facilitate the achievement of the Commission’s tasks, which consist in particular … in ensuring that the provisions of the Treaty and the measures taken by the institutions pursuant thereto are applied’,49 but there is also an obligation on the Commission to provide assistance to the Member States. For instance, in the State aid context, the CJEU has held that ‘the Commission must respond as quickly as possible to requests from national courts’ for clarification concerning the application of Article 108(3) TFEU.50 The principle of sincere cooperation, including the Commission’s responsibility to support and assist national courts, is also embodied in the Procedural Regulation51 and the Enforcement Notice, which set out two different support mechanisms that are intended to be ‘practical and user-friendly’.52 The first support mechanism is the ability of the national court to request that the Commission transmit relevant information in its possession.53 Such information might include, for example: information concerning a pending Commission procedure, in particular whether the measure at issue before the national court is being considered concurrently by the Commission (and if so, the status of the Commission’s assessment); or factual data, statistics, market studies and economic analysis.54 These examples are merely illustrative, as the Commission considers that it has a duty to provide ‘whatever information the [national court] may seek’.55 The second support mechanism is the ability of the national court to request an ‘opinion’ from the Commission concerning the application of State aid rules.56 Such opinions are not legally binding on the national court but are intended to provide guidance on particular factual, economic or legal issues.57 For instance, the Enforcement Notice lists the following examples of the subject matter that an opinion might address: whether a particular measure qualifies as ‘aid’ within the meaning of

48 Case C-214/07 Commission v France, EU:C:2008:619, para 45; Case C-415/03 Commission v Greece, EU:C:2005:287, para 42; Case C-348/93 Commission v Italy, EU:C:1995:95, para 17. 49 Case C-33/90 Commission v Italy, EU:C:1991:476, para 18. See also: DEI and Commission (n 9) para 105; PGE (n 7) para 33; Deutsche Lufthansa (n 1) para 41. 50 SFEI and Others (n 6) para 50, with reference to Zwartveld (n 18) paras 17–18. See also: Case C-334/99 Germany v Commission, EU:C:2003:55, para 49; Case C-234/89 Delmitis, EU:C:1991:91, para 53; and Case T-353/94 Postbank v Commission, EU:T:1996:119, para 65. 51 Procedural Regulation (n 13) Art 29(1). 52 ibid. See also: Enforcement Notice (n 2) para 78. 53 See also: Enforcement Notice (n 2) paras 82–88. 54 ibid, para 83. 55 ibid, para 86, with reference to: Postbank (n 50) para 64; and Zwartveld (n 18) paras 16–22. 56 Procedural Regulation (n 13) Art 29(1). See also: Enforcement Notice (n 2) paras 89–96; SFEI and Others (n 6) para 50; Zwartveld (n 18) paras 17 and 18. 57 Enforcement Notice (n 2) para 93.

Private Enforcement of State Aid 367 Article 107(1) TFEU; whether the requirements of the Block Exemption Regulation are met; whether a certain aid measure falls under a specific aid scheme that has been notified and approved by the Commission; whether ‘exceptional circumstances’58 exist such that the national court is prevented from ordering the full recovery of aid granted to a beneficiary; assistance as regards the calculation of interest; and assistance in the context of a claim for damages.59 The Procedural Regulation also provides the possibility for the Commission to submit, on its own initiative, written observations to national courts in cases where ‘the coherent application of Article 107(1) or Article 108 TFEU so requires’.60 The Commission can also make oral observations, provided the national court grants permission.61 For the purpose of preparing observations in national court proceedings, the Commission may request that the national court transmit documents at the court’s disposal that are ‘necessary for the Commission’s assessment’.62

III. PRIVATE ENFORCEMENT OF EU STATE AID LAW IN PROCEEDINGS BEFORE NATIONAL COURTS

As noted in the introduction to this chapter, true private enforcement of State aid, ie, actions brought before national courts, initiated by one competitor against another, is not commonplace. To the extent that such actions are brought, the nature of the proceedings will vary significantly from one Member State to the next, due to the principle of procedural autonomy discussed above. Nonetheless, certain commonalities are readily identifiable. Below, we consider the parties typically involved in such actions, and the type of relief that is often sought.

A. Parties i. Plaintiffs The plaintiff in a private State aid enforcement action is typically a competitor of the alleged aid beneficiary or other third party affected by the aid measure. The larger the body of prospective plaintiffs entitled to initiate a case before national courts, the higher the effectiveness, and prevalence, of private enforcement of State aid rules. In this regard, it is essential to address two interrelated questions. First, who supposedly benefits from the directly effective rights enshrined in Article 108(3) TFEU? Second, to what extent do the national laws of the Member States, concerning legal standing, facilitate or impede such parties from bringing their case?

58 59 60 61 62

This is discussed further in section III.B.ii below. Enforcement Notice (n 2) para 91(a)–(f). Procedural Regulation (n 13) Art 29(2). ibid, Art 29(2). ibid.

368 Georg M Berrisch and Brian R Byrne As regards the first question, the case law of the CJEU provides guidance. According to the Court, it is ‘important to protect parties affected by the distortion of competition caused by the grant of the unlawful aid’ (emphasis added).63 Therefore, it seems uncontroversial to conclude that direct competitors of the beneficiary are, at least as a matter of EU law, entitled to benefit from the direct effect of Article 108(3) TFEU.64 However, the extent of the protection is not limited to competitors.65 In the Streekgewest case, the Court stated that: [A]n individual may have an interest in relying … on the direct effect of … Article [108(3) TFEU] not only in order to erase the negative effects of the distortion of competition created by the grant of unlawful aid, but also in order to obtain a refund of a tax levied in breach of that provision. In the latter case, the question whether an individual has been affected by the distortion of competition arising from the aid measure is irrelevant to the assessment of his interest in bringing proceedings. The only fact to be taken into consideration is that the individual is subject to a tax which is an integral part of a measure implemented in breach of the prohibition referred to in that provision (emphasis added).66

Strictly speaking, the Court’s statement in Streekgewest would appear to benefit only claimants subject to a tax which is an integral part of a State aid measure; in other words, a tax the proceeds of which are earmarked specifically to finance the alleged aid (referred to as a ‘hypothecated’ tax). However, it is foreseeable that eager plaintiffs, who are neither competitors, nor payers of a hypothecated tax, would seek to rely on Streekgewest to argue that the extent of the protection afforded by Article 108(3) TFEU is much broader. In any event, it is clear that the clarity afforded by EU case law, as regards who precisely can rely on Article 108(3) TFEU, is limited. Therefore, in light of the principle of procedural autonomy, national rules on legal standing are critical. Hence, it is necessary to address the second question referred to above, ie, the extent to which national law rules facilitate or hamper private enforcement actions. Domestic procedural laws concerning legal standing vary across the EU, as do their interpretation and application by national courts. In the context of private State aid enforcement, this has led to significant inconsistencies in the case law of different Member States, but also within the same Member State.67 If the applicable

63

CELF I (n 21) para 38. eg, in Belgium, competitors of the aid recipient have brought admissible claims based on Art 108(3) TFEU (see Commercial Court of Brussels, Ryanair v Brussels Airlines [29 April 2015] A 2014/52.655 and President of the Commercial Court of Brussels, Breda Fucine Meridionali v Manoir Industries [13 February 1995] (1995) Journal des tribunaux—Droit européen 72). (For more details on the Breda case, see also, 2006 Enforcement Study (n 3) 87.) 65 eg, the 2006 Enforcement Study (n 3) 268, refers to Greece, where creditors of the recipient of unlawful State aid may also challenge a relevant administrative act, since it may affect their legal interests, for example, by creating a false image of the creditworthiness of the beneficiary (see Decision 3910/1988 of the 4th Chamber of the Hellenic Council of State). 66 Streekgewest (n 41) para 19. 67 The 2006 Enforcement Study (n 3) 323, refers, for example, to Italy, where a court of appeal stated that a claimant does not have standing in national proceedings concerning the implementation of a negative Commission decision if it is not directly affected by the decision, even if it had a material interest which coincided with the interest underlying the Commission decision (case of the Court of Appeal of Cagliari Exol SpA v Nuova Cartiera di Arbatax SpA [21 July 1999]). Another Italian court, however, had 64

Private Enforcement of State Aid 369 thresholds to establish legal standing are set too high, the private enforcement of State aid will be inhibited. Indeed, the Commission has identified legal standing as a hurdle contributing to a low level of private State aid actions overall.68 Moreover, overly burdensome standing rules might violate the right to effective judicial protection under EU law69 and/or the principle of effectiveness (discussed above in section II.C.ii).70 ii. Defendants In general terms, State aid litigation before national courts can feature two different categories of defendant: —



The entity that adopted the measure at issue: The notification and standstill obligation pursuant to Article 108(3) TFEU lies with the Member State, which includes not only the central government but also any entity that adopts measures that can qualify as aid. This can include private entities if their actions can be imputed to the State. Therefore, actions by a competitor of the alleged aid beneficiary, or another third party which has been affected by the measure, alleging a violation of Article 108(3) TFEU, will be brought against the Member State, ie, the entity that adopted the alleged aid measure. These cases may also involve claims for damages. Beneficiary: As a matter of EU law—Article 108(3) TFEU does not impose any specific obligations on a recipient of aid.71 Therefore, if the action by a competitor, or other third party affected by an alleged aid measure is based solely on a violation of Article 108(3) TFEU, it cannot be brought against the beneficiary

previously taken the opposite position on the same question when hearing an action for unfair competition brought by the competitor of a cargo ferry service that had received State aid (case of the Court of Appeal of Cagliari Exol SpA v Nuova Cartiera di Arbatax SpA [21 July 1999] and case of the Court of First Instance of Genova Grandi traghetti di navigazione SpA v Viamare di navigazione Spa and Finmare SpA [26 April 1993], see the 2006 Enforcement Study (n 3) 326). 68

2006 Enforcement Study (n 3) 44. Streekgewest (n 41) para 18. 70 For a detailed discussion of the interaction between legal standing requirements and the principle of effectiveness, in the context of Dutch case law, see A Metselaar, ‘Who Can Invoke State Aid Law before National Judges? That Floating Question of Legal Interest in the Case Law of Dutch Courts’ (2014) 13 European State Aid Law Quarterly 250. Metselaar refers to two particular cases of interest. In one case, a Dutch court ruled that certain housing corporations had no standing to challenge subsidies granted to other housing corporations, because they did not qualify as ‘competitors’ and consequently, according to the court, lacked the requisite interest (see decision of the Administrative Jurisdiction Division of the Council of State Vogelaarwijken [6 February 2013] NL:RVS:2013:BZ0794). In the other case, a Dutch court ruled that a squatters’ organisation confronted with an eviction notice, after the property it was occupying was sold below market price following a tender procedure, could invoke Art 108(3) TFEU, since it participated in the tender procedure by offering to buy the property in question for the price of €1.25 and was therefore considered a competitor of the purchaser (see decision of the Appellate Court of The Hague Gemeente Leiden v Koppenhinksteeg [18 February January 2010] NL:GHSGR:2010:BL7630). 71 Joined Cases C-442/03 P and C-471/03 P P&O European Ferries (Vizcaya) v Commission, EU:C:2006:356, para 103; SFEI and Others (n 6) paras 73–74; Case T-308/00 Salzgitter v Commission, EU:T:2013:30, para 163. 69

370 Georg M Berrisch and Brian R Byrne of the alleged aid.72 However, that does not exclude the possibility that such actions are brought on the basis of provisions of national law, such as non-contractual liability.73 iii. Interveners In addition to the main parties (ie, plaintiff and defendant), national court procedures in a particular Member State might allow other interested parties to participate by way of an intervention. For instance, in a case where a competitor has brought an action against the Member State for a violation of Article 108(3) TFEU, it might be possible for: (i) the beneficiary to intervene in support of the Member State; and/or (ii) another competitor to intervene in support of the plaintiff competitor. For the beneficiary, in particular, such an intervention may have advantages and disadvantages, depending on the applicable national law provisions. A typical benefit is the ability to influence the outcome of the proceedings and to ensure the best possible defence, in particular in cases where the aid character of the measure is disputable. A possible downside is that, through the intervention, the ruling in the dispute between the competitor and the State will be binding on the beneficiary as well. For example, under German law in a case between a competitor of the alleged aid beneficiary and the Member State, if the court concludes that the measure constituted unlawful aid and orders the Member State to recover the aid from the beneficiary, the finding as regards the aid character of the measure would be binding on the beneficiary if it intervened in the court proceedings. Thus, if the Member State, in compliance with the court judgment, seeks to recover the aid, the beneficiary would be precluded from contesting the aid character of the measure. If, however, the beneficiary had not intervened in the case, the findings concerning the aid character of the measure would not be binding on it.74

B. Remedies As noted throughout this chapter, national courts are obliged under EU law to safeguard the rights of individuals against the possibility that Member State authorities will disregard Article 108(3) TFEU.75 However, because national courts 72 See, eg, Commercial Court of Brussels Ryanair v Brussels Airlines [29 April 2015] AR 2014/52.655, in which the Court rejected Ryanair’s application for a cease and desist order against Brussels Airlines stating that only Member States can violate the State aid provisions and, as a consequence, a cease and desist order which is addressed to undertakings cannot, in itself, be granted on the basis of those provisions. 73 See Enforcement Notice (n 2) para 55. See also: 2006 Enforcement Study (n 3) respectively 66 and 87, which refer to the following examples: Supreme Court of Austria Spa Gardens, judgments of [16 July 2002] and [4 May 2004]; and President of the Commercial Court of Brussels Breda Fucine Meridionali v Manoir Industries [13 February 1995] (1995) Journal des tribunaux—Droit européen 72 (For more details on the Breda case, see also 2006 Enforcement Study (n 3) 87.) In the Breda case, the Court decided that the recipient of unlawful aid commits an act of unfair competition and granted a cease and desist order requested by the claimant. 74 See ss 66–71 (‘Nebeninterventionʼ) of the German Code of Civil Procedure (ZPO). 75 PGE (n 7) para 31; Deutsche Lufthansa (n 1) para 28; Transalpine Ölleitung in Österreich (n 1) para 38; van Calster and Cleeren (n 6) para 75.

Private Enforcement of State Aid 371 enjoy procedural autonomy in discharging that obligation, the specific nature of the relief available to claimants, and/or the ability of claimants to avail of a particular type of relief, ‘depend[s…] on the legal remedies provided for under domestic law’ (emphasis added).76 Therefore, the potential remedies available to private claimants before national courts vary, depending on the particular Member State. Nonetheless, four prominent remedy categories are identifiable and merit further consideration: the suspension of ongoing aid; the recovery of aid that has already been granted; interim measures regarding suspension and recovery; and damages. Irrespective of the domestic law of the Member State and the particular remedies sought, all private claimants are confronted with the same task, namely to demonstrate to the national court that: (i) the disputed measure constitutes ‘aid’ within the meaning of Article 107(1) TFEU; and (ii) the Member State did not notify that measure to the Commission.77 i. Suspension of Ongoing Aid Measures The notion of an ‘ongoing aid measure’ can take many forms. It could be, for instance: a contract between a public authority and a private company, the terms of which allegedly constitute aid; or a selectively beneficial tax regime. In such cases, the first interest that a private claimant will have is to ensure that the disbursement of the aid is suspended. The claimant will therefore request that the national court orders the discontinuation of the measure. Depending on the Member State, the suspension might take the form of a ‘cease and desist’ order, as national legislation in several Member States allows claimants to bring specific actions seeking such orders on the basis of unfair competition law principles.78 If the national court accedes to a claimant’s request for suspension, the Member State must terminate the implementation of the measure until such time (if ever) that the Commission approves the measure following notification by the Member State.79 A ‘suspension’ of ongoing aid in this context is ordered in the final judgment of the national court. This remedy should not be confused with interim measures, discussed below, which might be adopted by the national court to suspend the disbursement of alleged aid, pending the outcome of the national court proceedings.

76 Transalpine Ölleitung in Österreich (n 1) para 46. See also: DEI and Commission (n 9) para 100; Klausner Holz (n 39) para 35; i-21 Germany (n 39) para 57; Köbler (n 39) para 46. 77 As noted earlier, national courts are competent only to determine whether ‘aid’ has been granted in violation of Art 108(3) TFEU, not to determine whether that aid is compatible with the common market. 78 See, eg, s 1 of the Austrian Act Against Unfair Competition (UWG), on the basis of which the Supreme Court granted a cease and desist order in the Spa Gardens case (Supreme Court Spa Gardens judgments of [16 July 2002] and [4 May 2004], see above n 73). Belgium also allows for a specific action under which the President of the Commercial Court can grant a cease and desist order (see s XVII of the Code of Economic Law). 79 See, eg, President of the District Court of Groningen Essent Kabelcom BV v Gemeente Appingedam [3 September 2004] NL:RBGRO:2004:AQ8920, which was a case brought by a competitor of the beneficiary of aid granted by the municipality of Appingedam. The municipality, which was funding a project for the provision of internet access, was ordered by the Court to cease all activities relating to the project until the European Commission declared that the measure did not constitute State aid or that the aid was compatible (see the 2006 Enforcement Study (n 3) 375).

372 Georg M Berrisch and Brian R Byrne ii. Recovery of Aid Already Granted If the claimant successfully establishes that the Member State violated Article 108(3) TFEU, the national court is, in principle, obliged to order the recovery of any aid already paid. According to the CJEU, the reason for this somewhat automatic obligation to recover the aid is that Member States would otherwise be encouraged to disregard the prohibition laid down in Article 108(3) TFEU.80 The recovery will be ordered at the conclusion of the national court proceedings.81 The national court might order that the beneficiary repay the aid directly to the granting authority or, alternatively, into an escrow account pending the Member State’s satisfaction of the notification and standstill obligation under Article 108(3) TFEU. In ‘exceptional circumstances’82 the national court can refrain from ordering the recovery of non-notified aid; however, the threshold is high.83 To determine whether such ‘exceptional circumstances’ exist, national courts should assess ‘whether a diligent businessman ought to have realised that the measures in question constituted aid which could be granted only in accordance with the procedure laid down in Article [108(3) TFEU]’.84 Factors weighing against a recovery order (ie, factors suggesting that even a diligent businessman would not have realised the measure was unlawful aid) might include: (i) difficulty in identifying/verifying that a measure is aid; (ii) indications from the Commission that the measure is not problematic (eg, declining to open a formal investigation); and (iii) excessive delays in the Commission reaching a decision about a particular measure.85 A precise assurance from the Commission regarding the lawfulness of the disputed measure would also likely amount to an ‘exceptional circumstance’ justifying non-recovery of the aid.86

80 OTP Bank (n 1) para 76; CELF I (n 21) para 40; Xunta de Galicia (n 34) para 63; SFEI and Others (n 6) para 69; FNCE and Others (n 1) para 16. 81 Therefore, this remedy should not be confused with ‘interim recovery orders’ discussed below (see section III.B.iii below). 82 OTP Bank (n 1) para 72; Residex Capital IV (n 9) para 35; CELF II (n 23) para 36; CELF I (n 21) para 42; SFEI and Others (n 6) para 70; Case C-5/89 Commission v Germany, EU:C:1990:320, para 16. See also: Procedural Regulation (n 13) Art 16(3). 83 OTP Bank (n 1) para 77; Unicredito Italiano (n 4) para 104; Demesa and Territorio Histórico de Álava (n 4) para 45; Germany v Commission (n 50) para 42; Case C-24/95 Alcan Deutschland, EU:C:1997:163, para 25; Commission v Germany (n 82) para 14. See also: Enforcement Notice (n 2) para 32. 84 See Advocate General Opinion in Case C-39/94 SFEI and Others, EU:C:1995:445, para 75. See also: Unicredito Italiano (n 4) para 104; Demesa and Territorio Histórico de Álava (n 4) para 44; Germany v Commission (n 50) para 41; Alcan Deutschland (n 83) para 25; and Commission v Germany (n 82) para 14. For example, the 2006 Enforcement Study (n 3) 102, refers to the Belgian Idealspun case in which the Commercial Court rejected Idealspun’s claim that it had legitimate expectations regarding its entitlement to retain the aid since, according to the Court, a diligent businessman would have known that the Belgian State had not complied with the standstill obligation under Art 108(3) TFEU. (Commercial Court of Kortrijk, Gimvindus and Flemish Region v Idealspun, De Clerck and Others [20 September 1994] Case No 1310/90.) 85 See AG Opinion in Case C-39/94 SFEI and Others (n 84) para 76. 86 Enforcement Notice (n 2) para 33 with reference to Belgium and Forum 187 (n 4) para 147. See also: Case C-369/09 P ISD Polska and Others v Comission, EU:C:2011:175, para 123; Case C-82/98 P Kögler v Court of Justice, EU:C:2000:282, para 33; and Case C-111/86 Delauche v Commission, EU:C:1987:562, para 24.

Private Enforcement of State Aid 373 iii. Interim Measures Considering that national court proceedings might be lengthy, private claimants must have the ability to seek interim measures from the national court to safeguard their interests pending the national court’s final ruling.87 For instance, the national court’s final ruling might be deferred to allow for: (i) a response from the CJEU following a request for a preliminary ruling; (ii) a Commission opinion; or (iii) further information from the Commission. The Commission considers that national courts are ‘very well placed’ to adopt interim measures due to their ability to act quickly and their proximity to the facts.88 Indeed, the nature and suitability of interim measures will depend on the facts of a particular case. In its Enforcement Notice, the Commission envisages, in broad terms, two different scenarios that the national courts might encounter. In this first scenario, described by the Commission as ‘straightforward’, ‘unlawful aid has not yet been disbursed, but … there is a risk that such payments will be made during the course of national court proceedings’.89 In that case, the logical interim measure would be an order preventing the disbursement pending the outcome of the national court proceedings.90 In the second scenario, which is more complex, the alleged aid measure has already been disbursed. According to the Commission, the national court should consider in such cases whether an ‘interim recovery’ order is required to ‘at least terminate the anticompetitive effects of the aid on a provisional basis’ pending the national court’s final judgment.91 The Enforcement Notice provides the following guidance for national courts, as regards the appropriate test and procedure for interim recovery orders: Where, based on the case law of the [EU] courts and the practice of the Commission, the national judge has reached a reasonable prima facie conviction that the measure at stake involves unlawful State aid, the most expedient remedy will, in the Commission’s view and subject to national procedural law, be to order the unlawful aid and the illegality interest to be put on a blocked account until the substance of the matter is resolved. In its final judgment, the national court would then either order the funds on the blocked account to be returned to the State aid granting authority, if the unlawfulness is confirmed, or order the funds to be released to the beneficiary.92

The Commission also considers that interim recovery orders are effective in cases involving concurrent proceedings before the national court and the Commission,93 a topic discussed further in section IV below. iv. Actions for Damages With an action seeking suspension of an alleged aid measure and an order obliging the Member State to recover sums already paid, the claimant can achieve a result

87 88 89 90 91 92 93

SFEI and Others (n 6) para 52. Enforcement Notice (n 2) para 57. ibid, para 58. ibid. ibid, para 60. ibid, para 61. ibid, para 62.

374 Georg M Berrisch and Brian R Byrne whereby its competitor, the beneficiary of the aid, no longer enjoys the benefit of the aid. An action for damages offers the plaintiff the prospect of financial compensation beyond the removal of the advantage that the beneficiary of the aid enjoyed. Therefore, the ability to obtain damages incentivises private enforcement, arguably contributing to a more robust State aid regime overall. In the antitrust sphere, for example, the growing prevalence of private damages claims against cartel participants has undoubtedly increased deterrence. Therefore, by analogy, the possibility that the Member State and/or the beneficiary might be liable to pay damages following a violation of Article 108(3) TFEU should encourage Member States to notify, and/or beneficiaries to insist on the notification of, potential aid measures to the Commission. In practice, however, private claimants face significant challenges when seeking to obtain damages following a violation of State aid law. Below, we consider: (i) the legal framework for bringing an action for damages against the Member State; (ii) the extent to which a claimant might be able to seek damages from the beneficiary; and (iii) the specific practical difficulties that private claimants must overcome. a. Member State Liability It is a well-established principle of EU law that Member States will, in certain circumstances, be obliged to compensate private parties for loss and damage caused as a result of violating EU law.94 According to the CJEU, this principle is ‘inherent in the system of the Treaty’.95 In the Brasserie du pêcheur and Factortame judgment, the CJEU laid down the following three-part test to determine whether a private claimant has a right to damages from a Member State following an EU law infringement:96 (i) the rule of law infringed must be intended to confer rights on individuals; (ii) the breach must be sufficiently serious; and (iii) there must be a direct causal link between the breach and the damages sustained by the claimant. Applying this test in the State aid context, the first condition is relatively easy to satisfy, as there is no question that Article 108(3) TFEU confers rights on individuals (as discussed extensively in this chapter). As regards the second condition, relevant factors to determine whether the breach is sufficiently serious include, among others: the clarity and precision of the rule of EU law; the measure of discretion left to the Member State; and whether the infringement was intentional/voluntary.97 Accordingly, the second condition is likely to be satisfied in most cases because Article 108(3) TFEU imposes a clear obligation that affords no discretion to the Member State. Thus, it is the third condition, namely the requirement to establish damages and a causal link between the damages and the violation of Article 108(3) 94 Brasserie du pêcheur and Factortame (n 44) para 20; Joined Cases C-6/90 and C-9/90 Francovich and Bonifaci v Italy, EU:C:1991:428, para 33. See also: Case C-173/03 Traghetti del Mediterraneo v Italy, EU:C:2006:391, para 41; EISA (n 44) para 109; Case C-268/15 Ullens de Schooten, EU:C:2016:874, para 41; Case C-441/14 DI, EU:C:2016:278, para 42; Joined Cases C-501/12 to C-506/12, C-540/12 and C-541/12 Specht and Others, EU:C:2014:2005, para 98; Joined Cases C-178/94, C-179/94, C-188/94, C-189/94 and C-190/94 Dillenkofer and Others v Germany, EU:C:1996:375, para 20. 95 Brasserie du pêcheur and Factortame (n 44) para 31; Francovich and Bonifaci v Italy (n 94) para 35. 96 Brasserie du pêcheur and Factortame (n 44) para 51. See also: Enforcement Notice (n 2) para 45. 97 Brasserie du pêcheur and Factortame (n 44) para 56.

Private Enforcement of State Aid 375 TFEU, that is likely to prove the most problematic from the claimant’s perspective, as discussed further below. In addition to the possibility of bringing an action for damages against the Member State, on the basis of EU law (ie, the three-part test set out above), private claimants might also have the possibility to sue the Member State for damages on the basis of national law. This prospect has been recognised consistently by the CJEU in relation to violations of Article 108(3) TFEU.98 However, whether or not damages are actually available in a particular case depends on the law of the Member State in question, which will also dictate the legal basis for the action.99 The only example, of which the authors are aware, of a successful damages action under national law against a grantor of aid is the Corsica Ferries case.100 b. Liability of the Beneficiary The CJEU has held that the notification and standstill obligation pursuant to Article 108(3) TFEU applies to Member States only; it does not impose a specific obligation on the recipient of aid.101 Therefore, there is no basis under EU State aid law for the beneficiary to incur liability, vis-a-vis a third party, for failing to verify whether an aid measure it received had been duly notified to the Commission.102 Nevertheless, the CJEU has left the door open for a beneficiary to be sued for damages under provisions of national law concerning non-contractual liability. The CJEU articulated this prospect in the following terms: If, according to national law, the acceptance by an economic operator of unlawful assistance of a nature such as to occasion damage to other economic operators may in certain circumstances cause him to incur liability, the principle of non-discrimination may lead the national court to find the recipient of aid paid in breach of Article [108(3) TFEU] liable.103

It remains to be seen the extent to which this, somewhat theoretical, possibility of holding the beneficiary liable for damages can actually be exploited by private claimants in practice. The authors are not aware of any cases where a beneficiary has been ordered by a national court to pay damages to a third party because it received aid granted in violation of Article 108(3) TFEU. c. Practical Difficulties Associated with Damages Claims Private claimants seeking damages following a violation of Article 108(3) TFEU face some significant practical challenges. These challenges were highlighted in the 2009 Update of the 2006 Study on the enforcement of State aid rules at national level,104 98 CELF I (n 21) paras 53 and 55; Case C-334/07 P Commission v Freistaat Sachsen, EU:C:2008:709, para 54; Case C-384/07 Wienstrom, EU:C:2008:747, para 29; Transalpine Ölleitung in Österreich (n 1) para 56; SFEI and Others (n 6) para 75. 99 This leads to significant variations across the EU. See Enforcement Notice (n 2) para 44. 100 In that case, a French court ordered the State entity that had implemented unlawful aid to a ferry company to pay over €84 million in damages to a competitor of the aid beneficiary. Administrative Tribunal of Bastia (TA Bastia), Corsica Ferries [23 February 2017], Case No 1500375. 101 Vizcaya (n 71) para 103; SFEI and Others (n 6) paras 73–74; Salzgitter (n 71) para 163. 102 SFEI and Others (n 6) para 74. 103 ibid, para 75. 104 2009 Update (n 3).

376 Georg M Berrisch and Brian R Byrne which noted that ‘national courts remain reluctant to award monetary damages to competitors of the beneficiary’.105 First, according to the 2009 Update, the ‘main obstacle’ facing private claimants is the ‘lack of a clear legal basis under national law’106 and the diverging approaches that Member States take when dealing with damages claims. A second obstacle facing claimants is the requirement to establish a causal link between the violation of Article 108(3) TFEU and economic harm suffered by the claimant. On this point, the 2009 Update refers to the need for the claimant ‘to show how its market share would have developed had the aid not been granted to its competitor’.107 In its Enforcement Notice, the Commission also focuses on the requirement to establish a causal link between the unlawful grant of aid and economic harm. It refers to the ‘easier’ case of unlawful aid enabling the beneficiary to win a particular contract or business opportunity at the claimant’s expense; in that instance, the national court should be able to calculate the revenue that the claimant was likely to generate under the particular contract.108 More difficult questions arise, however, when the claimant asserts a loss of market share as a result of the unlawful aid. The guidance offered by the Commission is that the national court should ‘compare the claimant’s actual income situation (based on the profit and loss account) with the hypothetical income situation had the unlawful aid not been granted’.109 Unfortunately, however, this guidance is of limited practical value because it offers no insight into how one would calculate the ‘hypothetical income situation’ to which the Commission refers.

IV. OBLIGATIONS OF NATIONAL COURTS IN CASES OF CONCURRENT PROCEEDINGS BEFORE THE COMMISSION OR THE EU COURTS

As noted earlier, the Commission and the national courts have overlapping responsibilities insofar as both are competent to assess the existence of ‘aid’. However, their respective competences also differ in two key respects: only the Commission is competent to assess the compatibility of aid; and only the national courts are competent to order the recovery of aid on the sole ground that the aid was granted in violation of the notification and standstill obligation. The combination of overlapping and distinct responsibilities/competences of the national courts and the Commission raises important and difficult questions in cases of concurrent proceedings, ie, cases where both the national court and the Commission are considering the same alleged aid measures at the same time. This happens, for example, where a competitor of the beneficiary brings a case before the national court and lodges a complaint with the Commission, prompting the Commission to investigate the measure at issue. Similar issues also arise where the subject matter of a national court case is also at issue in a case pending before the EU Courts. 105 106 107 108 109

ibid, 2. ibid, 4. ibid, 4. Enforcement Notice (n 2) para 49(b). ibid, para 49(c).

Private Enforcement of State Aid 377 A. Concurrent Proceedings before the Commission In cases where an alleged aid measure is the subject of both a case pending before a national court and a Commission State aid investigation, the scope of the national courts’ obligation to safeguard the directly effective rights afforded by Article 108(3) TFEU will vary, depending on the particular stage of the investigation.110 Accordingly, we consider below three important milestones in Commission State investigations and how they impact the obligations imposed on the national courts, specifically: (i) the Commission’s initiation of an informal investigation; (ii) the Commission’s opening of a formal investigation; and (iii) the Commission’s adoption of a decision at the conclusion of its investigation. Importantly, as noted above, neither the initiation of the ‘informal’ nor ‘formal’ investigation phases release the national court from its duty to safeguard the individual rights associated with Article 108(3) TFEU.111 According to the CJEU, national courts have a duty to ensure that unlawful aid ‘does not remain at the free disposal of the recipient during the period remaining until the Commission makes its decision’.112 Therefore, national courts are not permitted to stay proceedings pending the outcome of the Commission’s investigation because, according to the CJEU, a ‘decision to stay proceedings would … amount to maintaining the benefit of aid during the period in which implementation is prohibited, which would be inconsistent with the very purpose of Article [108(3) TFEU] and would render that provision ineffective’.113 i. Informal Investigation Started During the informal investigation stage, there is no specific obligation on the national court that would alter its duty to uphold the individual rights granted by Article 108(3) TFEU. In other words, the national court must, as discussed earlier in this chapter, carry out its own assessment of whether the measure can be classified as ‘aid’ within the meaning of Article 107(1) TFEU and, if so, it must draw the ‘appropriate conclusions’.114 ii. Commission has Opened the Formal Investigation Initially, the Commission took the position that decisions opening a formal investigation under Article 108(2) TFEU (so-called ‘opening decisions’) are mere procedural decisions and that, therefore, applications by the Member State concerned or beneficiaries of the measures at issue seeking the annulment of opening decisions are inadmissible. However, the CJEU ruled that the opening decision in relation to an alleged aid measure already implemented and classified by the Commission as

110 111 112 113 114

Deutsche Lufthansa (n 1) para 33; SFEI and Others (n 6) para 53. Deutsche Lufthansa (n 1) para 32; CELF II (n 23) para 27; SFEI and Others (n 6) para 44. Deutsche Lufthansa (n 1) para 31; CELF II (n 23) para 30. See also: Klausner Holz (n 39) para 25. CELF II (n 23) para 31. Deutsche Lufthansa (n 1) paras 34–35.

378 Georg M Berrisch and Brian R Byrne ‘new aid’ entails independent legal effects, because it produces an immediate, certain and sufficiently binding effect on the Member States to take all necessary measures to safeguard the rights of parties affected by an unlawful implementation of an aid measure. This necessarily alters the legal implications of the measure under consideration and the legal position of the beneficiaries, particularly as regards the suspension of the measure in question.115 The CJEU added that an opening decision might be invoked before a national court called upon to draw all the consequences arising from the infringement of the standstill obligation (eg, suspend further payments to beneficiaries and order the recovery of payments already made).116 Should a national court grant such a remedy, it would, at a minimum, result in a delay of further payment. A delay in payment of alleged aid might have irreversible financial consequences on beneficiaries, who count on receiving the payments. According to the Court, these consequences could not be eradicated by a subsequent decision finding the aid compatible with the common market.117 The CJEU noted that, even without suspension of the measure and recovery of payments, the doubt of the lawfulness of the measure may lead beneficiaries to refuse new payments or to hold necessary sums as provision in case repayment of the alleged aid is ordered. Thus, in their relations with those beneficiaries other businesses may take account of the uncertainty of the beneficiaries’ legal and financial situation.118 The CJEU also noted that this legal effect of opening decisions arises irrespective of the wording which the Commission uses in the opening decision. A decision of the Commission to open a formal investigation under Article 108(2) TFEU implies that in the Commission’s view, even if this is only a preliminary assessment, the measure constitutes aid and was granted in violation of Article 108(3) TFEU.119 The numerous judgments by the CJEU and, subsequently, the General Court declaring actions against opening decisions admissible eventually prompted the Commission to change its position and to adopt the view that if the Commission has opened a formal investigation into measures that are the subject of parallel proceedings before a national court, the national court may no longer decide on the aid character of the measures but instead must assume that the measures constitute aid and, as the case may be, order the suspension of the measures and/or the recovery of moneys already paid. The Commission expressed this view in a response pursuant to point 3.2 of the Enforcement Notice120 to questions from two German courts, the Higher Regional Court Koblenz (OLG Koblenz) and the Schleswig-Holstein Higher Regional Court (OLG Schleswig), which had to decide on cases brought by

115

Case C-400/99 Italy v Commission, EU:C:2005:275, para 59. ibid. 117 Case C-312/90 Spain v Commission, EU:C:1992:282, para 22; Case C- 47/91 Italy v Commission EU:C:1994:358, para 28. 118 Case C-400/99 Italy v Commission (n 115) para 59; Case T-517/12 Alro v Commission, EU:T:2014:890, para 63. 119 Case C-400/99 Italy v Commission (n 115) paras 54–58. 120 See Enforcement Notice (n 2) paras 89–96. 116

Private Enforcement of State Aid 379 Lufthansa and Air Berlin regarding alleged aid to Ryanair at Frankfurt-Hahn airport and Lübeck airport, respectively. Both courts then submitted a request for a preliminary ruling to the CJEU.121 The Commission’s new position raised many concerns. First, the finding of a violation of the standstill obligation of Article 108(3) TFEU requires that the measure at issue constitutes aid as defined in Article 107(1) TFEU, because Member States are only obliged to notify aid and not measures that might constitute aid. However, in cases where the aid character of the measure is disputed, an opening decision often does not contain a definitive finding as to the aid character of the measure. In fact, in many cases it leaves certain aspects of the aid assessment open, stating that they require further investigation.122 Second, there are no effective remedies against an opening decision. According to the General Court, an applicant seeking annulment of an opening decision must show that the Commission committed a manifest error when finding that it needed to investigate in further detail whether the measure at issue constitutes aid.123 In practical terms, this is virtually impossible.124 Third, it ran counter to Article 13(2) of the Procedural Regulation,125 which provides that the Commission can, after having opened the formal investigation but before adopting a final decision, require the Member State to recover money already paid, provided the Commission concludes, and shows, that the measure clearly constitutes aid. Such a decision could also be subject to an appeal before the General Court. The CJEU dealt with the Lufthansa case first. It essentially sided with the Commission, but its answer to the German court created some uncertainty. It was worded as follows: Where, in accordance with Article 108(3) TFEU, the European Commission has initiated the formal examination procedure under Article 108(2) TFEU with regard to a measure which has not been notified and is being implemented, a national court hearing an application for the cessation of the implementation of that measure and the recovery of payments already made is required to adopt all the necessary measures with a view to drawing the appropriate conclusions from an infringement of the obligation to suspend the implementation of that measure. To that end, the national court may decide to suspend the implementation of the measure in question and order the recovery of payments already made. It may also decide to order provisional measures in order to safeguard both the interests of the parties concerned and the effectiveness of the European Commission’s decision to initiate the formal examination procedure.

121

Case C-27/13 Flughafen Lübeck, EU:C:2014:240 and Deutsche Lufthansa (n 1). This was also the case with respect to the opening decisions at stake in the two cases before the German courts, where the Commission explicitly stated that it needed to further investigate whether the measure complied with the Market Economy Operator (MEO) principle. 123 Case T-461/12 Hansestadt Lübeck v Commission, EU:T:2014:758, para 42; Joined Cases T-269/99, T-271/99 and T-272/99 Diputación Foral de Guipúzcoa and Others v Commission, EU:T:2002:258, para 49. 124 Case C-524/14 P Commission v Hansestadt Lübeck, EU:C:2016:971, brought by the city of Lübeck against a subsequent opening decision, constitutes a rare example of a successful challenge. However, that case succeeded on the basis of a well-defined legal argument and not a challenge of the Commission’s discretion, eg, in the application of the MEO principle. 125 Procedural Regulation (n 13). 122

380 Georg M Berrisch and Brian R Byrne Where the national court entertains doubts as to whether the measure at issue constitutes State aid within the meaning of Article 107(1) TFEU or as to the validity or interpretation of the decision to initiate the formal examination procedure, it may seek clarification from the European Commission and, in accordance with the second and third paragraphs of Article 267 TFEU, it may or must refer a question to the Court of Justice of the European Union for a preliminary ruling.126

First, the CJEU did not directly answer the question of the referring court as to whether a national court is bound by the Commission’s legal view in an opening decision on the aid character of the measure.127 While the first point of the CJEU’s response may indeed suggest that the CJEU considers that the characterisation of a measure as aid in an opening decision binds the national court, and that the national court therefore does not, and cannot, rule on the aid character, the third point of the response speaks against such an interpretation. There, the CJEU explicitly states that the national court can seek a preliminary ruling from the CJEU if it has doubts as to the aid character of the measure. This strongly suggests that the national court can, and must, rule on the aid character of the measure, because according to established case law a national court may only seek a preliminary ruling on questions that are relevant for its ruling.128 Second, the use of the word ‘may’ in the second point of the response suggests that in any event the national court has discretion as to whether and what measures it orders. Further uncertainty was caused by the CJEU’s reasoning in the judgment. For example, paragraph 40 of the judgment states: On the other hand, even if in its final decision the Commission were to conclude that there were no aid elements, the preventive aim of the State aid control system established by the TFEU … requires that, following the doubt raised in the decision to initiate the formal examination procedure as to the aid character of that measure and its compatibility with the internal market, its implementation should be deferred until that doubt is resolved by the Commission’s final decision (emphasis added).129

Given that doubts cannot be created retroactively, this statement suggests that doubts created by an opening decision cannot be relied upon to order the recovery of money paid prior to the opening decision. This would imply that in order to order the recovery of such moneys, the national court would have to find that the measures constitute aid. This was an important issue in the two German cases because almost all measures at issue had been implemented prior to the adoption of the opening decision. After issuing its judgment in the Lufthansa case, the CJEU asked the OLG Schleswig whether it wished to maintain its request for a preliminary ruling. Referring to some ambiguities in the Lufthansa ruling, the OLG Schleswig answered affirmatively. The CJEU, by an order of the Grand Chamber, confirmed its Lufthansa ruling,

126

Deutsche Lufthansa (n 1) operative part. ibid, para 18. 128 Case C-152/03 Ritter-Coulais, EU:C:2006:123, para 15; Case C-314/01 Siemens and ARGE Telekom [2004] ECR I-2549, para 35; Case C-167/01 Inspire Art [2003] ECR I-10155, para 45; Case C-390/99 Canal Satélite Digital [2002] ECR I-607, para 19. 129 Deutsche Lufthansa (n 1) para 40. 127

Private Enforcement of State Aid 381 although its order in the Lübeck case only contained the first two points of the Lufthansa judgment.130 These two rulings caused considerable unrest among beneficiaries of alleged aid measures with respect to which the Commission had opened a formal investigation, and resulted in an unprecedented number of appeals against opening decisions.131 Despite the doubts as to the precise meaning of the Lufthansa judgment and the order in the Lübeck case, and despite entertaining considerable doubts as to the merits of these rulings, the OLG Schleswig eventually held that it had to accept the judgment and ruled that it was bound by the characterisation of the measures as aid in the opening decision.132 Its judgment was appealed to the German Supreme Court (BGH). In its ruling of 9 February 2017, the BGH set aside the judgment of the OLG Schleswig. It ruled that, while national courts must take into account the findings on the aid character of the measure in the opening decision, they are not obliged to blindly follow them. If they have doubts about the aid character of the measure, for example, because information was not considered in the opening decision, they can ask the Commission for clarification and, if the Commission’s answer is unsatisfactory, request a preliminary ruling from the CJEU. Moreover, even if, following that, they must consider that the measure constitutes aid, they are not always obliged to order recovery. Instead, they must take into account the interests of all parties involved, as well as other circumstances, including the delay since the opening decision, and, in particular, pay due regard to the principle of proportionality.133 This ruling has given more flexibility to the national courts. It allows for an effective enforcement of the State aid regime while at the same time safeguarding the rights of all parties involved. The case has been referred back to the Regional Court Kiel (Landgericht Kiel), which is the court of first instance in Germany in these proceedings. However, it is unlikely that the Landgericht Kiel will apply the criteria set by the BGH because,

130

Flughafen Lübeck (n 121). eg, approximately 60 cases were brought against the opening decision concerning the German Renewable Energy Act. See, eg, Case T-301/14 Michelin Reifenwerke v Commission, EU:T:2014:747; Case T-300/14 Fricopan Back v Commission, EU:T:2015:336; Case T-298/14 Erbslöh Aluminium v Commission, EU:T:2015:349; and Case T-172/14 Stahlwerke Bous v Commission, EU:T:2015:402, challenging Commission Decision C (2013) 4424 final of 18 December 2013 on the aid scheme SA.33995 (2013/C) (ex 2013/NN) (implemented by Germany for the support of renewable electricity and reduced EEG-surcharge for energy-intensive users) ([2014] OJ C37/73 and [2014] OJ C250/15). The overall number of new State aid cases brought before the General Court increased from 54 cases in 2013, to 148 cases in 2014. In 2015, the number of new State aid cases dropped down to 73, which is an upward trend compared with the period from 2010 to 2013. See the Annual Report of the Court of Justice 2015—Judicial Activity (Luxemburg, Publications Office of the European Union, 2016) 167–170. Also available at: curia.europa.eu/jcms/upload/docs/application/pdf/2016-08/rapport_annuel_2015_activite_ judiciaire_en_web.pdf. 132 The procedure was a so-called ‘action by stages’ (Stufenklage) under German law and the OLG Schleswig therefore only ordered the airport to provide information allowing to determine the alleged benefit that Ryanair received under the contested measures. The OLG Koblenz never ruled on the Lufthansa case because, on 1 October 2014 the Commission decided that the measures at issue in this case did not constitute aid. See Frankfurt-Hahn Airport (Case SA.21121) Commission Decision (EU) 2016/789 [2014] OJ L134/46 implemented by Germany concerning the financing of Frankfurt Hahn airport and the financial relations between the airport and Ryanair. 133 German Supreme Court (BGH) Flughafen Lübeck [9 February 2017] I ZR 91/15. 131

382 Georg M Berrisch and Brian R Byrne on 7 February 2017, the Commission finally issued its final decision in the Lübeck case finding, among others, that the various measures that were the subject of the litigation in Germany did not constitute aid.134 It is also interesting to note that in two recent cases, brought after the CJEU’s Lufthansa judgment, where beneficiaries of alleged aid measures challenged an opening decision, the General Court, following the Commission’s submissions, dismissed the actions as inadmissible.135 It held that the applicants did not have a legal interest in the annulment of the opening decisions, because the measures had already been implemented. This seems to support the view that an opening decision cannot have the effect of obliging a national court to order the recovery of moneys already paid, because otherwise the beneficiary of the aid would have an interest in seeking the annulment of the opening decision. It is also interesting that the Commission’s position in these cases contradicted its position in the Lufthansa and the Lübeck case, where the Commission argued that a national court, when deciding on a request for a recovery order, must allow the order if the Commission had adopted an opening decision in respect to the measures. As a practical matter, one can note that the position of plaintiffs in private litigation is significantly strengthened once the Commission has opened a formal investigation, although the precise effect depends on the specific findings on the aid character of the measure in the opening decision. However, at least in Germany, national courts will not necessarily find in favour of the plaintiff if the Commission has adopted an opening decision. It also remains to be seen how the Commission will react to the latest judgment of the BGH and whether it will seek an opportunity to bring the matter again before the CJEU. iii. Commission has Adopted a Final Decision In cases of concurrent proceedings, it is possible that the Commission will adopt its final decision while the case is still pending before the national court. The consequences of the Commission’s decision, for the national court proceedings, will vary depending on the nature of the decision itself. There are three distinct possibilities, each of which is considered below: (i) a decision finding that the measure does not constitute aid;136 (ii) a decision finding that the measure constitutes aid compatible with the common market;137 or (iii) a decision finding that the measure constitutes aid that is incompatible with the common market.138 a. No Aid Once the Commission makes a determination as to the aid character of a particular measure, that determination is binding on national courts. Therefore, in the event

134 Commission, ‘Commission clears several public measures in favour of Lübeck airport and airlines’ (Press Release of 7 February 2017) MEX/17/231. 135 Case T-129/13 Alpic v Commission, EU:T:2014:895; Alro (n 118). 136 Procedural Regulation (n 13) Art 9(2). 137 ibid, Arts 9(3) and 9(4). 138 ibid, Art 9(5).

Private Enforcement of State Aid 383 that the Commission concludes that the disputed measure does not constitute ‘aid’ within the meaning of Article 107(1) TFEU, the national court is precluded from making a finding to the contrary, for example, the national court cannot issue a judgment declaring that the Member State violated Article 108(3) TFEU for failing to notify the measure.139 b. Compatible Aid Following a Commission decision finding that the disputed measure constitutes compatible aid, the national court is not obliged, under EU law, to order full recovery of the aid.140 It is obliged, however, to order the beneficiary to pay interest with respect to the period of unlawfulness.141 The rationale for this obligation is that the implementation of the aid, without observing the notification and standstill obligation, violates Article 108(3) TFEU.142 Thus, the aid is ‘unlawful’ and, during the period of unlawfulness, the beneficiary has an undue advantage in two forms: first, it avoids the payment of interest that would otherwise have been due if it borrowed the money from the market; second, the competitive position of the beneficiary improves vis-a-vis other market operators.143 Although EU law does not impose an obligation on the national court to order full recovery of the aid, the national court may nonetheless make such an order within the framework of its domestic law, without prejudice to the Member State’s right to re-implement it subsequently.144 It may also be required under national law to uphold claims for damages.145 c. Incompatible Aid In the event that the Commission finds that the disputed measure constitutes incompatible aid, the decision adopted by the Commission will order recovery of the aid. That decision is binding on all organs of the Member State, including the national courts,146 which have an obligation not to apply any rule of national law that would prevent the immediate and effective execution of the Commission’s decision. Therefore, in the context of private enforcement and concurrent proceedings, the case pending before the national court typically becomes devoid of purpose on the basis

139

DEI and Commission (n 9) para 105; Deutsche Lufthansa (n 1) paras 36 et seq. Wienstrom (n 98) para 28; CELF I (n 21) para 46. 141 Wienstrom (n 98) para 29; CELF I (n 21) paras 52 and 55; Alro (n 118) para 41; Case T-674/11 TV2/Danmark v Commission, EU:T:2015:684, paras 83–84 (this case is currently under appeal before the CJEU); Procedural Regulation (n 13) Art 16(2); Enforcement Notice (n 2) para 40(b). 142 OTP Bank (n 1) para 76; CELF I (n 21) para 40; Xunta de Galicia (n 34) para 31; SFEI and Others (n 6) paras 67 and 69; FNCE and Others (n 1) para 16. 143 CELF I (n 21) para 51; Joined Cases T-115/09 and T-116/09 Electrolux and Whirlpool v Commission, EU:T:2012:76, para 67. 144 Wienstrom (n 98) para 29; CELF I (n 21) para 53; Transalpine Ölleitung in Österreich (n 1) para 56. See also: Procedural Regulation (n 13) Art 16(3). 145 CELF I (n 21) para 53; Transalpine Ölleitung in Österreich (n 1) para 56; SFEI and Others (n 6) para 75. 146 Case C-69/13 Mediaset, EU:C:2014:71, para 23; Köbler (n 39) para 32; Case C-249/85 Albako v BALM, EU:C:1987:245, para 17. See also: Recovery Notice (n 11) paras 45 and 72. 140

384 Georg M Berrisch and Brian R Byrne that the Commission has stipulated the remedy and the Member State, including its national courts, are bound to ensure that the decision is implemented.147

B. Pending Appeals against Commission Decisions Even if a beneficiary challenges a Commission State aid recovery decision before the EU courts, this does not postpone the Member State’s obligation to recover the aid.148 Therefore, the beneficiary may find itself fighting on two fronts simultaneously: one before the EU courts; and the other before the national authorities in the Member State tasked with securing the recovery. As noted earlier, the forum for the latter will often be the national courts. As a result, national courts might be asked to order recovery from a beneficiary that is concurrently challenging the legal basis for that recovery at the EU courts. In such cases, the Commission ‘may also accept, … a provisional implementation of the decision, … eg, the payment of the full amount of unlawful and incompatible aid into a blocked account pending the outcome of the EU Court proceedings’.149 The Member State is required, however, to submit to the Commission, for its approval, ‘a justification for the adoption of such provisional measures and a full description of the provisional measure envisaged’.150

147 Scott (n 41) paras 59 and 60; Joined Cases C-346/03 and C-529/03 Atzeni and Others [2006] ECR I-1875, para 31; Alcan Deutschland (n 83) para 34; Case C-188/92 TWD Textilwerke Deggendorf GmbH v Germany [1994] ECR I-833, paras 17–18; Case C-465/93 Atlanta Fruchthandelsgesellschaft and Others (I) v Bundesamt für Ernährung und Forstwirtschaft [1995] ECR I-3761, para 51; Joined Cases C-143/88 and C-92/89 Zuckerfabrik Süderdithmarschen and Zuckerfabrik Soest v Hauptzollamt Itzehoe and Hauptzollamt Paderborn, EU:C:1991:65, para 33. See also: Procedural Regulation (n 13) Art 16(3); Recovery Notice (n 11) paras 44 et seq; Enforcement Notice (n 2) paras 63–69. 148 According to Art 278 TFEU, actions for annulment brought under Art 263 TFEU before the EU courts do not have suspensory effect and it is only in exceptional circumstances that the EU courts may order suspension of the contested act. See also: Case C-63/14 Commission v France, EU:C:2015:458, para 47; Case C-213/85 Commission v Netherlands, EU:C:1988:39, para 21; Case T-812/14 R BPC LUX 2 and Others v Commission, EU:T:2015:119, para 17; Case T-468/08 R AES-Tisza kft v Commission, EU:T:2008:621, para 13; and Recovery Notice (n 11) para 25. 149 Recovery Notice (n 11) para 70. 150 ibid.

14 Germany FALK SCHÖNING AND CLEMENS ZIEGLER

I. INTRODUCTION

S

UPPORT FOR RENEWABLE energy projects (RES support) and the German Energiewende, including the Atomausstieg,1 as well as the long-term exit from coal mining and coal-fired power plants have been the most debated issues over the past few decades in the energy sector in Germany. Most of these issues have been the subject of EU Commission State aid decisions, with some also subject to judgments by the EU courts, including in the form of preliminary rulings by the CJEU. There are only a few relevant national court decisions in Germany linked to these preliminary rulings that are relevant from a State aid perspective and hardly any as regards the Atomausstieg and coal. That said, there are some other interesting German court decisions, in particular regarding RES support. In addition, EU State aid law has also been discussed in a number of German court decisions on grid-related matters. Those will be dealt with in a separate section of this chapter. The chapter will also provide a short overview of available German State aid measures in the energy area. Furthermore, the possibility of complaints by competitors against illegal State aid measures has been discussed in the Federal Supreme Court’s (BGH) 2011 judgment on the Frankfurt-Hahn airport,2 as well as, most recently, in the BGH’s 2017 judgment regarding the Lübeck airport.3 This subject, which is also relevant for the energy sector, has already been discussed in chapter thirteen.

1 The German decision for a long-term exit from the use of nuclear energy for electricity production (Atomausstieg) was first taken in 2000 under Chancellor Schröder and his social democrats–green coalition government and, after extensions granted to the nuclear power producers by Chancellor Merkel’s liberal–conservative government in 2010, ultimately reactivated by Chancellor Merkel shortly after the 2011 nuclear disaster in Fukushima, Japan. 2 See BGH, judgment of 10 February 2011—I ZR 136/09, EuZW 2011, 440. 3 See BGH, judgment of 9 February 2017—I ZR 91/15 Flughafen Lübeck (full text not yet available); Press Release of 9 February 2017, available at: www.juris.bundesgerichtshof.de/cgi-bin/rechtsprechung/ document.py?Gericht=bgh&Art=pm&Datum=2017&Sort=3&nr=77354&pos=0&anz=18, 2.

388 Falk Schöning and Clemens Ziegler II. NATIONAL (AND EU) DECISIONS RELATING TO RES SUPPORT

As regards RES support measures in Germany, a summary of the development of European decisional practice regarding the State aid element of ‘granted through State resources’ in the ECJ’s PreussenElektra judgment of 20014 can be found in chapter one above.5 As will be shown below, the decisional practice of German courts thereafter (and up to now) assumes that none of the successor schemes to the Stromeinspeisungsgesetz (StrEG) discussed in PreussenElektra involved State aid, including the Erneuerbare Energien-Gesetz (EEG) 2012, the law on renewable energy of 2012.6 As mentioned above,7 according to the German StrEG, private electricity distributers had to purchase certain amounts of electricity from renewable energy sources at State-fixed minimum prices and to partially pay the difference between those prices and the market electricity prices. According to the ECJ, the purchase obligation, while constituting an advantage, did not involve a direct or indirect transfer of State resources, that is to say, no State aid was involved.8 The ECJ held that ‘the allocation of the financial burden arising from that obligation for those private electricity supply undertakings as between them and other private undertakings cannot constitute a direct or indirect transfer of State resources either’.9 The BGH held that the successor scheme to the StrEG, the Gesetz über den Vorrang Erneuerbarer Energien (EEG 2000—Law on the priority for renewable energy sources)10 essentially contained the same purchase obligation as the StrEG and therefore did not involve a direct or indirect transfer of State resources.11 This also applies to the successor schemes under the EEG 200412 and the EEG 2006,13 as for example held by the BGH in 2014.14 For the 2012 version of the EEG, the EEG 2012,15 the OLG (Court of Appeal) Hamm held that it did not contain a parafiscal levy, because it was limited to a system of payments exclusively between private entities. The Court considered it to be

4

Case C-379/98 PreussenElektra, EU:C:2001:160. See ch 1, sections II.B.iii.a and b. Gesetz zur Neuregelung des Rechtsrahmens für die Förderung der Stromerzeugung aus Erneuerbaren Energien (Law revising the legal framework for the promotion of electricity production from renewable energy) of 28 July 2011 BGBl 2011 I (Federal Official Gazette 2011 pt I) 1634. 7 See ch 1, section II.B.iii.a. 8 PreussenElektra (n 4) para 58. 9 ibid, para 60. 10 BGBl 2000 I (Federal Official Gazette 2000 pt I) 306. 11 Judgment of 11 June 2003—VIII ZR 160/02—BGHZ 155, 141–66, para 39. 12 Gesetz zur Neuregelung des Rechts der Erneuerbaren Energien im Strombereich, BGBl 2004 I (Federal Official Gazette 2004 pt I) 1918. 13 Erstes Gesetz zur Änderung des Erneuerbare-Energien-Gesetzes vom 7. November 2006, BGBl 2006 I (Federal Official Gazette 2006 pt I) 2550. 14 Judgment of 6 May 2015—VIII ZR 56/14—BGHZ 155, 141–66, para 34. The Commission’s opening decision of 18 December 2013 in Support of renewable electricity and reduced EEG surcharge for energy-intensive users (Case SA.33995) [2014] OJ C37/73 pursuant to Art 108(2) TFEU against Germany, however, referred to the EEG 2012, which was not the subject of that proceeding. Therefore, the Federal Supreme Court did, in this judgment, not discuss whether State aid was involved in the EEG 2012. See ibid, para 35. 15 See n 7 above. 5 6

Germany

389

a mere price regulation. There was no flow of payment to the State sector, hence the State did not have any control over the funds.16 The BGH confirmed this finding:17 the fact that the electricity utilities had to reimburse the electricity distributors, through the EEG-Umlage (EEG-Surcharge), for the difference between the costs due to the to-be-purchased RES-electricity, was not a parafiscal charge with a financing function for the public sector.18 The BGH stressed that this was different from the so-called Kohlepfennig,19 for which the Federal Constitutional Court (BVerfG) had decided that a charge levied from electricity consumers to support the use of German black coal in electricity production could not be designed as a parafiscal charge, because the electricity consumers did not have a special financing responsibility for the support of black coal.20 The EEG 2012 imposed any service, purchase and payment obligations exclusively on entities of private law. Also, the exemption of energy-intensive undertakings (EIUs) and railways from the EEG-Umlage did not lead to the relevant financial means entering the public domain. While the legislative aim of the EEG-Umlage was to achieve a sustainable energy supply in the interest of climate and environmental protection, the exemption for EIUs and railways aimed at the reduction of electricity costs of these EIUs and railways, to maintain their international and intermodal competitiveness.21 Even where the State significantly interferes with additional regulations of these private relationships, it does not turn an admissible price regulation into an inadmissible special parafiscal charge.22 Although the OLG Hamm and the BGH did not expressly state that they therefore also did not consider the EEG 2012 to contain State aid, this is the necessary legal conclusion. According to PreussenElektra, minimum price regulations that are limited to organising the flow of financial means between private entities cannot constitute a direct or indirect transfer of State resources.23 On the other hand, parafiscal charges that either flow into a State-administered fund or are otherwise subject to State control, also through delegated administration, have been considered to constitute State aid.24 This jurisprudence of the German courts still applies, even after the Commission’s 2014 decision finding State aid in financial support under the German EEG 2012 and the General Court’s (GC) 2016 judgment refusing to annul that decision.25 16

Judgment of 14 May 2013—I-19 U/12, juris, paras 28–30. Judgment of 25 June 2014—VIII ZR 169/13—BGHZ 201, 355-363, paras 12–25. 18 ibid, paras 13–14. 19 ibid, para 16. 20 BVerfG (Federal Constitutional Court), decision of 11 October 1994—2 BvR 633/86 (BVerfGE 91, 186–207). 21 BGH (n 17) para 16–18. 22 ibid, paras 24–25. 23 PreussenElektra (n 4) paras 59–62. 24 Case C-206/06 Essent Netwerk Noord et al, EU:C:2008:413, paras 68–73; see also J Kröger, ‘Die EEG-Umlage ist keine Sonderabgabe—zugleich Anmerkung zu OLG Hamm, Urt. v. 14.5.2013—19 U 180/12’ (2013) 9 Zeitschrift für Umweltrecht 480, 482–83; G Manssen, ‘Die Zukunft der EEGUmlage—weiter auf verfassungswidrigen Wegen?’ (2012) 11 Energiewirtschaftliche Tagesfragen 50. 25 Both of these decisions (Commission Decision (EU) 2015/1585 of 25 November 2014 on the aid scheme SA.33995 (2013/C) (ex 2013/NN) ([2015] OJ L250/122) and Case T-47/15 Germany v Commission, EU:T:2016:281 upholding it) have been discussed in ch 1, sections II.B.iii.b. 17

390 Falk Schöning and Clemens Ziegler The OLG (Court of Appeal) Hamburg held, on 5 July 2016, that the EEG 2012, according to PreussenElektra, also does not contain State aid.26 The Court expressly referred to the Commission’s different 2014 decision and the GC’s 2016 judgment denying annulment of that decision,27 but without entering into the discussion about their different assessment of the element of State resources in the EEG 2012.28 This judgment has been appealed to the BGH.29 The judgment of the OLG Hamburg is noteworthy. A national court would be expected to give deference to a final Commission decision finding State aid following a formal investigation procedure and, at the very least, a judgment by the GC declining an application to annul such decision. In 2012, the CJEU had established that Commission decisions opening the formal investigation procedure required national courts to adopt all the necessary measures with a view to drawing the appropriate conclusions from an infringement of the obligation to suspend the implementation of that measure. To that end, national courts may decide to suspend the implementation of the measure in question and order the recovery of payments already made. They may also decide to order provisional measures in order to safeguard both the interests of the parties concerned and the effectiveness of the Commission’s decision to initiate the formal examination procedure. Where they entertain doubts as to whether the measure at issue constitutes State aid within the meaning of Article 107(1) TFEU or as to the validity or interpretation of the decision to initiate the formal examination procedure, national courts may seek clarification from the Commission and, in accordance with the second and third paragraphs of Article 267 TFEU, as interpreted by the Court, they may or must refer a question to the Court for a preliminary ruling.30

If even mere Commission opening decisions can have such far-reaching consequences,31 a final Commission decision finding State aid and a GC judgment declining to annul that same decision should all the more. That said, in this case the question of whether the measure constituted State aid or not did not make a practical difference. The judgment of the OLG Hamburg only dealt with the EEG-Surcharge as such that the Commission and GC had considered to be compatible State aid, not with the exemption for EIUs that the Commission and GC found to be illegal State aid.32 Similarly, the most recent decision by the BGH of 9 February 2017 stated that a national court did not have an absolute obligation to give deference to the Commission’s preliminary assessment in a decision opening the formal investigation procedure. The judgment sets out the same limitations as contained in the Deutsche Lufthansa judgment of the CJEU of 2013 just mentioned, but does not put

26

OLG Hamburg, judgment of 5 July 2016—9 U 156/15—, juris, para 88. ibid. ibid. 29 BGH, VIII ZR 156/16. 30 Case C-284/12 Deutsche Lufthansa AG v Flughafen Frankfurt-Hahn GmbH, EU:C:2013:755, paras 42–44. 31 It is debatable if already Commission opening decisions should have such implications. See the discussion in ch 13. 32 OLG Hamburg, judgment of 5 July 2016 (n 26). 27 28

Germany

391

in question the general rule that in principle, deference has to be given even to opening decisions. According to the BGH, should the national court have doubts, it may send an inquiry to the Commission or request a preliminary ruling from the CJEU … Should the Commission continue to consider the measure to be State aid, the national court, however, does not find its arguments to be convincing, it can request a preliminary ruling from the CJEU.33

As regards the question of deference to Commission State aid decisions by national courts, the BVerwG stated that the Commission decision not to raise objections at the end of a preliminary investigation pursuant to Article 4(3) of the Procedural Regulation34 did not release a national court from the obligation to itself assess whether the measure in question qualified as State aid or not. This would apply in any event where the Commission did not express whether it deemed the measure to constitute State aid. It would, however, also apply where the Commission came to the conclusion that there was aid, but nevertheless decided not to raise objections without initiating the formal investigation procedure pursuant to Article 4(4) of the Procedural Regulation.35 This would follow from the mere fact that the Commission’s decision pursuant to Article 4(3) of the Procedural Regulation was only based on a preliminary assessment. However, in the main proceedings a national court would have to decide on the basis of a final assessment. While it could use the Commission’s preliminary assessment as a starting point, it could not limit itself to it. Rather, it would have to use all available evidence. Exceptionally, in the area of temporary legal protection, where there may not be sufficient time for a full assessment, it may be sufficient for a national court to only rely on the Commission’s preliminary assessment.36 Additionally, Commission decisions based on Article 4 of the Procedural Regulation are issued without the involvement of the recipient of a possible aid measure.37 Moreover, according to the BVerwG, the fact that the General Court of the EU had rejected an application for annulment against the Commission decision did not lead to the increased reliability on the Commission’s assessment. Although the recipient of the measure had been intervening in the procedure, the General Court did not assess the question of whether there was an aid measure at all.38 Finally, the BVerwG also stated that a national court would not be bound by the Commission’s decision because of the general principles of Union law.39 33 BGH, Press Release regarding judgment of 9 February 2017—I ZR 91/15 (n 3) 2 (own trans). See also the discussion of this judgment in ch 13. 34 Council Regulation (EU) 2015/1589 of 13 July 2015 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union (codification) [2015] OJ L248/9 (although the BVerwG, ratione temporis, was referring to the old Procedural Regulation (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). 35 BVerwG, judgment of 26 October 2016—10 C 3/15, para 23. 36 ibid, para 24. 37 ibid, para 25. 38 ibid, para 27. In this regard, it stated that the scope of examination by national courts was not limited by (i) the general principle of legality of EU acts (paras 28–29), nor (ii) the EU principle of loyalty pursuant to Art 4(3) TEU (paras 31–33). 39 BVerwG, judgment of 26 October 2016 (n 35) paras 28 ff.

392 Falk Schöning and Clemens Ziegler III. NATIONAL DECISIONS RELATING TO COAL AND NUCLEAR

A. BVerwG Case Regarding Greenhouse Gas Emission Allowances A 2012 judgment of the Federal Administrative Court (BVerwG) regarding a reduction of emission allowances40 contains a good example for the absence of selectivity due to the nature of the system argument that has been discussed in chapter one above:41 the plaintiff, an energy supply company running a coal-fired power and heat plant, opposed the reduction of emission allowances and requested an additional allocation of approximately two million allowances.42 The reduction in these emission allowances was based on the German law regarding the national allocation plan for greenhouse gas emission allowances for the allocation period from 2008 to 2012 (ZuG 2012)43 and only applied to energy supply companies. The BVerwG held that because the reductions in question had not been granted through State resources, they could not be considered to be imputable to the State and therefore did not violate Article 108(3) TFEU. In addition, there was no selectivity. Even if the measure in question only applied to energy supply companies but not to industrial plants or industrial power plants, the BVerwG considered this unequal treatment to be justified based on the nature of the system.44 According to the BVerwG, the emissions trading system served primarily to deliver on the obligations the EU assumed in the Kyoto Protocol. However, according to recital 5 of the Emission Allowance Directive,45 the primary goal of contributing to ‘fulfilling the commitments of the European Community and its Member States more effectively’, had to be achieved subject to certain secondary goals, ‘through an efficient European market in greenhouse gas emission allowances, with the least possible diminution of economic development and employment’.46 According to the BVerwG, the national implementing measures had to ensure compliance with the primary obligations while maintaining achievement of the secondary goals. In that regard, already the primary goal, but more so the secondary goal, commanded a differentiation between energy and industrial plants, because they were subject to different competitive conditions on the market. As the BVerwG found, the operators of energy production plants are primarily producing for the German market and to a certain extent for markets abroad that are also subject to the emissions trading system. In contrast, the operators of industrial power plants typically compete on the world market, insofar as they are subject to the emissions trading system. Thus, according to the BVerwG, they are only to a very limited 40

BVerwG, judgment of 10 October 2012—7 C 11/10—, juris. See ch 1, section II.B.iv.b ‘Measures Justified by the Nature of the System’. 42 BverwG, judgment of 10 October 2012 (n 40) para 17. 43 §§ 4(3), 19, 20 Zuteilungsgesetz 2012 vom 7. August 2007 (BGBl I (Federal Official Gazette 2007 pt I) 1788), das zuletzt durch Artikel 4 Absatz 32 des Gesetzes vom 18. Juli 2016 (BGBl I (Federal Official Gazette pt I) 1666) geändert worden ist. 44 BVerwG, judgment of 10 October 2012 (n 40) paras 19–24. 45 Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC [2003] OJ L275/32. 46 BVerwG, judgment of 10 October 2012 (n 40) para 26. 41

Germany

393

extent in a position to pass through the costs for the acquisition of emission allowances for pay. Their exemption thus also aimed at reducing the risk of these companies transferring their production into States that impose less stringent climate protection measures.47 The BVerwG found this view confirmed also by the fact that the differentiation between energy production and industrial plants was not a particularity of the German allocation rules. Among others, Article 10a(12) of the Emission Allowance Directive as amended by Directive 2009/29/EC48 also continues to foresee a free allocation of emission allowances to the industrial power plants, whereas, aside from certain exceptions, the allocation of emission allowances to energy power plants is only foreseen against payment. For those reasons, an early and slightly higher financial burden through reduced allocations had to be seen as inherent to EU law. The BVerwG concluded that this was justified by the nature of the system and therefore not selective.49 The energy supply company then submitted a constitutional complaint to the BVerfG, because the BVerwG had not submitted a request to the CJEU for a preliminary ruling in this case.50 The BVerfG refused to accept the constitutional complaint for judgment, because it did not see a violation of the claimed rights, in particular not the right to a judge appointed by law. It referred to the EJC’s CILFIT judgment of 1982, in which the ECJ had defined the conditions under which a national court of final instance had to request a preliminary ruling from the ECJ.51 The BVerfG did not find those conditions to be met. In particular, there were no indications that the BVerwG deliberately deviated from the jurisprudence of the CJEU as regards the interpretation of the notion of State aid.52

B. FG Hamburg Request for a Preliminary Ruling Regarding a Levy on Nuclear Fuel In a request for a preliminary ruling, the Finanzgericht (Fiscal Court) Hamburg (FG Hamburg) discussed national legislation that levies a duty on the use of nuclear fuel for the commercial production of electricity.53 The FG Hamburg stated that the CJEU had repeatedly made clear that in relation to State aids that are granted through exemption from a levy, the debtor of the levy could not claim that the exemption

47

ibid, para 27. Directive 2009/29/EC of the European Parliament and of the Council of 23 April 2009 amending Dir 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community [2003] OJ L 140/63. 49 BVerwG, judgment of 10 October 2012 (n 40) paras 28–29. 50 BVerfG, decision of 6 September 2016—1 BvR 1305/13—, juris. 51 Case C-283/81 CILFIT, EU:C:1982:335, para 21. 52 According to the BVerfG, the mere fact that the BVerwG did not agree with the complainant’s opinion as regards EU law cannot be seen to constitute a deliberate deviation from the jurisprudence of the CJEU. On the contrary, the BVerwG had discussed the exact same CJEU rulings that the complainant also cited. However, it considered the rulings to be consistent with them and applied the standards found in them to the case at hand. BVerfG, decision of 6 September 2016 (n 50) paras 10–12. 53 FG Hamburg, request for preliminary ruling of 19 November 2013—4 K 122/13—, juris. 48

394 Falk Schöning and Clemens Ziegler of other undertakings from the levy constitutes State aid in order to avoid payment of that levy or to claim repayment.54 In such cases, the fiscal measure against the debtor of the levy does not violate the EU State aid rules. Instead, it is the exemption from the levy that is granted to some others, actually also debtors. The distortion of competition in such cases therefore has to be eliminated by also imposing the same amount of the levy on the illegally exempted debtors.55 The FG Hamburg then referred to an exception that the CJEU had made in cases of so-called asymmetrical taxation.56 Asymmetrical taxation means that only one of two categories of undertaking that are competing against each other is taxed. The CJEU had decided that an economic operator could claim illegality of a levy that is supposed to be a State aid measure, aiming at recovery of amounts paid based on that levy, even in cases where the levy and the alleged State aid measure are inseparably linked to one and the same fiscal measure.57 The FG Hamburg then stated that it was not clear whether this jurisprudence had also to be applied to the levy on nuclear fuel, in connection with the non-imposition of levies on other means to produce electricity, and if both of them could be seen as inseparable elements of one and the same fiscal measure. On the one hand, operators of nuclear power plants were the only undertakings on which the nuclear fuel tax was imposed; on the other hand, they are competing against other undertakings producing electricity. However, since, different from other undertakings producing electricity, only operators of nuclear power plants use nuclear fuel to produce electricity, the FG Hamburg doubted whether the taxation of nuclear fuels only can be seen as a case of asymmetrical taxation in which the CJEU allows claims for a tax exemption due to a violation of the State aid rules.58 The CJEU has answered the FG Hamburg’s questions in its preliminary ruling in Case C-5/14,59 by confirming that Article 107 TFEU did not preclude legislation that levies a duty on the use of nuclear fuel for the commercial production of electricity. In particular, only the use of nuclear fuel in electricity production produces nuclear waste, not the use of other means to produce electricity. Therefore, the tax in question cannot be considered to be a selective measure.60

IV. NATIONAL DECISIONS ON GRID-RELATED ISSUES

EU State aid law continues to be cited by national courts in Germany in the context of grid-related disputes.

54 FG Hamburg, para 238, citing to CJEU, Case C-526/04 Laboratoires Boiron SA v Union de recouvrement des cotisations de sécurité sociale et d’allocations familiales (Urssaf) de Lyon, assuming the rights and obligations of the Agence centrale des organismes de sécurité sociale (ACOSS), EU:C:2006:528, para 30. 55 ibid. 56 ibid, para 239, citing to CJEU, Case C-526/04 Laboratoires Boiron SA (n 54) para 33. 57 ibid, citing to CJEU, Case C-526/04 Laboratoires Boiron SA (n 54) paras 43–46. 58 ibid, para 240. 59 Case C-5/14 Kernkraftwerke Lippe-Ems GmbH v Hauptzollamt Osnabrück, EU:C:2015:354. 60 ibid, paras 69–82.

Germany

395

A. OLG München Case Regarding Piping Rights In a case before the OLG (Court of Appeal) München of 2000,61 the defendant, an energy utility, had an easement over a plot of land that allowed it to build, operate and maintain a gas pipeline. The defendant had also laid additional tubes into the plot which contained telecommunications wires. The plaintiffs requested an injunction against the laying of further tubes for telecommunications wires.62 The placement of telecommunications wires was based on German legislation favouring the expansion of telecommunications infrastructure.63 Although the plaintiffs had asked for it, the OLG München refused to request a preliminary ruling from the ECJ, because it was convinced that the measure was not State aid under Article 107(1) TFEU. In this case, the advantage was not imputable to the State because the State did not suffer any financial loss. According to the OLG München, in line with the jurisprudence of the ECJ, the advantage enjoyed by utilities that dispose of certain piping rights cannot be seen as an advantage in the meaning of Article 107(1) TFEU, even if the piping rights represented a financial benefit for the utilities concerned.64

B. OLG Düsseldorf Case Regarding Remuneration for Redispatch Measures Another grid-related area where State aid arguments have been discussed in German courts is the remuneration for balancing services such as redispatch measures. Redispatch measures have become more important in Germany in the context of the decommissioning of eight nuclear power plants in March 2011 (following the nuclear disaster in Fukushima, Japan). They intervene in the operation of a power plant to act against the overcharge of a grid element that threatens network security. In an electricity redispatch, the overcharge of a grid element is counteracted by the reduction of the electricity fed into that overcharged grid element, with additional electricity fed in by the power plant behind the overcharged grid element.65 The OLG Düsseldorf had to decide about the application of a compensation scheme for redispatch measures drawn up by the German Federal Network Agency.66 According to that compensation scheme, a service compensation would be paid to operators of plants if redispatch measures accounted for more than 10 per cent of the amount of electricity fed into the grid during the year before.67 However, the German Federal Cartel Office, as well as the OLG Düsseldorf, found agreements between Transmission System Operators (TSOs) and plant operators based on that compensation scheme to be anticompetitive, because they amounted to a contractual limitation of electricity output. The compensation mechanism provided

61 62 63 64 65 66 67

OLG München, judgment of 18 October 2000—20 U 2503/00—, juris. ibid, paras 2–10. ibid, para 38. ibid, para 65. OLG Düsseldorf, decision of 28 April 2015—VI-3 Kart 332/12 (V)—, juris, para 9. ibid, para 4. ibid, para 15.

396 Falk Schöning and Clemens Ziegler an incentive to operate power plants as little as possible on a market basis: a marketbased operation is usually done as soon as marginal costs are covered, whereas an operation under redispatch measures leads to compensation of the entire cost base of the power plant, ie, including fix cost.68 In the context of discussing the legality of agreements based on the compensation mechanism, the OLG Düsseldorf held, based on the EU Commission’s State aid decision approving the UK electricity capacity mechanism,69 that, provided the design of the relevant scheme was compatible with EU State aid rules, there would at least not be a general concern with the payment of compensation for the mere availability of capacity.70

C. OLG Düsseldorf Case Regarding Calculation of Network Charges In another case, the OLG Düsseldorf did not see a violation of EU State aid rules in a certain methodology the German Bundesnetzagentur, the Federal Network Agency, used in order to calculate whether electricity-intensive users could benefit from a (considerable) reduction of network charges.71 The provision on which this possible reduction of network charges is based, § 19(2)(2) of the German StromNEV (the electricity charges regulation), is currently subject to a formal investigation procedure by the European Commission which was opened on 6 March 2013.72 In this opening decision, the Commission discusses the system established by the German StromNEV, under which TSOs have to compensate Distribution System Operators (DSOs) for their financial losses due to the exemption of electricity-intensive users from network charges. Under this system, both TSOs and DSOs are allowed to pass through their financial losses due to the reduced network charges for electricity-intensive users73 to final consumers. The Commission ultimately took the same position as in its 2014 decision regarding the EEG-Umlage/RES-surcharge:74 it considers this system to constitute operating aid for the exempted electricity-intensive users.75 The OLG Düsseldorf did not mention the Commission’s opening decision at all. Its result, of refusing to extend the advantage of the allegedly illegal aid to the complainant

68

ibid, para 250. GB Capacity Mechanism (Case SA.35980) [2014] OJ C 348/5. This decision is still subject to two applications for annulment before the GC—T-788/14 MPF Holdings v Commission [2015] OJ C65/38, and T-793/14 Tempus Energy and Tempus Energy Technology v Commission [2015] OJ C81/21. 70 OLG Düsseldorf, decision of 28 April 2015 (n 65) para 252. 71 OLG Düsseldorf, decision of 15 July 2015—VI-3 Kart 83/14 (V)—, juris. 72 Commission Decision State aid SA.34045 (2012/C) (ex 2012/NN)—Exemption from network charges for large electricity consumers (§ 19 StromNEV) [2013] OJ C128/43. This procedure has now lasted over four years. Yet, it is unknown to the authors when the Commission’s final decision can be expected. 73 The TSOs’ financial losses stem on the one hand from their own, direct financial losses due to reduced transmission network charges from energy-intensive users and on the other hand from the compensation obligation regarding the DSOs’ financial shortfalls due to lost distribution network charges. In addition, since 2011, DSOs also have to pass through their financial losses from the exemption to final consumers, but then to transfer those proceeds back to the TSOs. See Commission Decision (n 72) recitals 7–18. 74 See n 25 above and corresponding text under section II above, 389. 75 ibid, recitals 98–99. 69

Germany

397

as a potential additional beneficiary of a possibly illegal State aid measure (based on the view underlying the Commission’s opening decision), is ultimately consistent with the CJEU’s Lufthansa judgment mentioned above.76 The Bundesnetzagentur’s methodology for calculating whether a user qualifies for the network charge reduction for electricity-intensive users that the complainant attacked, exclusively considered the physical path of electricity flow, ie, based on a fictitious grid usage between the grid connection point of the final consumer and a suitable electricity production plant.77 The complainant claimed, among other arguments, that the physical path methodology granted an advantage that the grid operator would not have received under normal market conditions.78 The OLG Düsseldorf did not agree. In this case, there was a suitable, equivalent and market-based remuneration, ie, there was no advantage. According to the OLG Düsseldorf, the calculation methodology using the physical path was exactly aiming at giving more weight to the idea of service and consideration. The German StromNEV (the electricity grid charges regulation) and the Federal Network Agency considered that a network charges reduction corresponding to the costs that a final consumer would have saved in case a direct grid connection was built was an appropriate consideration for its continued connection to the general supply network.79 The OLG Düsseldorf finally held that, where it was not possible to determine a market value for the service provided by a baseload customer, it should be fair to assume that the value that the service has for the baseload customer itself should be in reasonable proportion to the service delivered by it.80

V. GERMAN STATE AID MEASURES IN THE ENERGY AREA

As will be shown, in particular in the energy sector, block-exempted aid plays a predominant role in Germany, and Germany, unsurprisingly, provides the most blockexempted aid for environmental protection and energy within the EU. Based on the European Commission’s statistics (up to and including 2014) on block-exempted aid, the category of aid for ‘Environmental protection and energy saving’ is the most important one among all aid categories in Germany:81 for example, the share of environmental and energy aid in all block-exempted aid in Germany rose from below 39 per cent in 2011 to nearly 52 per cent in 2014.82 From 2011 to 2014, the absolute amounts of block-exempted aid granted in Germany for environmental and energy aid represented a share in the range between 48 per cent and 57 per cent of

76

See under section II above, 390–391. OLG Düsseldorf, decision of 15 July 2015—VI-3 Kart 83/14 (V)—, juris, paras 1 and 9. 78 ibid, para 41. 79 ibid, paras 108–10. 80 ibid, para 111. 81 Tables, graphs and maps showing, on a yearly basis until 2014, the total amount of block-exempted aid and aid for environmental protection and energy saving, and the GDP percentages those represent, respectively, are available at: www.ec.europa.eu/eurostat/tgm_comp/table.do?tab=table&plugin=1&lang uage=en&pcode=comp_bex_sa_01 and www.ec.europa.eu/eurostat/tgm_comp/table.do?tab=table&plug in=1&language=en&pcode=comp_bex_sa_02. 82 Own calculation based on the numbers available from the first link provided in n 81 above. 77

398 Falk Schöning and Clemens Ziegler the total block-exempted aid in this category in the EU, compared with a German share of total block-exempted aid in the EU in the range of 30 per cent to 44 per cent during the same period.83 For each year from 2012 until 2014, Germany spent the third largest percentage of GDP on environmental protection and energy saving State aid measures out of the 28 EU Member States, preceded by Finland and Sweden.84 Therefore, we will have a closer look at what should be the most important blockexempted energy-related aid categories in Germany. In accordance with the General Block Exemption Regulation (GBER),85 the following support measures are available in the energy sector in Germany:86 (i) investment aid for energy efficiency measures; (ii) investment aid for building-related energy efficiency measures; (iii) investment aid for high-efficiency combined heat and power projects; (iv) RES support in the form of investment aid; (v) RES support in the form of operating aid; (vi) operating aid for RES support in small-scale plants; (vii) investment aid for energy efficient district heating and district cooling; (viii) investment aid for building and expansion of energy infrastructures; and (ix) support for environmental studies, including energy audits. The conditions for a number of those measures are summarised below.

A. Investment Aid for Energy Efficiency Measures (Articles 38 and 4(1)(s) GBER) Eligible costs for investment aid for energy efficiency measures are the extra investment costs necessary to achieve a higher level of energy efficiency. Costs that are not directly linked to the improvement of energy efficiency are not eligible. The aid intensity is capped at 30 per cent of the eligible costs. For support to small enterprises, the aid intensity can be increased by 20 percentage points; for aid provided to mediumsized enterprises by 10 percentage points. In assisted areas pursuant to Article 107(3) (c) TFEU, the aid intensity can be increased by 5 percentage points. The notification threshold is €15 million per undertaking and investment project.87

B. Investment Aid for High-Efficiency Combined Heat and Power Projects (Articles 40 and 4(1)(s) GBER) Investment aid is only available for newly installed or refurbished capacity. Pursuant to Directive 2012/27/EU of the European Parliament and the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and

83

See n 81 above. See the numbers on block-exempted aid for environmental protection and energy saving in the second link provided in n 81 above. 85 Commission Regulation (EU) 651/2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty [2014] OJ L187/1. 86 Bundesministerium für Wirtschaft und Energie (Federal Ministry for the Economy and Energy), ‘Handbuch über staatliche Beihilfen. Handreichung für die Praxis von BMWi-EA6 [2016]’, available at: www.esf.de/portal/SharedDocs/PDFs/DE/Sonstiges/handbuch-beihilfe-bmwi.pdf?__blob=publication File&v=1, 64 ff. 87 ibid, 64 f. 84

Germany

399

2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC,88 a new cogeneration unit needs to reduce the primary energy consumption compared with separate generation. The improvement or the refurbishment of an existing cogeneration unit needs to result in additional savings of primary energy compared with the baseline situation. The eligible costs shall be the extra investment costs for the equipment needed for the installation to operate as a high-efficiency cogeneration unit, or the extra investment cost to achieve a higher level of efficiency when an existing installation already meets the high-efficiency threshold. The aid intensity cannot exceed 45 per cent of the eligible costs. For aid measures for small enterprises, the aid intensity can be increased by 20 percentage points; for aid provided to medium-sized enterprises by 10 percentage points. In assisted areas pursuant to Article 107(3)(c) TFEU, the aid intensity can be increased by 5 percentage points. The notification threshold is €15 million per undertaking and investment project.89

C. RES Support in the Form of Investment Aid (Articles 41 and 4(1)(s) GBER) Investment aid measures for the production of bio fuels are only exempted from the notification obligation if the investments aim at the production of sustainable bio fuels from raw materials other than food crops. Investment aid measures for the refurbishment of existing plants to produce bio fuels from food crops into production plants for modern bio fuels are exempted as long as the production of bio fuels from food crops is reduced in proportion to the new capacity. Funding is not available for: — — —

Bio fuels which are subject to a supply or blending obligation. Hydropower plants that do not correspond to Directive 2000/60/EC of the European Parliament and the Council.90 No aid can be granted or paid after the production plant has started to operate; the support does not depend on the production output.

The eligible costs are the additional investment costs necessary to promote the production of energy from renewable sources. Costs that are not directly related to achieve a higher level of environmental protection are not eligible. The notification threshold is €15 million per undertaking and investment project. The aid intensity cannot exceed the following amounts: —

45 per cent of the eligible costs if those are calculated based on Article 41(6)(a) or (b) of the GBER.91

88 Directive 2012/27/EU of the European Parliament and the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directive 2004/8/EC and 2006/32/EC [2012] OJ L315/1. 89 ‘Handbuch über staatliche Beihilfen. Handreichung für die Praxis von BMWi-EA6’ (n 86) 65 f. 90 Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy [2000] OJ L 327/1. 91 Commission Regulation (EU) 651/2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty [2014] OJ L187/1.

400 Falk Schöning and Clemens Ziegler —

30 per cent of the eligible costs if those are calculated based on Article 41(6)(c) of the GBER. — For aid measures for small enterprises, the aid intensity can be increased by 20 percentage points; for aid provided to medium-sized enterprises by 10 percentage points. — In development areas pursuant to Article 107(3)(c) TFEU, the aid intensity can be increased by 5 percentage points. — If the aid is granted through an open tender process based on clear, transparent and non-discriminatory criteria, the aid intensity can reach up to 100 per cent of the eligible costs.92

D. RES Support in the Form of Operating Aid (Articles 42 and 4(1)(v) GBER) Aid measures are granted based on clear, transparent and non-discriminatory criteria in a tender process that is open to all producers of electricity from RES, in a non-discriminatory manner. The tender can be technology-specific if a tender open to all producers would lead to a suboptimal result that could not even be corrected through the design of an open procedure. The aid is granted each year for up to 5 per cent of the planned new capacity for the production of electricity from RES. The support is granted as a premium in addition to the market price at which the producers sell their electricity directly on the market. No aid is granted for electricity produced at negative prices. Certain small-sized plants benefit from specific simplifications, for example, waiver of a tender process. Aid can be granted only as long as the RES plant has not yet been fully depreciated based on generally accepted accounting principles. Any investment aids have to be deducted from the operating aid. The notification threshold is €15 million per undertaking and investment project or €150 million per year if it is granted after a tender process.93

E. Operating Aid for the Promotion of Energy from Renewable Sources in Small-Scale Installations (Articles 43 and 4(1)(v) GBER) Aid is only available for plants for the production of energy from RES with an installed capacity below 500 kW. However, wind generators are eligible up to an installed capacity of less than three MW or less than three generation units, as well as plants for the production of bio fuel with an installed capacity of less than 50,000 tons per year. For the calculation of these maximum capacities, smallscale installations with a common connection point to the electricity grid will be considered as one installation.

92 93

‘Handbuch über staatliche Beihilfen. Handreichung für die Praxis von BMWi-EA6’ (n 86) 66 f. ibid, 67.

Germany

401

Aid is only granted for plants producing sustainable bio fuels other than foodbased fuels. Operating aid to plants producing food-based bio fuels that started operation before 31 December 2013 and are not yet fully depreciated shall be exempted but in any event no later than 2020. Aid shall not be granted for bio fuels that are subject to a supply or blending obligation. The aid per unit of energy shall not exceed the difference between the total levelised energy production cost from the source in question and the market price of the form of energy concerned. The maximum rate of return used in the levelised cost calculation shall not exceed the relevant swap rate plus a premium of 100 basis points. The notification threshold is €15 million per undertaking and investment project or €150 million per year if it is granted after a tender process.94

F. Investment Aid for Energy Efficient District Heating and District Cooling (Articles 46 and 4(1)(w) GBER) The eligible costs for the production plant shall be the extra costs needed for the construction, expansion and refurbishment of one or more generation units to operate as an energy efficient district heating and cooling system compared to a conventional production plant. The investment shall be an integral part of the energy efficient district heating and cooling system. The aid intensity: — Cannot exceed 45 per cent of the eligible costs for the production plant. — Can be increased by 20 percentage points for aid measures for small enterprises and by 10 percentage points for aid provided to medium-sized enterprises. — Can be increased by 5 percentage points in assisted areas pursuant to Article 107(3)(c) TFEU. The eligible costs for the distribution network shall be the investment costs. The aid amount for the distribution network shall not exceed the difference between the eligible costs and the operating profit. The operating profit shall be deducted from the eligible costs ex ante or through a clawback mechanism. The notification threshold amounts to €20 million per undertaking and investment project.95

G. Investment Aid for Construction or Upgrade of Energy Infrastructure (Articles 48 and 4(1)(x) GBER) The aid must be granted for energy infrastructures in assisted areas. The energy infrastructure continues to be subject to tariff and access regulation in accordance with the internal energy market rules. Eligible costs are the investment costs. The aid amount cannot exceed the difference between the eligible costs and the operating

94 95

ibid, 67 f. ibid, 69.

402 Falk Schöning and Clemens Ziegler profit of the investment. The operating profit shall be deducted from the eligible costs ex ante or through a clawback mechanism. Aid for investments in electricity or gas storage projects or in oil infrastructures shall not be exempt from the notification obligation pursuant to Article 48 GBER. The notification threshold amounts to €50 million per undertaking and investment project.96

VI. CONCLUSION

The above discussion confirms that the most important guidance regarding EU State aid law in Germany is to be found in State aid decisions of the European Commission and in the jurisprudence of the European courts. That being said, German courts may at times, as shown in the OLG Hamburg decision regarding the EEG 2012 case97 or in the BVerwG decision of October 2016 mentioned above,98 even be prepared to go their own way and deviate from Commission decisions and General Court judgments. That said, overall, when State aid law questions become relevant in German court procedures, the courts generally apply principles of EU State aid law, and overall, they do so in strong reliance on the jurisprudence of the CJEU. Not least considering the challenges ahead in the energy sector in Germany and in the EU, the Energiewende, including Atomausstieg as well as the long-term exit from coal mining and coal-fired power plants, the various available State aid measures summarised above will continue to play an important role in the energy sector in Germany.

96 97 98

ibid, 70. See under section II above, 390. See under section II above, 391.

15 France LILIANA ESKENAZI

I. INTRODUCTION AND BACKGROUND

T

HIS COUNTRY-SPECIFIC CHAPTER on France is a showcase for a number of key State aid topics in the energy sector discussed in the previous chapters. The selected case law reported in this chapter provides a detailed illustration of those topics, as well as some insights as to how the relevant issues arose, what the national legal background and the factual context were and, where relevant, how the national courts and authorities drew the legal and practical consequences of EU decisions. The first section is dedicated to the sensitive issue of regulated prices for electricity and gas. While the main focus is naturally on the State aid analysis, there is also an attempt to consider this topic in the wider perspective of the implementation of successive EU energy packages in France, as well as the national case law relevant for this assessment. The second section discusses State aid for renewable energies, as illustrated by the landmark Vent de Colère case. It is not a recast of the substantive analysis of the Court of Justice (CJEU) preliminary ruling (already discussed in detail in chapter two), but an attempt to shed more light on how the case originated in the context of a complex French support scheme for renewable (wind) energies and how, eventually, the French authorities had to comply with the CJEU ruling and the Commission’s State aid decisions that further assessed the support scheme and its financing mechanism. Finally, the third section focuses on capacity mechanisms. France is among the first Member States to have introduced a capacity mechanism and was part of the sample in the Commission’s State aid sector inquiry. This section however puts in the spotlight two case-specific Commission decisions—the first declaring the French capacity mechanism compliant with State aid rules after the French government designed remedies to address the Commission’s concerns, and the second authorising, after an in-depth investigation, a support scheme for the construction of a new gas-fired power plant in Brittany.

II. THE PERSISTENT ISSUE OF REGULATED ENERGY PRICES IN FRANCE

By virtue of the Internal Energy Market legislation (Directive 2003/54/EC and 2003/55/EC, replaced by Directives 2009/72/EC and 2009/73/EC), Member States

404 Liliana Eskenazi must ensure that consumers are free to buy electricity and gas from the supplier of their choice as from 1 July 2007. Even if it is not stated explicitly in the Directives, their purpose and general scheme appears to require that the price for the supply of gas and electricity is determined solely by the operation of supply and demand. The Directives are designed to achieve progressively a total liberalisation of the market for national gas and electricity with suppliers totally free to deliver their products to all consumers. Therefore, the persistence of regulated tariffs may impede new entrants. In spite of this framework, ever since the theoretical opening of the energy market, France has maintained regulated prices on the sale of both electricity and gas, arguably to protect the consumers and ensure security of supply. While market prices are proposed by all suppliers, regulated tariffs (often set at a level below the market price) could only be proposed by a strictly limited number of suppliers (essentially the incumbents EDF and Engie, formerly GDF–Suez). The competition and general EU law issues arising from regulated prices were analysed at both EU and national level, by each of the European Commission, the French Competition Authority (FCA), the national courts and even the CJEU. Among these, the European Commission’s analysis in the regulated electricity tariffs case is most relevant to the State aid analysis. Nevertheless, we will also shortly address a couple of other cases to put into context all recent developments leading to what may become the progressive disappearance of regulated prices in France.

A. State Aid Analysis of the French Regulated Tariffs for Electricity There have been numerous cases at national level challenging various aspects of the regulated electricity tariffs, including their compatibility with general competition law (eg, abuse of dominance, insofar as the State may have encouraged such abuse), or simply with the pricing formulae set out by the legislator.1 As a consequence, the ministerial orders setting out the level of the regulated prices have been annulled several times by the administrative jurisdictions. However, a challenge to the very principle of the existence of such regulated prices has never succeeded at national level on such grounds—so far. It was the European Commission that, following the opening of several procedures—failure to fulfil obligations (a procedure that was later closed) and a State aid investigation—finally concluded that the regulated electricity prices for professional consumers constitute State aid and should be phased out. i. Commission Decision of 12 June 2012 (SA.21918) The Commission opened a State aid case against the regulated tariffs of electricity in 2009.2 The opening of the case eventually led France to abolish, as of 1 July 2011,

1 The Conseil d’Etat annulled several ministerial orders since the regulated tariffs were set below the average total costs (coûts moyens complets) of the supply of electricity in contradiction with Law No 2000-108, which provides that regulated tariffs for electricity should be set above the average total costs of the operators. See, eg, CE 1 July 2010, No 321595 or CE 11 April 2014, No 365219. 2 Commission, ‘Press Release of 10 March 2009’ IP/09/376.

France

405

the so-called ‘return tariffs’ (a transitional mechanism allowing consumers that choose market prices to go back to the regulated tariffs under certain conditions), and set the abolition of the so-called ‘yellow’ and ‘green’ tariffs (regulated tariffs for consumers with subscribed power above 36 kVA, meaning essentially SMEs and large companies) by 1 January 2016.3 a. Scope of the Decision At the time of the Commission decision, the regulated tariffs were segmented by user categories into ‘tariff options’: —

— —

‘Blue tariff’: applying to consumption sites with subscribed power less than or equal to 36kVA, which concerns, in practice, household (residential) customers and small professionals. ‘Yellow tariff’: applying to consumption sites with subscribed power between 36 and 250kVA, which concerned, in practice, SMEs. ‘Green tariff’: applying to consumption sites with subscribed power greater than 250kVA, which concerned, in practice, large companies and industrial sites.

The State aid case did not cover the regulated tariffs applicable to households and small businesses, but concerned the ‘non-residential’ tariffs and the prolongation of the system of ‘return tariffs’ until June 2010 and its extension to new beneficiaries. The level of return tariffs was set by ministerial decree, by reference to the level of the standard regulated tariff which would be applicable to a consumer with the same characteristics and which had not exercised its eligibility right, and slightly higher than the regulated price.4 Electricity suppliers which supply customers at the return tariff following a request made by them benefited from compensation if they were unable to generate or procure the quantities of electricity necessary at a price below the ‘supply’ component of the return tariff. The compensated costs were calculated on the basis of accounts kept by the suppliers according to rules laid down by the French Energy Regulatory Commission (CRE). The compensation was financed by the revenue from two contributions imposed by the State: (i) a portion from the contribution to the public service of electricity (contribution au service public de l’électricité (CSPE)); and (ii) a contribution payable by electricity generators operating facilities with a total installed capacity exceeding two gigawatts. This contribution was capped at 3€/MWh in 2008. The yields from these two compulsory contributions were collected by the Caisse des Dépôts et Consignations, a State-owned special financial institution entrusted with public service obligations. A provision from a 2008 law5 enabled any final consumer with a consumption site already supplied with electricity under the return tariff system to continue to

3

Law No 2010-1488 of 7 December 2010 (Loi NOME). eg, the decree of 3 January 2007 set the level of return tariffs: for the yellow tariffs, 20% more than the standard regulated tariff; for the green tariff, 23% more than the standard regulated tariff. 5 Art 166 of Law No 2008-776. This provision also enabled any final consumer to submit an application up to 30 June 2010, whereas previously the applications to benefit from the return tariffs system could no longer be accepted after 1 July 2007. 4

406 Liliana Eskenazi benefit from return tariffs for this site until 30 June 2010, whereas originally the final consumer could benefit from them only for a two-year period ending no later than 30 June 2009. Ultimately, however, no final consumer could benefit from the return tariffs system after 30 June 2010. b. State Aid Analysis of the Commission State Resources and Imputability to the State Both the standard tariffs and the return tariffs mechanisms were implemented by laws and regulations passed by the French State, and the level of the tariffs was set by ministerial decree. Therefore, the condition of imputability to the State was met. The standard tariffs were financed by the resources of EDF which sold electricity to its customers at a price below the price which would result from the free functioning of the market. Since the State held an 84.4 per cent stake in EDF on 31 December 2010, EDF’s resources were deemed State resources, meaning that the sums financing the standard tariffs are State resources. As explained above, the return tariffs were financed by the revenue from two contributions imposed by the State. The revenue from the contributions were to be paid to a public entity designated by the State: the Caisse des Dépôts et Consignations. The return tariffs were therefore also financed by State resources. Economic Advantage The Commission analysed whether the (non-residential) standard and return tariffs allowed their beneficiaries to procure electricity at a more advantageous price than that which would prevail in their absence, ie, market prices. The regulated standard and return tariffs were (systematically each year for the standard tariffs and the green return tariffs, and on average for the yellow return tariffs) lower than the market prices. Therefore, the application of the tariffs conferred an economic advantage on the beneficiary undertakings. The fact that a vast majority of eligible customers have chosen to retain the standard tariffs or to benefit from the green and yellow return tariffs was indicative. Selectivity The tariff measures were deemed selective since they favour electricity-consuming undertakings as compared with those using fossil fuels. Moreover, such tariffs in fact favoured the enterprises with larger electricity consumption, since the advantage they derive from them necessarily increases with the level of electricity consumption. In addition, the irreversibility of exercising the right of eligibility between the market tariffs and the regulated tariffs contains an obvious element of selectivity, and the application of the criteria consisting of dates has the additional effect of limiting the benefit from the tariffs to certain enterprises, by excluding others. Therefore, the Commission concluded that the yellow and green standard regulated tariffs and the return tariffs are selective.

France

407

Effect on Competition and Trade between Member States Thousands of undertakings benefiting from these tariffs in the industrial and services sectors in France operate in markets which are fully open to competition within the internal market, and therefore affect competition and trade between Member States. Conclusion on the Existence of State Aid The Commission concluded that the yellow and green standard regulated tariffs and the yellow and green return tariffs constitute State aid. Since the two tariff systems were implemented without prior notification, they were deemed unlawful. c. Compatibility Analysis The Commission acknowledged that, on the basis of Article 107(3)(c) TFEU, a State aid can be justified for a limited time insofar as it has as its object or effect to accompany and facilitate the successful achievement of the electricity market liberalisation process where market forces alone prove to be insufficient to achieve this. In this case, it therefore assessed whether the aid can contribute to the successful achievement of the liberalisation. Since the French authorities entered into the commitment to put an end to the regulated tariffs from 1 January 2016 (whereas the return tariffs had already been abolished in July 2011), the State aid was deemed to facilitate a gradual transition to a genuinely competitive market. The Commission recognised the need for the aid, and particularly the restriction of the freedom to fix prices, due to the situation and characteristics of the French market (with a near monopoly situation of EDF that could apply excessive tariffs for a long period after liberalisation or squeeze out its competitors due to its available large financial resources). The Commission also recognised the proportionality of the aid. Finally, the Commission acknowledged that the aid did not, under the reform brought by Law No 2010-1488 implementing the commitments undertaken by France, affect trade to an extent contrary to the common interest. Therefore, the Commission considered that the aid measure implemented by France by means of regulated tariffs for the sale of electricity and regulated transitional market adjustment tariffs for large and medium-sized consumers is compatible with the internal market, subject to the commitments undertaken by France. d. Commitments Undertaken by France Pursuant to the Commission’s decision,6 the French authorities committed (i) to phase out the regulated tariffs for non-household customers by 31 December 2015, and (ii) in parallel, to introduce an obligation for EDF to sell to its competitors at retail level (alternative electricity suppliers) part of its nuclear electricity (up to 100 TWh) at a regulated price. The obligation to sell nuclear electricity is imposed until 31 December 2025. It is intended to allow competition to fully emerge, given that—at

6

Art 2 of the decision.

408 Liliana Eskenazi the time of the decision—EDF benefited from a comparative advantage due to the fact that it could sell its nuclear electricity at prices lower than the (then) market prices. The Commission’s decision further imposed the following rules concerning the price of the regulated access to existing nuclear power (called ARENH) for alternative suppliers: —





The ARENH price is examined (but not necessarily modified) every year and should reflect the economic conditions for electricity generation for the duration of the mechanism. The level of the ARENH price should not exceed €42/MWh and should not evolve as long as the calculation method for its determination has not entered into force. Such calculation method should be submitted to the Commission for prior approval.

Finally, the Commission’s decision also imposed on France the obligation that all decisions taken by the French authorities after the summer of 2012 in relation to the regulated tariffs for electricity should allow the progressive reduction—as compared with 2012, and then for each year as compared with the previous one—of the difference between the cost addition (of EDF’s costs) and the regulated tariffs. ii. Follow-Up: The Setting Up and Functioning of the Regulated Access to Existing Nuclear Power (ARENH) The ARENH mechanism was laid down by Law No 2010-1488 of 7 December 2010 (Loi NOME) and Decree No 2011-466 of 28 April 2011. These provisions have been incorporated into the French Energy Code.7 A draft decree dealing with the substance of the mechanism, and in particular the price calculation method, has been in the pipeline for a long time. The FCA has already taken a position on it in a formal opinion in 2014. However, as per the Commission’s State aid decision (see II.A.i above), such method should first receive the Commission’s approval before being implemented. At the time of writing this chapter, the decree has still not been published. There are several possible explanations for that. We focus below on a tentative reasoning from a State aid law perspective. Currently, the ARENH price cannot be deemed as such to constitute incompatible State aid, because the mechanism was introduced precisely as a condition for developing competition on the French market in the context of the progressive abolition of the regulated tariffs. The mechanism results from the Commission’s State aid decision of 2012 and, as long as it complies with the conditions set out in this decision (see II.A.i.d above), it should not be deemed contrary to competition law.

7 Arts L 336-1 to L 336-10, and L 337-13 to L 337-16 of the French Energy Code. On 30 December 2015, a new decree on the ARENH was published (Decree No 2015-1823). However, its purpose has only been to incorporate the existing mechanism into the regulatory part of the Energy Code (now Arts R 336-1 and following).

France

409

This should continue to be the case as long as the ARENH price is not below cost and does not lead to a State subsidy (through EDF, which is a State-owned company) to customers who would benefit from lower costs for buying nuclear electricity.8 Conversely, if the price is above cost, EDF could arguably cross-subsidise its market activities. However, this will not constitute a State aid issue, because the ARENH price is paid by private undertakings and therefore does not commit State resources (though it might raise general antitrust concerns, as the FCA has warned in some of its opinions).9 From this perspective, and assuming that currently the ARENH reflects correctly EDF’s costs, an increase in its level should in principle not raise State aid concerns. However, the structure and the determination of the relevant costs for this purpose is a complex issue, which probably explains why, so far, the Commission has still not taken a position on the French government’s proposed methodology for calculating (and increasing) the ARENH price.

B. (Lack of) State Aid Analysis of the French Regulated Prices for Gas Directive 2003/55/EC, replaced by Directive 2009/73/EC, imposes on Member States an obligation to ensure that consumers are free to buy gas from the supplier of their choice as from 1 July 2007. But those Directives also permit Member States to impose ‘public service obligations’ on undertakings operating in the gas sector, which may in particular concern the price of supply.10 In France, regulated prices for gas have always existed, in parallel to market prices (since the liberalisation). The privatisation of the incumbent, GDF–Suez (now Engie), drew specific attention to this issue. In a decision of 30 November 2006,11 the French Constitutional Council (Conseil constitutionnel) deemed the 2006 Energy Law implementing the 2003 EU Energy Directives to be essentially compliant with the French Constitution insofar as it transferred GDF to the private sector. However, it (partially) sanctioned the provisions relating to regulated gas (and electricity) prices, considering that they were incompatible with the objective of the market opening insofar as they were not limited to the ongoing contracts at the date of the market opening (1 July 2007) and not justified by precise public service requirements (as required by Article 106 TFEU), insofar as they concerned nonresidential customers. Ultimately however, the regulated prices were maintained, in this slightly modified format, allowing a de facto transitional period for the non-household regulated

8 The State subsidy would result from the fact that EDF—and consequently its main shareholder, the State—would perceive a price lower than costs and therefore forgo revenue. 9 eg, Opinion No 14-A-04 of 26 September 2014 on a draft decree amending decree No 2009-975 of 12 August 2009 regarding regulated tariffs for the sale of electricity; Opinion No 14-A-04 of 20 October 2014 on a draft decree amending decree No 2011-466 of 28 April 2011 fixing the modalities of the regulated access to historic nuclear electricity (ARENH) para 53. 10 Art 3(2). 11 Decision No 2006-543 DC of 30 November 2006—Act pertaining to the energy sector.

410 Liliana Eskenazi tariffs and the continuation of the household ones. As far as gas prices are concerned (but the situation for electricity was to some extent similar), it was not until the CJEU in Case C-121/15 (see II.B.iv below) declared them incompatible with EU law that they were kept alive, in spite of numerous proceedings at both EU and national level. i. Opinions of the FCA The FCA, in a couple of opinions on ministerial orders setting regulated gas tariffs, said that such tariffs have an unfavourable influence on the competitive process without making any positive contribution to the competitiveness of French companies or the purchasing power of households. The FCA therefore recommended that the regulated prices should be removed progressively, so as to allow the development of more competitive offerings, to the benefit of consumers.12 However, for obvious reasons, the FCA has not carried out any State aid law analysis. ii. Infringement Proceedings Initiated by the Commission Infringement proceedings against France (and some other Member States) had been opened already back in 2006 for failure to correctly implement the Energy Directives.13 The proceedings, however, did not give rise to a court action. It was only some six years later, on 31 May 2012, that the Commission called on France to revise its system of regulated prices for non-household users in order to bring its legislation in line with EU law.14 The Commission considered that, as regards non-household users, French legislation failed to comply with the compulsory time limit upon the application of regulated prices (the Energy Code does not provide for such time limit) and with the proportionality which should characterise all regulated tariff systems as specified in the Federutility judgment of the CJEU. Any such provision should be aimed at achieving a precise objective, and the legislation should not go beyond what is required to pursue the intended general economic interest. The Commission therefore considered France’s protection of prices for all non-household users, regardless of their size and situation, to be disproportionate. Following the Commission’s reasoned opinion, the relevant French legislation was modified to schedule the disappearance of regulated gas prices for non-residential customers within a period of between 19 June 2014 and 31 December 2015 (depending on the category of users).

12 Opinion No 13-A-09 of 25 March 2013 on a draft decree regarding regulated tariffs for the sale of natural gas. 13 See Commission, ‘Press Release of 4 April’ IP/06/430 and ‘Press Release of 12 December 2006’ IP/06/1768. 14 Commission, ‘Press Release of 31 May 2012’ IP/12/542. The Commission recalled again the basic principle of the Energy Directives, ie, that under EU internal energy market legislation, prices should primarily be determined by supply and demand and that, as a consequence, regulated prices impede new entrants from entering the market.

France

411

iii. Numerous Cases before the French Supreme Administrative Court (Conseil d’Etat) Meanwhile, the Conseil d’Etat annulled or suspended several ministerial orders (for reasons unrelated to State aid) since the regulated tariffs for gas were set at a level that did not cover the average total costs of the supply of natural gas, in contradiction to Article L 445-3 of the Energy Code that imposed this condition.15 In the most recent of these challenges, the Conseil d’Etat finally decided to refer a preliminary question to the CJEU as to the compatibility of the regulated gas prices with EU law, including Article 106 TFEU. iv. Preliminary Reference to the CJEU (Case C-121/15) and Follow-Up Judgment of the Conseil d’Etat In a judgment following the referral by the Conseil d’Etat of 15 December 2014,16 the CJEU held that the regulated tariffs in the French case are not the result of a free determination deriving from the interplay of supply and demand in the market. On the contrary, those tariffs are the result of a determination made on the basis of criteria imposed by the public authorities, which is outside the dynamics of market forces and therefore contrary to the EU Gas Directive. The Court concluded that such regulation is by its very nature an obstacle to the achievement of a competitive natural gas market, and that the obstacle exists even though suppliers can make competing offers at prices lower than the regulated prices. Assessing the French mechanism in light of the Federutility judgment, the Court deemed that security of supply and territorial cohesion are objectives of general interest which may justify State intervention in fixing the price of supply of natural gas. However, in this case, the permanent regulation of tariffs at national level, imposed only on certain undertakings in the gas sector (essentially the incumbent), might prove discriminatory and go beyond what is necessary. In its follow-up judgment delivered on 19 July 2017, the Conseil d’Etat, implementing the conclusions of the CJEU preliminary ruling, held—unsurprisingly—that the French legislative provisions setting up the regulated gas prices were incompatible with EU law. The Conseil d’Etat even deemed it unnecessary to examine all three of the criteria set out by the CJEU ((i) objective of general economic interest, (ii) proportionality and (iii) non-discriminatory, transparent and controllable nature), holding that the French regulated gas prices could not be justified by any of the objectives of general economic interest accepted by the CJEU and therefore constituted as such an obstacle to the achievement of a competitive natural gas market.17 More interestingly, there was a debate over the effects in time of the annulment of the decree implementing the regulated gas prices, which was challenged before the French court. The Conseil d’Etat acknowledged that incompatibility with EU law

15 16 17

eg, CE 10 July 2012, No 353356 or CE 20 November 2012, No 363572. CE 15 December 2014, No 370321. CE 19 July 2017, No 370321, para 15.

412 Liliana Eskenazi could only justify a time limitation of the effects of the annulment in very exceptional circumstances.18 Considering the circumstances of this case, namely the potential legal impact on more than nine million regulated price contracts with consumers, which could potentially be challenged, the Conseil d’Etat decided to limit the retroactive effect of the annulment in the name of the principle of legal certainty.19 However, it did not accept the request for a transitional period post-judgment.20 It is therefore at this stage unclear how the full transition to market prices will take effect in practice.

III. STATE AID TO THE RENEWABLE ENERGIES IN FRANCE

A. Background of the French Feed-In Tariffs (FIT) Case In France, to develop wind power, the State has implemented since 2000 a system of incentives, and in particular a purchase obligation. Electricity distributors must purchase electricity produced by wind energy producers who request it when wind installations meet certain conditions, at a purchase price set under a ministerial order. The electricity distributor passes the overcharge on to its customers through a contribution (CSPE) in a proportion which corresponds to the amount of electricity consumed.21 A French Ministerial Order dated 17 November 2008 (the 2008 Order) laid down the conditions of purchase of electricity produced by installations using wind energy. It established a fixed guaranteed purchase price of 8,2 eurocents kWh for a 10-year period (the contract being awarded for 15 years starting from the commission date of the installations), adjusted on the basis of the hourly labour cost index and of the producer price index.

B. Challenge of the FIT before the Conseil d’Etat and Preliminary Ruling of the CJEU (Case C-262/12 Vent de Colère) An action for cancellation of the 2008 Order was initiated before the Conseil d’Etat, by several groups of opponents, including the association Vent de Colère. The Conseil d’Etat, in a decision Association Vent de Colère! Fédération Nationale et autres of 15 May 2012,22 suspended the cancellation proceedings until the CJEU 18 The Conseil d’Etat referred to case C-379/15 Association France Nature Environnement, EU:C:2016:603. 19 CE 19 July 2017, No 370321, para 19. 20 In its judgment, the Conseil d’Etat even considered that the effects of the annulment could not go beyond 31 December 2015, the date on which the decree that was challenged in this case was repealed by another decree, which codified the regulatory provisions implementing the regulated gas prices into the Energy Code. 21 The possibility for wind energy producers to benefit from the incumbent (EDF)’s obligation to purchase the produced electricity, at a fixed price, results from Law No 2000-108 dated 10 February 2000 on the modernisation and development of the electricity public service. 22 CE 15 May 2012 Association Vent de Colère! Fédération Nationale et autres No 324852.

France

413

replied to a preliminary question as to whether the mechanism of compensation of additional costs related to the power purchase obligation should be considered as an intervention of the French State or as a mechanism including French State resources. The CJEU responded positively to this question. The Court found that the offset mechanism is attributable to the French State as it was established by legislation. As regards the State resources question, the Court observed that the sums intended to offset the additional costs arising from the obligations to purchase are available to the French authorities, as they are managed by the Caisse des Dépôts et Consignations (a French State-owned special financial institution), which acts as an intermediary in the management of those funds and under an express mandate from the French State to provide administrative, financial and accounting management services for the CRE (for a detailed analysis of the preliminary ruling see chapter two).

C. Consequences of the CJEU Preliminary Ruling i. Judgment of the Conseil d’Etat Annulling the FIT Decree On 28 May 2014, implementing the CJEU preliminary ruling, the Conseil d’Etat annulled the disputed 2008 Ministerial Order setting out the prices and conditions of the purchase obligation for wind energy.23 It also rejected a request that aimed at limiting the effects of the annulment in the time. It referred, in this regard, to the CJEU ruling, according to which, in the case at hand, there was no relevant element and in particular no risk of serious difficulties that could justify such temporal limitation.24 ii. Opinion of the Conseil d’Etat of 22 July 2015 in the Context of Follow-On Actions Requesting the Reimbursement of the CSPE In the context of a reimbursement claim made by a company, Praxair, before the Paris (first-instance) Administrative Court filed in 2011 (ie, before the CJEU ruling and the Conseil d’Etat judgment in Vent de Colère), Praxair sought a reimbursement of the CSPE it paid as a result of the FIT scheme, alleging that the CSPE was part of a mechanism that constituted illegal State aid and that was, consequently, subject to recovery. While the Paris Administrative Court rejected the claim, the Paris Administrative Court of Appeal decided to request the Conseil d’Etat’s opinion on whether the CSPE is an integral part of the State aid. The Conseil d’Etat, in its opinion of 22 July 2015, relied on the case law of the CJEU,25 according to which a tax is regarded as forming an integral part of an aid, if it is hypothecated to the aid, in the sense that the revenue from the charge has a direct impact on its amount. The Conseil d’Etat thus observed that the amount of

23 24 25

CE 28 May 2014 Association Vent de Colère! Fédération Nationale et autres No 324852. Case C-262/12 Vent de Colère and Others, EU:C:2013:85, paras 38–44. Case C-333/07 Société Régie Networks, EU:C:2008:764.

414 Liliana Eskenazi the aid, in other words of the obligation to buy the electricity produced in the onshore wind installations above market price, does not depend on the revenue from the CSPE. Hence, the Conseil d’Etat concluded that the CSPE cannot be regarded as forming an integral part of an aid. The Conseil d’Etat therefore confirmed the interpretation of the first-instance Administrative Court and indicated that the illegality of the State aid which was not notified to the Commission, does not result in the illegality of the CSPE. As a consequence of this opinion, the courts (including in the original Praxair case), as well as the CRE who was literally overwhelmed by more than 50,000 claims of the same nature, should refuse all request for reimbursement of the CSPE paid as a result of the FIT mechanism. iii. Further Judgment of the Conseil d’Etat of 15 April 2016 By a decision dated 15 April 2016, the Conseil d’Etat ordered the French State to recover the illegality interest related to the implementation of the FIT resulting from the cancelled 2008 Ministerial Order. It is interesting to note that the applicant Vent de Colère had to introduce a separate request for that purpose, since the Conseil d’Etat did not examine the recovery issue in the main proceedings that led to the annulment. In this instance, the Conseil d’Etat referred to the CELF case law of the CJEU, according to which where the State has implemented a non-notified aid, which was later deemed compatible by the European Commission, the national court did not need to order the recovery of the aid itself, however it had to order the repayment of the illegality interest.

D. Parallel State Aid Proceedings before the Commission (Case SA.36511) On 11 October 2013, ie, while the proceedings initiated by Vent de Colère before the Conseil d’Etat were still ongoing, and shortly before the CJEU delivered its preliminary ruling, but the outcome seemed rather clear and predictable, France notified to the Commission of its support scheme for the production of electricity from onshore wind installations. The Commission opened in-depth investigations on 27 March 2014. Its decision of that date is two-fold: on the one hand, it concluded that the FIT mechanism was compatible State aid and the case was therefore closed on this point; on the other hand, it expressed concerns regarding the financing mechanism and, in particular, the CSPE caps (tax reductions) for large consumers. i. Compatibility Decision on the FIT The Commission assessed the measure in light of section 3.1.6.2 of the 2008 Environmental and Energy State Aid Guidelines (the 2008 EEAG). It considered that the FIT allows producers of renewable energy to cover the additional production costs they face compared with traditional electricity generation. The Commission’s investigation also showed that the FIT only compensates for these additional costs and allows a reasonable return rate, in line with the EEAG. On this basis, the Commission concluded that the FIT scheme is compatible with EU State aid rules.

France

415

ii. Initiation of a Formal Investigation Procedure on the CSPE Caps (Tax Reductions for Large Energy Consumers) a. State Aid Analysis The Commission further examined the surcharge reduction with regard to State aid rules. To finance the FIT, every electricity consumer in France has to pay a uniform levy per kWh consumed (CSPE). However, the law provides for three exceptions: no CSPE on own consumption of auto-produced electricity below 240 GWh per year; no CSPE beyond €550 000 per consumption site per year; and CSPE capped at 0.5 per cent of their annual value added for industrial companies consuming at least 7 GWh per year. The Commission voiced its concerns that these three exceptions are selective, as they appear to give large electricity consumers a selective advantage and cannot be explained by the logic of the CSPE. Accordingly, the Commission cast doubts as to the compatibility of the CSPE caps with the single market. The Commission therefore invited France to provide more details concerning the objective pursued by each of the caps, the information allowing to determine whether the amount of the aid is proportionate to the aim pursued and why the potential positive effects of the aid counterbalance the possible distortion of the competition. b. Analysis Under the 2014 EEAG The possibility of such reductions to the surcharge was not foreseen under the 2008 EEAG, applicable during the proceedings in question. However, the Commission revised these guidelines on 9 April 2014 (the 2014 EEAG), including provisions allowing reductions for energy-intensive users under certain conditions. The 2014 EEAG now allow Member States to partially relieve cost burdens for a limited number of energy-intensive sectors or individual companies. Accordingly, the Commission will only approve aid that is limited to beneficiaries who are exposed to a competitive risk26 (undertaking belonging to the sectors listed in Annex 3, or alternatively companies with a high ‘electo-intensity’27 and which are active in a sector exposed to international trade).28 Moreover, Member States need to ensure that the choice of the aid beneficiaries is made on the basis of objective and nondiscriminatory criteria.29 Finally, the aid should be proportionate, in other words the aid beneficiaries must pay at least 15 per cent of the full renewable surcharge, provided that Member States have the possibility to further limit the amount of the renewable surcharges to be paid under certain circumstances.30 The 2014 EEAG apply retroactively to the assessment of reductions in the financing of renewables for energy-intensive users. Therefore, they apply to the State aid case in question. 26

2014 EEAG, pt 185. Electricity costs accounting for at least 20% of gross-value added. 28 A trade intensity of at least 4%, calculated as total trade of the sector with third countries, relative to the market in the EU. 29 2014 EEAG, pt 187. 30 ibid, pts 188–89. 27

416 Liliana Eskenazi iii. Modification of the CSPE to Comply with the State Aid Concerns: New Environmental Tax In order to comply with the State aid concerns, France has reformed the CSPE. The provisions governing the CSPE of the Energy Code were repealed by the Amending Finance Law of 201531 and incorporated into the existing tax on energy consumption (taxe intérieure sur la consommation finale d’éléctricité—TICFE).32 The scope of the tax is expanded in order to take into account all power consumption. Its level is fixed at 22.5 €/MWh.33 In addition, the Amending Finance Law of 2015 created a special allocation account (compte d’affectation special’ in the State budget to fund the support for the public service of electricity— Transition énergétique.34 In order to ensure compatibility with section 3.7 of the 2014 EEAG, more limited exemptions and tax reductions for energy-intensive consumers were created. Accordingly, electricity-intensive industrial companies and energy-intensive undertakings benefit from the reduced tariffs (from 0.5€/MWh to 7.5€/MWh).35 These tariffs have been already notified to the Commission and will enter into force once the Commission takes a decision to approve them.

IV. CAPACITY MECHANISMS

The French government put in place a whole new market-based capacity obligation mechanism,36 the purpose of which is to ensure security of supply. According to the Commission, which conducted an investigation, such a mechanism is liable to constitute State aid within the meaning of Article 107 TFEU, insofar as the money received in exchange of a capacity certificate would be an additional reward on top of electricity sales to the providers of electricity capacity.

A. Legislative and Regulatory Framework in France The legal, technical and financial aspects of the French capacity mechanism are set out in an order of 22 January 2015, pursuant to Article 2 of Decree No 2012-1405 of 14 December 2012.37 Law No 2010-1488 of 7 December 2010 introduced a

31

Art 5, III of the Amending Financial Law for 2015 No 2015-1786 of 29 December 2015. ibid, Art 14. 33 ibid, Art 14, I, B. 34 ibid, Art 5, I. 35 ibid, Art 14, I, C, (a). 36 See http://www.cre.fr/marches/marche-de-gros/marche-des-garanties-de-capacite. 37 See Decree No 2012-1405 relatif à la contribution des fournisseurs à la sécurité d’approvisionnement en électricité et portant création d’un mécanisme d’obligation de capacité dans le secteur de l’électricité; Ministerial Order (Arrêté) of 22 January 2015 relatif à la contribution des fournisseurs à la sécurité d’approvisionnement en électricité et portant création d’un mécanisme d’obligation de capacité dans le secteur de l’électricité, JORF No 0021 of 25 January 2015, 1143 text no 1. 32

France

417

new obligation on all electricity suppliers located in metropolitan France, and some network operators and end consumers to contribute to a security of supply scheme in proportion to the amount/volume of energy they or their clients consume (the certificate matches their energy portfolio). The capacity obligation serves as a kind of insurance that operators will always ensure the equilibrium between the instantaneous demand and the supply of electricity on the national grid.38 These rules are now codified in the Energy Code.39 More precisely, the suppliers’ capacity obligation (garantie de capacité) may be fulfilled through their own means (demand side: in-house through production plants when vertically integrated, or by means of demand response) or through the acquisition of capacity obligations on a decentralised market (supply side: capacity providers, other suppliers). Capacity providers (exploitants de capacité de production ou d’effacement) need to certify their capacity. This is the role of the TSO (RTE) which, in addition to certifying and issuing capacity certificates, administers the whole capacity mechanism. The capacity certificates can be traded over the counter or on specific platforms. The price of the capacity certificates is freely determined on the market and it is therefore not possible to know the price in advance. Besides, a financial penalty procedure is activated when there is a spread/imbalance between the ex ante certified capacities and the ex post effective amount of capacity released.

B. Opinions of the French Competition Authority (FCA) The FCA has been asked several times to give its opinion on the generation adequacy issues40 from a competition law perspective. These opinions fall outside the scope of this chapter, as (again, understandably) they do not address State aid issues. It is nevertheless worth pointing out that, in its Opinion No 13-A-25 of 20 December 2013, concerning the demand response mechanism detailed in Law No 2013-312 of 15 April 2013,41 the FCA considered that the financial reward in exchange for a demand response service should be notified to the Commission as it may be constitutive of State aid.

38 H Gaymard and C Valter ‘Rapport fait au nom de la Commission d’enquête sur les tarifs de l’électricité’ (2015) 2618 French National Assembly, 15: www.assemblee-nationale.fr/14/pdf/rap-enq/r2618.pdf. 39 Art L 335-1 and ff. 40 Opinion No 13-A-25 of 20 December 2013 on consumption interruption in the electricity sector, pts 85 et seq; see also Opinion No 12-A-09 of 12 April 2012 on a draft decree regarding the establishment of capacity mechanisms in the electricity sector; Opinion No 12-A-19 of 26 July 2012 on consumption interruption in the electricity sector. 41 Law No 2013-312 of 15 April 2013 aiming at preparing the transition towards a low-carbon energy generation system including a number of arrangements on water tariffs and wind turbines. Art L 123-1 and following the Energy Code, stemming from Law 2013-312, provides for the introduction of the reward allowing for the development of consumption interruptions for the community’s benefit. This reward will be financed by the CSPE, a tax paid by all electricity consumers in proportion to the volume of electricity consumed. Art L 271-1 provides that the reward setting methods will be detailed in a decree subject to the opinion in question.

418 Liliana Eskenazi C. Conseil d’Etat Judgment of 9 October 2015 and Aborted Preliminary Reference to the CJEU In an intermediate ruling of 9 October 2015, the French highest administrative court concluded that the capacity mechanism scheme did not infringe the State aid rules considering the lack of use of State resources, but further decided to refer the scheme for a preliminary ruling regarding its compatibility with the EU free movement rules.42 This case was, however, withdrawn from the Conseil d’Etat by the parties who initiated it and, as a consequence, the preliminary question was also removed from the CJEU’s registry.

D. Commission Investigation into the French Capacity Mechanism (Case SA.39621) i. Procedural Aspects As the French authorities did not notify the measure as required by Article 108(3) TFEU, the Commission sent a request for information to which the French authorities replied. Later however, France notified the measure as a public service obligation under Article 3(15) of Directive No 2009/72/EC.43 Nevertheless, the Commission decided to initiate a formal State aid investigation procedure and, in its Decision of 13 November 2015, it concluded (provisionally) that the measure may constitute State aid within the meaning of Article 107 TFEU. The Commission had concerns that the planned capacity mechanism might favour certain companies over their competitors and hinder the entry of new market players. ii. Commission Compatibility Decision of 8 November 2016 While the Commission confirmed its preliminary analysis that the French capacity mechanism does involve State aid, it finally concluded that such aid is compatible with the internal market, following some amendments proposed by the French government. a. Existence of State Aid State Resources The Commission considered, first, that the fact that the trade of certificates gives beneficiaries a market value and that these certificates were awarded for free by the French authorities is equivalent to forgoing public resources.44 Moreover, according to the Commission, this case was no different from the Romanian green certificates case.45

42

CE 9 October 2015, No 369417. Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity [2009] OJ L211/55. 44 French country-wide capacity mechanism (Case SA.39621) [2017] OJ L83/116, para 200. 45 ibid, para 203. 43

France

419

The Commission also considered the tender process provided in the capacity mechanism to be a possible direct State intervention on the market.46 Finally, some changes made by France to the capacity mechanism were to be qualified as State aid themselves, in particular the multi-year contracts due to the important role played by the State in this mechanism.47 Selective Advantage The Commission maintained its assessments and conclusions in its opening decision according to which the mechanism confers a selective advantage to the capacity operators. The mechanism was deemed selective as it targets capacity operators and no other sector of the economy.48 Effect on Competition and Trade between Member States The Commission also maintained its conclusion, according to which the measure opened to French metropolitan undertakings only is detrimental to competition and trade within the EU.49 Unlawful Aid The Commission considered that by implementing the capacity mechanism, while the Commission had not made a final decision on the alleged measure, France had violated Article 108(3) of the Treaty.50 b. Proposed Remedies In order to alleviate the Commission’s concerns, France agreed to modify the capacity mechanism in the following ways: — —



The new capacities, which are more competitive than existing capacities, can obtain certificates with a seven-year duration instead of the one-year duration. The French capacity mechanism will include foreign capacity providers, both generators and demand response operators, subject to the extended capacity of the interconnector at peak times. A series of measures will be introduced to prevent possible market manipulation.

Based on these changes, the Commission found that the remedies proposed by the French authorities adequately address concerns raised in its opening decision. The Commission also confirmed the positive features of the capacity mechanism already proposed in its original design. c. Compatibility Assessment: Section 3.9 EEAG Taking into account the different remedies proposed by the French authorities, the Commission concluded that the alleged measure complies with the EEAG. 46 47 48 49 50

ibid, para 204. ibid, para 205. ibid, para 207. ibid, para 142. ibid, paras 213–15.

420 Liliana Eskenazi Objective of Common Interest and the Need for State Intervention According to the Commission, the mechanism, the main objective of which is the security of supply, is necessary in France. It does not contradict the objective of the EEAG to phase out environmentally harmful subsidies,51 as it facilitates the demand management, increases the interconnection capacities, includes the contribution of renewable energy sources in the mechanism and gives preference to producers emitting insignificant amounts of carbon. Appropriateness of the State Intervention In its opening decision, the Commission raised numerous concerns regarding the measure’s ability to incentivise investment, as there are too many unknown variables such as the price of the certificates, the number of clients and price volatility. However, according to the Commission, the remedies proposed by the French authorities adequately addressed these points of concern, and the measure was therefore deemed appropriate according to its objectives.52 Incentive Effect The Commission maintained its assessment in the opening decision that the mechanism has an incentive effect. The scheme does encourage capacity providers to maintain capacity available during peak hours and suppliers to keep an adequate number of certificates to accommodate the needs of their client portfolio. The whole system is regulated by a financial penalty mechanism.53 Proportionality Taking into account the remedies proposed by the French authorities, the Commission found that the alleged mechanism is proportionate to its objective. As regards the overcompensation of consumption, the Commission concluded that the measures envisaged by the French authorities to help the suppliers better calculate their capacity obligations address the concern raised in its opening decision.54 Concerning the lack of transparency in price fixing, the Commission found that the remedies proposed by the French authorities, such as free access to the anonymised transaction registry for all operators and the strengthening of the organised auctions, will foster transparency in the market, increase the liquidity of the organised market and facilitate market surveillance by the regulator.55 Furthermore, the Commission found that the mechanisms proposed for the participation of foreign capacities and for the multi-year contractualisation of the new capacities is proportionate.56

51

ibid, paras 219–30. ibid, paras 231–54. 53 French country-wide capacity mechanism (Opening Decision) (Case SA.39621) [2016] OJ C46/35, paras 185–88. 54 French country-wide capacity mechanism (n 44) paras 256–58. 55 ibid, paras 259–69. 56 ibid, paras 270–74. 52

France

421

Finally, the Commission concluded that the remedies proposed by the French authorities address its concern that the incumbent EDF could easily manipulate the prices of capacities to its benefit.57 The proposed amendments aim, inter alia, (i) to organise more frequent capacity auctions to resolve the capacity retention risk and (ii) to regulate the free transfer of certificates, which will not be possible within a group, to avoid the margin squeeze risk. The proposed remedies also aim at creating transparent access to the (anonymous) registry of over-the-counter capacity certificate transactions.58 Avoidance of Undue Negative Effects on Competition and Trade The Commission concluded that the capacity mechanism does not distort competition. First, the Commission found that a longer contract duration for new capacities will facilitate the entry of new market players by giving sufficient investment certainty for new projects.59 Second, the Commission considered that differentiation between explicit and implicit demand response is necessary to allow an optimal participation of two types of capacity demand response.60 Third, the French authorities agreed to explicitly include and remunerate foreign capacities. The Commission also added that the remedies proposed by the French authorities to improve the visibility of the exchanges for all operators will rebalance information asymmetries between the dominant incumbent and its actual and potential competitors.61 Finally, the Commission noted that, for the new capacities, the French authorities intend to introduce the environmental criteria leading to give preference to producers with insignificant carbon emission.62

E. Commission Investigation into the Tender to Support the Construction of a New Gas-Fired Power Plant in Brittany (Case SA.40454) In parallel to the development of a country-wide capacity mechanism, France has launched a tender to support the construction of a new gas-fired power plant (Combined Cycle Gas Turbine—CCGT) in Brittany. The aim is to increase electricity generation capacity in this region, which due to its peninsular geography is not well connected with the rest of France.63 The Commission had concerns, in particular, that support is granted to only one type of technology and is not open to other potential solutions, for example, those also being implemented in Brittany (as planned in the pacte électrique Breton), namely other types of power generation, demand-side management,

57 58 59 60 61 62 63

ibid, paras 275–89. ibid, para 287. ibid, paras 290–91. ibid, para 292. ibid, paras 296–98. ibid, paras 299–300. See Commission, ‘Press Release of 13 November 2015’ IP/15/6077.

422 Liliana Eskenazi network extensions or storage solutions. At the stage of the preliminary decision, the Commission considered that there is a risk of creating a subsidy dependent market, where investors will develop projects only on the basis of public tenders granting State aid. Therefore, according to the Commission, the tender could risk exacerbating the adequacy problem in the long term.64 Ultimately, by a decision of 15 May 2017, the Commission deemed the measure to be in line with EU State aid rules because it is both necessary and proportionate to address the security of electricity supply concerns in Brittany. The Commission’s approval is, however, subject to the condition that the successful tenderer, CEB,65 will not sell output from the Landivisiau power plant through long-term contracts to any undertaking with a share of over 40 per cent of the French electricity-generation capacity market. i. Preliminary Assessment and Final Conclusion of the Commission that the Tender Constitutes State Aid and that the Altmark Criteria do not Apply According to the terms of the tender, the successful tenderer was to receive (for a period of 20 years starting when the plant would become active) financial incentives based on several factors such as: (i) the capacity generated; and (ii) surcharges specifically due to the geographical situation of the plant (overcharges due to gas transportation to the plant, interconnections, environmental regulation).66 The French authorities were of the opinion that the measure does not constitute State aid. Nonetheless, and for the sake of legal certainty, they notified it to the Commission. The Commission’s preliminary conclusion was that the measure is imputable to the State as the tender has been launched by the State (Ministry in charge of Energy). The measure was also considered by the Commission as entailing State resources as it will be passed on to retail price through the Contribution au Service public de l’Electricité (CSPE).67 Furthermore, the Commission was of the opinion that the measure creates a selective advantage that does not fulfil the Altmark exemption (the SGEI,68 the absence of overcompensation69 and the selection70 criteria were deemed not fulfilled). These findings were confirmed in the final decision.71 Regarding, in particular, the analysis of the Altmark criteria, the Commission considered that there could be no public service obligation (i) because the (electricity-supply) activity is already provided or could be provided satisfyingly by the market72 and (ii) because of the

64 65 66 67 68 69 70 71 72

ibid. Direct Energie–Siemens consortium and plant will be located in Landivisiau (Finistère). Tender for additional capacity in Brittany (Opening Decision) (SA.40454) paras 54–58. ibid, paras 61–65. ibid, paras 69–85. ibid, paras 88–89. ibid, paras 90–102. Tender for additional capacity in Brittany (SA.40454) [2017] OJ L235. ibid, para 143.

France

423

technological discrimination in the tender (which opted for the combined cycle technology).73 The non-discrimination criterion, provided for in Article 3(2) of the Electricity Directive 2009/72/EC has, according to the Commission, to be interpreted strictly. ii. Compatibility Assessment under Section 3.9 of the 2014 EEAG Since the Commission considered that there was no public service obligation, it concluded that the SGEI Framework did not apply. It therefore proceeded with assessing the measure in light of section 3.9 EEAG. Contrary to what it had considered provisionally in the opening decision,74 the Commission ultimately concluded that the support complied with the exemption framework, because it contributes effectively to the recognised objective of common interest (security of supply). The measure is deemed necessary and adequate to match an identified market failure.75 The Commission pointed out, in particular, that the recently approved French capacity mechanism was not sufficient to address the specific local demand in Brittany.76 Besides, the Commission approved the adequacy of the scheme as no alternative means (such as renewable energy sources, existing conventional production sites, the interconnection investment level or demand response) could attain efficiently the objective of security of supply.77 The incentive effect and the proportionality of the measure were then deemed compliant with sections 3.2.4 and 3.9.5 EEAG. As regards the avoidance of undue negative effects on competition and trade, in line with section 3.9.6 EEAG, the Commission examined various alternative technologies in order to conclude that the CCGT remained qualitatively the most efficient one. Accordingly, the measure is designed in such a way that makes it possible for any capacity which can effectively contribute to addressing the generation adequacy problem to participate.78 Finally, the Commission assessed the condition according to which the measure should not ‘unduly strengthen market dominance’.79 Insofar as the beneficiary is entitled to sell the electricity produced to EDF at a 5 per cent discounted rate, or has the possibility to conclude a tolling agreement (notably with EDF) allowing it to sell a fixed amount of the electricity produced, the Commission considered that the measure could reinforce EDF’s market power and distort the EU level playing field and trade within the internal market.80 EDF is still the dominant player on the French electricity market, with 83.5 per cent of the total electricity production and 89.4 per cent of the total capacity installed in France.81 EDF also has 86 per cent of

73 74 75 76 77 78 79 80 81

ibid, para 144. Tender for additional capacity in Brittany (Opening Decision) (SA.40454) paras 111–21. Tender for additional capacity in Brittany (n 71) paras 164–78. ibid, para 177. ibid, para 183. ibid, paras 197–205, referring, in particular, to para (232) EEAG. EEAG, para (233)(d). Tender for additional capacity in Brittany (n 71) paras 210–18. ibid, para 212.

424 Liliana Eskenazi the retail market. To counter the risk of further strengthening EDF’s dominance, the Commission therefore required the French authorities to take the necessary steps to ensure that the aid recipient does not, for the entire duration of the aid, sell the electricity of the power station to an operator which would have more than 40 per cent of electricity generation capacity on the French market, whether through a tolling agreement or a long-term sales contract at a price equal to 95 per cent of the market price. Subject to this commitment, the aid was ultimately deemed compatible.

16 Netherlands MARINUS WINTERS

I. INTRODUCTION

A. Introduction to the Electricity Wholesale and Generation Market in the Netherlands

T

HIS CHAPTER IS focused on the main State aid issues relevant to the Dutch market for the wholesale and generation of electricity. In the Netherlands, there have been no State aid issues with respect to the retail markets for electricity. In addition, it should be noted that there have been no State aid issues in relation to the gas market, except for an investigation that the European Commission (Commission) opened in 2009 in relation to an envisaged tax exemption on the purchase of natural gas by companies producing ceramic products.1 The Commission determined that this envisaged tax exemption was not in line with the internal market, which means that the Dutch government did not apply this exemption. Since this tax exemption was in fact State aid for the ceramics industry as opposed to the energy sector, this case will not be discussed in more detail.

B. The History of the Dutch Market: Electricity Wholesale and Generation In the 1990s, the previous Electricity Act obliged electricity producers to work closely together and to pool their costs through the Dutch Electricity Generation Board (Samenwerkende Elektriciteitsproductiebedrijven: Sep). Sep was the central planning and despatching entity and responsible for the procurement of fuel. Sep supplied electricity to the regional distribution companies at uniform prices: Sep was in fact a legalised cartel. Sep conducted a policy of fuel diversification and an environmental policy as well. This meant that Sep agreed on (i) long-term import contracts for electricity and gas, and developed (ii) district heating systems in various cities in the Netherlands, (iii) a nuclear power plant and (iv) a coal regasification plant. The additional costs

1 Tax Exemption from Environmental Taxes for Ceramic Products (Case C5/2009) Commission Decision 2010/402/EU [2010] OJ L186/32.

426 Marinus Winters of the development of these assets were invoiced to the distribution companies and therefore these costs were in fact socialised. In 1998 the Dutch government enacted the (current) Electricity Act 1998 (Elektriciteitswet 1998: Electricity Act), which incorporated the first European Electricity Directive 96/92/EC.2 When the Electricity Act entered into force the production of electricity was liberalised. The Electricity Act no longer contained a permit obligation for the production of electricity. The liberalisation meant that the generators were free to determine the price for the production of electricity, which means that they were no longer able to invoice the extra costs of the investments made by Sep. As a result, the activities/investments mentioned under (i)–(iv) above were stranded costs. The Ministry of Economic Affairs developed a Transition Act for the Electricity Generation Sector for these stranded costs. This transition Act led to several court cases, which will be further discussed below. As soon as the production market was liberalised the generators that were united in the Sep were privatised to foreign companies. Companies like Reliant, Electrabel and E.ON (currently Uniper) entered the Dutch market by the acquisition of electricity production companies that were (mainly) owned by Dutch provinces.

C. The History of Subsidisation in Renewable Electricity Since the liberalisation of the electricity market the Dutch government has used several systems to incentivise the use and production of renewable electricity. Initially the Dutch government incentivised the use of renewable electricity by a tax exemption on the sale of renewable electricity. The Dutch government also opened the retail market for the sale of renewable electricity to small consumers earlier than the retail market for conventional electricity. The retail market for renewable electricity was already open to competition as of 1 July 2001, while the retail market for conventional electricity (only) opened on 1 July 2004. The tax exemption and the liberalisation (of the market) for renewable electricity was in fact a huge success. Many customers purchased renewable electricity, which meant that the utility companies had to purchase the renewable electricity in surrounding countries in order to meet their obligations towards their customers. The downside of this success was that this system in effect created a ‘tax leakage’, since the production of renewable electricity outside the Netherlands was subsidised by the tax exemption in the Netherlands. As a result, in 2004 the Dutch government decided to subsidise the production of renewable electricity in the Netherlands by a subsidy per kWh of generated renewable electricity. This subsidy was called the Environmental Quality Electricity Production (Milieukwaliteit Elektriciteitsproductie: MEP-subsidy) and the MEP-subsidy was based on the Dutch Electricity Act (Elektriciteitswet 1998). EnerQ BV, a group company of the Dutch State-owned Transmission System Operator (TSO), TenneT TSO BV (TenneT), was responsible for the grants and payments of the MEP-subsidy. 2 Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L27/20.

Netherlands 427 The problem with the MEP-subsidy was that this subsidy was an ‘open-end’ system, which means that every generator meeting the requirements under the Electricity Act was eligible for the MEP-subsidy. Therefore, the Electricity Act did not contain a fixed budget for the MEP-subsidy each year. The MEP-subsidy was also a success, which meant that the MEP-subsidy was expensive for the Dutch government. Therefore, the Dutch government abruptly stopped the MEP-subsidy in August 2006 by setting the MEP-subsidy tariff at €0/kWh. This decision led to political turmoil and several claims from companies which had already invested in the development of renewable installations under the legitimate expectation that they would receive the MEP-subsidy.3 After the MEP-subsidy the Ministry developed the SDE-subsidy scheme. As explained below, the SDE-subsidy scheme funds the uneconomic difference between the levelised costs of energy and the market price. The SDE-subsidy scheme is still the current subsidy scheme to stimulate the development of renewable energy projects in the Netherlands.

II. CASE LAW IN RELATION TO STRANDED COSTS

A. Introduction As indicated in the introduction to this chapter, the main Dutch generators cooperated within Sep before the process of liberalising the market began. Following the implementation of the first Electricity Directive in the Dutch Electricity Act, the market for the production and wholesale of electricity opened in 1999. The investments of the Sep became stranded costs and the Dutch government enacted the Transition Act for the Electricity Generation Sector (Overgangswet elektriciteitsproductiesector),4 which dealt with these stranded costs. Pursuant to the Transition Act for the Electricity Generation Sector, the former shareholders of the Sep were jointly liable—pro rata to their participation in Sep—for the costs as a result of: — — — —

The development of an experimental coal regasification unit, Demkolec, until the installation was taken over. The costs for the repayment of a loan from the former owner of a nuclear power plant to the Sep. Import contracts of gas and electricity. The costs due to the obligations of Sep in relation to the development of the interconnector between the Netherlands and Norway.

3 In two cases the Trade and Appeals Tribunal (College van Beroep voor het bedrijfsleven) determined that the Dutch Ministry of Economic Affairs had acted in breach of the principle of legitimate expectations (vertrouwensbeginsel) by setting the MEP-subsidy on €0.00/kWh before the subsidy was even granted. See also CBb, 9 December 2009, AWB 08/532, 08/533 and 08/534 and CBb, 9 December 2009, AWB 08/433 and 08/434. 4 Overgangswet elektriciteitsproductiesector, 21 December 2000, Staatsblad 2000, no 607.

428 Marinus Winters Furthermore, the Transition Act for the Electricity Generation Sector determines that the Dutch State provided compensation for the stranded costs in relation to district heating and to costs in relation to the sale of the experimental coal regasification plant Demkolec. Until the end of 2000, all parties connected to the Dutch grid were obliged to pay a levy of €0.0053 (NLG 0.0117) per kWh in order to cover the above-mentioned costs (households were obliged to pay this amount to suppliers, large consumers were obliged to pay this amount to the grid operators). This is laid down in Clause 9 of the Transition Act for the Electricity Generation Sector. This compensation scheme was notified to the Commission on 16 October 1998. Almost three years later, on 25 July 2001, the Commission issued a decision in which it raised no objections to the measures of the Dutch government.5 According to the aforementioned decision, the Dutch government had originally also notified a measure for the import contracts of gas and electricity, but later withdrew this part of the notification. In June 2001, the Dutch government also withdrew the financing mechanism, since the Commission was of the opinion that the levy qualified as a parafiscal levy. The compensation for the stranded costs for the district heating projects and Demkolec were assessed on the basis of the Commission Communication relating to the methodology (Methodology) for analysing State aid linked to stranded costs. The Methodology was determined on the same day as the decision to raise no objections (25 July 2011). In this decision, the Commission assessed the elements of Chapter 3 of the Methodology, which in fact defines which costs are eligible for stranded costs. The Transition Act for the Electricity Generation Sector led to several disputes between the energy companies, which were shareholders in the Sep, and the Dutch State. State aid was an element in some of these proceedings. The State aid elements in these proceedings will be discussed below. The Transition Act for the Electricity Generation Sector originally also contained a priority rule for the import contracts of electricity, which meant that TenneT was obliged to allocate sufficient import capacity to the Sep or its legal successor and their counterparties. This priority access would enable the generators to fulfil their obligations under the long-term import contracts with mainly German and French companies. However, the decision of the Dutch energy regulator (at that time, the Nederlandse Mededingingsautoriteit (NMa), currently the Autoriteit Consument & Markt (ACM)) to determine the allocation rules for the import capacity on the interconnectors was appealed at the Trade and Appeals Tribunal (College van Beroep voor het bedrijfsleven) by several market parties. These market parties were of the opinion that the priority access rules were discriminatory. The Trade and Appeals Tribunal asked for a preliminary ruling, which led to the judgment of the Court of Justice of 7 July 2005, in which the Court decided that the priority access rules were discriminatory and therefore these priority access rules were not in line with the First Electricity Directive.6 Although this is an important judgment of the Court,

5 Measures in Favour of the Electricity Market for Stranded Costs (Case N597/1998) [2001] OJ C268/5. 6 Case C-17/03 VEMW and others, EU:C:2005:362.

Netherlands 429 State aid did not play a role in the preliminary proceedings at the Court of Justice or at national level.

B. Essent Netwerk As indicated above, the Transition Act for the Electricity Generation Sector stipulated that everyone connected to the grid had to pay a levy of €0.0053 /kWh during the second half of 2000. Aluminium Delfzijl BV (Aldel), which is a large electricity consumer, had to pay an amount of approximately €4.5 million to Essent Netwerk BV, the distribution system operator of Aldel. Aldel refused to pay this levy and therefore Essent Netwerk claimed these costs back from Aldel before the Court in Groningen. Aldel stated that the levy qualified as illegal State aid for all electricity producers, which were shareholder in Sep. The Court in Groningen asked for a preliminary ruling, in which it (ao) asked the Court of Justice, whether the levy was in line with Article 107(1) of the Treaty on the Functioning of the European Union (TFEU).7 The ECJ determined that Article 107 TFEU must be construed as meaning that the amounts paid to the designated company under Article 9 of the Transitional Act for the Electricity Generation Sector constitute ‘State aid’ for the purposes of that provision of the TFEU in so far as they represent an economic advantage and not compensation for the services provided by the designated company in order to discharge public service obligations.8 The Court in Groningen did not render a final judgment.

C. Demkolec In 1989, Sep decided to develop a coal regasification unit in Buggenum (in the southern part of the Netherlands). The name of this unit was Demkolec and this was an experimental unit, which was not cost-effective from an economic perspective. Sep decided to construct the coal regasification unit for environmental reasons. As indicated above, Demkolec became a stranded cost as a result of the liberalisation of the market. It was decided that BV Nederlands Elektriciteit Administratiekantoor (NEA), which is the legal successor of Sep, would sell Demkolec in an auction. Pursuant to Clause 7(b) of the Transition Act for the Electricity Generation Sector, the Dutch State would subsequently fund the difference between the book value of Demkolec and the sale value of Demkolec in the aforementioned auction. This was also approved by the Commission in its State aid decision of 25 July 2001.9 After the sale of Demkolec, the Dutch State paid €134 million to NEA, which equals the book value of Demkolec and the sale value. Subsequently, NEA and the Dutch State entered into a dispute about the question of whether the exploitation

7 8 9

Court Groningen, 22 June 2005, NJF 2005, 296. Case C-206/06 Essent Netwerk Noord and others, EU:C:2008:413. Measures in Favour of the Electricity Market (n 5).

430 Marinus Winters losses and the interests were part of the stranded costs as well. NEA and the Dutch State decided to start arbitral proceedings in relation to this issue. On 7 September 2005, the Netherlands Arbitration Institute (NAI) decided that the Dutch State was obliged to pay an additional amount of €38 million, since the Dutch State was obliged to pay all stranded costs to NEA. The NAI furthermore decided that this was in line with the aforementioned State aid decision. As a result of the arbitral award, the Dutch State paid an additional €46 million, which is the €38 million plus costs and interest. The Dutch State subsequently went to the Court of The Hague and claimed that the arbitral award of 7 September 2005 was null and void, since this arbitral award was in breach of the European State aid rules. Pursuant to Clause 1065 of the Dutch Code of Civil Procedure, a Dutch Court can nullify an arbitral award if the award is in breach of the public order. The Dutch State argued that NEA received compensation, which exceeded the maximum amount of €136 million as indicated in the approval decision of the Commission. This appeal of the Dutch State was rejected by the Court of The Hague on 1 August 2007.10 In its judgment, the Court carefully assessed the arguments of the Dutch State in light of the State aid decision of the Commission. In this respect, the Court decided that the amount of €136 million (as included in the approval decision of the Commission dated 25 July 2001) is not a maximum amount. When the Commission adopted the approval decision the State aid amount could not be determined, since the auction of the Demkolec plant had not yet taken place. Therefore, the Court decided that the range of €90 to €136 million as included in the State aid decision was a preliminary estimate or an estimate and could not be considered as a maximum State aid amount. The Court furthermore decided that the State aid decision of the Commission did not determine the date of the book value of the Demkolec plant either. According to the Court, there is no exact date for the book value in the State aid decision, since the auction had not yet taken place. The Court finally considered that the approved auction system of the Demkolec plant meant that the exploitation costs and interests were stranded costs, since these costs led to a lower sale price of the Demkolec plant. On the basis of these arguments, the Court of The Hague determined that the arbitral award of the NAI was not in breach of the State aid decision of the Commission in relation to stranded costs and therefore there was no reason to nullify the arbitral award of the NAI.

D. District Heating Projects As indicated in the introductory paragraph of this chapter, the Transition Act for the Electricity Generation Sector determines that the Dutch State would provide compensation for the stranded costs in relation to district heating projects. In this respect, the Ministry determined a specific regulation, which calculated the compensation for

10

Court of The Hague, 1 August 2008, Dutch State/NEA, NL:RBSGR:2007:BB1424.

Netherlands 431 the generators of district heating projects.11 The regulation determined the compensation on the basis of a model. Through this, the MEP-subsidy received by the generators would be deducted from the compensation. E.ON Benelux NV (currently Uniper Benelux NV) appealed this regulation at the Court of Arnhem and stated that (ao) this deduction of the MEP-subsidy was not line with the State aid decision of the Commission, since the MEP-subsidy was not meant to cover stranded costs.12 The MEP-subsidy had a different purpose. The Court did not accept this argument. In this respect, the Court referred to the Commission’s State aid decision of 25 July 2001 and paragraph 3.8 of the Methodology, in which the Commission stated that: Stranded costs must be valued net of any aid paid or payable in respect of the assets to which they relate. In particular, where a commitment or a guarantee of operation corresponds to an investment which is the subject of State aid, the value of the aid must be deducted from any stranded costs resulting from the commitment or guarantee.

On the basis of this paragraph in the Methodology, the Court decided that aid which is meant to cover other costs must be deducted from the compensation. Therefore, the Court did not accept Uniper’s argument.

III. SDE-SUBSIDY

A. Introduction As indicated in the introduction to this chapter, the generation of renewable electricity in the Netherlands is stimulated by the SDE-subsidy. This section will explain the SDE-subsidy scheme and will discuss the assessment of the SDE-subsidy scheme by the Commission under the State aid rules of Clause 107 of the TFEU.

B. Explanation of the SDE-Subsidy System i. General Principle In 2008, the Dutch government introduced the subsidy for the Stimulation of Renewable Energy (Stimulering Duurzame Energieproductie: SDE-subsidy). The general rules of the SDE-subsidy are laid down in the SDE-Decree.13 The SDEsubsidy is a subsidy, which is granted by the Ministry of Economic Affairs for each kWh of generated renewable electricity. The purpose of the SDE-subsidy is to fund

11 Regulation of the Ministry of Economic Affairs, 20 January 2005, no WJZ 4081042, Uitvoeringsregeling Overgangswet elektriciteitsproductiesector, Staatscourant 2005, 22. 12 Court Arnhem, 22 June 2009, E.ON Benelux N.V./Ministry of Economic Affairs, NL:RBARN: 2009:BJ1171. 13 Besluit stimulering duurzame energieproductie, Staatsblad 2007, 410. The SDE-Decree has been amended several times.

432 Marinus Winters the uneconomical part of the generation of renewable electricity, which owners of renewable installations (such as wind farms, PV parks, but also producers of biogas) will not earn back through the sale of the generated electricity on the market under a Power Purchase Agreement. In general, the cost price of the generation of renewable electricity is higher than the market price. The cost price for the generation of renewable electricity by most renewable installations such as a PV park or a wind farm is fixed. The market price varies over time. Under the SDE-subsidy scheme, the Ministry will pay the difference between the (fixed) cost price and the (variable) market price during the subsidy term (10–15 years depending on the technology). ii. The Determination of the Subsidy Amount The SDE-subsidy will be paid for each generated kWh of renewable electricity. The amount of generated kWhs will be determined by the number of guarantees of origin (garanties van oorsprong), which will be redeemed by the producer of renewable electricity. The actual subsidy amount per year will be determined by multiplying the number of guarantees of origin by the ‘corrected basis amount’. The corrected basis amount will be determined every year by the Ministry by decreasing the basis amount with the market price. The market price is based on the average annual APX price level, which will be multiplied by: — the correction factor for imbalance costs;14 and — the correction factor for profile costs.15 The basis amount might, in the future, also be corrected by the value of the guarantees of origin (garanties van oorsprong) as well as other future corrections set by the Dutch government and which could have a substantial influence on the difference between the average cost price of the generation of renewable electricity and the average market price (Clause 14, sub 1 of the SDE-Decree). This means that the SDE-subsidy scheme will avoid an overstimulation due to high revenues, which will be generated by the sale of guarantees of origin. The average annual APX price-level and the factors for imbalance costs and profile costs will be determined annually by the Ministry with retroactive effect. Each year before 1 April, the Ministry determines the average annual APX price level and the factors for imbalance costs and profile costs of the previous year. The Ministry will take an expert assessment (of, eg, the Energy Research Centre, the Netherlands) regarding the price level and account for the aforementioned factors in order to determine the amount, which will be deducted from the basis amount.

14 Generators of electricity have programme responsibility. Due to differences between the projected production of electricity and the actual production, generators may encounter imbalance costs, which will be invoiced by the Dutch Transmission Operator TenneT, since TenneT is required to incur costs to level out these differences. 15 Profile costs are costs (or revenues) resulting from the difference between the value of the generated electricity and the average market value of the electricity.

Netherlands 433 The SDE-subsidy also contains the base electricity price (basiselektriciteitsprijs). The base electricity price is applicable for the whole term of the SDE-subsidy. In the market, the base electricity price is also called ‘the floor’. The maximum subsidy per kWh is capped at the difference between the basis amount and the base electricity price, which means that if the correction amount is lower than the base electricity price, the subsidy recipient will (only) receive the difference between the basis amount and the base electricity price. The figure below (Figure 1) illustrates an example of the SDE-subsidy system. The vertical axis indicates the price (in €/kWh) and the horizontal axis indicates the term of the subsidy (in years). The light grey line is the basis amount, which is a fixed amount/kWh. The dark grey line is the market price, which is (of course) a variable price. The dotted line indicates the base electricity price (or the floor). As indicated above, the Ministry will determine the market price each year and will pay the difference between the light grey line and the dark grey line to the company. If the market price is below the ‘floor’ (as shown in the figure in the third year), the Ministry will pay the difference between the light grey line and the dotted line. If the market price is above the basis amount (as it is in the sixth year in the example below) the Ministry will not pay a subsidy, since the generator should be able to earn back the costs of the generation of electricity with the Power Purchase Agreement, which the generator has entered into with a purchaser of the electricity, such as a utility company. The SDE-Decree contains two forms of ‘banking’. Production capacity that is eligible for subsidisation but has not been realised in a certain year (underproduction) can be carried over to the next calendar year with the possibility of realising the capacity and corresponding subsidy in that next year. The subsidy recipient can furthermore request to extend the subsidy term by no more than one year, to offset this underproduction.

Figure 1: SDE-Subsidy System

434 Marinus Winters iii. The Allocation of Available Funds Each year, the Ministry determines which technologies of renewable energy are eligible for the SDE-subsidy. The technologies, which are eligible for the SDE-subsidy will be determined in the Regulation determination categories of renewable electricity production (Regeling aanwijzing categorieën duurzame energieproductie). This Regulation also determines for each technology the basis amount (basisbedrag). The basis amount is a fixed amount during the term of the subsidy. The basis amount represents the cost price for the generation of renewable electricity by the respective technology. The Ministry determines this basis amount with the help of expert analyses of, for example, the Energy Centre of the Netherlands. These expert analyses carry out independent research about the cost price of the various technologies. The available funds will be released in different phases. The basis amount increases for each phase. Each technology has a maximum basis amount, above which no subsidy will be granted. Cost-effective technologies with a low basis amount can apply earlier than the more expensive technologies. Therefore, there is a greater chance that a budget will be available for the cost-effective technologies than for technologies with a higher maximum basis amount. In order to create competition between different technologies, developers can also apply for a lower subsidy than the maximum basis amount for the technology in question. Such applications can be made in a ‘free category’. This allows them to tailor their subsidy application to their business case. The amount of subsidy applied for in the ‘free category’ is lower than the maximum phase amount and therefore the developers who apply in the free category, have a greater chance of receiving an SDE-subsidy. The allocation of available funds is different for the SDE-subsidy which will be granted to offshore wind farms. The SDE-subsidy for offshore wind farms will be tendered together with a permit to construct, operate and maintain the offshore wind farms. These permits will be granted for special designated areas in the Dutch part of the Exclusive Economic Zone in the North Sea. The tenderer with the lowest tender amount (which in fact represents the levelised cost of energy) will be granted the SDE-subsidy as well as the permit for the offshore wind farm. For these tenders, the Ministry determines a separate budget. In 2016 the Ministry organised two tenders for the development of four offshore wind farms with an installed capacity of approximately 350 MW each in the area known as Borssele. This area is located in the North Sea near the Province of Zeeland and is very close to the offshore wind farms in the Belgian Exclusive Economic Zone. These tenders were very competitive. DONG Energy won the first tender for the offshore wind farms in the areas known as ‘Borssele I and Borssele II’ with a tender amounting to €72,7/MWh.16 A consortium consisting of Shell, Eneco, Van Oord and Mitsubishi won the second tender for the wind farms in the areas known as ‘Borssele III and Borssele IV’. This consortium won the tender with a tender amount of €54.5/MWh.17

16 www.rijksoverheid.nl/actueel/nieuws/2016/07/05/windpark-borssele-goedkoopste-ter-wereld (in Dutch only). 17 www.rijksoverheid.nl/actueel/nieuws/2016/12/12/nederlandse-consortium-bouwt-tweedewindpark-borssele-nog-goedkoper (in Dutch only).

Netherlands 435 In relation to the allocation of the SDE-subsidy, it should finally be noted that the Minister of Economic Affairs issues a decision in view of the Dutch General Administrative Act when granting the SDE-subsidy. This means the SDE-subsidy is not an agreement with, for example, the TSO. Therefore, the developers are required to enter into a normal Power Purchase Agreement (or similar agreement) with an offtaker. If the market price is above the basis amount (or tender amount) the subsidy recipient will not receive any SDE-subsidy and therefore the generator of renewable electricity needs to rely on the Power Purchase Agreement in order to earn back its costs. Given the fact that the tender amounts for the development of offshore wind farms have decreased dramatically, it is conceivable that the operators of these wind farms no longer require the SDE-subsidy if the market price increases.

C. Assessment of the European Commission The SDE-subsidy is clearly a form of State aid in the sense of Clause 107 of the TFEU. In the Netherlands, there has never been a discussion, like the PreussenElektra case,18 of whether or not the SDE-subsidy is State aid and should be notified to the Commission. The Netherlands notified the SDE-subsidy scheme to the Commission in August 2007. The Netherlands notified the SDE-subsidy scheme as an amendment to the former subsidy scheme for the MEP-subsidy. The Commission approved the system of the SDE-subsidy on 21 December 2007.19 The Commission determined that the SDE-subsidy scheme is compatible with the common market, since it limits ‘the subsidy to the difference between the production costs of the renewable energy and the market price of the form of power concerned and [is] therefore in line with point 59 of the environmental aid guidelines’.20 Later amendments to the SDE-subsidy scheme have also been notified to and approved by the Commission. According to paragraph 160 of the Commission’s ‘Community guidelines on State aid for environmental protection 2008’,21 Member States must notify in advance any individual case of investment or operating aid granted under an authorised scheme when the aid is granted to renewable electricity installations in sites where the resulting renewable electricity generation capacity exceeds 125 MW. This enables the Commission to carry out a more detailed assessment of any substantial amounts of aid and to decide whether such aid is compatible with the common market. Following the ‘Guidelines on State aid for environmental protection and energy 2014–2020’, the threshold was increased to 250 MW unless the aid is granted on the basis

18

Case C-379/98 PreussenElektra, EU:C:2001:160. Stimulating Renewable Energy, Modification of the MEP (N 707/02) (Case N478/07) [2008] OJ C39/1. 20 Since this decision taken in 2007, the decision refers to the Community guidelines on State aid for environmental protection of 3 February 2001 [2001] OJ C37/5. 21 Commission, ‘Community guidelines on state aid for environmental protection’ [2008] OJ C82/01. 19

436 Marinus Winters of a competitive bidding process.22 When the aid is granted through a competitive bidding process, an individual assessment by the Commission is not necessary. In the past, the Netherlands notified several individual SDE-subsidy decisions due to this obligation in the Guidelines. The Commission did not raise objections on these notifications. Most of these decisions are still not published on the website of the Commission,23 with the exception of the decision in relation to Q10 Offshore Wind BV (currently known as the Luchterduinen offshore wind farm).24 In this decision, the Commission (ao) determined that the Netherlands provided credible information according to which the aid to Q10 Offshore Wind BV will be limited to the minimum. The Commission furthermore noted that, in particular: [T]he SDE system has been designed to only subsidise the extra production costs of producing renewable energy. The conclusion that the aid is kept to the minimum is further reinforced by the fact that a tender procedure was used for the purpose of granting the aid, compelling parties to submit a project proposal as competitive as possible, thus mitigating the risk of overstimulation. Moreover, financial data provided by the authorities confirms that the aid results in achieving a minimum rate of return as required for such projects.

As indicated above, SDE-subsidy decisions for offshore wind farms are tendered in combination with a permit. This system has been in place since 2016 and this amendment has been notified to the Commission as well. In its decision in relation to this modification of the SDE-subsidy scheme of 7 April 2015, the Commission explicitly confirmed that individual subsidy decisions granted to offshore wind farms no longer require an individual assessment, since the SDE-subsidy decisions for offshore wind farms are granted in a competitive bidding process.25

D. Comparison of the Guidelines Environmental Protection and Energy 2014–2020 The aforementioned decision of the Commission in relation to the modification of the SDE-system contains an assessment of the SDE-subsidy scheme on the basis of the Commission’s ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG). In this decision, the Commission determined that the SDE-subsidy scheme is compatible with the common market. The Commission assessed the SDEsubsidy scheme on the basis of the general compatibility provisions in Chapter 3.2 of the EEAG and the specific compatibility criteria for operating aid granted for electricity from renewable energy sources. The Commission considered that: —

The SDE-subsidy scheme aims at an objective of common interest in accordance with Clause 107(3)(c) of the TFEU. The aim of the notified aid is to help the Netherlands achieve the renewable energy targets set by the EU as part of its

22 Commission, ‘Guidelines on State aid for environmental protection and energy 2014–2020’ [2014] OJ C200/1, para 20. 23 Aid for offshore wind farm Buitengaats (Case SA.31961), Decision of 22 November 2011, not yet published; Westermeerwind (Case SA.32859), Decision of 21 December 2011, not yet published. 24 Q10 Offshore Wind BV (Case SA.34742) [2013] OJ C1/7. 25 Modification of Dutch SDE+ RES scheme (Case SA.39399) [2015] OJ C234/1.

Netherlands 437 2020 strategy, as well as its domestic longer-term goals to increase the share of energy generated from renewable sources. — There is a need for State aid, since the Netherlands has explained that without the aid it will not meet the EU 2020 target to supply at least 14 per cent of energy from renewable sources in 2020. According to paragraph 116 of the EEAG, the Commission presumes that achieving the targets is appropriate to provide aid, provided that all other conditions of the EEAG are met. — The SDE-subsidy scheme has an incentive effect, since no subsidies will be granted to projects on which work has already started (point 50 of the EEAG). This does not apply to biomass projects, but the Commission considers that the requirement of point 50 of the EEAG is met as long as the cost of producing energy from biomass is higher than the energy market price after plant depreciation. — The SDE-subsidy scheme is proportionate. In this respect, the Commission considers that the subsidies will be allocated to the beneficiaries via a competitive bidding process, where successful participants will receive a level of subsidy based on the phase in which they made the bid for their project. This is also explained above. The risk of overcompensation is further limited by the technology-specific subsidy caps based on an acceptable 7.8 per cent rate of return. The scheme also complies with point 124 of the EEAG, since the subsidy is in the form of a variable premium on top of the reference price for electricity. Dutch producers have standard balancing responsibilities and no subsidy will be paid for hours in which the APX price is negative (for at least six consecutive hours). The Commission furthermore considered that point 126 of the EEAG requires aid granted from 1 January 2017 to be granted through a competitive bidding process, provided that the bidding process can be limited to particular technologies. The Netherlands has justified the separate tenders for offshore wind on the basis of the longer-term potential of offshore wind and the need to achieve diversification. Offshore wind farms are significantly more expensive than the projects of other renewable energy producers. — The Commission finally determined that the Netherlands complies with the transparency requirements (points 104–06 of the EEAG), the financing requirements (point 29 of the EEAG) and the non-cumulation requirements (section 3.2.5.2 of the EEAG). As a result, the Commission considered that the notified SDE-subsidy scheme is an appropriate instrument necessary to pursue an objective of common interest and that therefore the SDE-subsidy scheme is compatible with the internal market pursuant to Clause 107(3)(c) of the TFEU.

IV. CASE LAW IN RELATION TO THE DEVELOPMENT OF PROJECTS

In recent years there have been several cases in the Netherlands in which people living in the surrounding area of a new planned renewable project, like a wind farm, tried to block the development of these new projects with the argument that this project was not financially viable, since these projects were subsidised.

438 Marinus Winters In spatial planning cases, affected parties that wish to appeal a zoning plan can argue that the financial feasibility of the zoning plan is not assured. A zoning plan is a policy document of, for example, the municipality, that provides for the spatial planning of a certain region. These plans can be adopted by the municipality (the municipal council), the province, or the central government (the Ministry). Affected parties who disagree with the plan can oppose it by appealing to the court. Affected parties are people living in the direct vicinity who are affected by the new spatial developments and whose enjoyment of their property is impaired. To get the plan off the table they can, among other things, raise doubts about the financial feasibility of the plan by arguing that unlawful State aid is being granted. Reclaiming the aid would block the realisation of the plan. The affected parties must first deem it likely that there is State aid that can be reclaimed, and second that the Ministry/ municipal council could reasonably have foreseen in advance that, as a consequence, the plan cannot be realised within the planning period without unlawful State aid being granted. It is only if the affected parties manage to make these two arguments seem likely that their arguments can result in the annulment of the zoning plan.26 In existing cases, the court first assessed whether the Commission had given its view on the existence of State aid. If the Commission decided that there is no unlawful State aid, there are no grounds to assume that the Ministry/municipal council should have assumed that the financial feasibility of the plan would not be sufficiently ensured in view of the State aid. The same applies if the actual decision to grant a subsidy must still be approved by the Commission but the draft decision has already been approved, or if the Commission sees no reason to conduct an investigation into a breach of State aid regulations.27 Reliance on financial feasibility will not succeed in these cases. If the Commission has not expressed its view on the existence of State aid, the court will assess, whether the existence of State aid is probable. The court will subsequently examine whether the financial feasibility is sufficiently ensured if the aid would be reclaimed. The court will also assess this if there are no reasons to assume that State aid is involved or if the subsidy must still be approved by the Commission,28 should it later become apparent that there was indeed State aid. The financial feasibility is sufficiently ensured if it is likely that the undertaking, after the aid has been reclaimed, will still continue to realise the plan. Matters considered here are, for instance, a statement from the beneficiary undertaking that it will also continue with the plan after aid has been reclaimed, whether the undertaking is able to acquire other funds,29 or a situation where the realisation of the plan is already at an advanced stage.30 These circumstances contribute to the likelihood that the undertaking will not abandon the plan, which means that the financial feasibility

26 Dutch High Administrative Court (Afdeling bestuursrechtspraak van de Raad van State), 8 February 2012, NL:RVS:2012:BV3215, para 2.76.4.2. 27 ibid, para 2.76.4.1 28 ibid, para 2.76.4.2. 29 ibid. 30 Dutch High Administrative Court (Afdeling bestuursrechtspraak van de Raad van State), 22 October 2008, NL:RVS:2008:BG1152, para 2.6.

Netherlands 439 is ensured. And also, in the case where the undertaking has to withdraw, financial feasibility only starts to play a role if it is unlikely that there are other market parties that could continue the realisation of the plan. Those market parties could in an amended form and/or way—one that fits within the plan—take over the realisation of the zoning plan.31 If it appears that the financial feasibility is sufficiently guaranteed by other means, a final assessment of the existence of State aid is not necessary. When a court has doubts about the financial feasibility of the zoning plan, as a result of the existence of State aid, the court will examine whether the Commission has already expressed its opinion on the existence of State aid. If not, it will consider whether State aid is likely, and assess whether the plan would also be realised without State aid. Only once it has been deemed likely that State aid does exist and that the zoning plan cannot be realised within the planning period without that aid, can this lead to the opinion that the Ministry/municipal council should reasonably have foreseen in advance that the plan is not financially feasible and, if this is invoked in court, can result in the annulment of the plan. If the financial feasibility can be sufficiently guaranteed by other means, the court does not have to express its opinion on the existence of State aid.

V. CASE LAW IN RELATION TO COAL TAX

In the Netherlands, the Environmental Taxes Act is applicable (Wet belastingen op milieugrondslag). Pursuant to the Environmental Taxes Act, the supply of electricity and gas is subject to taxes. The Environmental Taxes Act determines that coal is also subject to taxes, provided that coal, which is used for the generation of electricity is exempted from taxes, since the supply of electricity is already subject to taxes. In 2012, the Dutch government decided to abolish the exemption for taxes on coal used for the generation of electricity. This decision was part of a wider package of additional taxes and cost-cutting measures. The Dutch government took these measures in order to meet the obligations of the Netherlands in the Stability and Growth pact of the eurozone, which means that the budget deficit cannot exceed 3 per cent of GNP. The abolishment of the exemption for coal meant that the generators of coal-fired power plants (together) were obliged to pay an amount of approximately €115 million per year in environmental taxes. All generators of the coal-fired power plants appealed their tax declarations for the environmental taxes. In these appeals, the generators mainly argued that the abolishment of the tax exemption is in breach of Article 14 of Directive 2003/96/EC restructuring the Community framework for the taxation of energy products and electricity,32 since the abolishment of the exemption has not taken place on the basis of environmental reasons, but for budgetary reasons.

31

ibid, para 2.76.4.2. Council Directive 2003/96/EC of 27 October 2003 restructuring the Community framework for the taxation of energy products and electricity [2003] OJ L283/15. 32

440 Marinus Winters In a judgment of the Court of Appeal in The Hague it is stated that one of the generators also argued that the abolition of the exemption qualifies as illegal State aid.33 In this respect, the generator referred to the judgment of the Court of Justice in the Laboratoires Boiron case.34 In this case, the French Supreme Court asked for a preliminary ruling from the Court of Justice in relation to a tax on direct sales of medicines by pharmaceutical laboratories. This French tax regime meant that sales of medicines by wholesale distributors were not subject to this tax. The Court of Justice decided that it should be accepted that an economic operator such as Laboratoires Boiron may plead that the charge on direct sales by pharmaceutical laboratories is unlawful, for the purposes of applying for reimbursement, on the grounds that it amounts to an aid measure. However, in the Laboratoires Boiron case, the Court also decided that the charge and the alleged aid measure constitute two elements of one and the same fiscal measure, which are inseparable. If the revenue from the tax on direct sales leads to an overcompensation of the wholesale distributors, the tax on direct sales qualifies as State aid. The aforementioned judgment of the Court of Appeal in The Hague did not specifically mention which companies or industries will be supported by the abolishment of the tax exemption for coal, but it should be assumed that the relevant generator meant that gas-fired power plants would benefit. In this respect, the Court of Appeal decided that the revenues of the tax on coal will not be used to maintain the tax exemption on other energy products for the generation of electricity. Therefore, the tax exemption does not qualify as illegal State aid. This generator appealed this judgment to the Dutch Supreme Court (Hoge Raad) and this case is still pending at the Supreme Court.

33 34

Court of Appeal The Hague, 16 October 2015, NL:GHDA:2015:3415. Case C-526/04 Laboratoire Boiron, EU:C:2006:528.

17 Austria MORITZ AM ENDE AND JUDITH GRIMM*

I. INTRODUCTION

T

HE CASE LAW of the Austrian courts on State aid and energy is, to a large extent, related to the Law on the rebate of energy taxes (Energieabgabenvergütungsgesetz—EAVG). Originally introduced in 1996, and reformed numerous times thereafter, the law has been a constant source of disputes about who is entitled to the tax rebate—or rather, whether the exclusion of certain groups from the entitlement to the rebate is (or was) lawful. Because of this, virtually all the cases concern constellations in which claimants applied to receive the tax rebate, rather than aiming at the prevention of aid being paid out to a competitor. The starting point for each of the ensuing legal arguments was a breach of the standstill obligation pursuant to (now) Article 108(3) TFEU. Accordingly, the Austrian case law on State aid and energy is, basically, an in-depth investigation of the meaning of the standstill obligation. To highlight the evolutionary nature of this case law, we will present it in chronological order.

II. THE INTRODUCTION OF THE ORIGINAL EAVG AND THE ADRIA WIEN PIPELINE CASE

The EAVG in its original version, as introduced in 1996, essentially provided for an energy tax rebate for energy-intensive undertakings engaged in the ‘manufacture of goods’.1 The provision of services was excluded from the benefit of the tax rebate (as were energy taxes not exceeding 0.35 per cent of the net production value). The introduction of the law had not been notified to the European Commission (Commission) as a State aid measure. The Austrian authorities—on the basis of an informal opinion of the Commission—had taken the view that the law was not (sufficiently) selective to qualify as a State aid measure.2

*

The authors would like to thank Thomas Starlinger for his valuable comments. cf s 2(1) EAVG as in Law on structural adjustment 1996 (Strukturanpassungsgesetz 1996), Law No 201/1996 [1996] Federal Law Gazette. 2 See Cases B2251/97 and B2594/97 [1999] Constitutional Court, VfSlg 15450/2001. 1

442 Moritz Am Ende and Judith Grimm The limitation of the tax rebate to manufacturing undertakings was challenged by service providers. Eventually, the Verfassungsgerichtshof (Constitutional Court) was summoned by Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke, whose applications for the tax rebate had been rejected because they were not active in the manufacturing of goods. The Constitutional Court had its doubts as to whether the differentiation between those undertakings producing goods and undertakings providing services was sufficient to render a measure selective for the purposes of State aid law and thus referred corresponding questions to the European Court of Justice (ECJ). By a judgment of 8 November 2001 in Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke (Adria-Wien), the ECJ replied that: [N]ational measures which provide for a rebate of energy taxes on natural gas and electricity do not constitute State aid within the meaning of Article [107 TFEU] where they apply to all undertakings in national territory, regardless of their activity.

But, on the other hand: [N]ational measures which provide for a rebate of energy taxes on natural gas and electricity only in the case of undertakings whose activity is shown to consist primarily in the manufacture of goods must be regarded as State aid within the meaning of Article [107 TFEU] (emphasis added).3

As a result, a number of things happened. On 13 December 2001, the Constitutional Court rendered its judgment in the reference case: the Court reasoned that undertakings which are active in the manufacturing of goods had been granted the rebate lawfully. It found that only the part of the law which ‘limited’ the reimbursement of energy taxes to undertakings which are primarily active in the manufacturing of goods constituted State aid. The Court concluded that the legislator had not been allowed to implement the restriction without approval from the Commission, and considered that the standstill obligation also prevented itself from applying it. To deny the rebate to undertakings not active in the manufacturing of goods on the basis of an ‘obviously inapplicable’ provision would amount to an arbitrary application of the law, which in turn would infringe upon the right to equal treatment under the law as enshrined in the Constitution. As a result, the Constitutional Court annulled the contested decision by which the application for the tax rebate had been rejected.4 The Commission, too, responded to the ECJ’s judgment by revising its previous position on whether the EAVG constituted aid. On 22 May 2002, it adopted a decision addressed to Austria by which it first, expressed its ‘regret’ that Austria had introduced aid in contravention of (then) Article 88(3) EC and second, held that the

3

Case C-143/99 Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke, EU:C:2001:598. Case B 2251/97 [2001] Constitutional Court; see for a discussion of the judgment, F Sutter ‘Energieabgabenvergütung als unzulässige Beihilfe; Innerstaatliche Folge aus Durchführungsverbot’ [2002] Österreichisches Anwaltsblatt 235; see also M Potacs, ‘Energieabgabenvergütung—unzulässige Beihilfe’ [2002] Österreichische Zeitschrift für Wirtschaftsrecht 82. 4

Austria

443

aid had been compatible with Article 87(3)(c) EC and Article 4(c) ECSC from 1 June 1996 to 31 December 2001.5 Last but not least, the Austrian legislator amended the EAVG in October 2002, so that the tax rebate applied to all businesses, independent of their field of activity, to also include providers of services in the rebate scheme.6 In light of the prospective EU harmonisation in the field of the taxation of energy producers and electricity entering into force in 2003, the applicability of the law was initially limited to cases occurring in the year 2002, but it was later extended to also apply to the following years.7 However, the tax procedures at the outset of the Adria-Wien cases were not closed yet. Indeed, the tax authorities again rejected the applications of service providers, now arguing on the basis of the aforementioned Commission decision that the aid scheme had, in the meantime, been notified to and approved by the Commission. Again, service providers turned to the Constitutional Court, only this time their request to set the tax ruling aside was rejected. The Constitutional Court pointed out, first, that its competence to assess the lawfulness of the legislative procedure was limited to infringements of the Constitution, and that an infringement of the standstill obligation pursuant to Article 88 EC (now Article 108 TFEU) did not render a law unconstitutional. Second, it considered that the standstill obligation was not ‘obviously’ opposed to an application of the EAVG any longer after the Commission had approved the aid scheme. Hence, the application of the law by the tax authorities was not considered arbitrary, and in turn no infringement of the constitutional right to equal treatment under the law could be established.8

III. WHAT STANDSTILL MEANS: TRANSALPINE ÖLLEITUNG AND WIENSTROM

A. Transalpine Ölleitung However, the judgment of the Constitutional Court was not the end of the proceedings. In accordance with the applicable procedure, the case was referred to the Verwaltungsgerichtshof (Supreme Administrative Court). Unlike the Constitutional Court, which only addresses infringements of the Constitution, the Supreme

5 Energy Tax Rebate (Case NN 165/2001) [2002] OJ C164/4; In detail see also T Jaeger, ‘Durchführungsverbot und rückwirkende Beihilfegenehmigung, Eine Kritik der österreichischen Rechtsprechung zu rechtswidrig durchgeführten, mit dem Gemeinsamen Markt vereinbaren Beihilfen’ [2003] Zeitschrift für Verwaltung 645. 6 Amendment of the Law on the rebate of energy taxes (Änderung des Energieabgabenver gütungsgesetzes), Law No 158/2002 [2002] Federal Law Gazette. 7 Amendment of the Law on the rebate of energy taxes (Änderung des Energieabgabenvergütungsgesetzes), Law No 71/2003 [2003] published in the Federal Law Gazette. The law was further amended, Law No 92/2004 [2004] Federal Law Gazette. 8 Case B 1348/02 [2002] Constitutional Court.

444 Moritz Am Ende and Judith Grimm Administrative Court assesses the lawfulness of a decision in full, which includes the application of the State aid standstill obligation. By a decision of 12 August 2004, the Supreme Administrative Court therefore referred questions to the ECJ, essentially asking whether the retroactive approval of the aid scheme by the Commission also led to an ex post facto regularisation (Heilung) of the infringement of the standstill obligation. The reference in the case (C-368/04 Transalpine Ölleitung) was seemingly still based on the assumption—in line with the Adria-Wien judgment—that only the limitation to manufacturing undertakings had rendered the EAVG selective.9 This assumption however, which appears to have been common ground in the Adria-Wien procedure,10 was not necessarily valid any longer. In April 2003, the Commission had initiated proceedings against Austria regarding the amended EAVG. In March 2004, the Commission adopted a decision classifying the EAVG as State aid, arguing that already the threshold of 0.35 per cent of the net production value rendered the EAVG selective because it effectively benefited only energyintensive undertakings.11 The Commission found that the EAVG as in force until end of 2003 constituted unlawful aid incompatible with the common market, and ordered Austria to take the necessary steps for the amendment of the law as promised by the Austrian authorities.12 This decision had not been challenged by Austria and had become final. Against this background, the ECJ rendered its judgment Transalpine Ölleitung. While the Court did not explicitly address the question whether the focus on energyintensive undertakings already rendered the EAVG selective, it clearly operated on the basis that this was the case. The Court denied the question whether the Commission decision of 2002 had led to an ex post facto regularisation of the infringement of Article 88 EC (now Article 108 TFEU); but more importantly, it also stressed that the national court ‘must not adopt a measure which would have the sole effect of extending the circle of recipients of the aid’.13 As a consequence, the Supreme Administrative Court turned down the appeals from service providers pending before it. It did so on the grounds that Community law stood against an unlawful extension of the circle of recipients of the aid.14 Moreover, with regard to appeals where applicants claimed that they were entitled to the tax rebate (or a higher tax rebate) because they were undertakings with a focus on manufacturing, the Supreme Administrative Court dismissed these appeals as unfounded, too, because the aid scheme had been introduced in contravention of Article 88 EC (now Article 108 TFEU) and the decisions of the Commission did not have the effect of regularising this illegality.15

9

Case 2003/17/0001 [2004] Supreme Administrative Court, referral decision. cf Case C-143/99 Adria-Wien Pipeline, Opinion of AG Mischo, EU:C:2001:250, points 29–31. cf Energy Tax Rabate 2002–2003 (Case C33/2003) Commission Decision 2005/565/EC [2004] OJ 2005 L190/13, para 46 et seq. 12 cf ibid, Arts 1 and 3. 13 Adria-Wien Pipeline (n 3) para 58. 14 Case 2006/17/157 [2006] Supreme Administrative Court. 15 Case 2004/05/0274 [2007] Supreme Administrative Court; Case 2004/17/0078 [2007] Supreme Administrative Court. 10 11

Austria

445

B. Wienstrom The latter question came to the fore again shortly thereafter, this time in a dispute related to the Green Electricity Act (Ökostromgesetz—ÖG’). The ÖG provides support for heat and power plants (CHP plants). During the relevant years, the support was financed by means of a uniform surcharge on the quantity of electricity supplied to end customers, irrespective of actual consumption by those end customers of combined heat and power electricity. That surcharge was transferred to Energie-Control GmbH, the public body responsible for paying assistance to CHP plants. Austria had not notified the measure, arguing, with reference to the ECJ’s PreussenElektra judgment,16 that the measure did not constitute aid. The Commission disagreed with that assessment and as a consequence had opened proceedings. By a decision of 4 July 2006, the Commission decided not to raise objections against the measure after Austria had retroactively amended it so that imports of so-called ‘green’ electricity were effectively exempted from the surcharge.17 Otherwise, the Commission again merely expressed its ‘regret’ that Austria had failed to notify the measure.18 Wienstrom, an operator of CHP plants, challenged decisions which fixed the amounts of support which it was supposed to receive for the years 2003 and 2005, arguing that it should have received an even higher support. The Supreme Administrative Court found that the relevant provisions of the ÖG clearly constituted illegal aid; indeed, the Supreme Administrative Court considered this to be an acte claire within the meaning of the CILFIT19 case law. The Supreme Administrative Court wondered if the breach of the standstill obligation meant that it had to disapply the provisions on which Wienstrom based its application. It therefore referred questions to the ECJ, noting that applying those provisions in breach of the standstill obligation would not ‘extend the circle of aid recipients’, and that the dispute also did not involve third parties whose interests might be concerned by the implementation of the illegal aid. It further wondered whether it made a difference if the contested decision had been adopted before or after the Commission decision, considering that the aid measure, ie, the ÖG, had in any event been adopted before the Commission decision.20 The ECJ replied that the standstill obligation did not prevent the Supreme Administrative Court from applying the relevant provisions. It pointed out that while the aid was illegal, the exact consequence of that illegality depended on whether the Commission considered the aid compatible with the common market or not. In the first case, only the interest in respect of the period of unlawfulness would be repayable upon request by the Commission or a third party concerned.21 Full repayment of the aid might be required only in the second case. Therefore, once the

16

Case C-379/98 PreussenElektra, EU:C:2001:160. Support of renewable Electricity Production (Case NN 162/A/2003 and N 317/A/2006) [2006] OJ C221/8, para 74. 18 ibid, para 61. 19 Case C-283/81 CILFIT, EU:C:1982:335. 20 Case 2004/05/0274 [2007] Supreme Administrative Court, referral decision. 21 Case C-384/07 Wienstrom, EU:C:2008:747, para 29. 17

446 Moritz Am Ende and Judith Grimm Commission had found that the aid was compatible, Community law would not preclude the disbursement of the principal of the aid even though the aid measure had been introduced illegally.22 In the following, the courts also had to decide on two actions brought by the Austrian railway operator ÖBB: one against the refusal to grant it, on an exceptional basis, the EAVG tax rebate for the years 1996–2001; the other against the obligation to pay the CHP surcharge for the years 2003–06. The first action was dismissed by the Supreme Administrative Court on the grounds that the decision of the Commission had not covered aid to service providers; hence, other than in the Wienstrom case, the aid which ÖBB claimed to be entitled to had not been approved by the Commission.23 The second action was dismissed by the Constitutional Court: it pointed out that the ECJ had clarified in Wienstrom that the ÖG had been retroactively approved by the Commission, therefore also the levying of the surcharge was considered not to be precluded by the standstill obligation any longer.24

IV. GOING GBER: THE 2010 REFORM OF THE EAVG AND THE DILLY’S WELLNESSHOTEL CASE

A. The Reform of the EAVG and its ‘Notification’ by the Authorities In December 2010, the Austrian legislator enacted yet another reform of the EAVG to achieve the result it had always wanted: limitation of the energy tax rebate to energy-intensive undertakings with a focus on manufacturing (the EAVG reform or the EAVG 2011).25 This time, however, the legislator also wanted to ensure compliance with the State aid rules: the application of the revised provisions for periods after 31 December 2010 was explicitly made subject to ‘approval by the European Commission’. The travaux preparatoires for the EAVG 2011 read: The application of the amended provision is subject to the approval of the amended provision. That amendment shall enter into force in respect of energy use after 31 December 2010. Applications from service providers for periods after 31 December 2010 shall therefore no longer be accepted. If the amendment is approved by the Commission as authorised State aid, the statutory restriction to production undertakings shall apply from 1 January 2011, with the result that, after that date, service providers will no longer be entitled to the energy tax rebate for energy used. If the amendment is not approved by the Commission, the current legal situation shall remain unchanged and both production undertakings and service providers shall be entitled to an energy tax rebate.

The task of ensuring the ‘approval by the European Commission’ fell to the Federal Ministry of Finance. Rather than embarking on the procedure to notify the EAVG

22

ibid, para 30. Case 2008/17/0138 [2009] Supreme Administrative Court. Case A 10/08 [2009] Constitutional Court. 25 Law accompanying the budget (Budgetbegleitgesetz), Law No 111/2010 [2010] Federal Law Gazette. 23 24

Austria

447

reform according to Article 2 of Regulation 659/1999 and Regulation 794/2004—possibly as a simplified notification pursuant to Article 4(2)(c) of Regulation 794/200426— the Ministry considered that there was an easier way to get an ‘approval by the Commission’: in its view, an approval by the Commission, broadly speaking, already existed in the form of the General Block Exemption Regulation No 800/2008 (GBER 2008), and in particular in the form of Article 25 GBER 2008 on aid in the form of reductions in environmental taxes.27 Hence, the Ministry of Finance decided to simply forward a summary of the information regarding the EAVG reform to the Commission, by using the template form provided in Annex III to the GBER 2008, which it did on 7 February 2011. Regarding the duration of the scheme, the Ministry submitted ‘1.2.2011–31.12.2013’ in the respective field of the template. Nonetheless, several service providers, in particular operators of hotels, applied for tax rebates for periods later than 31 December 2010, and when their applications were turned down by the tax authorities, they went to court.

B. National Proceedings Prior to the Dilly’s Wellnesshotel Reference In several of the early cases, the appeal authority of the first instance, the Federal Finance Court,28 decided that the authority had been wrong to refuse the tax rebate for January 2011, but had acted lawfully regarding the periods afterwards: the Federal Finance Court considered that the reform had indeed been ‘approved’, but only as of February 2011.29 The first instance judgments to overturn the original decision regarding the month of January were appealed by the tax authorities to the Supreme Administrative Court. The authorities argued that, for the purposes of the EAVG 2011, it mattered only if the aid measure had been approved. The Supreme Administrative Court rejected the appeals. It held that the term ‘subject to the approval by the Commission’ regarding the applicability of the EAVG reform had to be understood as requiring a ‘positive decision’ by the Commission. The Supreme Administrative Court found that such a decision did not exist (at least) for the month of January: even if a mere information of the Commission would be sufficient pursuant to the GBER-procedure, a positive

26 The authors do not know whether the other requirements of Art 4 of Reg 794/2004 had been met, such as the submission of annual reports for the existing aid scheme pursuant to Art 4(3) Reg 794/2004. 27 For a discussion on the applicability of aid in the form of reductions in environmental taxes under the criteria of Art 25 GBER see T Bieber, ‘Energiesteuerbegünstigungen als Staatliche Beihilfen’ [2012] Europäische Zeitschrift für Wirtschaftsrecht 257; M Laudacher, ‘Energieabgabenvergütung und AGVO’ [2014] BundesFinanzGerichtjournal 428. 28 At the time, the first instance appeal authority was the Independent Finance Senate (Unabhängiger Finanzsenat, UFS) which, while being technically an appeal body within the tax administration, fulfilled all the substantive criteria of an independent ‘court’ for the purposes of Art 267 TFEU. In 2013, the tax appeal procedure was reformed and the appeal body also formally became a part of the judiciary. To reflect this, the Independent Finance Senate became the Federal Finance Court (Bundesfinanzgericht, BFG). For the sake of simplicity, the authors will refer to the first instance tax appeal body as ‘Federal Finance Court’ throughout this chapter. 29 eg, Case RV/0188-I/12 [2012] UFS, Außenstelle Innsbruck.

448 Moritz Am Ende and Judith Grimm decision did, in any event, not exist for the month of January 2011. Incidentally, the Supreme Administrative Court further noted that the EU law requirements were also not met regarding the notification pursuant to the GBER procedure for the period after 1 February 2011.30 The service providers, on their part, also appealed the first instance judgments. Rather than appealing their cases to the Supreme Administrative Court in order to argue that the EAVG reform had been implemented contrary to EU State aid law, a number of them lodged their appeals with the Constitutional Court in order to argue that the EAVG reform was unconstitutional to begin with. The service providers submitted a passionate plea to the Constitutional Court that the distinction between energy-intensive service providers and energy-intensive production undertakings was contrary to the principle of equal treatment (and hence unconstitutional). Eventually however, those claims were dismissed by the Constitutional Court. State aid law does not seem to have been debated in the procedure before the Constitutional Court—the Court’s summary of the arguments of the applicants makes no reference to any State aid issues. The Constitutional Court addressed the issue nonetheless, but limited itself to some preliminary remarks on the matter, noting that ‘Union law concerns against the EAVG apparently do not exist’, and that the requirements of Article 25 GBER 2008 have ‘apparently been met’. Referring to the earlier judgment of the Supreme Administrative Court, the Constitutional Court considered that it could therefore be left open for the purpose of the proceedings before it whether the exclusion of the service providers took effect pursuant to national law ‘already on 1st of January or only on 1st of February 2011’.31 Shortly thereafter, it was the Supreme Administrative Court’s turn once again. This time, the issue was whether the tax rebate had been legally denied for the period of February to December 2011; specifically, whether the EAVG reform could be considered ‘approved’ by the Commission for this period. After a detailed analysis of the State aid rules in general, the GBER 2008 in particular, and the reservation in the EAVG regarding its applicability, the Supreme Administrative Court came to the conclusion that the publication of a State aid measure in the Official Journal of the European Union pursuant to Article 9 GBER 2008 ‘could be regarded a form of “approval by the European Commission” within the meaning of Section 4 para 7 EAVG’. Whether or not all the requirements for the application of the GBER 2008 had been met was considered as irrelevant for the purpose of the entry into force of the EAVG reform by the Supreme Administrative Court. Finally, the Supreme Administrative Court noted that the Constitutional Court had already dismissed any constitutional concerns regarding the reform. With this, the Supreme Administrative Court dismissed the appeals.32

30

Case 2012/17/0175 [2012] Supreme Administrative Court. Case B 321/12 [2012] Constitutional Court. Case 2012/17/0469 [2013] Supreme Administrative Court; cp for a discussion of the judgment B Ludwig, ‘VwGH bestätigt das Inkrafttreten des EnergieabgabenvergütungsG idF BudBG’11 ab 1.2.2011’ (2013) 9 Finanz Journal 290. 31 32

Austria

449

C. The Reference Case: Dilly’s Wellnesshotel This could have been the end of the story. However, yet another appeal to the Supreme Administrative Court had been lodged, this time by Dilly’s Wellnesshotel, concerning a refusal of the tax rebate for the entire year 2011. Since the contested decision was unlawful regarding the month of January, the Supreme Administrative Court annulled it in its entirety and referred the case back to the Federal Finance Court Linz. The Supreme Administrative Court made it clear however that it considered the claims regarding the rest of the year to be unfounded—in line with its previous judgment.33 The Federal Finance Court Linz was however not convinced of the lack of relevance of the applicability of the GBER 2008. Rather than granting the rebate for the month of January and refusing the remainder of Dilly’s application, the Federal Finance Court Linz referred a number of questions to the ECJ. First, whether reliance on Article 25 GBER 2008 infringed EU law in case of non-compliance with various formal obligations of chapter I of the GBER 2008. Second, whether a State aid scheme based on Article 25 GBER 2008 must also promote the environment according to Article 17(1) GBER 2008. Third, whether the limitation of the energy tax rebate to 10 years, as required in Article 25(3) GBER 2008, must be contained in the aid measure itself (which it was not), or if it is sufficient if it can be inferred from the exemption notice.34 More specifically to the first question, the Federal Finance Court Linz had identified three obligations which had not been complied with: (1) There was no reference in the EAVG to the GBER 2008 or, for that matter, to the publication reference in the Official Journal, as required by Article 3(1) GBER 2008; (2) The summary of the information regarding the aid measure at issue was not forwarded to the Commission within 20 working days ‘following the entry into force of an aid scheme’, as required by Article 9(1) GBER 2008; (3) The text of the aid scheme at issue was not published on the internet in accordance with Article 9(2) GBER 2008, since the internet address notified to the Commission did not enable access to the text concerned.35 The ECJ considered it sufficient to address the first question only, and it further limited its answer to the consequences of the failure to incorporate an express reference to the GBER in the aid scheme. In this regard, the ECJ held that this obligation ‘does not amount to a mere formality, but is mandatory in nature, so that the failure to fulfil that condition precludes an exemption from being granted under that regulation’.36 The ECJ also noted that a subsequent, second notification in the course of

33

Case 2013/15/0053 [2013] Supreme Administrative Court. Case C-493/14 Dilly’s Wellnesshotel, EU:C:2016:577; and Case RE/5100001/2014 [2014] BFG, Außenstelle Linz; see R Grabner, ‘Energieabgabenvergütung für Dienstleistungsbetriebe: Antwort des EuGH in der Rechtssache C-493/14, Dilly’s Wellnesshotel’ (2016) 7–8 BFG Journal 276. 35 cf the information submitted in Energieabgabenrückvergütung, Gesetzesnovelle BGBl. I Nr 111/2010 (Case SA.32526) [2011] OJ C288/21. 36 Dilly’s Wellnesshotel (n 34) para 51. 34

450 Moritz Am Ende and Judith Grimm 2014 (submitted for the years 2015–20 with regard to the implementation of the new Block Exemption Regulation No 651/2014)37 could not compensate for the failure to refer to Regulation No 800/2008 in the EAVG for the period concerned (ie, the year 2011).38

D. Follow-Up on the Preliminary Ruling in Dilly’s Wellnesshotel Taking into account the ECJ’s answer, the Federal Finance Court Linz arrived at the conclusion that the EAVG reform had not been ‘approved’ within the meaning of the EAVG 2011 for the purpose of adjudicating Dilly’s application for the year 2011. It considered that both the standstill obligation of Article 108(3) TFEU, as well as the Austrian legislator’s intention not to infringe upon that provision (as expressed in the Austrian standstill clause in Section 4(7) EAVG 2011), militate in favour of such an interpretation of the EAVG. More fundamentally, the Federal Finance Court Linz also considered that the lack of any reference to the GBER procedure in both the EAVG 2011 as well as its travaux preparatoires, and the resulting, obvious shortcomings regarding compliance with the GBER, were strong indications that the Austrian legislator had never envisaged relying on the GBER 2008 in the first place, but had intended that a formal approval of the aid scheme in accordance with Regulation No 659/1999 should be sought. The Federal Finance Court Linz also noted that until 30 September 2011, the day of the publication of the information communicated to the Commission in the Official Journal, even tax experts who were actively researching the legal basis for the State aid assessment of the EAVG reform had been unable to establish any link to the GBER 2008. It therefore considered that the legal position of the tax authorities was also incompatible with basic principles of the rule of law, such as the requirement that legal norms must be clear, ie, intelligible, consistent and workable. To the Federal Finance Court Linz, the application of the EAVG 2011 was so obviously in breach of the standstill obligations (both national and EU) that it even amounted to arbitrary conduct by the authorities, and hence an infringement of the constitutional right to equal treatment of all citizens before the law within the meaning of the jurisprudence of the Constitutional Court.39 In sum, the Federal Finance Court Linz concluded that the term ‘approval by the Commission’ in the EAVG 2011 referred to the formal notification procedure. Since no such approval was present, it considered that the exclusion of the service providers from the tax rebate was not yet applicable—certainly not for the year 2011—and that the application by Dilly’s Wellnesshotel for the energy tax rebate therefore had to be granted for the entire year 2011.

37

Energieabgabenvergütung für Produktionsbetriebe (Case SA.40192). Dilly’s Wellnesshotel (n 34) para 44. 39 Case RV/5100360/2013 [2016] BFG, Außenstelle Linz, on the last point with reference to Case B 2251/97 (n 2). 38

Austria

451

E. The Final Outcome is Still Open At the time of the finalisation of this chapter, the final outcome of the EAVG saga remains to be seen. The Federal Ministry of Finance has appealed the judgment of the Federal Finance Court Linz to the Supreme Administrative Court once more, and public servants in the ministry have publicly announced that they consider the case law of the Supreme Administrative Court, according to which the EAVG reform was applicable as of February 2011, to still be valid.40 The Supreme Administrative Court has now referred yet another set of questions to the ECJ—the saga continues.41

40 D Pfau, ‘EuGH Rs Dilly: Phyrrussieg oder neue Chance auf Energieabgabenvergütung für Dienstleistungsbetriebe?’ (2016) 12 Österreichische Steuerzeitung 478. 41 Decision of 14 September 2017 in Case Ro 2016/15/0041; registered in Luxembourg as Case C-585/17 Dilly’s Wellnesshotel.

428 Marinus Winters Furthermore, the Transition Act for the Electricity Generation Sector determines that the Dutch State provided compensation for the stranded costs in relation to district heating and to costs in relation to the sale of the experimental coal regasification plant Demkolec. Until the end of 2000, all parties connected to the Dutch grid were obliged to pay a levy of €0.0053 (NLG 0.0117) per kWh in order to cover the above-mentioned costs (households were obliged to pay this amount to suppliers, large consumers were obliged to pay this amount to the grid operators). This is laid down in Clause 9 of the Transition Act for the Electricity Generation Sector. This compensation scheme was notified to the Commission on 16 October 1998. Almost three years later, on 25 July 2001, the Commission issued a decision in which it raised no objections to the measures of the Dutch government.5 According to the aforementioned decision, the Dutch government had originally also notified a measure for the import contracts of gas and electricity, but later withdrew this part of the notification. In June 2001, the Dutch government also withdrew the financing mechanism, since the Commission was of the opinion that the levy qualified as a parafiscal levy. The compensation for the stranded costs for the district heating projects and Demkolec were assessed on the basis of the Commission Communication relating to the methodology (Methodology) for analysing State aid linked to stranded costs. The Methodology was determined on the same day as the decision to raise no objections (25 July 2011). In this decision, the Commission assessed the elements of Chapter 3 of the Methodology, which in fact defines which costs are eligible for stranded costs. The Transition Act for the Electricity Generation Sector led to several disputes between the energy companies, which were shareholders in the Sep, and the Dutch State. State aid was an element in some of these proceedings. The State aid elements in these proceedings will be discussed below. The Transition Act for the Electricity Generation Sector originally also contained a priority rule for the import contracts of electricity, which meant that TenneT was obliged to allocate sufficient import capacity to the Sep or its legal successor and their counterparties. This priority access would enable the generators to fulfil their obligations under the long-term import contracts with mainly German and French companies. However, the decision of the Dutch energy regulator (at that time, the Nederlandse Mededingingsautoriteit (NMa), currently the Autoriteit Consument & Markt (ACM)) to determine the allocation rules for the import capacity on the interconnectors was appealed at the Trade and Appeals Tribunal (College van Beroep voor het bedrijfsleven) by several market parties. These market parties were of the opinion that the priority access rules were discriminatory. The Trade and Appeals Tribunal asked for a preliminary ruling, which led to the judgment of the Court of Justice of 7 July 2005, in which the Court decided that the priority access rules were discriminatory and therefore these priority access rules were not in line with the First Electricity Directive.6 Although this is an important judgment of the Court,

5 Measures in Favour of the Electricity Market for Stranded Costs (Case N597/1998) [2001] OJ C268/5. 6 Case C-17/03 VEMW and others, EU:C:2005:362.

Netherlands 429 State aid did not play a role in the preliminary proceedings at the Court of Justice or at national level.

B. Essent Netwerk As indicated above, the Transition Act for the Electricity Generation Sector stipulated that everyone connected to the grid had to pay a levy of €0.0053 /kWh during the second half of 2000. Aluminium Delfzijl BV (Aldel), which is a large electricity consumer, had to pay an amount of approximately €4.5 million to Essent Netwerk BV, the distribution system operator of Aldel. Aldel refused to pay this levy and therefore Essent Netwerk claimed these costs back from Aldel before the Court in Groningen. Aldel stated that the levy qualified as illegal State aid for all electricity producers, which were shareholder in Sep. The Court in Groningen asked for a preliminary ruling, in which it (ao) asked the Court of Justice, whether the levy was in line with Article 107(1) of the Treaty on the Functioning of the European Union (TFEU).7 The ECJ determined that Article 107 TFEU must be construed as meaning that the amounts paid to the designated company under Article 9 of the Transitional Act for the Electricity Generation Sector constitute ‘State aid’ for the purposes of that provision of the TFEU in so far as they represent an economic advantage and not compensation for the services provided by the designated company in order to discharge public service obligations.8 The Court in Groningen did not render a final judgment.

C. Demkolec In 1989, Sep decided to develop a coal regasification unit in Buggenum (in the southern part of the Netherlands). The name of this unit was Demkolec and this was an experimental unit, which was not cost-effective from an economic perspective. Sep decided to construct the coal regasification unit for environmental reasons. As indicated above, Demkolec became a stranded cost as a result of the liberalisation of the market. It was decided that BV Nederlands Elektriciteit Administratiekantoor (NEA), which is the legal successor of Sep, would sell Demkolec in an auction. Pursuant to Clause 7(b) of the Transition Act for the Electricity Generation Sector, the Dutch State would subsequently fund the difference between the book value of Demkolec and the sale value of Demkolec in the aforementioned auction. This was also approved by the Commission in its State aid decision of 25 July 2001.9 After the sale of Demkolec, the Dutch State paid €134 million to NEA, which equals the book value of Demkolec and the sale value. Subsequently, NEA and the Dutch State entered into a dispute about the question of whether the exploitation

7 8 9

Court Groningen, 22 June 2005, NJF 2005, 296. Case C-206/06 Essent Netwerk Noord and others, EU:C:2008:413. Measures in Favour of the Electricity Market (n 5).

430 Marinus Winters losses and the interests were part of the stranded costs as well. NEA and the Dutch State decided to start arbitral proceedings in relation to this issue. On 7 September 2005, the Netherlands Arbitration Institute (NAI) decided that the Dutch State was obliged to pay an additional amount of €38 million, since the Dutch State was obliged to pay all stranded costs to NEA. The NAI furthermore decided that this was in line with the aforementioned State aid decision. As a result of the arbitral award, the Dutch State paid an additional €46 million, which is the €38 million plus costs and interest. The Dutch State subsequently went to the Court of The Hague and claimed that the arbitral award of 7 September 2005 was null and void, since this arbitral award was in breach of the European State aid rules. Pursuant to Clause 1065 of the Dutch Code of Civil Procedure, a Dutch Court can nullify an arbitral award if the award is in breach of the public order. The Dutch State argued that NEA received compensation, which exceeded the maximum amount of €136 million as indicated in the approval decision of the Commission. This appeal of the Dutch State was rejected by the Court of The Hague on 1 August 2007.10 In its judgment, the Court carefully assessed the arguments of the Dutch State in light of the State aid decision of the Commission. In this respect, the Court decided that the amount of €136 million (as included in the approval decision of the Commission dated 25 July 2001) is not a maximum amount. When the Commission adopted the approval decision the State aid amount could not be determined, since the auction of the Demkolec plant had not yet taken place. Therefore, the Court decided that the range of €90 to €136 million as included in the State aid decision was a preliminary estimate or an estimate and could not be considered as a maximum State aid amount. The Court furthermore decided that the State aid decision of the Commission did not determine the date of the book value of the Demkolec plant either. According to the Court, there is no exact date for the book value in the State aid decision, since the auction had not yet taken place. The Court finally considered that the approved auction system of the Demkolec plant meant that the exploitation costs and interests were stranded costs, since these costs led to a lower sale price of the Demkolec plant. On the basis of these arguments, the Court of The Hague determined that the arbitral award of the NAI was not in breach of the State aid decision of the Commission in relation to stranded costs and therefore there was no reason to nullify the arbitral award of the NAI.

D. District Heating Projects As indicated in the introductory paragraph of this chapter, the Transition Act for the Electricity Generation Sector determines that the Dutch State would provide compensation for the stranded costs in relation to district heating projects. In this respect, the Ministry determined a specific regulation, which calculated the compensation for

10

Court of The Hague, 1 August 2008, Dutch State/NEA, NL:RBSGR:2007:BB1424.

Netherlands 431 the generators of district heating projects.11 The regulation determined the compensation on the basis of a model. Through this, the MEP-subsidy received by the generators would be deducted from the compensation. E.ON Benelux NV (currently Uniper Benelux NV) appealed this regulation at the Court of Arnhem and stated that (ao) this deduction of the MEP-subsidy was not line with the State aid decision of the Commission, since the MEP-subsidy was not meant to cover stranded costs.12 The MEP-subsidy had a different purpose. The Court did not accept this argument. In this respect, the Court referred to the Commission’s State aid decision of 25 July 2001 and paragraph 3.8 of the Methodology, in which the Commission stated that: Stranded costs must be valued net of any aid paid or payable in respect of the assets to which they relate. In particular, where a commitment or a guarantee of operation corresponds to an investment which is the subject of State aid, the value of the aid must be deducted from any stranded costs resulting from the commitment or guarantee.

On the basis of this paragraph in the Methodology, the Court decided that aid which is meant to cover other costs must be deducted from the compensation. Therefore, the Court did not accept Uniper’s argument.

III. SDE-SUBSIDY

A. Introduction As indicated in the introduction to this chapter, the generation of renewable electricity in the Netherlands is stimulated by the SDE-subsidy. This section will explain the SDE-subsidy scheme and will discuss the assessment of the SDE-subsidy scheme by the Commission under the State aid rules of Clause 107 of the TFEU.

B. Explanation of the SDE-Subsidy System i. General Principle In 2008, the Dutch government introduced the subsidy for the Stimulation of Renewable Energy (Stimulering Duurzame Energieproductie: SDE-subsidy). The general rules of the SDE-subsidy are laid down in the SDE-Decree.13 The SDEsubsidy is a subsidy, which is granted by the Ministry of Economic Affairs for each kWh of generated renewable electricity. The purpose of the SDE-subsidy is to fund

11 Regulation of the Ministry of Economic Affairs, 20 January 2005, no WJZ 4081042, Uitvoeringsregeling Overgangswet elektriciteitsproductiesector, Staatscourant 2005, 22. 12 Court Arnhem, 22 June 2009, E.ON Benelux N.V./Ministry of Economic Affairs, NL:RBARN: 2009:BJ1171. 13 Besluit stimulering duurzame energieproductie, Staatsblad 2007, 410. The SDE-Decree has been amended several times.

432 Marinus Winters the uneconomical part of the generation of renewable electricity, which owners of renewable installations (such as wind farms, PV parks, but also producers of biogas) will not earn back through the sale of the generated electricity on the market under a Power Purchase Agreement. In general, the cost price of the generation of renewable electricity is higher than the market price. The cost price for the generation of renewable electricity by most renewable installations such as a PV park or a wind farm is fixed. The market price varies over time. Under the SDE-subsidy scheme, the Ministry will pay the difference between the (fixed) cost price and the (variable) market price during the subsidy term (10–15 years depending on the technology). ii. The Determination of the Subsidy Amount The SDE-subsidy will be paid for each generated kWh of renewable electricity. The amount of generated kWhs will be determined by the number of guarantees of origin (garanties van oorsprong), which will be redeemed by the producer of renewable electricity. The actual subsidy amount per year will be determined by multiplying the number of guarantees of origin by the ‘corrected basis amount’. The corrected basis amount will be determined every year by the Ministry by decreasing the basis amount with the market price. The market price is based on the average annual APX price level, which will be multiplied by: — the correction factor for imbalance costs;14 and — the correction factor for profile costs.15 The basis amount might, in the future, also be corrected by the value of the guarantees of origin (garanties van oorsprong) as well as other future corrections set by the Dutch government and which could have a substantial influence on the difference between the average cost price of the generation of renewable electricity and the average market price (Clause 14, sub 1 of the SDE-Decree). This means that the SDE-subsidy scheme will avoid an overstimulation due to high revenues, which will be generated by the sale of guarantees of origin. The average annual APX price-level and the factors for imbalance costs and profile costs will be determined annually by the Ministry with retroactive effect. Each year before 1 April, the Ministry determines the average annual APX price level and the factors for imbalance costs and profile costs of the previous year. The Ministry will take an expert assessment (of, eg, the Energy Research Centre, the Netherlands) regarding the price level and account for the aforementioned factors in order to determine the amount, which will be deducted from the basis amount.

14 Generators of electricity have programme responsibility. Due to differences between the projected production of electricity and the actual production, generators may encounter imbalance costs, which will be invoiced by the Dutch Transmission Operator TenneT, since TenneT is required to incur costs to level out these differences. 15 Profile costs are costs (or revenues) resulting from the difference between the value of the generated electricity and the average market value of the electricity.

Netherlands 433 The SDE-subsidy also contains the base electricity price (basiselektriciteitsprijs). The base electricity price is applicable for the whole term of the SDE-subsidy. In the market, the base electricity price is also called ‘the floor’. The maximum subsidy per kWh is capped at the difference between the basis amount and the base electricity price, which means that if the correction amount is lower than the base electricity price, the subsidy recipient will (only) receive the difference between the basis amount and the base electricity price. The figure below (Figure 1) illustrates an example of the SDE-subsidy system. The vertical axis indicates the price (in €/kWh) and the horizontal axis indicates the term of the subsidy (in years). The light grey line is the basis amount, which is a fixed amount/kWh. The dark grey line is the market price, which is (of course) a variable price. The dotted line indicates the base electricity price (or the floor). As indicated above, the Ministry will determine the market price each year and will pay the difference between the light grey line and the dark grey line to the company. If the market price is below the ‘floor’ (as shown in the figure in the third year), the Ministry will pay the difference between the light grey line and the dotted line. If the market price is above the basis amount (as it is in the sixth year in the example below) the Ministry will not pay a subsidy, since the generator should be able to earn back the costs of the generation of electricity with the Power Purchase Agreement, which the generator has entered into with a purchaser of the electricity, such as a utility company. The SDE-Decree contains two forms of ‘banking’. Production capacity that is eligible for subsidisation but has not been realised in a certain year (underproduction) can be carried over to the next calendar year with the possibility of realising the capacity and corresponding subsidy in that next year. The subsidy recipient can furthermore request to extend the subsidy term by no more than one year, to offset this underproduction.

Figure 1: SDE-Subsidy System

434 Marinus Winters iii. The Allocation of Available Funds Each year, the Ministry determines which technologies of renewable energy are eligible for the SDE-subsidy. The technologies, which are eligible for the SDE-subsidy will be determined in the Regulation determination categories of renewable electricity production (Regeling aanwijzing categorieën duurzame energieproductie). This Regulation also determines for each technology the basis amount (basisbedrag). The basis amount is a fixed amount during the term of the subsidy. The basis amount represents the cost price for the generation of renewable electricity by the respective technology. The Ministry determines this basis amount with the help of expert analyses of, for example, the Energy Centre of the Netherlands. These expert analyses carry out independent research about the cost price of the various technologies. The available funds will be released in different phases. The basis amount increases for each phase. Each technology has a maximum basis amount, above which no subsidy will be granted. Cost-effective technologies with a low basis amount can apply earlier than the more expensive technologies. Therefore, there is a greater chance that a budget will be available for the cost-effective technologies than for technologies with a higher maximum basis amount. In order to create competition between different technologies, developers can also apply for a lower subsidy than the maximum basis amount for the technology in question. Such applications can be made in a ‘free category’. This allows them to tailor their subsidy application to their business case. The amount of subsidy applied for in the ‘free category’ is lower than the maximum phase amount and therefore the developers who apply in the free category, have a greater chance of receiving an SDE-subsidy. The allocation of available funds is different for the SDE-subsidy which will be granted to offshore wind farms. The SDE-subsidy for offshore wind farms will be tendered together with a permit to construct, operate and maintain the offshore wind farms. These permits will be granted for special designated areas in the Dutch part of the Exclusive Economic Zone in the North Sea. The tenderer with the lowest tender amount (which in fact represents the levelised cost of energy) will be granted the SDE-subsidy as well as the permit for the offshore wind farm. For these tenders, the Ministry determines a separate budget. In 2016 the Ministry organised two tenders for the development of four offshore wind farms with an installed capacity of approximately 350 MW each in the area known as Borssele. This area is located in the North Sea near the Province of Zeeland and is very close to the offshore wind farms in the Belgian Exclusive Economic Zone. These tenders were very competitive. DONG Energy won the first tender for the offshore wind farms in the areas known as ‘Borssele I and Borssele II’ with a tender amounting to €72,7/MWh.16 A consortium consisting of Shell, Eneco, Van Oord and Mitsubishi won the second tender for the wind farms in the areas known as ‘Borssele III and Borssele IV’. This consortium won the tender with a tender amount of €54.5/MWh.17

16 www.rijksoverheid.nl/actueel/nieuws/2016/07/05/windpark-borssele-goedkoopste-ter-wereld (in Dutch only). 17 www.rijksoverheid.nl/actueel/nieuws/2016/12/12/nederlandse-consortium-bouwt-tweedewindpark-borssele-nog-goedkoper (in Dutch only).

Netherlands 435 In relation to the allocation of the SDE-subsidy, it should finally be noted that the Minister of Economic Affairs issues a decision in view of the Dutch General Administrative Act when granting the SDE-subsidy. This means the SDE-subsidy is not an agreement with, for example, the TSO. Therefore, the developers are required to enter into a normal Power Purchase Agreement (or similar agreement) with an offtaker. If the market price is above the basis amount (or tender amount) the subsidy recipient will not receive any SDE-subsidy and therefore the generator of renewable electricity needs to rely on the Power Purchase Agreement in order to earn back its costs. Given the fact that the tender amounts for the development of offshore wind farms have decreased dramatically, it is conceivable that the operators of these wind farms no longer require the SDE-subsidy if the market price increases.

C. Assessment of the European Commission The SDE-subsidy is clearly a form of State aid in the sense of Clause 107 of the TFEU. In the Netherlands, there has never been a discussion, like the PreussenElektra case,18 of whether or not the SDE-subsidy is State aid and should be notified to the Commission. The Netherlands notified the SDE-subsidy scheme to the Commission in August 2007. The Netherlands notified the SDE-subsidy scheme as an amendment to the former subsidy scheme for the MEP-subsidy. The Commission approved the system of the SDE-subsidy on 21 December 2007.19 The Commission determined that the SDE-subsidy scheme is compatible with the common market, since it limits ‘the subsidy to the difference between the production costs of the renewable energy and the market price of the form of power concerned and [is] therefore in line with point 59 of the environmental aid guidelines’.20 Later amendments to the SDE-subsidy scheme have also been notified to and approved by the Commission. According to paragraph 160 of the Commission’s ‘Community guidelines on State aid for environmental protection 2008’,21 Member States must notify in advance any individual case of investment or operating aid granted under an authorised scheme when the aid is granted to renewable electricity installations in sites where the resulting renewable electricity generation capacity exceeds 125 MW. This enables the Commission to carry out a more detailed assessment of any substantial amounts of aid and to decide whether such aid is compatible with the common market. Following the ‘Guidelines on State aid for environmental protection and energy 2014–2020’, the threshold was increased to 250 MW unless the aid is granted on the basis

18

Case C-379/98 PreussenElektra, EU:C:2001:160. Stimulating Renewable Energy, Modification of the MEP (N 707/02) (Case N478/07) [2008] OJ C39/1. 20 Since this decision taken in 2007, the decision refers to the Community guidelines on State aid for environmental protection of 3 February 2001 [2001] OJ C37/5. 21 Commission, ‘Community guidelines on state aid for environmental protection’ [2008] OJ C82/01. 19

436 Marinus Winters of a competitive bidding process.22 When the aid is granted through a competitive bidding process, an individual assessment by the Commission is not necessary. In the past, the Netherlands notified several individual SDE-subsidy decisions due to this obligation in the Guidelines. The Commission did not raise objections on these notifications. Most of these decisions are still not published on the website of the Commission,23 with the exception of the decision in relation to Q10 Offshore Wind BV (currently known as the Luchterduinen offshore wind farm).24 In this decision, the Commission (ao) determined that the Netherlands provided credible information according to which the aid to Q10 Offshore Wind BV will be limited to the minimum. The Commission furthermore noted that, in particular: [T]he SDE system has been designed to only subsidise the extra production costs of producing renewable energy. The conclusion that the aid is kept to the minimum is further reinforced by the fact that a tender procedure was used for the purpose of granting the aid, compelling parties to submit a project proposal as competitive as possible, thus mitigating the risk of overstimulation. Moreover, financial data provided by the authorities confirms that the aid results in achieving a minimum rate of return as required for such projects.

As indicated above, SDE-subsidy decisions for offshore wind farms are tendered in combination with a permit. This system has been in place since 2016 and this amendment has been notified to the Commission as well. In its decision in relation to this modification of the SDE-subsidy scheme of 7 April 2015, the Commission explicitly confirmed that individual subsidy decisions granted to offshore wind farms no longer require an individual assessment, since the SDE-subsidy decisions for offshore wind farms are granted in a competitive bidding process.25

D. Comparison of the Guidelines Environmental Protection and Energy 2014–2020 The aforementioned decision of the Commission in relation to the modification of the SDE-system contains an assessment of the SDE-subsidy scheme on the basis of the Commission’s ‘Guidelines on State aid for environmental protection and energy 2014–2020’ (EEAG). In this decision, the Commission determined that the SDE-subsidy scheme is compatible with the common market. The Commission assessed the SDEsubsidy scheme on the basis of the general compatibility provisions in Chapter 3.2 of the EEAG and the specific compatibility criteria for operating aid granted for electricity from renewable energy sources. The Commission considered that: —

The SDE-subsidy scheme aims at an objective of common interest in accordance with Clause 107(3)(c) of the TFEU. The aim of the notified aid is to help the Netherlands achieve the renewable energy targets set by the EU as part of its

22 Commission, ‘Guidelines on State aid for environmental protection and energy 2014–2020’ [2014] OJ C200/1, para 20. 23 Aid for offshore wind farm Buitengaats (Case SA.31961), Decision of 22 November 2011, not yet published; Westermeerwind (Case SA.32859), Decision of 21 December 2011, not yet published. 24 Q10 Offshore Wind BV (Case SA.34742) [2013] OJ C1/7. 25 Modification of Dutch SDE+ RES scheme (Case SA.39399) [2015] OJ C234/1.

Netherlands 437 2020 strategy, as well as its domestic longer-term goals to increase the share of energy generated from renewable sources. — There is a need for State aid, since the Netherlands has explained that without the aid it will not meet the EU 2020 target to supply at least 14 per cent of energy from renewable sources in 2020. According to paragraph 116 of the EEAG, the Commission presumes that achieving the targets is appropriate to provide aid, provided that all other conditions of the EEAG are met. — The SDE-subsidy scheme has an incentive effect, since no subsidies will be granted to projects on which work has already started (point 50 of the EEAG). This does not apply to biomass projects, but the Commission considers that the requirement of point 50 of the EEAG is met as long as the cost of producing energy from biomass is higher than the energy market price after plant depreciation. — The SDE-subsidy scheme is proportionate. In this respect, the Commission considers that the subsidies will be allocated to the beneficiaries via a competitive bidding process, where successful participants will receive a level of subsidy based on the phase in which they made the bid for their project. This is also explained above. The risk of overcompensation is further limited by the technology-specific subsidy caps based on an acceptable 7.8 per cent rate of return. The scheme also complies with point 124 of the EEAG, since the subsidy is in the form of a variable premium on top of the reference price for electricity. Dutch producers have standard balancing responsibilities and no subsidy will be paid for hours in which the APX price is negative (for at least six consecutive hours). The Commission furthermore considered that point 126 of the EEAG requires aid granted from 1 January 2017 to be granted through a competitive bidding process, provided that the bidding process can be limited to particular technologies. The Netherlands has justified the separate tenders for offshore wind on the basis of the longer-term potential of offshore wind and the need to achieve diversification. Offshore wind farms are significantly more expensive than the projects of other renewable energy producers. — The Commission finally determined that the Netherlands complies with the transparency requirements (points 104–06 of the EEAG), the financing requirements (point 29 of the EEAG) and the non-cumulation requirements (section 3.2.5.2 of the EEAG). As a result, the Commission considered that the notified SDE-subsidy scheme is an appropriate instrument necessary to pursue an objective of common interest and that therefore the SDE-subsidy scheme is compatible with the internal market pursuant to Clause 107(3)(c) of the TFEU.

IV. CASE LAW IN RELATION TO THE DEVELOPMENT OF PROJECTS

In recent years there have been several cases in the Netherlands in which people living in the surrounding area of a new planned renewable project, like a wind farm, tried to block the development of these new projects with the argument that this project was not financially viable, since these projects were subsidised.

438 Marinus Winters In spatial planning cases, affected parties that wish to appeal a zoning plan can argue that the financial feasibility of the zoning plan is not assured. A zoning plan is a policy document of, for example, the municipality, that provides for the spatial planning of a certain region. These plans can be adopted by the municipality (the municipal council), the province, or the central government (the Ministry). Affected parties who disagree with the plan can oppose it by appealing to the court. Affected parties are people living in the direct vicinity who are affected by the new spatial developments and whose enjoyment of their property is impaired. To get the plan off the table they can, among other things, raise doubts about the financial feasibility of the plan by arguing that unlawful State aid is being granted. Reclaiming the aid would block the realisation of the plan. The affected parties must first deem it likely that there is State aid that can be reclaimed, and second that the Ministry/ municipal council could reasonably have foreseen in advance that, as a consequence, the plan cannot be realised within the planning period without unlawful State aid being granted. It is only if the affected parties manage to make these two arguments seem likely that their arguments can result in the annulment of the zoning plan.26 In existing cases, the court first assessed whether the Commission had given its view on the existence of State aid. If the Commission decided that there is no unlawful State aid, there are no grounds to assume that the Ministry/municipal council should have assumed that the financial feasibility of the plan would not be sufficiently ensured in view of the State aid. The same applies if the actual decision to grant a subsidy must still be approved by the Commission but the draft decision has already been approved, or if the Commission sees no reason to conduct an investigation into a breach of State aid regulations.27 Reliance on financial feasibility will not succeed in these cases. If the Commission has not expressed its view on the existence of State aid, the court will assess, whether the existence of State aid is probable. The court will subsequently examine whether the financial feasibility is sufficiently ensured if the aid would be reclaimed. The court will also assess this if there are no reasons to assume that State aid is involved or if the subsidy must still be approved by the Commission,28 should it later become apparent that there was indeed State aid. The financial feasibility is sufficiently ensured if it is likely that the undertaking, after the aid has been reclaimed, will still continue to realise the plan. Matters considered here are, for instance, a statement from the beneficiary undertaking that it will also continue with the plan after aid has been reclaimed, whether the undertaking is able to acquire other funds,29 or a situation where the realisation of the plan is already at an advanced stage.30 These circumstances contribute to the likelihood that the undertaking will not abandon the plan, which means that the financial feasibility

26 Dutch High Administrative Court (Afdeling bestuursrechtspraak van de Raad van State), 8 February 2012, NL:RVS:2012:BV3215, para 2.76.4.2. 27 ibid, para 2.76.4.1 28 ibid, para 2.76.4.2. 29 ibid. 30 Dutch High Administrative Court (Afdeling bestuursrechtspraak van de Raad van State), 22 October 2008, NL:RVS:2008:BG1152, para 2.6.

Netherlands 439 is ensured. And also, in the case where the undertaking has to withdraw, financial feasibility only starts to play a role if it is unlikely that there are other market parties that could continue the realisation of the plan. Those market parties could in an amended form and/or way—one that fits within the plan—take over the realisation of the zoning plan.31 If it appears that the financial feasibility is sufficiently guaranteed by other means, a final assessment of the existence of State aid is not necessary. When a court has doubts about the financial feasibility of the zoning plan, as a result of the existence of State aid, the court will examine whether the Commission has already expressed its opinion on the existence of State aid. If not, it will consider whether State aid is likely, and assess whether the plan would also be realised without State aid. Only once it has been deemed likely that State aid does exist and that the zoning plan cannot be realised within the planning period without that aid, can this lead to the opinion that the Ministry/municipal council should reasonably have foreseen in advance that the plan is not financially feasible and, if this is invoked in court, can result in the annulment of the plan. If the financial feasibility can be sufficiently guaranteed by other means, the court does not have to express its opinion on the existence of State aid.

V. CASE LAW IN RELATION TO COAL TAX

In the Netherlands, the Environmental Taxes Act is applicable (Wet belastingen op milieugrondslag). Pursuant to the Environmental Taxes Act, the supply of electricity and gas is subject to taxes. The Environmental Taxes Act determines that coal is also subject to taxes, provided that coal, which is used for the generation of electricity is exempted from taxes, since the supply of electricity is already subject to taxes. In 2012, the Dutch government decided to abolish the exemption for taxes on coal used for the generation of electricity. This decision was part of a wider package of additional taxes and cost-cutting measures. The Dutch government took these measures in order to meet the obligations of the Netherlands in the Stability and Growth pact of the eurozone, which means that the budget deficit cannot exceed 3 per cent of GNP. The abolishment of the exemption for coal meant that the generators of coal-fired power plants (together) were obliged to pay an amount of approximately €115 million per year in environmental taxes. All generators of the coal-fired power plants appealed their tax declarations for the environmental taxes. In these appeals, the generators mainly argued that the abolishment of the tax exemption is in breach of Article 14 of Directive 2003/96/EC restructuring the Community framework for the taxation of energy products and electricity,32 since the abolishment of the exemption has not taken place on the basis of environmental reasons, but for budgetary reasons.

31

ibid, para 2.76.4.2. Council Directive 2003/96/EC of 27 October 2003 restructuring the Community framework for the taxation of energy products and electricity [2003] OJ L283/15. 32

440 Marinus Winters In a judgment of the Court of Appeal in The Hague it is stated that one of the generators also argued that the abolition of the exemption qualifies as illegal State aid.33 In this respect, the generator referred to the judgment of the Court of Justice in the Laboratoires Boiron case.34 In this case, the French Supreme Court asked for a preliminary ruling from the Court of Justice in relation to a tax on direct sales of medicines by pharmaceutical laboratories. This French tax regime meant that sales of medicines by wholesale distributors were not subject to this tax. The Court of Justice decided that it should be accepted that an economic operator such as Laboratoires Boiron may plead that the charge on direct sales by pharmaceutical laboratories is unlawful, for the purposes of applying for reimbursement, on the grounds that it amounts to an aid measure. However, in the Laboratoires Boiron case, the Court also decided that the charge and the alleged aid measure constitute two elements of one and the same fiscal measure, which are inseparable. If the revenue from the tax on direct sales leads to an overcompensation of the wholesale distributors, the tax on direct sales qualifies as State aid. The aforementioned judgment of the Court of Appeal in The Hague did not specifically mention which companies or industries will be supported by the abolishment of the tax exemption for coal, but it should be assumed that the relevant generator meant that gas-fired power plants would benefit. In this respect, the Court of Appeal decided that the revenues of the tax on coal will not be used to maintain the tax exemption on other energy products for the generation of electricity. Therefore, the tax exemption does not qualify as illegal State aid. This generator appealed this judgment to the Dutch Supreme Court (Hoge Raad) and this case is still pending at the Supreme Court.

33 34

Court of Appeal The Hague, 16 October 2015, NL:GHDA:2015:3415. Case C-526/04 Laboratoire Boiron, EU:C:2006:528.

19 Greece ANTONIS METAXAS

I. INTRODUCTION

T

HE PROVISIONS OF EU State aid law have an ever increasing scientific and practical importance due to the broad conceptual approach that is reflected in the definition of the ‘State aid’ notion in Article 107(1) TFEU. Their increased effect is particularly accentuated during times of severe financial crisis. In view of this commonly acknowledged fact, it may come rather as a surprise that the jurisprudence of the Greek courts in this area is sparse and, unfortunately, often characterised by a rather vague understanding and analysis of the core legal notions of EU State aid law as shaped by the established practice of the European Commission and the case law of the EU courts.1 The following analysis aims to present an indicative sample of selected recent judgments of the Greek courts as well as evaluate these empirical findings given the role of national courts as the main pillar of decentralised enforcement of the normative provisions of EU State aid law.

II. VIOLATIONS OF THE NOTIFICATION AND THE STANDSTILL OBLIGATION OF ARTICLE 108 TFEU

A. Preliminary Remarks As is well known, the national courts are under the obligation not only to safeguard the enforcement of State aid procedural rules but also to protect the legal rights of the third parties, who are affected by a possible breach of the notification and the standstill obligation2 according to Article 108(3) TFEU. It is settled case law of

1 See A Metaxas and E Sgouridou, ‘State Aid: The Effective Application of EU State Aid Procedures: From a Plan to Grant Aid to the Recovery of Illegal Aid—The Role of National Law and Practice’ in P Nemitz (ed), The Effective Application of EU State Aid Procedures: The Role of National Law and Practice (The Hague, Kluwer Law International, 2007). 2 It is to be noted that all State aid measures fulfilling the conditions of the Block Exemption Regulation, as amended after the codification of Regulation (EU) 2015/1588 of the Council, escape the notification obligation.

476 Antonis Metaxas the CJEU that this provision is directly applicable,3 which means that the affected parties may challenge its violation directly before the national courts.4 In case the national judge establishes an infringement of these procedural obligations, all the legal effects, which are provided for in the national law concerning the invalidity of the legal act that contains the State aid element,5 are to occur and all the relevant procedures, including the recovery of the illegally granted State aid,6 are to be applied aiming to the substantial implementation of the State aid provisions.7 For this reason, the finding of the illegality of the contested measure, originating in the breach of the notification obligation, as established by Article 108(3) TFEU, is independent from the assessment on the compatibility of such a measure with the internal market.8 The national courts may only determine whether the constituent elements of the State aid notion set by Article 107(1) TFEU are met. Even in case the Commission determines, at a later stage, that the measure is compatible with the internal market, the initial unlawfulness cannot be reversed or retroactively cured.9 Nevertheless, the Commission has limited powers protecting the competitors (or the ‘third parties’ in a broader sense)10 against the grant of illegal State aid, since it cannot order the recovery of the State aid only based on the breach of the notification obligation. This reflects the settled case law of the CJEU11 and underlines the importance of the role of the national courts so as to guarantee the protection of the third parties negatively affected by the illegal granting of State aid.12 3

See C-120/73 Lorenz GmbH v Bundesrepublik Deutschland and others, EU:C:1973:152. See Ul Soltesz, ‘Der Rechtsschutz des Konkurrenten gegen gemeinschaftsrechtswidrige Beihilfen vor nationalen Gerichten’ (2001) 12 Europäische Zeitschrift für Wirtschaftsrecht 202; H.-G. Kamman, ‘Verfahrensrechtlicher und gerichtlicher Individualrechtsschutz im EG-Beihilfekontrollrecht aus der Sicht der Praxis’ in C Nowak and W Cremer, Individualrechtsschutz in der EG und der WTO (Baden-Baden, Nomos Verlag, 2002) 161; G Karydis, Illegal State Aid and Legal Protection of the Concerned Third Parties (Nomiki Vivliothiki, 2013) 144. 5 For the invalidity of the contractual clauses granting the illegal State aid see the German Cassation Court’s ruling (Bundesgerichtshof), Case V ZR 314/02 of 4 April 2003 and Case XI ZR 53/03 (2004) 8 Europäische Zeitschrift für Wirtschaftsrecht 252. On the latest relevant jurisprudence of the Austrian Courts, see B Haslinger, ‘Austria: Bank Burgenland—The Austrian Supreme Civil Court has the last say’ (2015) 1 European State Aid Law Quarterly 3; M Pechstein, ‘Nichtigkeit beihilfengewährender Verträge nach 93 III 3 EGV’ (1998) 16 Europäische Zeitschrift für Wirtschaftsrecht 495. 6 For an extensive analysis of the legal problems that arise in the framework of the execution of recovery decisions, see A Metaxas, Grundfragen des europäischen Beihilferechts (Baden-Baden, Nomos Verlag, 2003) 162 ff. 7 See C-354/90 Fédération nationale du commerce extérieur des produits alimentaires and others v France (FNCE), EU:C:1991:440. 8 EU Commission, Notice on the enforcement of State aid law by national courts [2009] OJ C85/1, para 30. 9 See C-39/94 SFEI and others, EU:C:1996:285, para 67. 10 The beneficiary of the State aid, his competitors as well as any other person concerned in the present case (eg, organisations—associations for the environmental protection or for the consumer’s protection), who could be considered as ‘third parties’ in the broad sense, constitute the notion of the ‘concerned parties’ mentioned in Art 108(2) TFEU. See also G Kotsiras, The Recovery of State Aid According to the EU Law and the Greek Legal Order (Athens, Nomiki Vivliothiki, 2015) 107; L Gyselen, ‘La transparence en matière d’ aides d’ état: Les droits des tiers’ (1993) 29 Cahiers de droit européen 417. 11 See C-301/87 France v Commission (Boussac), EU:C:1990:67; see also C-142/87 Belgium v Commission (Tubemeuse), EU:C:1990:125. 12 See S Werner in F-J Säcker and F Montag (eds), European State Aid Law: A Commentary (Munich, Beck, 2016) 1546 on the duty of the national courts to protect concerned parties as an EU law principle; Karydis (n 4) 151. 4

Greece 477 B. Case Law of the Greek Courts i. Case Law on Retroactive Measures Adopted Against RES Producers As mentioned above, according to the provisions of Article 107(1) TFEU, it is clear that the obligation of the national court to invalidate the illegally granted State aid measure is activated eo ipso, while, on the contrary, the compatibility of the measure with the internal market remains beyond the control of the national judge. The Greek case law, being remarkably limited on this issue, is well illustrated in two important judgments that are examined below. In Judgment No 2406/2014 of the Greek Council of State (Chamber B), the court was called upon to rule if the so-called ‘solidarity levy’, which was imposed on the RES (Renewable Energy Sources) producers’ turnover for the years 2012–14,13 complied with respective provisions of Greek constitutional law, as well as provisions and principles of EU law. This retroactive measure neither existed nor was known at the time when the relevant investments were launched and financed or when the Power Purchase Agreements between the RES producers and the Market Operator were executed. This measure allegedly aimed at combatting the existing deficit of the RES Account although a series of independent scientific studies had shown that this deficit was of a systemic character and caused, inter alia, by structural distortions of the Greek electricity market that provided economic advantages both to local fossil-fuel energy producers and electricity suppliers. The applicants, thus, claimed that this selective levy, apart from infringing a series of sector-specific EU energy law provisions and policy principles, constituted an illegal State aid measure, conferring an unlawful economic advantage to local fossil-fuel energy producers and energy suppliers, as they were explicitly excluded by the legislator from contributing to the reduction of this systemic deficit of the Greek electricity market; furthermore no notification of these legislative measures to the EU Commission took place prior to their implementation. The Council of State ruled that the allegation regarding the State aid qualification of the levy and the breach of the notification obligation was inadmissible, since, inter alia, ‘the consideration regarding the compatibility of a State aid measure with the internal market falls under the exclusive competence of the European Commission’. Exactly the same approach and almost the same wording were adopted in the Judgment No 3068/2015 of the Piraeus Administrative Court of Appeals. In Greece, renewable energy generation is mainly promoted through a guaranteed feed-in tariff system, whose main rules are established by Law No 3468/2006. Thus, Law No 4254/2014 (the so-called ‘New Deal’ Law)14 entered into force in April 2014 aiming, according to its explanatory memorandum, at the elimination of the deficit of the Greek Electricity Market Operator’s RES Account (RES Special Account).

13

Law No 4093/2012, Art 1 (I)(2). Law No 4254/2014 “Measures for the support and development of the Greek economy, in the context of the implementation of Law No 4046/2012, and other provisions of law” (Government Gazette No 85/7.04.2014). 14

478 Antonis Metaxas More specifically, the provisions of this above-mentioned law unilaterally imposed: (a) a sharp downward redefinition of the so far guaranteed feed-in tariffs, thereby amending the terms of the pre-existing contracts between the Market Operator and the RES producers; and (b) a retroactive cut in the amount due to RES producers for the energy produced during the year 2013, having the form of a lump sum. RES producers challenged these measures before the Greek competent courts as an illegally granted State aid measure, since, again, market players not only directly causally linked to the structural problem the legislator intended to address but actually benefiting from it were selectively excluded from their burden to contribute financially to its solution. Furthermore, again the legislative provisions in question were not notified to the Commission, thus violating the provisions of Article 108(3) TFEU. It is essential to underline that the inadmissibility invoked by both abovementioned courts due to lack of competence of the national judge ‘to assess the compatibility of State aid measures’ is unlawfully stored, as this reasoning disregards the notional and legal distinction between the State aid qualification and compatibility assessment and, consequently, the division of competences between the national courts (judging on the notional nature of a measure and its State aid qualification) and the Commission (assessing, apart from the State aid qualification, the compatibility of a State aid measure with the internal market). In fact, the applicants in this legal dispute never actually asked the Greek courts in question to assess the compatibility of the contested measures, but solely to draw the legal consequences that, according to standard case law of the CJEU, arise out of the invalidity of national acts that contain the State aid element and were executed in breach of the notification and standstill obligation. In order to do so, the only thing the national judge was expected to do was to decide if the notional and legal preconditions of a State aid within the meaning of Article 107 paragraph 1 TFEU were met in the context of the concrete contested measures. Thus, as already mentioned, the applicants did not ask the court to assess the compatibility of the contested measures, as the above judgments claim, since: (a) this lies outside their competence and (b) the incompatibility finding is not a prerequisite for triggering the legal consequences of an eventual breach of Article 108 paragraph 3 TFEU. The only prerequisite for that is the categorisation of a given measure as State aid (or not) based on the concrete criteria laid down in Article 107 paragraph 1 TFEU and specified in the established practice of the Commission and the case law of the EU courts. If the national judge thinks that there are notional ambiguities as regards the above categorisation in the sense of the acte clair principle, then he could (or ‘should’ in the case of the Council of State being a Supreme Court) request a preliminary ruling before the CJEU according to Article 267 TFEU. Consequently, both judgments follow a rather problematic doctrinal approach that results in depriving the affected third parties from enjoying an effective judicial protection, something that according to the settled case law of the CJEU falls under the exclusive competence of the national courts. In any case, the above rulings are of great significance for the depiction and legal assessment of the scope of competence, rights and obligations, as well as the leeway for discretion of the national judges, in case they need to make an interpretation of the national law provisions in light of the respective EU rules and regulations.

Greece 479 ii. The Alouminion Case The Alouminion case, which is set out below, demonstrates the important role of the national courts in the effective implementation of the State aid control system.15 Alouminion SA is a large enterprise, established in Greece, which operates in the production of aluminium and constitutes one of the major stakeholders in Europe in this field. The company was founded in 1960 and the Greek State granted it some specific privileges, including the right to be supplied with electricity at preferential tariffs by the (then) monopolistic supplier Public Power Corporation SA (PPC SA).16 The preferential tariffs scheme, that Alouminion SA benefitted from, was examined in 1992 by the European Commission. The latter reached the conclusion that such a scheme did not constitute a ‘State aid measure’ as provided for in Article 107(1) TFEU.17 PPC SA unilaterally decided to terminate its contract18 with Alouminion SA and, following this, started imposing non-preferential tariffs to Alouminion SA as of 1 April 2006. The latter contested the contract’s termination before the competent Athens Court of First Instance. According to the Interim Measures Order of the Single-Member Court of First Instance, issued on 5 January 2007, the contract’s termination effects were suspended ex nunc, meaning that the preferential tariffs were to be reapplied to Alouminion SA by PPC SA until the issuance of the Ordinary Proceedings Judgment of the Court. PPC challenged the aforementioned Interim Measures Order of the Single-Member Court of First Instance before the MultiMember Court of First Instance, which declared the contract terminated ex nunc via its Interim Measures Order, issued on 6 March 2008. In 2008, the Commission received a complaint about the alleged State aid measure granted to Alouminion SA in the form of the preferential tariffs paid by the latter during the period between the two above-mentioned Interim Measures Orders, claiming that this granted an economic advantage to the recipient compared with the usual tariff for large industries. In 2010, the Commission initiated the relevant formal investigation procedure which, one year later, resulted in its consideration that the privileges assigned to Alouminion SA indeed constituted a State aid measure incompatible with the internal market.19 Following this, Alouminion SA disputed the Commission’s decision before the General Court of the EU that was called upon to examine, inter alia, the nature

15

See C-590/14 P DEI and Commission v Alouminion tis Ellados, EU:C:2016:797. The Public Power Corporation SA was at the time the sole electricity producer and supplier in the market, wholly owned by the Greek State, and is also today the dominant power producer and electricity supply company in Greece. 17 Commission Decision SG (92) D/867 of 23 January 1992 (Case NN 83/91). 18 The agreement provided for the possibility of unilateral termination upon two-year prior written notice of the co-contractor. 19 Aluminum of Greece (Case SA.26117), European Commission Decision 2012/339/EU [2012] OJ L166/83: according to this decision (a) the preferential tariff for Alouminion SA constitutes an economic advantage compared with the price paid by industries in comparable situation; (b) there is use of State resources as PPC SA is a State-owned enterprise; (c) Alouminion SA only benefits from this measure, which proves its selective character; (d) there is distortion of competition as Alouminion SA operates in a relevant market of products widely traded among Member States. 16

480 Antonis Metaxas and the legal implications of the national courts’ judgments. The applicant claimed that the first Interim Measures Order, issued on 5 January 2007 did not amend substantially the initial regime, while the Commission argued that—since PPC SA stopped providing Alouminion SA with preferential tariffs once the contract was terminated—any provision of reduced electricity tariffs to Aluminion SA, which took place after the contract’s termination, should be considered as new State aid, since the provided State aid was extended due to the first Interim Measures Order, issued on 5 January 2007. Thus, according to the Commission, such an extension constituted ipso facto a new State aid, because it caused an alteration to the duration of the initial provisions, which was also taken into account by the Commission in its decision approving the initial regime in 1992. The General Court ruled in favour of the applicant but the Court of Justice upheld the appeal against the General Court’s ruling. In fact, both the General Court and the Court of Justice agreed on the fact that the extension of the duration of a measure constitutes a new aid, but disagreed on the impact of the national Interim Measures Order, issued on 5 January 2007. Did they constitute a simple suspension of the State aid provisions or an alteration of the State aid regime? For the purposes of our analysis, the main point of interest about this judgment is related to the notification and standstill obligation, as referred to in Article 108(3) TFEU, prohibiting the grant of new aid by Member States without notifying the Commission and obtaining authorisation respectively. In fact, the General Court considered that the consequence of an Interim Measures Order cannot be equalled to the granting of a State aid, otherwise the national courts would have the disproportionate duty to notify the Commission of any measure affecting the implementation or interpretation of a contract that might have an impact on the functioning of the internal market. On the contrary, the CJEU notes that the State aid control system is built on the essential and complementary roles of both the Commission and the national courts. The national courts, even in the context of interlocutory proceedings, must behave according to their duty for sincere cooperation as referred to in Article 4(3) TEU20 and declare as illegal any measure not subjected to the preliminary examination procedure as referred to in Article 108 (3) TFEU. In the same framework, it must be added that, aside from the Commission’s ex officio right to file either its written observations regarding the implementation of Articles 107(1) and 108(3) TFEU, even automatically, before the national courts during the pendency of the relevant proceedings or its verbal observations following the national courts’ respective authorisation, the latter have the right to request the Commission to provide them either with all the information it acquires and/or with its opinion on the issues concerning the implementation of EU State aid law, that way achieving a unified interpretation of the latter and, thus, a unified treatment of

20 ‘Pursuant to the principle of sincere cooperation, the Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks which flow from the Treaties’. Please also refer accordingly to Art 29 of the European Council Reg No 2015/1589 imposing the undertaking of all the measures needed in order to avoid conflicts between the judgments of the national courts and the subsequent decisions of the Commission on the compatibility of the State aid measures with the internal market.

Greece 481 all the cases arising respectively.21 At this point, it comes up as a matter for discussion in the relevant literature, if—and to what extent—the interference of the Commission during the pendency of the proceedings before the national courts entails a further limitation of their procedural autonomy.22 One has to underline the fact though, that the national procedural autonomy of the Member States has in any case, according to the standard case law of the CJEU, clear delimitations in view of the necessity to safeguard the effective, optimum and harmonised implementation of EU law provisions.23

III. RECOVERY OF ILLEGAL AID: THE KEY MECHANISM TO REMEDY DISTORTIONS OF COMPETITION

A. Preliminary Remarks In order to ensure the effectiveness of the enforcement of the State aid prohibition, the Commission is exclusively competent to decide whether a Member State must abolish or alter a State aid measure, in case it is incompatible with the internal market.24 Practically, this implies the obligation of the State to recover the unlawful aid amounts already paid out,25 as referred to in Article 108(2)(a) TFEU, which is broadly interpreted by the CJEU in order to consistently include the recovery of an aid under the notion of ‘abolishment or alteration’, considering that such a recovery effectively restores the status quo ante of the competition environment in the affected relevant market(s).26 Before the entry into force of Council Regulation No 2015/1589,27 the recovery decision was considered as an act falling, in principle, under the discretion of the Commission.28 The above Regulation introduced an important modification whereby, in case of the incompatibility of a State aid measure with the internal market, the Commission is, as a general rule, obliged to issue a decision obliging the Member State to recover the aid granted. The recovery is enforced by means of provisions and procedures of national law, provided that such provisions ‘allow the immediate and effective execution of the Commission decision’, as referred to

21

Werner (n 12) 1529 ff; Karydis (n 4) 222. Werner (n 12) 1550. 23 See A Metaxas, ‘Recovery Obligation and the Limits of National Procedural Autonomy’ (2007) 6 European State Aid Law Quarterly 407. 24 See C-404/97 Commission v Portugal EU:C:2000:345. 25 See L Ghazarian, ‘Recovery of State Aid’ (2016) 2 European State Aid Law Quarterly 228. 26 Tubemeuse (n 11) para 66; See C-169/95 Spain v Commission EU:C:1997:10; See A Evans, European Community Law of State Aid (Oxford, Oxford University Press, 1997) 443; J Schwarze, ‘Deutscher Landesbericht’ in J Schwarze (ed), Das Verwaltungsrecht unter europäischem Einfluß (Baden-Baden, Nomos Verlag, 1996) 157. 27 European Council, Regulation No 2015/1589 of 13 July 2015 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union [2015] OJ L248 (Procedural Regulation). 28 See C-310/85 Deufil v Commission, EU:C:1987:96, para 24; see Metaxas, Grundfragen (n 6) 168. 22

482 Antonis Metaxas in Article 16(3) of the Procedural Regulation.29 The wording of the subject provision illustrates the well-established case law of the CJEU.30 The latter refers to the relevant recovery provisions and procedures laid down in the national law in order to enforce the recovery decision of the Commission. However, the national law provisions apply only to the extent that the effective enforcement of the recovery as well as the substantial restoration of undistorted competition conditions are not compromised. Within this framework, issues arise when the national provisions either are unable to result in the effective implementation of the Commission decision or provide legal means to the recipients to effectively block the recovery procedure. Consequently, when a Member State fails to comply with the recovery decision enforcing it effectively (eg, within the time set by the Commission), the Commission, or each Member State involved, may refer it directly to the CJEU requesting it to declare the breach of the corresponding TFEU provisions, according to Article 28 of the Procedural Regulation and in derogation of the general procedure referred to in Article 108(2) TFEU.31

B. The Opinion of the Greek Legal Council of State (NSK) In 2009, the Commission received a complaint filed by a Greek private casino claiming that the Greek authorities had introduced discriminatory measures in favour of certain public casinos, ie, a levy of an 80 per cent rate on the price of admission tickets for all casinos and two regulated admission fees at €6 and €15 for publicly and privately owned casinos respectively. On 6 July 2010, the Commission initiated a formal investigation procedure falling under Article 108(2) TFEU. More specifically, before the opening of the relevant market in 1994, only three casinos existed in Greece, all operating as State-owned service clubs of the Greek Office of Tourism (EOT). In accordance with the Law No 2206/1994, six private casinos were licensed and started operating, remitting to the State 80 per cent of the price of admission tickets, which were set at drachmas 5,000 (approximately €15) by a Ministerial Decision. On the contrary, the operation of the public casinos was governed by a pre-existing lex specialis, thus excluded from the scope of the aforementioned Law No 2206/1994. Once they were privatised, though, they were licensed according to the provisions of Law No 2206/1994.32

29

See C-243/10 Commission v Italy, EU:C:2012:182, para 36. Tubemeuse (n 11) para 61. 31 Werner (n 12) 1536; Karydis (n 4) 155. 32 In fact: [T]he requirement to remit 80% of the price of admission tickets did not apply to the casinos of Mont Parnès and Corfu until 2003 … as concerns the casinos of Thessaloniki and Rhodes, this requirement became applicable upon issuance of their license under the Law 2206/1994, namely since 1995 in the case of Thessaloniki, and since 1996 in the case of Rhodes. As concerns the price of admission tickets, it has remained at EUR 6 in the case of Mont Parnès and Thessaloniki until present (the date of the 2011 Commission Decision), for Corfu until its privatization in August 2010, and for Rhodes until its privatization in 1999. However, both the requirement to remit 80% and the price of admission tickets of EUR 15 applied to other casinos as of 1995 and have been in practice applied as such’. 30

Greece 483 According to the Commission, since any advantage conferred is assessed according to its effects and, in the subject case, the 80 per cent levy was to be paid to the State on the basis of a lower price for public casinos, a discriminating advantage, similar to a reduction in the tax base, had been granted to those casinos. That advantage arose as the ‘joint effect of a uniform admission tax applied to unequal regulated prices of admission tickets’. It was imputable to the State and financed by State resources as the discriminatory measure had led to a loss of tax revenue for the State, equivalent to the use of State resources according to Article 107(1) TFEU. The selectivity condition was also considered to be met, as the different casinos (ie, those publicly and privately owned) were in a comparable factual and legal situation and the concerned measure was not justified by the nature of the Greek tax system as the Greek authorities did not manage to demonstrate a financial reason justifying the lower price set. Finally, the distortion of competition was also confirmed since the measure had strengthened the position of the beneficiary casinos compared with potential competitors in the intra-European trade, taking into account that international hotel groups are often the operators in this sector. The Commission concluded the existence of incompatible operating State aid, declared illegal as non-notified under Article 108(3) TFEU.33 Since the recovery competence of the Commission is limited to a 10-year period, as of the day the illegal aid is awarded, and no reason for waiving recovery had been found valid, the Commission ordered the immediate and effective recovery (including interest) by the Greek authorities of any aid awarded as of 21 October 1999—10 years before the transmission of the complaint to the Greek State. Following the EC Decision, the NSK has issued an opinion on the prescription of an established claim originating in the State authorities’ obligation of recovery of the granted State aid.34 In this opinion NSK notes that, according to Article 22 of Law No 4002/2011, any incompatible State aid is recovered upon a disbursement invitation addressed to the beneficiary of the aid; in case of non-compliance, the authority in charge of the recovery asks the competent tax service to ‘officially establish’ the claim (via an official document) and levy the amount due. Only this official announcement of the claim marks the beginning of the 20-year prescription set in Greek law.35 Thus, via the above-mentioned opinion, NSK notes that this long prescription period should not be considered as discharging the public authorities from their obligation to pursue the recovery of the illegally granted State aids without delay, according to Article 16(3) of the Procedural Regulation. In spite of this statement, it is difficult to consider that such a prescription ensures the effective enforcement of the Commission decision without compromising the substantial restoration of undistorted competition.

33 Aid to certain Greek Casinos (Case SA.28973) Commission Decision 2011/716/EU [2011] OJ L285/25. 34 Opinion No 171/2014. 35 Actually starting after the closing of the respective financial year of the claim, according to Art 86 of Law 2362/1995.

484 Antonis Metaxas It is well worth noting that in cases when the exact amount to be recovered is specified in the Commission recovery decision, the conditions36 of the Greek law provisions for the issuance of a ‘payment order’37 should be considered to be met. Issuing such an enforceable security entails that, in fact, while the Commission decision spawns the State’s obligation for recovery of the illegally granted aid, the latter shall not necessarily be reimbursed directly to the State, as, due to the payment order, it could (alternatively) be reimbursed to the specific legal entity having granted the aid.38

IV. CONCLUDING REMARKS

The crucial role reserved to national courts to act as a decentralised Union judge ensuring the prompt and effective protection of the individual rights deriving from EU law is indeed of major importance. The CJEU has explicitly stated that the role of the national courts is to ensure the full effectiveness of EU rules and provisions and the protection of the rights they confer on individuals. The function of the national judge disposes of an additional ‘systemic’ dimension, since the regulatory scope of the very same EU rules conferring these rights (systemschützende Funktion) is also strengthened through the effective protection of the individual EU law rights (individualschützende Funktion). This dual function reflects the broader ‘strategic’ role of the national courts to consolidate the regulatory influence of EU law within the national legal order of the Member States.39 In the above analysis, the importance of the role of the national judge and the extent of his awareness concerning the obligation to intervene, so as to protect the rights of individuals affected by breaches of the EU State aid law provisions, is highlighted through a targeted reference to a selected, indicative sample of the State aid case law of Greek courts. Unfortunately, the overall assessment of the existing case law is not satisfying, both from a quantitative as well as from a qualitative (due to the dogmatic inconsistencies that are often traced in many judgments) point of view. Private parties in Greece very rarely challenge the distortion of competition caused by illegally granted State aid to their competitors and also rarely seek legal redress before the national courts (eg, actions for damages). This may be linked to a limited awareness of State aid law in Greece, both within the business community as well as among the legal professions—not excluding the judges—something that is clearly reflected in the small number of relevant court decisions.

36 ie, according to Arts 623 et seq of the Greek Civil Procedure Code an established amount, proven by public or private authentic documentation, is required—among others—for the issuance of a ‘payment order’. 37 A ‘payment order’ is one of the enforceable securities that the Greek law order provides for the collection of the receivables. 38 See S Stamatopoulos and K Papanikolaou, ‘Payment Order for Recovery of State Aid by Private-law Corporations’ (in Greek) [2013] EΠολΔ 492. 39 See A Metaxas, ‘Reflections on the Specific Nature of the EU Legal Order’ (2016) 3 Efimerida Dioikitikou Dikaiou 346.

20 Belgium WIM VANDENBERGHE

I. INTRODUCTION

W

HILST ALL (APPELLATE) court judgments in civil proceedings—which may involve State aid arguments—are, in principle, public, Belgium lacks a disclosure system which publishes court decisions, except for judgments from the highest courts, which are systematically published. It will thus come as no surprise that this country overview is constrained largely to decisions taken by the European Commission in State aid matters in the energy industry in Belgium, and to only a handful of national court judgments.

II. OFFSHORE WIND SUBSIDY SCHEME

The initial aid scheme to support the development of offshore wind power was notified to the Commission by the Belgian government in 2001. The basic features of the scheme were as follows: Belgium’s federal energy regulator would grant a renewable energy certificate (REC) to producers of offshore renewable energy for every MWh generated. Eligible offshore renewable energy producers were those holding both a domain concession and a ‘guarantee of origin’ certificate. The producers would then have to sell their RECs in the regional certificates markets of Belgium.1 In addition, the Belgian transmission system operator (TSO), Elia, was obliged to buy RECs at a guaranteed (minimum) price determined by the federal State from producers that requested Elia to do so. Elia could subsequently sell the purchased RECs on the regional certificates markets. In the event that Elia would sell these certificates on the market at a loss, Elia was mandated to collect a surcharge on its general transmission tariffs to compensate for the loss. The Commission decided that the scheme (both the granting of green certificates as well as the minimum price for these certificates) did not constitute State aid within the meaning of Article 107(1) TFEU. There was no transfer of State resources, given that the scheme is funded by a separate transmission surcharge paid by

1

Both the Flanders region and the Walloon region had developed certificates markets.

486 Wim Vandenberghe Elia’s customers.2 In deciding that no State resources were involved, and thus no State aid, the Commission relied on the (back then recent) PreussenElektra judgment.3 Over the years, the Belgian legal framework, which underlies the subsidy scheme, has been modified. These changes do not appear to have been notified to the Commission, most likely because the Belgian government deemed notification as unnecessary in view of the Commission’s initial finding of ‘no aid’.4 This line of reasoning may, however, be contested. First, the Commission no longer applies a restrictive interpretation of the PreussenElektra case law, which supported the finding of ‘no aid’ in 2002. Instead, it applies more recent CJEU case law which considers that the decisive element in order to assess whether resources are public (even if their initial origin is from a surcharge paid by private companies) is the degree of State intervention/control in the design and financing of support measures.5 This criterion also applies to funds stemming from a surcharge paid by private entities, ie, Elia’s customers. The Commission decided that the unnotified amendments to a Romanian aid scheme approved in 2011, which was similar to the Belgian offshore wind subsidy scheme, did constitute State aid.6 In its decision, the Commission explicitly stated that the presence of ‘State resources’ needs to be reviewed in light of the evolution of CJEU case law. Secondly, substantial modifications have been made to the initial Belgian subsidy scheme over the years, including changes to the minimum price formula and the length of time for Elia to have to purchase the RECs.7 2 Régime fédéral belge de soutien aux énergies renouvables (Case N 14/2002) [2002] OJ C309/04. A more elaborate analysis of why the offshore wind subsidy scheme does not constitute aid can be found in the Commission’s decision regarding the Flemish green certificates system (Case N 550/2000 [2002] OJ C330/3), and the updates in 2005 regarding the Combined Heat and Power certificates (Case N 608/04 [2005] OJ C240/20), and the 2006 decision regarding more intensive support for photovoltaic energy (Case N 254/06 [2006] OJ C314/80). In its 2005 assessment of the Flemish combined heat and power certificates scheme, the Commission concluded that there was no transfer of State resources involved and hence no aid (Case N 608/04 [2005] OJ C240/20). The green certificates system in the Walloon region was assessed equally concluding that no aid was involved (Case N415a/2001 [2002] C30/14). 3 The Court of Justice of the European Union (CJEU) held that: ‘In this case, the obligation imposed on private electricity supply undertakings to purchase electricity produced from renewable energy sources at fixed minimum prices does not involve any direct or indirect transfer of State resources to undertakings which produce that type of electricity’ (Case C-379/98 PreussenElektra, EU:C:2001:160, para 59). 4 Similarly, the Flemish green certificates system has been modified over the years (see n 1). These modifications have been challenged, including before Belgium’s highest administrative law court (Raad van State/Conseil d’Etat). One of the challenges was based on EU State aid law in that the new system does constitute State aid. The Court, however, summarily dismissed this argument in its 2012 ruling by referring to the approval decision from the Commission 12 years ago, in case N 550/2000 that the system constitutes ‘no aid’ (NV Wattplus/NV Essent Belgium v Flemish Region, Case 221.374, judgment of 13 November 2012, available at: www.raadvanstate.be). 5 Case C-206/6 Essent, EU:C:2008:413, para 66; Case T-139/09 France v Commission, EU:T:2012:496, paras 63 and 64; Case C-262/12 Association Vent de Colère!, EU:C:2013:851; Case C-275/13 Elcogas, EU:C:2014:2314. 6 Case SA.37177 [2015] OJ C343/1. 7 Eg, the Programme Law of 26 December 2013 introduced certain reductions of the surcharge in function of the consumption level by means of an ex-post reimbursement by the CREG. The Raad van State/ Conseil d’Etat considered this tax reduction to constitute State aid (DOC 53 3258/001, 66–94). Recent decisions from the Commission confirm this position: Case SA.34045 (2013/C) (ex 2012/NN); Case SA.33995 (2013/C) (ex 2013/NN); Case SA.38632; Case SA.36511 (2014/C) (ex 2013/NN). Also see, s 3.7 of the Commission’s Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C200/01.

Belgium

487

Belgium finally notified the modifications to its scheme in 2016, but the Commission still qualified the updated scheme as State aid.8 It rejected Belgium’s argument that the amended support scheme constitutes ‘no aid’ because no material changes were made to the initial framework.9 In contrast to the 2001 decision, the Commission now considers that the REC regime is not a certificates scheme based on market prices, since producers are not selling green certificates on the regional market. Instead, the offshore producers are granted a premium on top of the market price in the form of a minimum price of certificates paid by the TSO. The REC regime thus does not constitute a certificates scheme where the price is determined in the certificates market. Applying the Vent de Colère judgment,10 the Commission argued that the financial flows stemming from the REC regime are under the strict control of the Belgian State. This includes the statutorily defined calculation method of the RECs’ minimum price, the levelised cost of energy (LCOE) set by the Ministry of Energy and the also statutorily defined surcharge that Elia imposed on its customers.11 The Commission subsequently assessed the REC regime, and two individually notified projects, under the Environmental and Energy State Aid Guidelines (EEAG),and concluded that the support scheme and projects are compatible with the internal market pursuant to Article 107(3)(c) TFEU.

III. NUCLEAR PHASE-OUT SUPPORT

The Belgian Nuclear Phase-Out Act12 not only contains a moratorium for the construction of new nuclear power plants (NPPs), it also restricts the lifetime of all existing NPPs to 40 years as follows: i. ii. iii. iv. v. vi. vii.

Doel 1 (433 MW): 15 February 2015 Tihange 1 (962 MW): 1 October 2015 Doel 2 (433 MW): 1 December 2015 Doel 3 (1006 MW): 1 October 2022 Tihange 2 (1008 MW): 1 February 2023 Doel 4 (1038 MW): 1 July 2015 Tihange 3 (1046 MW): 1 September 2025

Lifetimes can, however, be extended by the Belgian federal government, notably to face security of supply concerns.13 In November 2015, the government reached an

8

Case SA.45867 (2016/N) [2017] OJ C36/1. Belgium also notified two individual aid measures (Rentel wind project and Norther wind project) which were based on the modified support scheme. 10 Association Vent de Colère! (n 5) paras 20–21. 11 The minimum price is the difference between the LCOE and the reference electricity market price. 12 Nuclear Phase-out Act of 31 January 2003, Belgisch Staatsblad/Moniteur belge (Belgian Official Gazette) 28 February 2003. 13 These concerns were highlighted during the prolonged closure of nuclear reactors Tihange 2 and Doel 3 in 2012, following signs of metal degradation, raising fears about their safety. 9

488 Wim Vandenberghe agreement with ENGIE Electrabel to extend the lifespan of nuclear reactors Doel 1 and Doel 2 with an additional 10 years until 2025.14 In 2013, the government had already given a 10-year extension for Tihange 1 until 2025.15 In return, the companies committed to investing around €1.3 billion to safeguard the continued operation of the three NPPs through design upgrades, employee training and other measures. The companies would receive financial compensation if Belgium either closes the plants before the end of the 10-year period, modifies the level of nuclear tax to be paid by the owners or changes other economic parameters of the agreements. In Belgium’s view, nuclear energy requires a long-term commitment and these guarantees were necessary to secure the investment by the companies. Several Benelux-based renewable energy producers filed a complaint in 2016 with the Commission arguing that the lifetime extension and the conditions thereof constitute unlawful State aid. According to the complainants, the nuclear companies will receive an annual cash flow of approximately €420 million for the next 10 years, which equates to an internal rate of return of around 30 per cent. They argue that the nuclear tax that the companies need to pay each year is disproportionate to the profits made. Thus, the Belgian State is foregoing revenues. In addition, the rights to operate the NPPs for an extended period were granted on an exclusive basis to ENGIE Electrabel and EDF Belgium without prior tendering. This would distort competition on the highly concentrated Belgian electricity market in view of ENGIE Electrabel’s market share and the fact that the NPPs concerned are amortised and have low variable costs with high production stability (given their high position in the merit order). The complainants further allege that the measure has a negative effect on the investment climate in energy markets and more particularly for alternative energy generation, and even leads to the closure of gas-fuelled power plants. The Commission decided on 17 March 2017 that the aid is compatible with the internal market.16 It concluded that Belgium has sufficiently demonstrated that the measures avoid undue distortions of the Belgian energy market. They include an obligation on ENGIE Electrabel to sell on regulated electricity markets each year a volume equivalent to its share of the annual production of Tihange 1, Doel 1 and Doel 2. This should ensure liquidity on Belgian electricity markets and help increase competition between electricity suppliers. On this basis, the Commission has approved the measures under EU State aid rules. The above-mentioned nuclear tax that NPP operators have to pay to the Belgian State was also challenged under State aid law before the national courts.17 14 Law dated 28 June 2015 amending the Nuclear Phase-out Act, which lays down the principle of the lifetime extension, provided the Belgian State and ENGIE Electrabel reach an agreement by 30 November 2015. This agreement was signed between the two parties on 30 November 2015. 15 Tihange 1 NPP is co-owned by ENGIE Electrabel and EDF Belgium. The extension was stipulated in the law of 18 December 2013 (amending the Nuclear Phase-out Act) whereas the terms and conditions of the extension were detailed in a tripartite agreement of 12 March 2014 between the Belgian State, ENGIE Electrabel and EDF Belgium. 16 European Commission, Press Release of 17 March 2017, available at: www.europa.eu/rapid/pressrelease_IP-17-662_en.htm. The actual decision is not available yet. 17 The tax is calculated in function of the profits and costs of producing electricity from the NPPs. For details on the calculation methodology, see CREG, report (F)150312-CDC-1407, 12 March 2015, available at: www.creg.info/pdf/Studies/F1407NL.pdf.

Belgium

489

This challenge was actually made by the operators themselves before Belgium’s Constitutional Court. The claim was that non-NPP energy producers (ie, fossil fuel and renewables) benefit from State aid as they—in contrast to NPP operators—do not have to pay a similar nuclear tax.18 In its judgments,19 the Court rejected the State aid claim in a very succinct way by ruling that not all State aid criteria were fulfilled. In particular, the Court held that the nuclear tax is not a selective measure, as the NPP operators are in a different factual and legal situation compared with non-NPP energy producers in terms of the objective pursued by the nuclear tax.

IV. FROM ELECTRICITY MARKET LIBERALISATION TO CAPACITY MECHANISMS

Following the First Electricity Directive,20 Belgium adopted a transitional regime for the electricity market to offset the financial impact of commitments imposed on electricity producers and distributors in a pre-liberalised world.21 This transitional regime was analysed under State aid law. Such analysis gave the Commission an opportunity to specify the criteria under which the derogation in Article 107(3)(c) TFEU was applicable to aid-constituting stranded costs,22 in light of the methodology for analysing State aid linked to stranded costs.23 The Commission distinguished between three components of the transitional regime. The first concerned the dismantling of experimental nuclear sites for which electricity producers and the Belgian federal State had been jointly responsible since 1990, six years before the First Electricity Directive was adopted. The Commission considered that the compensation received by Electrabel (now ENGIE Electrabel) and SPE (now EDF Luminus) satisfied the criteria in points 4.1–4.3 of the methodology and therefore qualifies as permissible stranded costs. However, the Commission had doubts regarding the pension scheme for employees in the electricity sector given the non-specific nature of the commitments given by Electrabel and SPE to their employees. All companies covered by the collective agreement for the electricity and gas sectors, including new entrants, were subject to the same obligations. After the Commission had opened a formal investigation, Belgium withdrew its notification with regard to this part of the transitional regime.24 The third component of the measure concerned the promotion of renewable energy sources and energy conservation, financed by a price payable by electricity

18 Other arguments were filed, including breach of several constitutional law principles such as nondiscrimination, non-retroactivity and legitimate expectations. 19 Constitutional Court, decision 17 July 2014, Case 106/2014; Constitutional Court, decision 17 September 2015, Case 114/2015, available at: www.const-court.be. 20 Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L027/20. 21 This was allowed for under Art 24 of the First Electricity Directive. 22 Case C 31/2002 (ex N 149/2000) [2002] C222/2. 23 Commission Communication of 26 July 2001 relating to the methodology for analysing State aid linked to stranded costs; Commission letter SG (2001) D/290869 of 6 August 2001. 24 Case C 31/2002 (ex N 149/2000) [2005] OJ C175/8.

490 Wim Vandenberghe end-consumers that was higher than the market price. Since the facts were similar to those of a British case,25 the Commission reproduced its earlier analysis and found, in accordance with the PreussenElektra judgment, that the measure did not involve any aid element. While capacity mechanisms are very topical at the moment, generation support schemes have already been in place following the Second Electricity Directive.26 More recently, the Belgian Energy Ministry also introduced various generation support actions supposedly needed to safeguard electricity security of supply in the short, medium and long term.27 These actions mainly concerned tendering for new capacity and building a strategic reserve. Contrary to other EU Member States, Belgium did not formally notify these measures, even though they were highly questionable under State aid law.28 In 2014, Belgium launched a tender to attract investment in 700–900 MW of open-cycle gas turbine (OCGT) or combined-cycle gas turbine (CCGT) capacity. The tender was challenged before Belgium’s highest administrative law court. Shortly after, in early 2015, the tender was abandoned. The Court judgment therefore only ruled that the legal action no longer had an object.29 However, it does refer to the reasons why the Belgian Energy Ministry abandoned the tender. Interestingly, this seems to have been done in view of State aid issues. The measure would not have been compatible with the internal market because it is not technology neutral and would have a significant negative impact on competitive market dynamics.

25

Case N 661/1999 [2002] OJ C113. For an overview, see FE Gonzalez-Diaz, ‘EU Policy on Capacity Mechanisms’ in L Hancher, A de Hauteclocque and M Sadowska (eds), Capacity Mechanisms in the EU Energy Market: Law, Policy, and Economics (Oxford, Oxford University Press, 2015). 27 Commission Staff Working Document accompanying the Final Report of the Sector Inquiry on Capacity Mechanisms, COM(2016) 752 final. 28 W Vandenberghe and R Gonne, ‘Belgium’, in L Hancher, A de Hauteclocque and M Sadowska (eds), Capacity Mechanisms in the EU Energy Market: Law, Policy, and Economics (Oxford, Oxford University Press, 2015) 253–55. 29 Raad van State, decision of 30 June 2015, Case 231.827 T-Power v Belgian State. 26

21 Spain JOSE LUIS BUENDÍA AND MIGUEL ÁNGEL BOLSA

I. INTRODUCTION

T

HE AIM OF this chapter is to provide an overview of a number of topics related to State aid law in the energy sector within the Spanish jurisdiction. With that aim, the chapter will particularly focus on several national court cases that, based on State aid principles, have contributed to shape the current (and maybe future) structure of the Spanish electricity system. In order to do that, section II will start with a brief overview of how the Spanish national electricity system is designed and the legal framework that currently regulates it, with a brief mention of its evolution. This serves as background information to help the reader understand the subsequent sections. Section III will be devoted to the main focus of this chapter, and the most important topic in Spain within the scope of this book: the so-called ‘Spanish renewables saga’, or, in other words, the various investigations in different forums or from different institutions of the Spanish subsidy scheme in favour of renewable sources of energy, and its subsequent reforms by the Spanish government. This regime and its reforms are, at the time of writing this chapter, under investigation by the Spanish national courts, by the European Commission and by international arbitration tribunals, as will be explained. The chapter will conclude, in section IV, by considering additional smaller but highly topical issues of current importance for Spanish electricity legislation, such as the ‘social bond’ adopted by the Spanish Parliament (a regulated discount for disadvantaged consumers) which has also been challenged before national courts, and the regime on charges to self-consumption of energy, colloquially also known as the ‘sun tax’, which has also been the source of controversy as a consequence of which its legal regime may be subject to further reforms in the future.

II. OVERVIEW OF THE NATIONAL LEGAL FRAMEWORK OF THE ELECTRICITY SECTOR IN SPAIN

Prior to 1997, the electricity sector in Spain was made up of public companies with legal monopolies acting on the most important activities of the sector.

492 Jose Luis Buendía and Miguel Ángel Bolsa Spain introduced its first liberalising Electricity Sector Law in 1997 (Law 54/1997),1 by which EU Directive 96/922 was transposed into Spanish legislation. This legislation aligned the Spanish electricity system with those from other EU Member States and set the foundation for the modern legal framework for electricity in Spain. Seventeen years later, after a number of partial reforms, and in order to further consolidate the liberalisation of the sector, Spain adopted the new Electricity Sector Energy Law (Law 24/2013)3 in 2013, which is currently in force, and which focuses on the economic and legal unbundling of the different market players exercising the various activities within the system, in line with EU Directive 2009/72.4

A. Regulated and Liberalised Activities The Spanish Electricity Sector laws of 1997 and 2013 distinguish between regulated and liberalised activities. The transport of energy is strongly regulated, and it includes the transmission of energy through high-tension networks (transport of energy from producers to low-tension networks) and the distribution of energy through low-tension networks (from high-tension networks to consumers): —

The transmission of energy remains as a legal monopoly, with Red Eléctrica Española (REE) (20 per cent publicly owned) as the unique transmission system operator (single TSO model). — The distribution of energy, meanwhile, is a regulated activity carried out by a number of big electricity companies. Transmission and distribution are activities remunerated by regulated charges set by national legislation and implementing acts. Electricity networks must therefore be open to any supplier that pays a regulated network poll or access charge. These costs are ultimately passed on to consumers in the regulated electricity bill they pay. This remuneration is used by REE and the distributors to invest in the construction, maintenance, development and operation of the transmission and distribution infrastructure. On the other hand, the production of energy and the supply of energy (the purchase of electricity from producers and its provision to consumers) are liberalised activities: —

1

The supply of energy has traditionally been carried out by distributors in Spain, and provided at regulated tariffs. However, the Electricity Sector Law of 2013 introduced for the first time the total liberalisation of the energy supply sector with the possibility for small, independent suppliers (not belonging to vertically

Law 54/1997 of 27 October 1997 on the Electricity Sector. Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L27/20. 3 Law 24/2013 of 26 December 2013 on the Electricity Sector. 4 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] L211/55. 2

Spain



493

integrated corporate groups) to enter the market and give the opportunity to consumers to switch to unregulated prices set by suppliers. Meanwhile, the production of (ordinary) energy is remunerated on the basis of the wholesale price established in electricity auctions between producers and suppliers (the so-called ‘pool’ of energy). Without prejudice to that, the production of energy based on certain (renewable) sources is remunerated, exclusively or partially, on the basis of public subsidies.5

The economic and technical management of bids and offers of electricity in the wholesale market is carried out by an electricity market operator called OMIE (Operador del Mercado Ibérico de Electricidad), which acts in both Spain and Portugal. Finally, the whole electricity system is supervised by the Spanish independent electricity authority, formerly the stand-alone CNE (Comisión Nacional de la Energía), which, since 2013, is embedded within the National Authority for Markets and Competition (Comisión Nacional de los Mercados y la Competencia, or CNMC).6

B. Income and Expenses of the System The electricity system involves all the operators of the different electricity markets (production, transport and consumption). The system is financially closed: it comprises a number of income sources and several regulated expenses which have to be covered by the income of the system. If the expenses and the income of the system do not perfectly match, then a deficit or a surplus is generated by the system. The main source of income of the electricity system is the revenue resulting from the regulated tariffs paid by consumers of electricity (customers of the system). Although the regulated bill paid by consumers has been reformed several times in recent years, in general terms, the regulated tariff paid by consumers in Spain can be divided into two main elements: on one side, the market-price remuneration for electricity producers (ie, the cost for suppliers of procuring such energy in the wholesale market or ‘pool’ which, as we mentioned, is formed on the basis of periodic auctions) and, on the other side, a number of additional charges that are destined to cover the so-called ‘regulated expenses of the system’. These regulated expenses are established either by law or by the independent electricity market supervisor and include, inter alia: remuneration of network operators (remuneration for the costs of building, maintaining and operating the transmission and distribution networks); compensation for stranded costs to certain companies;7 subsidies to producers of

5 As will be explained below, before the 2013 reform of the electricity system, renewable energies were remunerated exclusively on the basis of public subsidies, regardless of the wholesale price of energy resulting from auctions. Following the reform, renewable energies are sold in the pool at market price, and the remuneration for its production is complemented by public subsidies. 6 Law 3/2013 of 4 June 2013 creating the National Commission for Markets and Competition. 7 Set by Transitory Provision 10 of 1997 Electricity Sector Law (following Arts 24(1) and 24(2) of Directive 96/92). Examined by the European Commission in its State aid Decision Scheme for competition transition costs (Case NN49/1999) [2001] C268/5. Later reformed by Law 50 of 1998 (securitisation) and by Royal Decree-Law 2 of 2001 (return to its original regime).

494 Jose Luis Buendía and Miguel Ángel Bolsa energy on the basis of specific sources, particularly renewable sources (subsidies to renewables) or indigenous coal;8 subsidies to companies operating in the Spanish territories outside the Iberian Peninsula (Balearic Islands, Canary Islands, Ceuta and Melilla); subsidies to companies of the energy sector in financial difficulties subject to a viability plan approved by the government; costs for nuclear waste disposal; and capacity payments to certain installations in order to ensure the stability of supply to certain geographic areas within Spain. In addition, although strictly speaking they are not considered income sources for the system, there are several taxes applied to different market participants in the system (eg, producers) that are used to reduce the tariff deficit or accumulated debt in the system, when it exists, which are therefore also indirectly used as a last-resource income for the system.9 The Spanish system of regulated tariffs was comprehensively analysed in 2014 by the European Commission from a State aid point of view in two Decisions. In its Decision SA.21817,10 the Commission considered that the system did not grant State aid to customers (some of them are large companies) and in its Decision SA.36559,11 the Commission found no State aid to distributors (which, as explained, obtain their revenue from regulated tariffs).

III. RENEWABLE ENERGY SUPPORT SCHEME AND ITS REFORMS

As we anticipated in the introduction, the longest section of this chapter is aimed at analysing the problems surrounding the support scheme for renewables adopted by Spain in 2007, which was substantially reformed numerous times in the years that followed. The reforms (adopted through several legislative acts mostly during the period from 2011 to 2014) consisted of a significant reduction in the incentives for the production of electricity based on renewable energy sources, and affected the expected profitability of a very large number of national and international investors who had invested in the sector, attracted by the government incentives.

A. The Original Spanish Support Scheme for Renewable Energy and the Problem of the Tariff Deficit The support to renewable energy in Spain originally dates back to the 1980s with the establishment of incentives for hydraulic power plants. However, it was in 2007 when, in the context of the green energy and polluting emission objectives set by the European Union to its Member States, and in a time of

8

See Preferential dispatch of indigenous coal plants (Case N 178/2010) [2010] C312/5. See section III.B of this chapter. 10 Spanish Electricity Tariffs: consumers (Case SA.21817) Commission Decision 2014/456/EU [2014] OJ L205/25. 11 Spanish Electricity Tariffs: distributors (Case SA.36559) Commission Decision 2014/457/EU [2014] OJ L205/45. 9

Spain

495

economic boom in Spain, the government adopted Royal Decree 661/2007,12 which designed a very generous subsidy scheme to promote the generation of electricity based on renewable energy sources and to attract national and international companies to invest in the sector. In particular, this Royal Decree included a remuneration system for renewable electricity producers which was not market based but based on subsidies, priority access to the distribution network and a guarantee that all the energy produced would be fed into the system. The incentive regime was a complete success in terms of attracting investors and increasing the use of renewable energy. As a consequence of the new legal framework, the generation of electricity in Spain based on renewable energy doubled within a few years and reached a significant share of the energy mix. The technology which most benefited from the scheme was wind energy, along with others such as photovoltaic energy. However, the outstanding success of this scheme also resulted in a significant increase in the ‘regulated expenses’ of the electricity system. The electricity producers using renewable energy were remunerated outside the market, on the basis of subsidies designed to cover the whole cost of installations and to generate a significant profit for investors. Those subsidies were considered ‘permanent expenses of the system’, which meant that the remuneration of these producers came from the income of the system, particularly from the regulated tariffs paid by consumers. However, while the electricity system expenses increased exponentially, the electricity system income did not follow the same pattern. In fact, the government did not proportionately increase the regulated tariffs paid by consumers to match the increase in expenses, and due to the economic crisis in Spain, the revenues from electricity consumption even decreased. As a consequence of that, the deficit of the electricity system inevitably skyrocketed. As a provisional solution, Spain legislated to ensure that the big electricity producers operating in the country temporarily financed the deficit in the system (the gap between expenses and revenues). These companies (Iberdrola, Fenosa, Hidroeléctrica del Cantábrico, Endesa and Elcogás), already identified by Spanish legislation as being the companies receiving compensation for stranded costs,13 provided in advance the funds to the electricity system, in exchange for the right to be reimbursed in the future. The electricity system deficit, however, continued to increase and, not much later, a profound reform of the system proved to be unavoidable.

B. The Successive Reforms of the Scheme Spain then initiated a series of reforms of the regulated revenues and expenses of the electricity system, aimed at controlling and reducing the tariff deficit. 12

Royal Decree 661/2007 of 25 May on the regulation of electric energy under a special regime. These undertakings had invested in electricity production assets in the framework of a nonliberalised electricity sector. The Commission declared its decision of not raising objection to compensation for stranded costs in Spain on 25 July 2001 (IP/01/1079). 13

496 Jose Luis Buendía and Miguel Ángel Bolsa The first reform, in 2010, consisted of the adoption of Royal Decree-Law 14/2010, on the Establishment of urgent measures to amend the tariff deficit in the electricity sector,14 of 23 December 2010; and Royal Decree 1565/2010, on the regulation and modification of certain aspects relating to electricity production under the special regime.15 As mentioned before, it was soon clear that the existing subsidy scheme for renewables was unsustainable, and the same government which had adopted it, was forced to reform it. These reforms introduced a number of adjustments to the system: — First, producers started to be charged for access to transport networks. — Secondly, the new legislation established a legal limit on the number of hours that installations under the subsidised regime could operate (before that, it was not limited, and the subsidised installation had the guarantee of selling all the electricity they produced). — Thirdly, the new legislation limited to 25 the number of years that installations could benefit from the subsidised regime (originally no limit was established).16 — Finally, the reform imposed on subsidised installation several additional technical requirements relating to response mechanisms to protect the electricity system in the event of power failure in the network. These requirements were once again not established in the original design of the regime. It is important to highlight this reform in the analysis because it is the only one which, as will be explained below, was analysed by the first arbitration award adopted by the Stockholm Chamber of Commerce, after a complaint from international investors against the legislative reform. The second reform, in 2012, consisted of the adoption of Law 15/2012 of 27 December 2012, on tax measures to ensure the financial sustainability of the energy sector.17 After the general election of 2011, a new government took over in Spain. By that time, it was clear that one of the most urgent tasks for the newly elected government was to find a solution to the (at the time already quite substantial) electricity system deficit. Naturally, in order to reduce the deficit, it was possible to act by either increasing the income of the system or by reducing its expenses. The first option was initially chosen by the new government. By virtue of Law 15/2012, the government created a number of new taxes on different operators in the electricity system. The proceeds raised by these new taxes were considered a new source of income for the electricity system and therefore they were used to cover the regulated expenses of the system or to reduce the existing tariff deficit. The most 14 Royal Decree-Law 14/2010 of 23 October on the establishment of urgent measures to reduce the tariff deficit in the electricity sector. 15 Royal Decree 1565/2010 of 19 November on the regulation and modification of certain aspects relating to electricity production under the special regime. 16 In successive reforms, however, the time limitation was extended first to 28 and then to 30 years in total. In any case, the original RD of 2007 did not include any limit other than for technical reasons. 17 Law 15/2012 of 27 December on tax measures to ensure the financial sustainability of the energy sector.

Spain

497

important tax (in quantitative terms) was a 7 per cent new tax on electricity producers’ turnover, but in total the fiscal reform of the electricity system involved the creation or the increase in six different taxes (seven if including Law 24/2013, adopted two years later, which will be considered below). The list of taxes that were created or increased includes: —

— — —

— — —

New tax on electricity producers: Law 15/2012 established a new tax on electricity production installations, regardless of the technology they used (based on renewable sources of energy or not). The tax rate was set at 7 per cent, not on the profits of the installation but on its total turnover. New tax on nuclear waste production. The amount of the tax depended on the type of waste and its level of radiation. New tax on nuclear waste storage, to be paid in installations which store, against a remuneration, nuclear waste produced by other installations. New canon on the use of inland waters by energy producers profiting from an inland water concession. The canon was set at 22 per cent of the produced energy value. Reform of the tax on hydrocarbon fuels production. Increase in the tax on energy production based on coal. Finally, increase in the tax on energy consumption (originally introduced by Law 66/1997, increased later by Law 28/2014).

The most relevant aspect of this legislation is that it established that the revenues generated from the new and increased taxes were considered a new source of income for the electricity system and therefore they were deemed to contribute to cover the regulated expenses and to reduce the existent tariff deficit. The third reform, in 2013 was the adoption of Royal Decree-Law 9/201318 and the new Electricity Sector Law 24/2013. The tax increase of 2012 proved insufficient to control the deficit, and the government decided in 2013 to further reform the legislation, this time focusing on the expenses side. In order to reduce the electricity system’s regulated expenses, the government acted on the subsidies to renewable energy. These subsidies had become the largest and fastest-growing expense item of the system, and were a significant cause of the increase in the system’s deficit over the previous years. The government thus adopted, by the urgent legislative procedure, Royal DecreeLaw 9/2013. The provisions of this legal text were later included in the new Electricity Sector Law 24/2013. The new Royal Decree-Law (RDL) formally replaced Royal Decree 661/2007 (which had created originally the subsidy regime for renewables). RDL 9/2013 was followed by two important implementing legislative acts, namely the Royal Decree 413/2014,19 which specified the calculation methodology 18 Royal Decree-Law 9/2013 of 12 July on urgent measures to ensure the financial stability of the electrical system. 19 Royal Decree 413/2014 of 6 June on the regulation of energy production based on renewable sources, cogeneration and waste.

498 Jose Luis Buendía and Miguel Ángel Bolsa for the remuneration of renewable energy producers, and Industry Ministry Order 1045/2014,20 which implemented that methodology for the first-year calculation. The new legal text introduced a number of substantial reforms in the remuneration system for renewable energy producers. Unlike the original regime, the remuneration for renewable energy producers turned from a subsidy system exclusively cost-based and the guaranteed absorption by the system of all the electricity produced, towards a remuneration system based on the market price plus a prime to compensate that renewable energy technologies were not yet considered mature enough to compete on an equal footing against traditional, polluting sources of energy. In other words, as from the adoption of the new legislation, the electricity produced based on renewables, as for any other type of energy, would be put on sale at the energy wholesale market or ‘pool’ and the sale of this energy would no longer be guaranteed. If producers wanted to sell it, they needed to do so at market prices. The legislator, however, acknowledged that these green technologies had not yet reached a competitive point, and therefore it foresaw a prime or complementary compensation to allow the producers to (i) cover their costs and (ii) to obtain ‘reasonable profitability’. However, and unlike under the previous regime, according to the new legislation the costs of producers that were used to calculate the producers’ remuneration were calculated on the basis of a grid that determined the level of costs an ‘efficiently managed undertaking’ should bear, depending on a number of factors such as the technology used, the capacity or the level of production. This had the consequence of obliging renewable energy producers to match the ideal efficiency requirements established by law, or otherwise exit the market. On top of being able to cover those ‘efficiency costs’, the legislation was designed to provide the renewable energy producers with ‘reasonable profitability’. This profitability, however, was set at a level significantly lower than that guaranteed by the previous legislation. The reasonable profitability had to be calculated, according to the new legislation, on the basis of the yield of 10-year Spanish public debt plus 300 basic points. Following this methodology, the first-year ‘reasonable profitability’ was originally set at 7.4 per cent return on investment. This contrasted with the profitability expected under the previous legal regime, which could amount to up to 20 per cent return on investment. Finally, and crucially, the new remuneration system was not intended to apply only to new investments, but also to already existing investments. In particular, the ‘reasonable profitability’ was deemed to be calculated for the whole lifespan of investments already started. This implied that the reform affected the expected profitability of investments that had already been initiated. The new Electricity Sector Law (Law 24/2013), adopted only a few months later, fully incorporated the remuneration regime designed by RDL 9/2013. The reforms created their intended—but also unintended—consequences. On the one hand, their main objective was fulfilled: the tariff deficit has remained under

20 Industry Ministry Order IET/1045/2014 of 16 June on the remuneration methodology for certain energy production installations based on renewable sources, cogeneration and waste.

Spain

499

control and the system has even generated a surplus over the years 2014 to 2016. However, there was also an unintended but highly predictable result: the investment in renewable energies has decreased substantially in Spain and its participation in the energy mix has also been reduced.

C. Legal Challenges The successive reforms adopted by Spain created a huge controversy among renewable energy producers, who saw their expected profits fall from an approximate 20 per cent to 7 per cent. Many national and international investors reacted by challenging the reforms adopted by the Spanish government before national courts, in the case of the former, and before international arbitration forums, in the case of the latter. Even regional governments put forward legal challenges against the reformed national legislation before the Spanish Constitutional Court. i. Proceedings before the Spanish Constitutional Court Regional governments were also indirect beneficiaries of the subsidies to renewables, because these subsidies attracted national and international investors to underdeveloped rural areas in Spain, generating economic activity and job creation. As a consequence, some regional governments were hugely disappointed with the subsidy cuts adopted by the Spanish central government. Some of these regional governments went a step further and challenged the reforms before the Spanish Constitutional Court, alleging that the reforms infringed a number of provisions in the Spanish Constitution. The regional government of Murcia (a southeastern region of Spain) was the first regional government to file an application for the annulment of RDL 9/2013 before the Constitutional Court. Not much later, the regions of Navarre and Extremadura followed. The judgments were released between December 2015 and March 2016.21 All four of them are very similar in substance. A summary of their main legal arguments follows. a. Possible Infringement of the Principle of Non-Retroactivity of the Legislation and of the Protection of Legitimate Expectations of Investors The applicants considered that the new legislation undermined investors’ legitimate expectations because the remuneration system applied not only to possible new investments but also to already existing installations.

21 Judgment of the Spanish Constitutional Court 270/2015 of 17 December 2015 (Applicant: Regional Government of Murcia); Judgment of the Spanish Constitutional Court 19/2016 of 4 February 2016 (Applicant: Regional Government of Navarre); Judgment of the Spanish Constitutional Court 29/2016 of 18 February 2016 (Applicant: Regional Government of Navarre); Judgment of the Spanish Constitutional Court 61/2016 of 17 March 2016 (Applicant: Regional Government of Extremadura).

500 Jose Luis Buendía and Miguel Ángel Bolsa The Constitutional Court disagreed. It considered that the principle of legitimate expectations protected by Article 9(3) of the Spanish Constitution does not imply a ‘frozen’ or ‘immutable’ legal order, and it is perfectly legitimate for a government to legislate reforms of a previous regime. In this regard, the Constitutional Court highlighted that the Spanish legislator justified the (urgent) need for reform in the preamble of the new law, and that the necessity of the reform was furthermore predictable for any diligent investor given the upward trend of the tariff deficit. The Constitutional Court recalled, however, that the limit to retroactive effects included in the Spanish Constitution only prohibits those effects in ‘punitive’ legislation and in legislation ‘restrictive of individual rights’. As regards the second case, the Constitutional Court considered that unlawful retroactive effects would only be found if the legislation revoked profits already monetised by investors, but not if investors only suffered a decrease in their expected profits. b. Possible Infringement of the Principle of Legal Certainty The applicants also considered that the principle of legal certainty was violated because the law contained vague or unspecific terms such as ‘efficiently managed undertaking’ or ‘reasonable profitability’, which, however, played a hugely important role in order to calculate the new remuneration of renewable energy installations. The Constitutional Court however considered that those terms were objectively defined in the legal act itself. The term ‘reasonable profitability’, for example, was defined in relation to the Spanish public debt yield. In the case of ‘efficiently managed undertaking’, this expression was developed with a number of criteria such as the production, capacity and technology of the installations. c. Possible Infringement of the Energy Charter Treaty Anticipating some of the crucial discussions that were going to take place before international arbitration courts, regional governments had already put forward legal pleas before the Spanish Constitutional Court on the possible infringement of the Energy Charter Treaty, an international treaty that protects investors in the energy sector.22 As would later happen in the arbitration proceedings, applicants specifically referred to Articles 10 (‘stable, fair and transparent conditions for foreign investors’) and 13 (‘ban of expropriations or equivalent measures’) of the Energy Charter Treaty. In order to find a way of presenting this issue before the Spanish Constitutional Court, the applicants argued that, by violating the provisions of this treaty with the adoption of an RDL contrary to it, the government was also infringing the principle of legislative hierarchy set by the Constitution (given that under that principle, international treaties signed and ratified by Spain are hierarchically superior norms than ordinary national legislation).

22

Energy Charter Treaty of 17 December 1994.

Spain

501

The Constitutional Court however refused to examine this plea. The Court argued that it was not competent to examine such an indirect infringement of a constitutional principle, since it would be required first to prove that such contradiction between the treaty and the national legislation existed, which was not a competence of the Constitutional Court. d. Conclusion In conclusion, the Spanish Constitutional Court fully upheld the reforms adopted by the government and dismissed the applications for annulment submitted by regional governments. The Constitutional Court ruled on the merits as regards the pleas on constitutional legal principles (retroactivity, legal certainty) and refused the plea on the infringement of the Energy Charter Treaty. ii. Proceedings before the Spanish Supreme Court National investors, who in theory cannot sue their own country before international arbitration courts, had chosen to challenge the reforms adopted by the government before the Spanish Supreme Court. Since they do not have legal standing to challenge RDL 9/2013, they applied for the annulment of its implementing acts, namely the already mentioned Royal Decree 413/2014—which contained the questioned remuneration system—and the Industry Ministry Order 1045/2014. The Supreme Court had, at the time of writing this chapter, already adopted several judgments on the applications for annulment submitted by several national investors. All of them were resolved in equivalent terms, upholding the possibility that Spain may reform its subsidy regime and therefore dismissing the applications for annulment. In our analysis, we will follow the first judgment adopted by the Supreme Court on the matter, but the succeeding judgments followed the same structure and most times simply quoted the original judgment. Before starting, however, it is important to highlight that the Supreme Court judgments were not unanimously approved by the judges. They contained dissenting opinions on the issue of retroactivity, as will be explained below. a. Case 1259/2016 INCAM Ingeniería Civil, Alternativa y Medioambiental (Civil Alternative Environmental Engineering) (INCAM) is a producer of energy which owned a renewable energy installation, subject to the special subsidised regime, in the Spanish region of CastileLa Mancha. It was one of the first producers which applied for annulment of the implementing legislation of the electricity reform before the Supreme Court, and its judgment was the first to be released, on 1 July 2016.23 In the course of these proceedings, Spain decided to notify its subsidy regime to the European Commission as State aid, years after being implemented without

23

Judgment of the Spanish Supreme Court of 1 July 2016 Case 1259/2016 INCAM.

502 Jose Luis Buendía and Miguel Ángel Bolsa notification. This move by the Spanish State was believed to be a strategy to avoid the payment of compensation to both national and international investors, following the Micula doctrine of the European Commission (see below). The Spanish legal representation informed the Supreme Court of the said notification and argued that the proceedings before the Supreme Court should be suspended until the European institution expressed a view on the matter. The Supreme Court however did not accept that argument, and decided to continue the proceedings. The debate on the allegedly unlawful retroactive effects of the regime reform was richer before the Supreme Court than before the Constitutional Court that we explained above. This was the case because the Supreme Court already had the opportunity to examine not only RDL 9/2013 but also its implementing legislation, which included the specific new remuneration methodology for renewable energy installations. INCAM, as well as other investors that followed it, argued before the Supreme Court that the new legislation had clear retroactive effects, because the calculation of the remuneration—which affected existing installations—was based on data preceding the entry into force of the legislation, such as the installations’ initial investment and maintenance costs during the first years, or the lifetime of the installation (which was determined at the beginning of the investment). In other words, since the already mentioned ‘reasonable profitability’ had to be calculated for the entire lifespan of a certain installation, the new legislation affected the remuneration that that installation was going to receive not only from the entry into force of the legislation, but also since the starting point of the investment. It could even be the case, the applicants argued, that the ‘accumulated’ profits of certain installations by the date of entry into force of the new legislation had already surpassed the ‘reasonable profitability’ set by the new legislation, and that, in those cases, investors were even forced to pay back the ‘excessive’ remuneration received to the administration. Despite that, the Supreme Court fully followed the legal reasoning of the Constitutional Court by highlighting that the prohibition of retroactive effects included in the Spanish Constitution only applied to legislation which is punitive or restrictive of individual rights. The Supreme Court agreed with the Constitutional Court that prior to the legislative reform, investors could have had expectations of profits but, in no case, could those profits be considered as ‘individual rights’ already materialised. Those expected profits, when they were affected by the legislation could not be considered as acquired profits or profits already incorporated into the investors’ wealth, and therefore its reduction could not be considered a restriction of individual rights. The Supreme Court also clarified that the extreme situation alleged by the applicants, whereby an investor would be forced to reimburse profits to the administration, was mostly a theoretical scenario and could, in reality, never happen, because the new remuneration regime did not affect the remuneration that had already been paid to investors. This reasoning was, however, not shared by the dissenting opinion adopted by two of the judges (out of seven) of the Supreme Court.

Spain

503

The dissenting opinion considered that the new legislation did infringe the prohibition of retroactivity, as well as the principle of legal certainty and the legitimate expectations of investors. The dissenting judges agreed with the Constitutional Court in that RDL 9/2013 alone could not create retroactive effects, but disagreed with their colleagues of the Supreme Court in considering that the combination of RDL 9/2013 with its implementing measures of 2014 was able to generate unlawful retroactive effects. They reached this conclusion because it was this implementing regulation that established the details of the new remuneration system, and in doing so created the fiction that installations already in place were subject to the new regulation over the course of their complete lifespan, even if part of it had occurred when the new legislation had not yet been adopted. Only that way, according to the dissenting judges, was it possible to explain that the legislation aimed at providing existing installations (not only new installations) with a ‘reasonable profitability’ which was calculated ‘during their complete lifetime’. Moreover, the dissenting opinion considered that while it could be argued that a legislative reform was predictable given the tendency of the tariff deficit, in no case did that mean that such a far-reaching legislative reform could be expected by investors (and, in fact, actions carried out by them seemed to demonstrate that they clearly were not expecting such a reform). It seems however that the dissenting opinion did not address one of the key points of the judgment, which is the consideration that the reform of the remuneration system for renewables is ‘legislation which is punitive or restrictive of individual rights’. The main judgment highlighted that retroactive effects are only unlawful in that type of legislation. Strictly speaking, the judgment of the Supreme Court had not denied that such retroactive effects could be generated by the new remuneration system, but it considered that the retroactivity prohibition protected by the Spanish Constitution was not applicable to the legislation regulating the remuneration system for renewables. Finally, the Supreme Court also addressed several other pleas raised by the applicants, but in all cases dismissed the possible infringement of EU Directive 2009/28,24 or even the European Convention on Human Rights25 (which also prohibits expropriation). After this judgment, the Supreme Court adopted a large number of judgments and resolutions which resolved the legal challenges submitted by a significant number of mostly national investors. These proceedings were resolved in the same terms and with references to the judgment previously adopted, and include, inter alia, the judgment of 20 July 2016 Compañía Europea General Textil SL,26 and the judgment of 3 October 2016 TRADELDA.27

24 Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC (OJ L140/16). 25 European Convention on Human Rights of 4 November 1950. 26 Judgment of the Spanish Supreme Court of 20 July 2016 Case Compañia Europea General Textil SL. 27 Judgment of the Spanish Supreme Court of 3 October 2016 Case TRADELDA.

504 Jose Luis Buendía and Miguel Ángel Bolsa b. Elcogás Judgment Unlike the judgments explained in the previous sections of this chapter, the Elcogás judgment28 did not concern the subsidies to renewable energy producers, but other types of regulated expenses of the electricity system. As mentioned above, one of the other regulated expenses of the electricity system are the subsidies in favour of electricity installations subject to a viability plan approved by the government. In particular, according to the Spanish 2013 Electricity Sector Law, currently in force, the government can approve viability plans in favour of participant companies of the energy system of strategic importance if they suffer financial difficulties. The viability plans qualify the restructuring costs as ‘regulated expenses of the electricity system’, and therefore they grant the companies undergoing restructuring the right to obtain funds through the electricity system, which therefore would be paid ultimately by consumers through their electricity bills. Elcogás was a company which owned a coal plant located in the town of Puertollano, in the Spanish region of Castile-La Mancha. The installation was going through financial distress in 2006. The government considered the plant to be of strategic importance and reacted approving a 10-year viability plan which included public subsidies to the plant. In 2010, however, in the middle of the economic crisis and with the tariff deficit uncontrolled, the government suspended the support to Elcogás. In particular, Industry Ministry Order 3353/2010,29 which had to set the regulated costs of the electricity system for 2011, did not include the expected remuneration for Elcogás. Elcogás reacted by challenging the government’s decision before the Supreme Court by applying for the annulment of that Order, considering that it was contrary to Article 17(1) and the Final Provision 20th of the 2013 Electricity System Law (which was the legal basis for subsidies in favour of installations under a viability plan). Alternatively, Elcogás demanded compensation for the remuneration eliminated from that Order. In the course of those proceedings, Spain argued that those payments should be considered as State aid, and since they had not been notified to the European Commission, their use would amount to a breach of the standstill obligation included in Article 108 TFEU. As a result of this, Spain argued that it had not only the right but, in fact, the obligation to withdraw support to the company (eighth legal ground of the judgment). The Supreme Court then took the decision to refer a preliminary question to the European Court of Justice (ECJ) asking whether the resources used by the electricity system in Spain (which are paid by consumers of electricity) had to be considered ‘State resources’ according to Article 107 TFEU. The ECJ replied to the Spanish Supreme Court by its Order of 22 October 2014 (Case C-275/13),30 where it explained that the Vent de Colère doctrine was directly

28

Judgment of the Spanish Supreme Court of 27 February 2015 Case Elcogás. Ministerial Order ITC/3353/2010 of 28 December setting electricity system tariffs and primes for the year 2011. 30 Order of the European Court of Justice of 22 October 2014 Case C-275/13 Elcogás SA v Administración del Estado and Iberdrola SA, ECLI:EU:C:2014:2314. 29

Spain

505

applicable to the case and thus the support to Elcogás indeed constituted State aid. The ECJ explained that despite the private origin of the funds (the electricity consumers), the fact that the funds were channelled through the electricity system by a public entity (the former national energy independent authority) was enough to consider such funds to be ‘State resources’ within the meaning of Article 107 TFEU. After receiving the ECJ preliminary ruling, the Supreme Court then decided in April 2013 to carry out a complete State aid analysis of the payments received by Elcogás (sixth and subsequent legal grounds of the judgment), and found that the payments were selective and provided an advantage to the company, with the effect of distorting competition and affecting trade. The Supreme Court included (twelfth and thirteenth legal grounds) an implicit objection to the ‘low standard’ required, according to the ECJ case law, to fulfil the criteria of distortion of competition/trade effects between Member States, but reminded that a national court can do little more than simply applying the mentioned case law, which the Supreme Court considered clear enough. In any case, the Supreme Court asserted that the Elcogás plant’s importance was clearly beyond the local community and that the energy produced in that plant was pumped into the national electricity system that, at the same time, transcended the Spanish national border given the energy exchanges with Portugal and France. The importance of this judgment relies on the fact that the costs of the Elcogás viability plan were remunerated, like the subsidies to renewables, from the electricity system. It therefore set an important precedent for the State aid analysis that the European Commission is carrying out over the renewables subsidies. In particular, the ECJ has by its Order in the Elcogás case asserted that the funds received by the Spanish electricity system coming from consumers, given that they are channelled through a public institution (the CNMC), qualify as ‘State resources’. The Supreme Court itself realised the importance of the Elcogás ECJ preliminary ruling for the analysis of subsidies for renewables, deeming it much more important in quantitative terms than payments for viability plans and, in a revealing paragraph, it considered: ‘this is not the scenario to highlight the relevance of this judgment as regards other types of remuneration to energy producers’ (eleventh legal ground). That said, the analysis on selectivity and the existence of advantage towards subsidies to renewable energy could potentially be different and the Supreme Court explicitly stated that it could not analyse the possible compatibility of subsidies with the internal market. Finally, the Supreme Court considered that it was clear that Spain had not notified the measure at the time it was required to do so and therefore did not respect the standstill obligation. In those circumstances, the Supreme Court considered that Spain had rightly withdrawn the support to Elcogás. As regards the demand for compensation by Elcogás, interestingly, the Supreme Court, in relation to the international arbitration proceedings, considered that the granting of compensation to Elcogás could not be upheld because the only ground for compensation would have been the withdrawal of a regime that had been qualified as illegal for not having been notified to the European Commission for State aid analysis. Therefore, the Supreme Court dismissed the request for compensation by Elcogás.

506 Jose Luis Buendía and Miguel Ángel Bolsa iii. International Arbitration Proceedings a. Introduction One of the most important consequences of the subsidy scheme adopted by Spain in 2007 was that many international investors were attracted by the conditions and decided to invest in renewable energy production installations in Spain. The volume of foreign investment attracted exceeded every forecast made by the government, and the share of renewable energy in the Spanish energy mix substantially increased. It could be argued, however, that the scheme ended up being a victim of its own success, and it was precisely the overwhelming increase in investments in renewables that made the regulated costs of the electricity system skyrocket and the deficit of the system unmanageable, creating the need to adjust the incentives downwards. However, following the successive reforms of the Spanish subsidy scheme carried out by different Spanish governments, many national and international investors saw their profit expectations decrease and reacted, in the case of international investors, by suing Spain before the international arbitration systems designated by the investor protection international treaties signed by Spain, particularly the Energy Charter Treaty of 1994. It is important to highlight that, according to the Energy Charter Treaty, an international element is required to start international arbitration proceedings (ie, in theory, a national company cannot litigate against its own country before an international arbitration tribunal in this context). However, some Spanish corporate groups have also made use—with the acquiescence of arbitrators—of these mechanisms using non-Spanish subsidiaries or associated companies. The three international arbitration systems that have been used to sue Spain pursuant to the Energy Charter Treaty are: (i) the International Centre for Settlement of Investment Disputes of the World Bank (ICSID); (ii) the United Nations Commission on International Trade Law (UNCITRAL) (where most of the complaints were submitted); and (iii) the Arbitration Institute of the Stockholm Chamber of Commerce (AISCC). By December 2016, more than 30 international complaints against Spain were reported to have been filed. Among them, only two arbitration awards had already been adopted, and so far only in the first case had the full text of the award been made public. The first two arbitration proceedings were submitted before AISCC. AISCC adopted its two first arbitration awards in January 2016 and July 2016, in both cases dismissing the complaints and therefore upholding Spain’s position (ie, considering that the legislative reforms did not infringe the Energy Charter). Before going into the content of the awards, however, it is important to highlight that none of the arbitration awards already adopted have analysed RDL 9/2013, which is the legal act that in appearance could be accused (without prejudging the final outcome of the proceedings) of having stronger retroactive effects. In particular, the first arbitration award, whose text was already public by the time of finishing this chapter and which we will analyse in the following lines, only analysed the first set of reforms, adopted by RDL 14/2010. The second arbitration

Spain

507

award, whose result was also known, but whose full text was not yet public, was also favourable to Spain but it only analysed the reforms of 2010 and 2012 (ie, it included the rise in taxes first adopted by the new Spanish government that took office in 2012). b. AISCC Arbitration Award of 21 January 2016: Charanne and Construction Investments v Spain Charanne and Construction Investments are two companies belonging to the Isolux corporate group. Isolux is a Spanish corporate group, but these two societies are based in the Netherlands and Luxembourg, respectively. Due to this circumstance, the first discussion before AISCC was about the admissibility of the complaint, since Spain argued that Isolux is a Spanish corporate group and therefore no society of this group should be allowed to sue Spain before an international arbitration court. The final award was released in January 2016, three years and nine months after the submission of the application. The arbitration award confirmed the admissibility of the application by these two companies. However, it dismissed the complaint on the merits, considering that the legislative reforms adopted by Spain did not infringe the Charter. Crucially, as we were suggesting above, the January 2016 arbitration award only analysed the possible infringement of the Energy Charter by Spain as regards the adoption of the 2010 legislative reform (namely RDL 14/2010 and a number of implementing acts, including RD 1565/2010). It is, however, important to highlight the scope of the arbitration award because, as already mentioned, the 2010 reform is allegedly the reform with the smallest degree of retroactive effects. The investor complaint was divided into two main pleas: On an Alleged Infringement of an Equitable and Fair Treatment towards International Investors (Article 10(1) Energy Charter) The arbitration award considered that the investors could not have the legitimate expectation that the legal regime in Spain would never be changed. In addition to recalling that a legal regime can never be considered ‘frozen’ or ‘immutable’, the arbitration award also highlighted that there was previous Spanish case law (which was public and therefore should have been known by the investors) allowing these types of legislative changes, and also because every economic projection of the electricity system deficit soon indicated that it was not under control and that adjustment reforms were going to be unavoidable. The original legislative framework itself included, in fact, the possibility of regime changes. In addition to that, the arbitration award emphasised that in no case can a legislative regime and the expected profits generated by it, be considered equivalent to a long-term contract or a formal commitment between a State and a group of investors. Against the retroactivity allegations of the complainants, the arbitration award considered that the reforms analysed were not retroactive, because they did not affect the period of the investment prior to them, but only the profits that investors

508 Jose Luis Buendía and Miguel Ángel Bolsa were likely to generate from the time of the reform onwards. It has already been highlighted, however, that the reform analysed by this arbitration award was only the one adopted in 2011, which was the less aggressive in retroactivity terms. Expropriation of Rights (Article 13 Treaty on the European Union) The arbitration award considered that it was not appropriate to qualify the reforms as an expropriation. The property of the installation was kept in investors’ hands. According to the arbitration court, even if the installation lost value, this could have happened for a number of reasons, including legislative reforms, and such loss in value could not be considered equivalent to a (forbidden) expropriation. The arbitration award dismissed the investors’ complaint by a majority of two votes out of the three members of the arbitration tribunal. Both the arbitrator appointed by Spain (Claus von Wobeser) as well as—crucially—the President of the arbitration tribunal (Alexis Mourre, President of the Stockholm Chamber of Commerce) voted in favour of Spain, whereas the arbitrator named by the investors (Guido Tawil) gave a dissenting vote. As explained, the award found no wrongdoing in Spain’s reform. It is interesting however that the arbitration award sets out that a legislative reform may infringe the Energy Charter if it carries the following three consequences: — —



First, a loss of almost the total value of the investment. Secondly, a non-justified fundamental legislative change that affects the very framework of the legal regime (eg, the total elimination of the right to receive public support by investors). Thirdly—and this could potentially be relevant for future proceedings— retroactivity is not found despite the fact that the reform affected ongoing investments, as long as the effect starts only from the moment of the reform itself. The arbitration award therefore apparently left open the possibility of finding unlawful retroactivity if a reform affects time periods preceding the moment the reform was adopted.

In order to analyse the potential relevance of this arbitration award, it is important to highlight first, that it was the first adopted arbitration award in the Spanish renewables saga. Secondly, it was elaborated and signed by arbitrators who are considered to have a high reputation in the sector. There are, however, a few reasons why the importance of this first arbitration award should not be overstated. First, arbitration awards do not formally set a binding precedent, especially as regards the different arbitration tribunals that will decide on the matter (and it is important to remember that most of the complaints were submitted not before AISCC but before UNCITRAL). Crucially, the first AISCC arbitration award did not analyse the reforms that took place in Spain in 2012 and—most importantly—the reform of 2013. c. Second AISCC Arbitration Award: Isolux Infrastructure Netherlands v Spain (July 2016) As in the first AISCC arbitration award, these arbitration proceedings were also initiated by a company belonging to the Isolux corporate group.

Spain

509

Following the same logic as the first proceedings, the complaint was admitted despite the fact that the complainant belonged to a Spanish corporate group. The main novelty of this proceeding was that it included in its analysis the Spanish reform of 2012. As has been explained, this reform focused on the income side of the system, and raised a number of new taxes to different participants of the electricity system, including a 7 per cent tax (over the annual turnover) on energy producers. The result of this arbitration award was announced in July 2016, and it was also favourable to Spain. Its full text had not been published by January 2017. d. Pending Arbitration Awards As we have explained, most of the complaints have been submitted before UNCITRAL. It has been reported that the number of complaints is close to 30. The first complaint before UNCITRAL was submitted in November 2011 by a group of 16 investors under the name of PV Investors. The typology of investors seeking compensation before UNCITRAL is very wide. It includes German public banks, investment or hedge funds from Japan, the US or the Middle East, and European electricity companies such as E.ON and RWE. e. Micula Doctrine Therefore, allegedly, the most interesting arbitration awards are still to come. That leaves a situation where, on the one hand, Spain has notified the subsidy regime to the European Commission, which will need to decide whether it constitutes State aid. In order to do that, the Commission will need to take into account a very important precedent, the Elcogás case, where the ECJ (and following its Order, the Spanish Supreme Court) had already decided that subsidies to companies subject to a viability plan funded by the same resources which are also used for the subsidies to renewables (coming from consumers who pay regulated tariffs, but channelled through a public institution) are considered State aid. In parallel, as we have explained, arbitration proceedings will decide whether the withdrawal of these subsidies by Spain constituted an infringement of international investor protection treaties. To make things a bit more complicated, this situation will need to be resolved against the background of the Commission’s Micula decision.31 In this Decision, the Commission prohibited Romania from complying with an international arbitration award which ordered the country to compensate a group of Swedish investors following the withdrawal of a Romanian subsidy scheme previously declared State aid by the European Commission itself. The Commission considered that complying with the arbitration award (which had found that by withdrawing an investment incentive scheme four years prior to its scheduled expiry, Romania had infringed a bilateral investor protection treaty between Romania and Sweden) by compensating the investors would circumvent the prohibition to grant State aid and therefore would impair the effectivity of State aid control.

31

Micula v Romania (ICSID Arbitration Award (Case SA.38517) [2015] OJ L232/43.

510 Jose Luis Buendía and Miguel Ángel Bolsa The Micula doctrine, mentioned above, seems also to have been followed by the Spanish Supreme Court which, in the Elcogás judgment, denied compensation for Elcogás precisely on those grounds. The Commission decision in Micula is currently under appeal before the GC: Cases T-624/15 (European Food and others v Commission)32 and T-704/15 (Micula v Commission).33

IV. FURTHER DEVELOPMENTS

The fourth and final section of this chapter is devoted to two recent developments in Spanish electricity legislation in relation to the remuneration of energy production that has been the source of a significant controversy, and which is likely to be reformed in the near future.

A. Social Bond The social bond is a regulated discount on the electricity tariff for specific groups of disadvantaged consumers. It was originally included in the Spanish legislation in 2012 but its current version dates back to 2014.34 The social bond grants a discount of up to 25 per cent on the regulated electricity tariff to certain groups of consumers, namely: (i) those who contract the minimum electricity power (below 3 kW); (ii) elderly people receiving low retirement pensions; (iii) large families; or (iv) households with all its members unemployed. In 2016, more than seven million people in Spain benefited from the social bond. While the social aim of the measure is uncontested, the financing mechanism to cover the approximately €200 million of annual costs that is generated has been the object of several challenges before the Spanish courts. In particular, Spanish legislation used to make the five largest electricity companies liable for bearing 94 per cent of the cost of the social bond, since regulation distributes this cost among the vertically integrated groups that are simultaneously present in the energy production, distribution and commercialisation markets. In 2015, these electricity groups challenged the financing mechanism of the social bond before the Spanish Supreme Court arguing that, as a social measure, the social bond should be financed by the public budget, not by them. In two judgments of 24 October 2016, the Supreme Court upheld the complaints from Endesa and E.ON and annulled the financing mechanism of the social bond35

32 Case T-624/15 European Food and Others v Commission. Application for annulment lodged on 6 November 2015. 33 Case T-704/15 Micula and Others v Commission. Application for annulment lodged on 28 November 2015. 34 Royal Decree 968/2014 of 21 November on the financing mechanisms for the social bond. 35 Judgments of the Spanish Supreme Court of 24 October 2016, STS 4526/2016 (ES:TS:2016:4526) and STS 4527/2016 (ES:TS:2016:4527).

Spain

511

on the basis of considering it incompatible with EU Directive 2009/7 on the electricity common market,36 which states that ‘public service obligations have to be clearly defined on a transparent and non-discriminatory basis’. In particular, the Supreme Court found it discriminatory that only certain companies within the electricity markets had to bear the costs of the social measure. The Supreme Court followed, without considering it necessary to raise a preliminary question, the ECJ case law on Anode (C-121/15),37 which analysed a similar situation in the gas sector. The Supreme Court found it discriminatory that: (i) there was no contribution to the social bond from the company in charge of the transmission of the energy (which in Spain, as we have explained, is a legal monopoly of the partially publicly owned REE); and that (ii) there was no contribution from companies present in the production, distribution or commercialisation markets alone which were not vertically integrated groups and were not present in all those markets simultaneously. The Supreme Court judgment also obliged the Spanish State to compensate these corporate groups for the costs they had borne thus far to finance the social bond (more than €500 million plus interests). This constituted the second time the Spanish Supreme Court annulled the financing regime of the social bond set by Spanish legislation for considering it discriminatory. In 2012, the Supreme Court had already questioned a different financing mechanism of the social bond (that time, the burden of financing the bond was placed exclusively on producers of energy).38 The Spanish government has already made public that it will appeal the Supreme Court judgment and a preliminary ruling before the ECJ cannot be ruled out. In the meantime, however, the government legislated to replace the annulled financing mechanism of the social bond (so it could be in operation for 2017) by Royal Decree-Law 7/2016.39 The government decided to extend the burden of financing the social bond to the remaining ‘independent’ energy suppliers. The government declared that this modification was enough to comply with the Supreme Court judgment. That, in practice, implied only a minimum change in the regulation, since these independent energy suppliers, on the basis of their market share, would only need to finance 6 per cent of the total costs of the social bond. Other proposals on the table consisted of either sharing the burden among all companies in the different electricity market segments, including the production, transmission and distribution, or even considering the costs of the social bond as new ‘regulated costs of the electricity system’, which would mean passing on the costs to all consumers through the charge of levies in their regulated bills.

36 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009, concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L211/55. 37 Judgment of the European Court of Justice of 7 September 2016 Case C-121/15 Association nationale des opérateurs détaillants en énergie (ANODE) v Premier Ministre and Others, ECLI:EU:C:2016:637. 38 Judgment of the Supreme Court of 7 February 2012 (STS 419/2010). 39 Royal Decree-Law 7/2016 of 23 December on the regulation of the financing mechanism of the social bond and other consumer protection measures in the electricity market.

512 Jose Luis Buendía and Miguel Ángel Bolsa B. Taxes on Self-Consumption of Electrical Energy According to Royal Decree 900/2015,40 consumers who use renewable sources of energy in order to generate energy for self-consumption are obliged to pay certain levies if they connect their systems to the electricity network (ie, if these domestic systems are not isolated, closed electrical networks). In particular, the RD sets two different types of charge. First, every self-consumption installation that is connected to the national electricity network is obliged to pay certain charges to contribute to the regulated expenses of the system. This connection to the general network is highly convenient and most of the time necessary, because it allows the domestic system to be backed up by the national network in any period in which the energy generated is insufficient to cover the required energy (renewable sources of energy have generally a more cyclical profile of energy production: photovoltaic energy generation substantially decreases during the night and eolic energy relies on wind force). In addition to this contribution to the general maintenance of the national electricity system, self-consumption electricity installations also have to remunerate the energy they consume as any other consumer. In other words, during the phases of energy deficit of these installations (during which the energy consumed is higher than the energy produced), these installations absorb energy from the national network, and they have to pay for it. The problem has arisen—and this is the main source of controversy in Spain regarding this legislation—due to the absence in the Spanish regulation of a number of mechanisms (present in other Member States’ equivalent regulations) that are used to reduce the charges paid by these energy self-consumption installations, and that they encourage their use by making it much more affordable. The first mechanism is a net-balance measure of the energy produced and consumed. While during the deficit phase these installations consume more energy than they produce (and thus they absorb that energy from the general network), during the so-called surplus phases the installation is able to generate more energy than it consumes. Since this energy is very difficult (and expensive) to store, the extra energy is generally fed into the general electricity grid. However, the Spanish regulation does not allow the owners of installations to calculate the net balance between the energy they absorb and the energy they introduce into the general network, only paying for the gap between them (if it exists). On the contrary, Spanish regulation does not allow the deduction of the energy feed into the grid from the energy absorbed from it, and therefore obliges these installations to pay for the whole gross amount of energy they consume from the national network during the deficit phases, and does not remunerate, in any way, the energy they introduce into the general network. The second restriction to self-consumption that Spanish regulation contains is the prohibition for self-consumption installations to associate with each other creating groups of installations that share an internal network where peaks and troughs of

40 Royal Decree 900/2015 of 9 October 2015 on the administrative, technical and economic regulation of production and self-consumption of electrical energy.

Spain

513

production and consumption could be compensated. According to the regulation, however, all self-consumption installations that are connected to the network have to be connected individually, and no groups of installations can create domestic networks. This regulation has been questioned both by opposition political parties as well as by regional governments, because instead of promoting (as happens in other countries) the use of these installations, it would be impairing its development and expansion. The government has argued, however, that if self-consumption installations do not pay a fair contribution to the general electricity network they use, other market participants (including consumers) would be cross-subsidising their activities. The regional government of Murcia, the Spanish region where most photovoltaic installations have been installed, adopted a regional regulation exempting a large number of small photovoltaic installations from paying certain charges to the electricity system.41 The regional measure was challenged by the central government before the Constitutional Court, but the measure is in force while the proceedings are pending.42 The Spanish regulation on self-consumption of energy seems to go in the opposite direction also from the legislative framework the European Commission is promoting at EU level. Indeed, in the framework of the revision process of the Renewable Energy Directive, the so-called ‘Clean Energy for all’ initiative, the Commission has advocated for ‘enabling consumers to self-consume renewable electricity without facing undue restrictions, and ensuring that they are remunerated for the electricity they feed into the grid’.43 Meanwhile, the central government has already committed with the opposition parties to modify the legal framework on self-consumption of energy in Spain, although the details of the future reform still remain unknown.

41 42 43

Art 20 bis of Regional Law 11 of 2015 on renewable energies. Order of the Constitutional Court of 19 May 2016. www.ec.europa.eu/energy/sites/ener/files/documents/technical_memo_renewables.pdf.

22 State Aid and the EEA SVEIN TERJE TVEIT AND JENS NAAS-BIBOW

I. THE SCOPE OF THE EEA AGREEMENT AND APPLICATION OF STATE AID RULES

A. Introduction

E

NERGY MATTERS ARE of considerable importance in the relations between the EU and the EEA EFTA States. First, the EEA EFTA States are part of the internal energy market by virtue of the EEA Agreement Article 24, in conjunction with Annex IV of the EEA Agreement. Second, the EU relies on imports of energy products from third countries and the EEA EFTA State Norway is the second biggest supplier of both oil and natural gas to the European Union, after Russia.1 The EEA Agreement entered into force on 1 January 1994 and allows and obliges the three EEA EFTA States, Norway, Iceland and Liechtenstein, to be part of the EU’s single market. In effect, Iceland, Liechtenstein and Norway apply most of the EU acquis, including legislation on the internal energy market and related flanking policies (competition, environment, consumer protection, R&D programmes etc). The EEA EFTA States are therefore fully integrated into the internal energy market and the EU State aid law is reflected in the EEA Agreement.2 The objective of the EEA Agreement is to establish a dynamic and homogenous European Economic Area based on common rules and equal conditions of competition … achieved on the basis of equality and reciprocity and an overall balance of benefits, rights and obligations for the Contracting Parties [ie, Norway, Iceland and Liechtenstein on the one hand and the EU Member states on the other hand].3

To achieve this objective, the EEA Agreement connects the supranational mechanisms of EU law with the autonomous law-making mechanisms in the individual EFTA State, thereby extending the body of law which governs the EU internal market to the EEA EFTA States in a continuous process so as to maintain an homogenous set of rules in the entire EEA market.4 In practice, this means that the EEA EFTA

1 In 2014, 31.6% of European natural gas imports were coming from Norway (37.5% from Russia). Source: Eurostat. 2 Note that the Third Energy Package is yet to be implemented in the EEA. 3 See the fourth recital of the preamble of the EEA Agreement. 4 C Baudenbacher, The Handbook of EEA Law (Cham, Springer, 2016) 86–87.

516 Svein Terje Tveit and Jens Naas-Bibow States have a duty to include new EU legislation that falls within the scope of the EEA Agreement (EEA relevant) unless they decide to use their right to veto the implementation of the EU legislation in the EEA Agreement.

B. The Principle of Homogeneity The principle of homogeneity is a cornerstone of the EEA Agreement and means that the same rules and conditions of competition apply to all economic operators within the EEA. This principle distinguishes the EEA Agreement from other international agreements.5 To maintain homogeneity, the EEA Agreement (ie, the relevant Annex of the EEA Agreement) is subject to continuous updates and amendments to ensure that EU legislation pertaining to the internal market is integrated into the national legislation of the EEA/EEA EFTA States. So far, more than 6,000 legal acts have been incorporated into the EEA Agreement.6 This makes the EEA Agreement an ever-evolving and dynamic agreement. The EEA EFTA States are not formally obliged to endorse and implement EEA relevant EU legislative acts, and may (in principle) veto the adoption of new EU legal acts into the EEA EFTA Countries (through a negative decision in the Joint Committee).7 In practice, the use of the veto right is a ‘nuclear option’ as it cuts off related single market rights (EEA Article 102). It has never been used by the EFTA Countries to date. Further, the EEA EFTA States have no legal duty to implement a body of EU law that is not EEA relevant. The question of whether the relevant body of the EU law is EEA relevant is therefore crucial when considering whether the principle of homogeneity applies. The starting point for deciding the EEA relevance is the legal basis of the relevant body of EU law in question. Legal acts that are related to the four freedoms and competition rules (Parts II–IV of the EEA Agreement) or fall within Part V, dealing with Horizontal Provisions Relevant for the Four Freedoms, are generally EEA relevant. If, on the other hand, the EU legal act covers a broader scope than that which is relevant to the four freedoms, the EEA Joint Committee will have to agree on which parts of the legal act should be categorised as EEA relevant and implemented. As an example, the EU legislative initiative to establish a common European energy policy was held by the EEA Joint Committee to fall outside the scope of the EEA Agreement. Consequently, the EEA Committee decided against incorporating measures to safeguard the security of oil and natural gas in the EEA Agreement.8 On the other hand, the corresponding directive dealing with electricity supply and infrastructure investment was included because of its clear relevance to the EEA market (Directive 2005/89/EC). 5 F Sejersted, F Arnesen, O-A Rognstad and O Kolstad, EØS-rett (Oslo, Universitetsforlaget, 2011) 87, 277, 279; T Bekkedal, Frihet, likhet og felleskap (Bergen, Fagbokforlaget, 2008) 137. 6 Sejersted et al (n 5) 193. 7 ibid, 191 f; F Arnesen, H-P Graver and U Sverdrup, ‘Vetoretten i EØS’ (2001) 2 Jussens Venner 115 f, available at www.sv.uio.no/arena/english/research/publications/arena-working-papers/2001-2010/ 2001/01_02.html. 8 Council Directive 2006/67/EC of 24 July 2006 imposing an obligation on Member States to maintain minimum stocks of crude oil and/or petroleum products [2006] OJ L217/8—or its predecessor—has not been implemented in the EEA.

State Aid and the EEA 517 More generally, with a few exceptions, all EU legislation forming European energy law is considered EEA relevant.9 The ‘First Energy Package’ of the liberalisation of the internal energy market was incorporated into the EEA in 1999 (for electricity) and 2001 (for gas). By means of a decision of the EEA Joint Committee amending Annex IV of the EEA Agreement, the EEA Contracting Parties also incorporated the provisions of the ‘Second Package’ of the liberalisation of the internal energy market in 2009. Notably, however, Annex IV to the EEA Agreement does not yet incorporate the five EU directives which form the Third Package of the liberalisation of the EU energy market.10

C. The Two-Pillar System In order to understand the institutions relevant to the EEA Agreement and the enforcement of the State aid rules in the energy sector, it is important to understand that the EEA Agreement is based upon two institutional pillars, EFTA and the EU: The Two-Pillar EEA Structure ICELAND LIECHTENSTEIN NORWAY

EEA COUNCIL

COUNCIL PRESIDENCY & EEAS

EFTA STANDING COMMITTEE

EEA JOINT COMMITTEE

EUROPEAN EXTERNAL ACTION SERVICE (EEAS)

EFTA SURVEILLANCE AUTHORITY

EUROPEAN COMMISSION

EFTA COURT

EUROPEAN COURT OF JUSTICE

COMMITTEE OF MPs OF THE EFTA STATES

EEA JOINT PARLIAMENTARY COMMITTEE

EUROPEAN PARLIAMENT

EFTA CONSULTATIVE COMMITTEE

EEA CONSULTATIVE COMMITTEE

ECONOMIC AND SOCIAL COMMITTEE

This diagram illustrates the management of the EEA Agreement. The left pillar shows the EFTA States and their institutions, while the right pillar shows the EU side. The joint EEA bodies are in the middle.

Figure 1: The Two-Pillar EEA Structure

9 Today, the EEA Agreement includes around 70 legal acts relating to energy (and more if directives relating to the environment are included); cf the Norwegian Energy Report to the Norwegian Parliament, Report No 25, 2015–2016. 10 The Third Package consists of the following five Directives: Common Rules for the internal market in Electricity Directive (2009/72/EC); Common Rules for the internal market in Natural Gas Directive (2009/73/EC); Regulation Establishing an Agency for the Cooperation of Energy Regulation (713/2009/EC); Regulation on Conditions for Access to the Network for Cross-Border Exchanges in Electricity (714/2009/EC); and Regulation on Conditions for Access to the Natural Gas Transmission Networks (715/2009/EC). The EEA Joint Committee adopted the third energy packages on 5 May 2017, but it is yet to be ratified by the Norwegian Parliament.

518 Svein Terje Tveit and Jens Naas-Bibow Figure 1 illustrates the management of the decision-making process of the EEA Agreement with the institutional EFTA pillar on the left side, the institutional EU pillar on the right side, and the coordinating EEA bodies sitting in between the two bodies of institutions.11 Both pillars have their own institutions and legal basis (the EFTA Convention and the EU treaties). The reason for the two-pillar system is that the EEA EFTA States have not transferred any legislative competences to the EEA institutions.12 They are unable, constitutionally, to accept direct decisions of the Commission or the European Court of Justice of the European Union (ECJ). To cater for this situation, the EEA Agreement establishes EEA EFTA bodies to match those on the EU side where decisions related to EEA legislation are normally taken by a majority vote. The practical implementation of EU legal acts in the EEA Agreement takes place through the EEA Joint Committee where all EEA relevant legislative acts passed by the EU institutions are discussed and subsequently adopted (unless vetoed).13 The EU legislative acts become binding in the EEA EFTA States after having been adopted by the EEA Joint Committee and implemented in the EEA Agreement pursuant to Articles 2(a) and 7 of the EEA Agreement.14 This means that the EEA EFTA Countries are committed to comply with the body of law emanating from the EU (if EEA relevant) even if the national decision-makers have not taken part in enacting the body of law.15 This raises a democratic legitimacy issue.16 The EU law and EEA law implementation process may be summarised as follows:

11 This figure is taken from EFTA’s publicly available webpage, available at: www.efta.int/eea/eeainstitutions.aspx. 12 Individuals and undertakings cannot rely on non-implemented EEA rules before national courts unless transformed into national law (eg, no ‘direct effect’), but the principle of effectiveness means that national courts will consider any relevant element of EEA law, whether implemented or not, when interpreting national law, cf Case E-04/01 Karl K Karlsson hf v The Icelandic State [2002] EFTA Ct Rep 240, para 28. 13 Art 102(1) EEA Agreement. 14 The internal procedures for the EU dealing with the EEA decision-making procedure, among others, are set out in more detail in Council Regulation (EC) 2894/94 of 28 November 1994 concerning arrangements for implementing the Agreement on the European Economic Area ([1994[OJ L20/11/, P 0006–0008). 15 In Case E-02/06 EFTA Surveillance Authority v The Kingdom of Norway (Waterfall case) [2007] EFTA Ct Rep 164, Norway argued that the special Norwegian regime for ‘reversion’ to the state of ownership of waterfalls developed for hydroelectric power production could not be challenged because it concerned the management of essential natural energy resources and therefore fell outside the scope of the Agreement. The Court refused this line of argument and stated that ‘unilateral expressions of understanding of the kind claimed to have been made by Norway and Iceland cannot constitute … circumstances [which lead to a different understanding of EEA law than of EU law]’. In the end, however, Norway changed the national regulation and limited private ownership of waterfalls, so the potential ‘democratic legitimacy’ issue was not properly tested. 16 Sejersted, Arnesen, Rognstad and Kolstad (n 5) 183 f; EO Eriksen, ‘Democracy Lost: The EEA Agreement and Norway’s Democratic Deficit’ in H Høibraaten and J Hille (eds), Northern Europe and the Future of the EU (Nordeuropa und die Zukunft der EU) (Berlin, Intersentia, 2011) 114.

State Aid and the EEA 519

Figure 2: EU Law/EEA Law—Implementation Process

D. The Scope of the EEA Agreement and the State Aid Rules The scope of the EEA Agreement is generally more limited than EU law.17 However, the EEA State aid law is closely modelled on EU State aid law and the EFTA Surveillance Authority (ESA) and the EFTA Court have broadly ensured that the State aid provisions of the EEA are homogenously applied and interpreted, in order to create a level playing field for the economic operators in the EEA. The criteria for assessing the existence of aid are set out in EEA Article 61(1) and mirror the criteria in TFEU Article 107(1) (State resources, economic advantage, undertakings, selectivity, distortion of competition and effect on trade). The criteria for assessing whether aid shall be or may be compatible (lawful) aid are set out in EEA Article 61(2) and (3) and similarly mirror the criteria in TFEU Article 107(2) and (3).18

17 The EEA Agreement does not cover the EU Common Agricultural and Fisheries Policies, the Customs Union, the Common Trade Policy, Common Foreign and Security Policy, Justice and Home Affairs or the Monetary Union. 18 Note that the EEA agreement does not reflect TFEU Art 107(3)(d) stating that ‘aid to promote culture and heritage conservation’ where such aid does not adversely affect trading conditions and competition may be compatible with the EEA Agreement. In practice, ESA has nevertheless taken the view that such measures may be taken into account on the basis of Art 61(3)(c), cf ESA Decision 496/13/COL of 11 December 2013 concerning the financing of the Harpa Concert Hall and Conference Centre (Iceland).

520 Svein Terje Tveit and Jens Naas-Bibow In relation to State aid, an important difference is that the EEA Agreement does not cover rules on the harmonisation of taxes.19 The VAT directive and other acts on taxation have therefore not been incorporated into the EEA Agreement. Tax differentiation is not unusual in the energy sector, and an important question is therefore to what extent EEA rules on State aid apply to such tax differentiation. The EFTA Court has—as the ECJ—drawn a distinction between the existence of a measure and its effects, and noted that although the tax system is not covered by the EEA Agreement, it ‘may have consequences that would bring it within the scope of application of Article 61(1) EEA’.20 Ultimately, the difference in scope may not therefore translate into a difference of substance, and the ESA and the EFTA Court have on several occasions dealt with State aid relating to taxation or VAT in the energy sector. First, in Decision 148/04/COL, ESA held that the exemption from tax on electricity consumption, which was applicable until 31 December 2003, for the manufacturing and mining industries constituted incompatible State aid pursuant to Article 61(1) EEA. ESA held that the measure was selective in nature, and that even if the derogation (exemption) was subsequently narrowed to only comprise electricity for production purposes, excluding the use of electricity in ‘administration buildings’, this narrowing of scope did not make it a general scheme incapable of constituting aid.21 Second, there are other non-energy related cases in which the ESA has reached the same conclusions.22

II. STATE AID PROCEDURE IN THE EEA

A. The Role and Competence of the EFTA Surveillance Authority (ESA) Pursuant to Article 62(1) of the EEA Agreement, the EFTA Surveillance Authority (ESA or the Authority) ensures that EEA rules are properly enacted and applied by the EEA EFTA States. The ESA has direct competence in questions regarding

19

Note that the EEA Agreement prohibits discrimination with regard to taxation (Arts 14 and 15). See EFTA Court Case E-06/98 The Kingdom of Norway v EFTA Surveillance Authority, para 34 referring to the European Court of Justice, Case C-173/73 Italy v Commission. The scope of the EEA Agreement on tax matters was also considered in the EFTA Court Case E-06/98, concerning the question of whether a scheme of regionally differentiated social security contributions in Norway involved State aid, [1999] EFTA Ct Rep 74, para 26. 21 The order in so far as the State aid and recovery questions were concerned, was later challenged before the EFTA Court in Joined Cases E-5/04, 6/04 and 7/04 Fesil et al v ESA [2005] EFTA Ct Rep 117. 22 In Decision 106/95/COL, ESA refused to approve a tax exemption for glass packaging from a basic tax on non-reusable beverage packaging under EEA Art 61(3)(c) (development of certain economic activities or areas), since the aid would benefit one particular competitor and was not linked to any initial investment, job creation or project and therefore constituted operating aid. The ESA noted that the exemption seemed to discriminate against imported products and that the environmental effects (eg, glass packaging being less damaging to the environment than other packaging materials) were not sufficiently demonstrated. Further, in Decision 193/14/COL, the ESA decided that the Icelandic VAT exemption on the import of servers and non-imposition of VAT on mixed services relating to data centres did not fall within the logic and nature of the Icelandic VAT system. The scheme was therefore considered selective and deemed as unlawful aid. 20

State Aid and the EEA 521 competition (including State aid and public procurement). This underlines the fact that the EEA EFTA States have conceded sovereignty.23 ESA’s powers in the area of State aid therefore generally mirror the extended competencies of the Commission.24 In effect, the EEA EFTA States are subject to the same strict State aid control as the EU Member States. EEA Article 62 states that all existing aid, as well as any plans to grant or alter State aid, shall be subject to constant review as to their compatibility with EEA Article 61 and stipulates that the ESA shall carry out this review pursuant to Protocol 26 to the EEA Agreement.25 As such, Article 5 of the Surveillance and Court Agreement gives the ESA the right to take decisions, provide recommendations, deliver opinions and issue notices or guidelines on matters dealt with in the EEA Agreement, including State aid. The caseload of the ESA stems from three main sources: (i) notifications from EEA EFTA States which must be seeking the ESA’s prior approval before implementing new aid;26 (ii) complaints from business or national authorities about existing unlawful aid; and (iii) ESA’s ex officio investigations. The State Aid Modernisation reform (SAM) and the General Block Exemption Regulation (GBER) are expected to gradually lead to more decentralised enforcement as the EEA EFTA States take on more of the ex ante assessments of aid and compatibility assessments (with the ESA continuing with a certain level of ex post control via monitoring). Given the importance of the energy sector in Norway, the ESA has been relatively active in State aid law enforcement in the area of energy, in particular relating to State support for green energy. In 2015, the ESA adopted 27 State aid decisions and opened 65 new State aid cases.27 Ten out of the 27 decisions concern energy and environmental aid.28 In 2016, the ESA adopted 24 State aid decisions of which 12 decisions are related to energy and environmental aid.29 Several of these decisions concern aid approved under the EEAG. The ESA has generally taken the view that operating aid, that is to say, aid intended to relieve an undertaking of the expenses which it would itself normally have had to bear in its day-to-day management or its usual activities, does not in principle fall within the scope of Article 61(3) of the EEA Agreement. The reasoning applied is that operating 23 Art 93 of the Norwegian Constitution, given in the constitutional assembly at Eidsvoll on 17 of May 1814 (Norw: Kongeriget Norges Grunnlov, LOV-1992-11-27-109). 24 EEA Art 62 does not grant ESA power to adopt block exemption regulations, cf TFEU Art 108(4). 25 As further set out in Pt I of Protocol 3 to the SCA. 26 The distinction between new and existing aid is equally relevant therefore under EEA law. 27 According to the ESA’s State aid register (19 Norwegian cases and 8 Icelandic cases). 28 Six of the Norwegian cases concern energy or environmental aid (Case 476/15/COL—aid to Titanium & Iron AS demonstration plant; Case 37/15/COL—aid to Hydro Aluminium S demonstration plant; Case 110/15/COL—aid to energy recovery system at Finnfjord AS in its ferrosilicon production plant; Case 150/15/COL—aid in the form of zero VAT rating for electric vehicles; Case 336/1 5/COL— aid schemes targeting different renewable energy technologies: (1) electric charging infrastructure for vehicles, (2) shore-side electricity supply for ships, (3) hydrogen refuelling infrastructure for vehicles, (4) biofuels refuelling infrastructure for vehicles, and (5) LNG refuelling infrastructure for ships); Case 478/15/COL—aid to CLIMATE Demo aid scheme intended to promote technology for CO2 capture, transport and storage (CCS) through development, pilot and demonstration projects). In relation to VAT rating for electric vehicles, the Norwegian government has recently announced in relation to the 2017 National Budget that this zero-rating will be prolonged until 2020. 29 According to ESA’s State aid register (17 Norwegian cases and 7 Icelandic cases), cf ESA State aid register.

522 Svein Terje Tveit and Jens Naas-Bibow aid generally distorts competition in the sectors in which it is granted, while nevertheless being incapable, by its very nature, of achieving any of the objectives of the exceptions foreseen in Article 61(3) EEA.30 As the authority for this view, the ESA has among others referred to the ECJ case law.31

B. The Role and Competence of the EFTA Court The EFTA Court has juridisdiction over EEA related matters pertaining to the three EEA EFTA States. More specifically, it is authorised to deal with direct actions (eg, infringement actions brought by the ESA against an EFTA State), the settlement of disputes between two or more EEA EFTA States, advisory opinions to the courts on the interpretation of EEA rules (not binding) and appeals concerning decisions taken by the ESA.32 In effect, the jurisdiction of the EFTA Court mainly corresponds to the jurisdiction of the ECJ over EU States. Significantly, the EFTA Court is not bound by the EEA EFTA States’ constitutions and the democratic principles set forth therein. The EFTA Court has currently dealt with 23 cases directly concerning State aid, 18 of which relate to applications for the annulment or partial annulment of the ESA’s State aid decisions.33 The other cases relate to legal actions against the ESA for failure to act on a State aid complaint; or against Iceland for failing to terminate an aid tax scheme34 and recover unlawful aid, and for failing to compensate Icesave depositors (lawful State aid).35 Finally, there is one State aid case responding to a request for advisory opinion.36 Only two of these 23 cases relate specifically to State aid in the energy sector.37 30 See, among others, ESA Decision 543/14/COL of 10 December 2014 to initiate the formal investigation procedure with regard to the Power Contract and the Transmission Agreement for the PCC Silicon Metal Plant at Bakki (Iceland). 31 Case C-156/98 Germany v Commission, EU:C:2000:467, paras 30–31; Case T-459/93 Siemens SA v Commission, EU:T:1995:100, para 48; and Case T-396/08 Freistaat Sachsen und Land SachsenAnhalt v Commission, EU:T:2010:297, paras 46–48. 32 See Art 31 of the Agreement between the EEA EFTA States on the Establishment of a Surveillance Authority and a Court of Justice (SCA): Art 32 SCA, Art 34 SCA and Art 36 SCA. 33 See Cases E-02/02 of 19 June 2003; E-1/13 of 27 January 2014; Joined Cases E-5/04, E-6/04 and E-7/04 of 21 July 2005; E-08/13 of 29 August 2014; E-2/94 of 21 March 1995; E-09/12 of 22 July 2013; E-06/98 of 20 May 1999; E-04/97 of 12 June 1998; E-1/12 of 11 December 2012; E-12/11 of 17 August 2012; Joined Cases E-10/11 and E-11/11 of 8 October 2012; Joined Cases E-17/10 and E-06/11 of 30 March 2012; E-14/10 of 22 August 2011; Joined Cases E-04/10, E-6/10 and E-07/10 of 10 May 2011; E-5/07 of 21 February 2008; E-09/04 of 7 April 2006; and E-4/97 of 3 March 1999. 34 See EFTA Court Case E-02/02 and Joined Cases E-5/04, E-6/04 and E-7/04 of 4 April 2013. 35 See Cases E-02/02 of 19 June 2003; E-1/13 of 27 January 2014; Joined Cases E-5/04, E-6/04 and E-7/04 of 21 July 2005; E-08/13 of 29 August 2014; E-2/94 of 21 March 1995; E-09/12 of 22 July 2013; E-06/98 of 20 May 1999; E-04/97 of 12 June 1998; E-1/12 of 11 December 2012; E-12/11 of 17 August 2012; Joined Cases E-10/11 and E-11/11 of 8 October 2012; Joined Cases E-17/10 and E-06/11 of 30 March 2012; E-14/10 of 22 August 2011; Joined Cases E-04/10, E-6/10 and E-07/10 of 10 May 2011; E-5/07 of 21 February 2008; E-09/04 of 7 April 2006; and E-4/97 of 3 March 1999 (all failure to act). Further, see Case E-06/09 Magasin- og Ukepresseforeningen v EFTA Surveillance Authority; Case E-2/05 of 24 November 2005; and Case E-16/11 EFTA Surveillance Authority v Iceland. 36 Case E-1/100 State Debt Management Agency v. Islandsbanki-FBA hf. 2000–2001 EFTA Ct Report 8, para 38, in which the EFTA Court confirmed that national courts have no jurisdiction to decide on the compatibility of State aid. 37 See Case E-02/02 of 19 June 2003 and Joined Cases E-5/04, E-6/04 and E-7/04 of 21 July 2005.

State Aid and the EEA 523 In Case E-02/02, Bellona Foundation and Technologien, Bau- und Wirtschaftsberatung, both active in environmental consulting, sought to challenge the ESA’s approval (as regional aid) of the Norwegian favourable tax depreciation rates for large-scale liquefied natural gas projects in Norway. An amendment to the Norwegian Petroleum Tax issued by the Norwegian government was designed to permit the Snøhvit liquefied natural gas project in the Barents Sea to go forward. The EFTA Court, however, ruled that neither of the applicants had shown that it was individually concerned by the decision and therefore declared the application as inadmissible. In Joined Cases E-5/04, E-6/04 and E-7/04, Fesil and others v ESA, the applicants sought to partially annul ESA’s Decision 148/04/COL in so far as it declared the Norwegian exemption from the tax on electricity which was applicable until 31 December 2003, for the manufacturing and mining industries, as incompatible State aid pursuant to Article 61(1) EEA and ordered recovery of this aid. The tax exemption in question was applicable until 31 December 2003. The applicants argued that the scheme was not selective, inter alia, because it applied only to a particular use of electricity (production process) and was, in principle, open to all undertakings carrying out this process. Recognising that there may be undertakings outside the manufacturing and mining sectors where there is a high consumption of electricity, and electricity is used for purposes similar to those exemptions, the Norwegian government argued that some degree of differential tax treatment must be justifiable by reference to practical and administrative considerations. The EFTA Court rejected the argument and held that the exemption benefited specific sectors and therefore was selective. The second plea, made by the applicants, alleged that the ESA had incorrectly concluded that the aid was incompatible with EEA Article 61(3). The applicants made the point that the ESA should have considered the energy tax directive (Directive/2003/96/EC), Article 2(4), exempting certain uses of energy products and electricity from the scope of the directive, including electricity, when they account for more than 50 per cent of the cost of the product. Second, the applicants argued that the ESA had failed to consider the possibility in the EEAG of granting operating aid in the absence of Community rules to temporarily protect international competitiveness.38 The EFTA Court noted that the ESA enjoys a wide level of discretion when assessing the compatibility of aid. As to the specific points made by the applicant, the EFTA Court concluded that the energy directive was not part of the EEA law and that the applicants had failed to show that the conditions for exemption under the EEAG were met—any lack of consideration of the aspect of international competitiveness was therefore not decisive. Interestingly, the EFTA Court further ruled that since Norway had agreed to propose appropriate measures to adjust existing aid schemes to comply with the new environmental guidelines, this had the effect of reclassifying existing aid as new aid. The EFTA Court thus ignored the Norwegian State’s argument that the legal consequences of a delayed implementation of the EEAG in any event would be limited to an enforcement action against the State and State liability.

38

Community guidelines on state aid for environmental protection [2001] OJ C37/3.

524 Svein Terje Tveit and Jens Naas-Bibow C. National Courts’ Role in Assessing Aid (Standstill Obligation) The national courts of the EEA EFTA States do not have the power to declare a State aid measure compatible with Article 61(2) or (3) of the EEA Agreement, but may be asked to intervene in cases where an EFTA State authority has granted aid without respecting the standstill obligation (unlawful State aid) and they play an important role in the enforcement of recovery decisions.39 A search in the Norwegian court database reveals a total of 63 results on ‘State aid’ in the period from 1994 to 2016, but the actual number of cases dealing with substantive State aid matters is significantly lower (if we include the EEA in the same search, the number of cases drops to 24).40 EEA State aid law has been invoked to set aside private agreements claiming invalidity,41 interpret public concessions,42 and interpret long-term power purchase agreements.43 In a decision of 17 December 2013 (Hydro-Søral), the Norwegian Supreme Court ruled that the obligation to recover the unlawful State aid only arises once the State/ grantor has adopted a recovery decision. This means that the limitation period does not run from the time of ESA’s final decision declaring the aid unlawful, but only later once the national authorities have ordered recovery. If the decision is interpreted literally, the decision entails that the standstill obligation is not relevant to the recovery of the aid, and that the standstill obligation does not in its own right establish an obligation to recover the aid. If interpreted this way, the decision appears to contradict the direct effect of the standstill obligation and the principle of efficient enforcement of EEA law, and it would mean that the standstill obligation would not serve as a legal basis for private damages claims, recovery or other legal effects related to the invalidity of the grantor’s aid decision. The standstill obligation is the only element in the EU (and EEA) State aid law that has direct effect.44 The direct effect and the principles of homogeneity, efficiency and equivalency mean that the national courts must ensure that undertakings are granted damages if they can prove economic losses as a consequence of unlawful State aid.45

39 Case C-199/06 CELF and Ministre de la Culture et de la Communication, EU:C:2008:79, para 38; Case C-17/91 Lornoy et al v Belgian State, EU:C:1992:514, para 30; and Case C-354/90 Fédération Nationale du Commerce Extérieur des Produits Alimentaires et al v France, EU:C:1991:440, para 14; and Art 14(1) in Pt II of Protocol 3 to the Surveillance and Court Agreement. 40 As at 6 July 2016. 41 An example of this is the decision by the Gulating Court of Appeal in Case LG-2008-16104. 42 The national power supplier company Statkraft argued that the concession that obliged Statkraft to deliver water from public waterfalls to the local community could not give third parties a right to damages for losses incurred due to periodic shortfalls in the water supply, since this would mean that the private parties were granted State aid, cf Gulating Court of Appeal judgment in Case LG-2015-150132. 43 In the Borgarting Court of Appeal Case LB-1998-1805, the biggest Norwegian energy generator, Statkraft, successfully argued that it was not obliged to deliver power under long-term power purchase agreements below market price (among others because this price would entail the granting of State aid). 44 See ECJ, Case 6/64 Flaminio Costa v ENEL Rec 1964, s 585 and Case C-284/12 Deutsche Lufthansa AG v Flughafen Frankfurt-Hahn GmbH, EU:C:2013:755, paras 29–31. 45 On this controversy, see Erling Hjelmeng’s article in the Norwegian journal Law and Order ‘Tilbakeføring av offentlig støtte—hva er situasjonen etter Hydro/Søral?’ (‘Recovery of Unlawful State Aid—what’s the situation after the Supreme Court judgment in Hydro/Søral?’) [2014] Lov og Rett 210.

State Aid and the EEA 525 In autumn 2007, the Confederation of Norwegian Business and Industry (NHO) and a Norwegian–Finnish joint venture burned lime producer, NorFraKalk, filed a complaint with the ESA about the discrimination between existing and new industry in the allocation of free emisssion allowances. Under the Norwegian Greenhouse Gas Emission Trading Act, the government would allocate allowances free of charge to onshore installations established prior to 31 December 2001 based on their emissions during the period from 1998 to 2001. The claimants claimed that the Norwegian Greenhouse Gas Emission Trading Act was in violation of the EEA Agreement provisions on State aid (Article 61) and the ban on imposing restrictions on the freedom of establishment (Article 31). In July 2008, it became clear that the ESA did not approve the Norwegian allocation plan. The ESA concluded that the allocation of allowances free of charge to the narrowly defined existing industries would result in improper discrimination in that it favours installations established before 2002. According to the wording of the Emissions Trading Directive, all enterprises that had been issued a complete allowance permit before the formal notification of the allocation plan was given (March 2008) were to be considered existing enterprises. Norway chose not to appeal ESA’s decision and revised the Norwegian allocation plan accordingly. In the Oslo Court of First Instance, Case TOSLO-2010-45497, the claimant, NorFraKalk, brought a follow-on suit against the Norwegian State claiming damages for wrongful implementation and application of the EU Emissions Trading Directive and unlawful granting of aid.46 The claimant argued that due to the adoption of the original allocation plan, the claimant had not been able to rely on free allowances and was thereby forced to compete on disadvantageous terms in the interim period until the government changed its allocation plan.47 The Norwegian government referred in the preparatory works to the original allocation plan, which prescribed that individual grants under the free allowance scheme were dependent on ESA’s approval of the national allocation plan. Consequently, the Norwegian government argued that no aid had been granted and no injury suffered. The Court ultimately held that there is no breach of the implementation prohibition if the State has ensured that the aid-granting body can only grant the aid after

46 The ECJ has confirmed that the implementation prohibition produces rights in favour of individuals, cf Case C-390/98 Banks, EU:C:2001:456, para 70, and Advocate General Tesauro stated in Case C-142/87 that ‘it is not clear whether the failure to give notice of aid proposed but not granted always involves an infringement of Community law … That is certainly the case if the measure granting the aid provides for its implementation or merely permits such implementation’. It is not clear, however, how a claimant can prove causation and/or quantify the loss suffered if no aid has actually been granted (no aid recipient and hence no causation between the behaviour of the aid recipient and the damage suffered, eg ‘downstream causation’), cf M Honoré and N Eram Jensen, ‘Damages in State Aid Cases’ (2011) 10(2) European State Aid Law Quarterly 265. 47 The Court, in principle, accepted that the claimant could have lost contracts to a Belgian competitor due to the inability to benefit from free emission quotas and thereby discount their quotation price. Curiously, the Court did not define a relevant market or assess to what extent the Norwegian claimant did in fact compete with the Belgian producer. Hence, the Court also did not make the finding of an infringement, loss and causation dependent on actual existence of aid, distortion of competition and effect on trade (which are prerequisites for finding unlawful aid), but simply noted that the claimant competed in a cross-border market where there could easily be an effect on competition and trade.

526 Svein Terje Tveit and Jens Naas-Bibow the ESA has cleared the aid (in this case, the allocation plan). This appears to be in line with the Commission’s previous reasoning and Advocate General Tesauro’s opinion in Case 142/87 Belgium v Commission.48

III. THE EEA ENERGY MARKET

A. Liberalisation Pre-EEA Agreement Norway was the first Nordic State to deregulate its electricity market, with the adoption of the Energy Act in 1990.49 The Energy Act came into force on 1 January 1991 and required that separate accounts be kept for network activities and generation and energy sales activities in order to prevent cross-subsidisation (accounting unbundling). It also required network owners to offer equal, non-discriminatory tariffs to electricity suppliers and end users. As a result of the deregulation, Statnett Marked, a liquid market place for the power industry, was established in December 1992.50 Prior to this, the Norwegian power system was organised by the Norwegian Power Pool, a membership organisation that operated as a service for Norwegian power producers.51 The establishment of Statnett Marked initiated the beginning of the Nordic power exchange,52 and the Nordic electricity market was formally opened up to regional competition with the establishment of Nord Pool ASA in 1996.53 Market liberalisation ensured less divergence in the price of electricity between different regions in Norway. It also promoted investment in new production capacity where it was most needed.54 The liberalisation of the electricity market has in many ways taken place in response to EU energy legislation. However, the liberalisation occurred in conjunction with a general shift in Norway towards decentralisation and the liberalisation of the Norwegian economy in general. Today, the position of the Norwegian government in energy-related matters is aligned to a great extent with that of the European Commission.

48 See Case C-142/87 Belgium v Commission, ECLI:EU:C:1990:125, para 6; Commission, XXIIIrd Report on Competition Policy (1993) para 397; and Commission, State Aid Manual of Procedures, s 5(2) (sub-para 3): ‘To avoid breaching this requirement [the implementation prohibition] when passing aid legislation, Member States can either notify the legislation while it is still at the drafting stage or, if not, insert a cl whereby the aid granting body can only grant the aid after the Commission has cleared the aid. Without such a cl the measure is considered as unlawuful (non-notified) aid’. 49 Finland liberalised its energy market in 1995; Sweden liberalised its energy market in 1996; Denmark started the liberalisation of its energy market in 1999 and opened its electricity market for all consumers in 2003. 50 H-A Bredesen and Terje Nilsen, ‘Power to the People—the first 20 years of Nordic power-market integration’ (Nord Pool Spot & NASDAQ OMX Commodities, 2013) 30. 51 The Norwegian Power Pool was established on 1 January 1971 and was to some extent continued under Statnett Marked. 52 See n 32 above. 53 For more information on Nord Pool, see section III.C.iv below. 54 Before, the electricity price differences had led to investments in new production capacity in areas with high prices. The result was that expensive investments in areas with high prices were preferred to less costly investments in the areas with cheaper prices.

State Aid and the EEA 527 B. The EEA and the Internal Energy Market As stated above, the EEA Countries, by signing the EEA Agreement, have already signed an international agreement which obliges them to implement and apply the Community acquis in the area of energy.55 The Norwegian Water Resources and Energy Directorate (NVE) is a member of the Council of European Energy Regulators (CEER), but has not been formally included in the Agency for the Cooperation of Energy Regulators (ACER) (pending the implementation of the Third Energy Package).56 The Norwegian government has been reluctant to submit NVE to the authority of ACER, since this would mean delegating the decision-making power to an agency under the EU in which Norway has no direct influence. The implementation of the Third Energy Package therefore raises a constitutional issue. One solution to this issue could be that the supervision of the implementation of ACER’s decisions is placed under the ESA’s (and the EFTA Court’s) competence. The ESA and the EFTA Court supervise the implementation of the EEA Agreement, and thus also any commitments to apply the acquis on energy. The Norwegian government has recently proposed amendments to the Norwegian Energy Act, allowing for other non-transmission system operator (TSO) interconnectors (eg, merchant cables).57 Currently, Section 4(2) of the Norwegian Energy Act states that only the TSO—the Norwegian TSO is Statnett SF—or a company where the TSO holds a decisive influence, may obtain a licence to build, own and operate an interconnector between Norway and other countries. Other companies are no longer eligible for a licence to build, own and operate an interconnector. Ownership restrictions could potentially entail favourable and discriminatory treatment of State-owned TSOs or State aid, if there is cross-subsidisation or otherwise an effect on the competition between Statnett SF and other market players on ancillary markets.58 The Third Energy Package also raises an issue relating to whether the Norwegian regulator is sufficiently independent from the Norwegian government, if it continues to be an entity organised under the Norwegian Ministry of Petroleum and Energy.59 55

See the EEA Agreement, Annex IV. EFTA Countries may participate in ACER similar to third countries subject to the fulfilment of the conditions set out in Art 31 of Reg 713/2009, in particular they must conclude an international agreement which also sets out the nature, scope and procedural aspects of the involvement of those countries in ACER: see Commission Staff ‘Working Paper on the Possibility of Neighbouring Countries and their Transmission System Operators to Participate in ACER and in the ENTSOs’ SEC(2011) 546 final (Brussels, 2011). 57 See the proposal from the Norwegian Energy and Environmental Committee of 11 October 2016 in Prop 98 L (2015–16). 58 The Norwegian Regulation could also be in breach of Regulation (EC) No 1228/2003 of 26 June 2003 on conditions for access to the network for cross-border exchanges in electricity, Art 7, Art 40 EEA on capital movements (limitation on investments in interconnectors) and Art 59(1) EEA read in conjunction with Art 54 EEA since it transfers a mandate to Statnett SF, establishing, inter alia, a conflict of interest within Statnett SF, pursuant to which exercising its monopoly will enable it to distort in its favour the equal conditions of competition between the various operators on the market, which in itself is an illegal abuse of a dominant position. 59 The proposed revision to the Norwegian Energy Act suggests that the Norwegian Regulator shall be free from any instructions from the Ministry in individual cases and not be an appeal board for the Regulator’s decisions. 56

528 Svein Terje Tveit and Jens Naas-Bibow C. The EEA EFTA States’ Energy Markets i. Norway a. Overview About 90 per cent of Norway’s electricity production is publicly owned by the government, county municipalities and local municipalities.60 Most of the electricity grid (transmission grid, regional grid and distribution grid) is publicly owned61 and the Norwegian energy sector is closely regulated by the government. Despite this strong public ownership, we have seen few State aid cases relating to the ownership structure in Norway.62 Norway’s electricity generation totalled 142 TWh in 2015, and the production stems mainly from renewable energy sources: 96 per cent from hydropower and the remaining 4 per cent from thermal and wind power.63 While in the EU the focus on building renewable energy has led to new challenges for the operation of the power supply system and stability in the supply and hence discussion on capacity mechanisms, these capacity mechanisms are less relevant for Norway as an energy-only market. 64 Hydropower production is volatile due to variations in rainfall which lead to unpredictable levels of catchment in the reservoirs. In addition, factors such as temperature, electricity prices and consumer behaviour will affect actual consumption and therefore lead to a shortage of electricity at times.65 This unpredictability and vulnerability has to a great extent been solved by relying on cross-border interconnectors. Norway’s neighbouring States produce energy from a variety of sources, such as renewable sources (ie, wind and hydro) as well as coal, gas, oil and uranium. These energy sources will normally be available when shortages in Norway’s reservoirs result in insufficient domestic electricity production. Thus, Norway depends upon its interconnectors to import electricity in order to ensure security of supply. However, interconnectors allow for both the import and export of electricity and in times of excess production, Norway exports electricity.66 b. Enforcement As Norway is an energy-only market there are no enforcement measures for capacity mechanisms in place. However, there are several governmental law enforcement 60

Official Norwegian Report of 30 November 2004, Hjemfall (NOU 2004:26) para 10.4.3. Official Norwegian Report of 3 July 1998, Energi- og kraftbalansen mot 2020 (NOU 1998:11) para 2.2.3. 62 The most important case related to the Waterfall (Case E-02/06), cf (n 15) above. 63 Statistical data available at Statistics Norway (Norw: Statistisk Sentralbyrå). 64 Commission, ‘An energy policy for Europe’ COM(2007) 1 final. 65 As Norway is situated in northern Europe, the temperature varies greatly between summer and winter, but also within each season as Norway has both ‘cold winters’ and ‘warm winters’. 66 Norway exports energy amounting to ten times its own energy consumption, whereas Great Britain and Germany import around 50% and over 60%, respectively, of their energy consumption. Hence, security of supply is strong in Norway and less in focus than in the EU. Norway’s electricity market is interconnected with Sweden, Denmark, Finland, the Netherlands and Russia. The total interconnector capacity for Norway today is around 5400 MW, and the government is currently negotiating with relevant authorities in Germany and the UK for a possible expansion of its transmission grid to these States. 61

State Aid and the EEA 529 agencies involved in the regulation of the energy markets; NVE, the Competition Authority and the Financial Supervisory Authority of Norway. These bodies cooperate to ensure compliance with the laws and regulation of Norway’s energy-only market. NVE is responsible for the supervision, inspection and reporting of the electricity market. The areas of particular importance are network regulation and tariffs, quality of supply, metering and settlement, billing, supplier switching, neutrality and non-discrimination, and, finally, the obligations and powers of the TSO, Statnett SF. Furthermore, NVE ensures compliance with the Electricity Certificate Act and with regulations establishing the Norwegian–Swedish electricity certificates market.67 The Electricity Certificate Act sets out different measures which can be taken in cases of non-compliance. This includes revocation of the right to receive electricity certificates as well as financial and criminal penalties.68 Appeals against decisions will be dealt with by the Norwegian Ministry of Petroleum and Energy, which is the final regulatory body for these matters.69 In addition, the Electricity Certificate Act gives Sweden the right to receive information when necessary to ensure enforcement of the rights and obligations under the Act.70 The Competition Authority has the overall responsibility for effective competition between the actors in the energy market, and Norway has entered into a Nordic agreement for cooperation in competition cases.71 In the area of State aid, however, the enforcement of the EEA State aid rules falls under the ESA. c. Regulatory Framework The rules related to the organisation and functioning of the electricity sector in Norway are set out in the Energy Act and the Energy Regulation which implement the EU Directive 2009/72/EC.72 The aim of the Energy Act and the Energy Regulation is to establish an efficient, competitive and open electricity market and to prevent discriminatory actions by network companies. The Energy Act and the Energy Regulation require a legal and functional separation between network activities, production, and sales of electricity for all vertically integrated companies.73 It is worth noting that a large part of the laws and regulations governing the Norwegian electricity sector has been broadly drafted and therefore lacks precision to a certain extent.74 As a consequence, the legal framework indirectly grants extensive powers to NVE, a Norwegian government agency responsible for regulating

67

For more detailed information regarding the electricity certificate market, see section IV.A.iii below. paras 24–29 of the Electricity Certificate Act of 24 June 2011. 69 ibid, para 23. 70 ibid. 71 Agreement between Denmark, Iceland and Norway on cooperation in competition cases. Please note that Sweden joined at a later stage and is therefore not mentioned in the title of the treaty. 72 cf The Norwegian Energy Act (Norw: Lov om produksjon, omforing, overføring, omsetning, fordeling og bruk av energi m.m.—lov 1990-06-29-50), the Norwegian Energy Reg (Norw: forskrift om produksjon, omforing, overføring, omsetning, fordeling og bruk av energi m.m.—FOR-1990-12-07-959) and Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity. 73 ie, undertakings engaged in both electricity distribution and production/supply of energy. 74 J Naas-Bibow and G Martinsen, Energiloven med kommentarer (Oslo, Gyldendal Norsk Forlag, 2011) 15. 68

530 Svein Terje Tveit and Jens Naas-Bibow the country’s water and energy resources and supervising the energy market.75 The Norwegian TSO, Statnett SF, is responsible for the operation of Norway’s transmission system as well as the interconnectors to Sweden, Finland, Russia, Denmark and the Netherlands. Ownership of the water resources is regulated through: (i) the rules related to licensing (the Industrial Concession Act (ICA));76 (ii) the doctrine of reversion, according to which the ownership of the waterfall reverts back to the State after a certain period; as well as (iii) the right of pre-emption, which gives the existing shareholders in a water resource a right to buy shares offered for sale before they are offered to the public.77 In 2008, the ICA and Waterfall Resources Act (WRA) were amended and the so-called consolidation model was made part of Norwegian law. The model was introduced in order to comply with the EFTA Court’s decision in the Waterfall case not to discriminate between private and public entities. The consolidation model consists of three main features: concessions for the acquisition of waterfalls can from now on only be granted to public entities, ie, entities with more than two-thirds public ownership; the sale of more than one-third of the shares in a company already holding a concession to private companies is not permitted; and the acquisition of waterfalls under reversion can only be made by public entities. The ownership rules and the consolidation model relating to Norwegian waterfalls have not been challenged under the EEA State aid rules, and the Norwegian government has taken the view that no State aid is involved. Given the changes to the ownership rules following the EFTA Court’s decision in the Waterfall case and in effect the abolition of private ownership (control), the issue of State aid is less imminent than before. Under the old regime, concessions for the acquisition of rights of ownership or use of a waterfall were granted for a limited duration of 60 years, after which all installations should be returned to the Norwegian State without compensation (the reversion rules). The reversion rules did not apply to public undertakings and so conferred a selective advantage on the public entities holders of public concessions. Under the new regime, the consolidation model in effect regulates the ability to acquire waterfalls, and is not in itself a grant of any economic advantage. Arguably, however, there could still be an element of aid in so far as the limitations on private undertakings’ ownership rights have effects on any ancillary market in which the private undertakings compete with the holders of public concessions.

75

NVE is organised as a division within the Ministry of Petroleum and Energy. The Norwegian Act regarding acquisition of waterfalls (Norw: Lov om erverv av vannfall mv.—LOV-1917-12-14-16). 77 In addition, Norway has several laws regulating the use of water which have an impact on the ways hydro production can be utilised. The Conservation Plans (Verneplan I–IV), the Common Plan for Water Systems (St. meld. nr. 60 (1991–92), the Industrial Licensing Act (Norw: Lov om erverv av vannfall, LOV1917-12-14-16), the Water Reg Act (Norw: Lov om vassdragsreguleringer, LOV-1917-12-14-17) and the Water Resources Act (Norw: Lov om vassdrag og grunnvann—LOV-2000-11-24-82) are all part of the legal framework regulating hydropower production. The Water Resources Act implements the EU Water Framework Directive (2000/60/EC). 76

State Aid and the EEA 531 ii. Iceland Iceland has attracted energy-intensive users since the creation of Landsvirkjun and its hydroelectric energy resources. The total generation of electricity in Iceland in 2014 was 18.120 GWh, of which Landsvirkjun generated around 71 per cent.78 Landsvirkjun is only active on the wholesale market for electricity, where its competitors are Orka náttúrunnar (Our Nature—ON) and HS Orka. Landsvirkjun’s customers are composed of seven energy-intensive users representing the purchase of 85 per cent of the company’s output (aluminium, ferrosilicon and aluminium foil industry), and six distribution companies, purchasing 13 per cent, whereas Landsnet, the TSO, purchases 2 per cent for electricity losses. The sale of electricity is made through directly negotiated contracts, and the energy-intensive users are connected to the transmission system directly. A large majority of all electricity is sold to a few customers, which all have concluded long-term power purchase agreements with the domestic power providers at different points in time. Furthermore, the Icelandic market is isolated from the rest of the world, as currently no interconnection exists.79 The abundant potential to produce electricity in Iceland and this isolation are assumed to be the main reasons for the differences in the price of electricity in Iceland and elsewhere in the EEA. iii. Liechtenstein Liechtenstein depends on energy imports at a rate of around 90 per cent and electricity imports of some 85 per cent. Electricity interconnectors exist with Switzerland and Austria, and around 50 per cent of the imports come from the Swiss company AXPO which also owns and operates the three 110 kv lines to Switzerland. The interconnectors with Austria are operated by the TSO of Vorarlberg. The national energy production consists mainly of hydropower, but also electricity and heat from wood and biogas. Both the electricity and gas sectors are run by Stateowned companies: Liechtensteinische Kraftwerke (electricity) and Liechtensteinische Gasversorgung (gas). iv. The Common Nordic Electricity Market (Financial Trading, Physical Trading and Trading of Electricity in the Wholesale Market and the Retail Market) The common Nordic electricity market and power exchange for the trading of electricity and power derivatives was founded in 1996 when Norway and Sweden

78 Source: Iceland National Energy Authority’s publicly available webpage: www.nea.is/thenational-energy-authority/energy-statistics/generation-of-electricity. 79 There have been discussions about an interconnector between Iceland and the UK on a long-term horizon, but this is preliminary and no decision has been taken, cf draft Ten-Year Network Development Plan 2014, prepared by ENTSO-E, European Network of Transmission System Operators for Electricity, 161–62.

532 Svein Terje Tveit and Jens Naas-Bibow established Nord Pool.80 Later Finland (1998), Denmark (1999), Estonia (2010), Lithuania (2012), Latvia (2013), the UK and Germany joined. In 2002, trading for physical electricity delivery was demerged into a separate daughter company, Nord Pool Spot AS,81 while financial trading continued to take place through Nord Pool until it was ultimately acquired by NASDAQ in 2010. The contracts for electricity traded at Nord Pool make reference to the transmission network on which the electricity must be delivered and the time frame during the day (peak load,82 baseload,83 and off-peak hours)84 or week (week day, weekend) when the delivery must take place.85 The duration of contracts vary. There are shortterm ‘spot’ contracts, which typically have a very short maturity of either (i) ‘on the same day’, also called ‘intraday’ (the contract will then generally specify the hour for the power to be delivered) or (ii) ‘next day’, also called ‘day-ahead’ delivery. These short-term contracts are mainly used for volume risk management (optimisation and balancing). Short-term contracts are usually physically settled, for example, relate to the trading of the underlying (electricity). Derivative contracts, including futures and options, generally have a longer tenor—these can be physically or financially settled. Financial trading in power derivatives is mainly used to hedge against the risk for electricity traders stemming from high variations in the price of electricity due to differences in, for example, rainfall and temperature. This trading takes place on the marketplace NASDAQ OMX Commodities. Physical electricity delivery can be in the form of wholesale or retail (directly to consumers).86 If trade is wholesale, it is arranged through bilateral contracts (less than 30 per cent) or on the electricity market Nord Pool Spot (more than 70 per cent). The wholesale market is regulated by NVE. Nord Pool Spot sets the prices for each bidding area.87 Furthermore, Nord Pool Spot sets the daily reference price for the Nordic region (the system price) calculated according to supply and demand bids for electricity in the Nordic market to be

80 Power derivatives trading relates to the buying and selling of electricity at one’s own risk and for one’s own account. This activity can take place bilaterally or be supported by brokers or (third partyoperated) trading venues. After the execution of the trade, power contracts are either cleared or remain uncleared and afterwards are either financially or physically settled. 81 Nord Pool Spot is owned by the Nordic TSOs. 82 Peak load contracts have their delivery between 9am and 8pm on each Monday, Tuesday, Wednesday, Thursday and Friday included in the delivery period. 83 Base load contracts have their delivery of the underlying asset from the first hour of the day until the last hour of all the days included in the delivery period. 84 Off-peak hours are the hours 1–8 am and 9–12 pm. 85 The unit of registration for base and peak load contracts is 1 Megawatt (MW). The nominal of the contract is determined in Megawatt hours (MWh). 86 Under MiFID I, the trade of wholesale energy products falling within the scope of the Regulation on Wholesale Energy Market Integrity and Transparency (REMIT) that are physically settled and not traded on a Regulated Market or Multilateral Trading facility are generally not considered to be ‘financial instruments’ and are consequently outside the clearing obligation and the margining requirements in Regulation 648/2012 of the European Parliament and the Council of 4 July 2012 on OTC derivatives, central counterparties and trade repositories, eg, ‘European Market Infrastructure Reg’ or ‘EMIR’ (REMIT carveout). 87 Bidding areas are geographical areas spanning an electrically unbroken part of the grid in the actual market area. Norway is currently divided into five price areas, Sweden into four, Denmark into two areas, and Finland, Estonia and Lithuania consist of one area each.

State Aid and the EEA 533 delivered the following day.88 In order to set the system price, the transmission network constraints are disregarded.89 If there are no constraints, the electricity prices in all the Nordic States (ie, all bidding areas) are equal to the system price. However, if there are constraints in the network, the Nordic electricity market subdivides into separate price areas. The area price in areas with excess electricity production will therefore be lower than the system price, whereas the area price in areas with insufficient electricity production will be higher than the system price. However, the goal is that electricity should be able to flow freely between areas. Market mechanisms ensure that the electricity runs from the area with a low price to the area with a high price, thus evening out the price differences. In addition to financial and physical trading, trading can take place on the regulation power market, which is directed by Statnett SF. The regulation power market was established to ensure a balance between supply and demand of electricity in real time. The balance in the regulation market is achieved by Statnett SF, which ensures that producers either buy back electricity or buy more electricity from the producers at a price below or above market price, respectively. Only producers with the ability to respond quickly are active participants in this market. Almost 90 per cent of the retail market is of household customers whereas the rest are commercial and industrial customers. The retail market is highly transparent and competitive as customers have full freedom to change their electricity supplier, and information regarding products and prices is available on the National Competition Authority’s webpage. Since the Nordic Energy market is operated pursuant to free market rules one would not expect major State aid issues and there are no ESA decisions or EFTA Court cases directly related to this market.

IV. STATE AID ENFORCEMENT IN THE EEA

A. Overall: General Trends i. Briefly on the Definition of State Aid As is the case in the EU, difficult questions around whether or not an aid element is present are seldom linked to the requirements of distortion of competition and the effect on trade between the Contracting Parties. As the electricity market is largely liberalised and there is a trade flow in energy products and electricity between the EEA EFTA States (eg, Norway imports and exports a certain percentage of its energy), there is normally a risk of distortion of competition in relation to other EEA undertakings if the other criteria for defining a measure as State aid are met. This is further demonstrated by the fact that various types of energy are traded on Nord Pool, a common framework between the Nordic countries.

88 Other factors, such as the price for electricity outside the Nordic market, are also decisive in setting the system price. 89 Transmission network constraints are also commonly known as bottlenecks.

534 Svein Terje Tveit and Jens Naas-Bibow ii. Environmental and Technology Aid Schemes in the EEA: The Norwegian Energy Fund In 2006, the ESA scrutinised the Norwegian Energy Fund, a support scheme for the production of renewable energy, such as wind, biomass and solar energy for the introduction of energy saving measures; and for the development of new energy technologies in the aforementioned fields.90 The scheme is operated by Enova, a State enterprise owned by the Norwegian State. Enova does not operate on any market and does not generate any income. Its principal task is to administer the Energy Fund, implement the support schemes and achieve the energy policy objectives set out by the Norwegian Parliament. The Energy Fund is subsidised by the Norwegian State budget and financed via a levy on the electricity distribution tariff. Enova can give investment support for energy-saving systems and production and use of renewable energy sources as well as initial investment aid for new energy technologies. Enova grants aid through programme chapters that are open for applications either throughout the year or through calls within certain dates. In its initial decision of 18 May 2005, ESA expressed doubts as to whether the investment support granted by the Energy Fund could be declared compatible under the then applicable Environmental Guidelines.91 The main concern was that the Energy Fund system might in certain instances lead to overcompensation, as the provisions regulating the Energy Fund did not contain precise limitations to ensure that the support does not exceed the difference between the market price of the energy concerned and the production costs.92 More specifically, the net present value (NPV) calculation as suggested by Enova in the notification did not contain sufficiently clear stipulations of the single components of the calculation method.93 Second, ESA noted that the scheme was not based on an ‘extra-cost’ approach as laid down in the environmental aid guidelines; among others it was not clear that the aid in relation to plant depreciation costs would be limited to investment costs only.94 The scheme was based on an NPV approach, meaning that renewable energy projects, which generally have higher investment costs than other technology projects, may obtain aid to reach an NPV of zero, which for the rational investor is the trigger point for when a project is realised. The Authority had concerns that projects that do not generate a positive cash flow could

90 In May 2005, the Authority opened the formal investigation procedure (see Decision 122/05/COL) because of doubts as to whether the Energy Fund as notified is compatible with the EEA State aid provisions. The Authority was, in particular, concerned that the subsidies granted by Enova might overcompensate certain renewable energy producers as well as certain energy-saving measures, and doubted further whether the grants would be limited to investment support. 91 Guidelines on the application and interpretation of Articles 61 and 62 of the EEA Agreement and Article 1 of Protocol 3 to the Surveillance and Court Agreement, adopted and issued by ESA on 19 January 1994 (the 1994 Guidelines) [1994] OJ L231/1, EEA Supps No 32, 1. 92 s D.1.3 (27) of the Environmental Guidelines (as part of the 1994 Guidelines) set out that investment aid must be limited to 40% of the extra investment costs necessary to meet the environmental objectives. 93 See point (54), third sub-para of (old) Environmental Guidelines. 94 The old guidelines point (54) allowed support to compensate for the difference between the production cost of renewable energy and the market price of the power concerned. However, the aid had to be limited to plant depreciation, which was understood as investment depreciation.

State Aid and the EEA 535 also be eligible, and that aid to reach an NPV value of zero could mean that the aid exceeded the actual investment cost. Also, since the NPV calculation suggested by Enova included a return on capital, ESA’s preliminary view was that support to be granted in excess of a zero NPV of the project will result in overcompensation. Third, the ESA questioned whether a flat discount rate (margin) of 7 per cent was correct for all industries, and whether the mechanism to establish the discount rate (by reference to government reports and pre-established ranges of risk premiums for certain industries) is sufficient to establish a rate which precludes overcompensation. Fourth, the ESA also had doubts as to whether there could be a cumulation of aid and whether eligible costs to be compensated were sufficiently clearly defined. Last, ESA also noted that the aid was not calculated according to methods which compare renewable energy production with the production of traditional energy.95 It was not clear whether energy technology unrelated to renewable energy sources could benefit from the scheme; and renewable energy support was generally treated more favourably than energy-saving measures, with up to 100 per cent of eligible costs being supported for the former (where necessary). The ESA ultimately approved the Energy Fund Scheme until 1 January 2011 under Article 61(3)(c) of the EEA Agreement in conjunction with the Authority’s Environmental Guidelines, subject to the Norwegian government’s proposed amendments to the scheme.96 Among others, the revised scheme was based on the ‘extra-cost’ approach in the guidelines and changed so that projects which have a negative discounted cash flow would not be eligible for aid. The Norwegian government also clarified the discount rates based on best practice financial methodologies and limited Enova’s involvement in the projects to the investment phase (once the project had received investment aid, it would not be eligible for further aid).97 The fund was later reorganised and reassessed by the ESA in 2012. Under the reorganised model, the fund comprised four separate programmes which were all approved by the ESA as compatible aid.98 The first two programmes, the renewable energy production programme and the energy-saving programme, were approved under the environmental aid guidelines. The third programme, the cogeneration and energy-efficient district heating and cooling programme, was approved directly on the basis of EEA Article 61(3)(c). In doing so, the ESA concluded that the aid was aimed at a well-defined objective of common interest (energy savings and in turn reduced CO2 emissions), was well designed to deliver that objective (the aid was calculated in accordance with the extra-cost method analogous to paragraphs 124–25 of the environmental aid guidelines), and was constructed so that the overall balance on competition was positive

95

See points D.1.3 (25) and D.1.7 (32) of the Environmental Guidelines. See Decision 125/06/COL of 3 May 2006, s 3.2.4 setting out the conditions subject to which the ESA approved the Energy Fund Scheme. Budget increases for the Fund were subject to Decision 536/09 of 16 December 2009 and 75/10/COL of 10 March 2010. A one-year prolongation was approved by Decision 486/10 of 15 December 2010. 97 s D.1.3 (25), ss D.1.6 (30) and (31), and s D.1.7 (32) of the Environmental Guidelines. 98 See ESA Decision 248/11/COL of 18 July 2011. 96

536 Svein Terje Tveit and Jens Naas-Bibow (the aid will be calculated net of the investment cost of a credible reference investment and net of operating benefits during the first five years). The fourth programme, the new energy technology programme supporting demonstration projects involving new energy technologies, was similarly approved directly on the basis of EEA Article 61(3). According to the ESA, the aid had a welldefined aim related to introducing new energy technologies to the market which would otherwise not be realised (market failure from long-term positive externalities not being sufficently taken into account in the investment decisions). The ESA further pointed to the fact that Enova, in order to grant individual aid, would assess, inter alia, whether the investment represented normal market behaviour, the level of risk and the increased environmental protection brought by the project; and the applicant would provide information about what actions they would have taken without the aid (counterfactual analysis). In the analysis of the third criteria with regard to any distortions of competition and effects on trade, the ESA noted that the aid recipients would be required to make the results of the project publicly available and as such be an input provider to other market participants in ancillary and downstream markets. The aid intensity under the Energy Fund scheme approved by the ESA in Decision 248/11/COL was fixed at a percentage of a project’s eligible cost (ranging from 50 to 80 per cent depending on the programme/supported activity and the size of the enterprise). The ESA therefore, later dealt with several cases concerning the approval of individual aid notifications for demonstration projects that exceeded the thresholds notified and approved in the general Energy Fund scheme for new energy technology programmes. In a decision of 26 May 2016, the ESA approved two individual aid notifications for demonstration projects related to energy emission reductions. First, the ESA approved a grant of up to NOK 280.5 million (approximately €29 million) to Alcoa Norway ANS to demonstrate a novel zero-emission Advanced Smelting Technology (AST) for aluminium production. Second, the ESA approved aid in the form of a non-reimbursable grant of NOK 380 million in nominal value (around €39 million) to Nikkelverk for a full-scale demonstration project in Kristiansand municipality (county of Vest-Agder), Norway. In both instances, the aid was granted by Enova under the new technology programme, which forms part of the Energy Fund scheme as approved by the Authority in July 2011 and had to be individually notified to the ESA for a detailed assessment because of the amount of aid. The ESA further assessed the appropriateness, incentive effect, proportionality and undue negative effects of the aid amount and further noted that the aid was needed to address the market failure related to long-term positive externalities such as knowledge spillovers stemming from the testing and deployment of new technologies not being sufficiently taken into account when profit-seeking undertakings make investment decisions. In line with the Norwegian authorities’ argument, the ESA assessed the two measures in line with Decision 248/11/COL as aid granted under the Energy Fund scheme, and ultimately declared the aid compatible with the functioning of the EEA Agreement on the basis of Article 61(3)(c) of the EEA Agreement.

State Aid and the EEA 537 iii. The Norwegian–Swedish Electricity Certificates Market (‘Green Certificates’) a. Background The main instrument to promote renewable power generation in Norway has been the Norwegian–Swedish green certificates market. The market was established in Norway in January 2012 with the signing of the agreement between Norway and Sweden on a common market for electricity certificates.99 The goal of the electricity certificates market is to finance the difference between the market price and the cost of new production in order to promote investment in renewable energy. The overall goal is to produce 28.4TWh of renewable energy by 2020.100 Currently, investments have taken place mainly in Sweden; in 2015 the normal year production for electricity generation plants included in the 28.4TWh target was 11.64TWh in Sweden and 2.27TWh in Norway.101 However, significant investments are expected in Norway in 2016. The electricity producers receive electricity certificates from Statnett SF (Norway) or the Swedish Energy Agency (Sweden), who are responsible for registering, issuing and deleting certificates. The certificates are given to the renewable energy producers according to their individual production,102 and the producers will later sell the certificates to electricity suppliers and certain electricity users. The certificates are given regardless of whether the power plant is located in Sweden or Norway and do not differ with the type of renewable energy source being used. This ensures compliance with the market investor principle (eg, that the investment in renewable energy generation takes place where profitability is highest). Power suppliers and certain electricity users are in turn legally obliged to purchase a set quota of certificates, but may choose to purchase them in Norway or Sweden.103 NVE and its Swedish counterpart set quotas for how many certificates end users need to purchase by kWh of electricity consumed. The costs incurred for purchasing the electricity certificates by the power suppliers ripple down to each single consumer in the form of an additional cost to their electricity bill. Entitled to electricity certificates are: (i) power plants based on renewable energy sources with a construction start after 7 September 2009; (ii) existing power plants expanding their production on a permanent basis with a construction start after 7 September 2009; and (iii) hydroelectric power stations with an installed capacity of up to 10 MW that had a construction start after 1 January 2004.

99 The treaty on electricity certificates has been implemented into Norwegian law with the enactment of the Electricity Certificate Act (Act-2013-12-13-126) and the Electricity Certificate Regulation (Reg 2011-12-16-1398). The treaty shall cease to apply on 1 April 2036. 100 The treaty, Art 2(1). 101 The Swedish and the Norwegian energy agencies’ annual report for 2015 on the Norwegian– Swedish electricity certificate market, table 3, at 33. 102 1 MWh equals one electricity certificate. 103 In general, the electricity users which have entered into bilateral supply agreements with the producers are obliged to purchase electricity certificates.

538 Svein Terje Tveit and Jens Naas-Bibow Supply and demand for electricity determines the price for the certificates. Both yearly demand for electricity and the set certificate quota for that year are taken into account in determining the demand. The quota of electricity certificates to be bought by the energy suppliers is given independently of the electricity certificates issued. Market mechanisms are then relied upon to ensure that the most economically efficient investments are taking place when the electricity certificate price is high. In short, large-scale investments in new energy production are likely to result in a greater number of certificates and will most likely lead to a lower price for each certificate. Conversely, fewer power plants under construction is likely to cause rising certificate prices until the prices reach a level where the energy market will attract new investors. The Norwegian government expressed its opinion in relation to the electricity certificate market in its non-paper of 16 May 2013: ‘the system is considered to be an effective use of one of the internal cooperation mechanisms under the renewable directive II’.104 However, market participants have questioned the effectiveness of the electricity certificate market and stressed that different rules for taxation of renewable electricity generation in Norway and Sweden create an uneven playing field and lead to inefficiencies.105 b. State Aid Implications of the Electricity Certificates Market Basically, two competing types of policy dominate as European countries implement enhanced investments in renewable energy: feed-in tariffs and green certificates. The European Commission had already decided in 2003 that the issuing of green certificates to producers did not constitute State aid.106 However, guaranteed prices for certificates during the first five years did constitute State aid, but was declared compatible with Article 87.3(c) EC Treaty (now Article 107.3 TFEU). In essence, the aid was necessary, low level and limited to a short period. Guaranteed prices for certificates expired in 2008 and there is no such element in the current system of a common certificate market for green electricity with Norway established from 1 January 2012. The scheme was declared compatible with EU State aid rules, being a nondiscriminatory (ie, not technology-specific) and market-based instrument financed through a State budget. The common Swedish and Norwegian market for the green certificates value chain starts with the issuer governments and the producers that then sell the certificates and electricity. Brokers trade certificates bilaterally, while electricity suppliers buy certificates. The costs are then passed on to the end users. The main advantages with a quota-based certificate scheme are: first, that it is a cost-effective instrument to fulfil renewable electricity targets; and second, that competition drives production costs down.107 104 Olje- og energiedepartment (2013) ‘Norwegian views on European energy issues’ (non-paper, 16 May 2013). 105 Thema Consulting Report 2014–26 on instructions from Energy Norway (Norw: Energi Norge): Group ‘Sertifikatkraft og skatt—oppdatering’ (May 2014). 106 See Commission Decision C(2003)382, final 9. 107 Regarding the cost effectivenes, this is arguably proven by the fact that renewable electricity production increased by 240%, corresponding to 13% of total Swedish electricity production in

State Aid and the EEA 539 ESA Decision 301/14/COL of 16 July 2014 amending for the ninety-eighth time the procedural and substantive rules in the field of State aid by adopting new EEAG 2014-2020 [2015/790]108 hence states that: (130) Contracting Parties may grant support for renewable energy sources by using market mechanisms such as green certificates. These market mechanisms allow all renewable energy producers to benefit indirectly from guaranteed demand for their energy, at a price above the market price for conventional power. The price of these green certificates is not fixed in advance, but depends on market supply and demand.

The certificates market is technology neutral and applies equally to bio power, sun power, hydropower and wind power generation. This is in line with the EEAG stating that the ESA will consider the electricity certificate system to be compatible with EEA Article 61(3) if Contracting Parties can provide sufficient evidence that such support (i) is essential to ensure the viability of the renewable energy sources concerned; (ii) does not, for the scheme in the aggregate, result in overcompensation over time and across technologies, or in overcompensation for individual less deployed technologies in so far as differentiated levels of certificates per unit of output are introduced; and (iii) does not dissuade renewable energy producers from becoming more competitive.

B. Upstream Aid (Energy Industry) i. Overview Given the accessibility of natural resources, the very high degree of renewable energy in overall energy production and consumption, and the concentration of hydropower, State aid questions in relation to the energy industry have been focused in particular on the ownership of natural resources and aid to support alternative renewable energy resources (wind-, bio- and sun-power generation). The main instruments to promote renewable power generation comprise the Norwegian–Swedish green certificates market (as presented above in section IV.A.iii), direct aid (see above in section IV.A.ii on Enova) but also tax depreciation rules.109 In the following, we will provide an overview of the key State aid cases and issues pertaining to such upstream aid. At the outset, it should be recalled, as stated above, that State aid is less of an issue in relation to the ownership of Norwegian waterfalls, since the introduction of

2011, whilst quota fulfilment was nearly 100% (77% first year). Regarding the impact of competition on production costs, see Eva Centeno Lopez, Ministry of Enterprise, Energy and Communications Sweden, presentation to workshop on review of the EEAG 12 April 2013: ‘Common Swedish Norwegian certificate market for renewable electricity’. 108 Notably, the guidelines correspond to the European Commission ‘guidelines on state aid for environmental protection and energy 2014–2020’ adopted on 9 April 2014 [2014] OJ C200/1. 109 (a) ESA has approved direct aid under the Norwegian Energy Fund scheme, cf ESA’s Decisions 487/10/COL, 488/10/COL and 490/10/COL; (b) the Norwegian system of a fixed mark-up per kWh of RES electricity produced—comparison to the feed-in tariff in EU Member States.

540 Svein Terje Tveit and Jens Naas-Bibow the consolidation model and the subsequent changes following the decision of the EFTA Court in the Waterfall case. However, in April 2016, the ESA notified Iceland of issues relating to the granting of natural resources to electricity producers raising issues under the State aid rules.110 In particular, the ESA proposed measures to ensure that electricity producers using public natural resources always pay the market price for those rights. The background for the proposed measures was that Iceland on numerous occasions had granted electricity producers, which had obtained licences to construct power plants, the rights to utilise public natural resources to generate hydroelectric or geothermal energy without any clear legal basis for granting the rights or without fixing a market-based remuneration. ii. Advantageous Tax Depreciation Rules for Wind Power Plants (Norway) The green certificates market has not boosted investment in wind power production in Norway as anticipated by the government and in May 2016 the Norwegian authorities notified more favourable depreciation rules for assets related to investments in new wind power plants to further support investments in wind power generation.111 In Norway, wind power plant assets are currently depreciated according to the general principles of the tax system. This means aiming for the period of depreciation to correspond with the expected economic lifetime of the assets. The notified measure entailed a straight-line depreciation for the assets for wind power plants over five years. The advantage for the beneficiaries consists in faster deprecation of investments in wind power plants compared with current depreciation rules and the increased present value of these deductions from taxable income. The new rules will only be applicable for investments made between 19 June 2015 and 31 December 2021. Only investments in wind power plants are eligible for the new depreciation rules. Other electricity producers cannot apply the measure even if they are otherwise in a similar factual and legal situation. The measure is therefore selective and notified as State aid by the Norwegian government.112 The proposed new rules are 110 See ESA Decision 150/16/COL of 6 July 2016 regarding amendment to the Norwegian Tax Act concerning changes in the depreciation rules for wind power plants. 111 See the Norwegian government’s Notification of new depreciation rules on windpower plants of 10 May 2016 (14/4159 SL LJ). 112 In the hearing preceding the adoption of the new depreciation rules, the industry (eg, Energy Norway et al) argued that the rules should be made technology neutral and apply to all types of renewable energy sources. If so, the question on selectivity would be less straightforward since the measure in effect created a level playing field between Norwegian and Swedish depreciation rules for renewable energy sources within the Norwegian–Swedish electricity market. On the other hand, the measure would still distinguish between renewable and non-renewable energy sources and the question is whether this distinction could be justified by the nature and overall structure of the Norwegian (and Swedish) tax system, cf Commission Decision in 2013 in Case SA.38672 Flexible depreciation of environmental investments (VAMIL)/sustainable stables and sustainable greenhouses, where the Commission noted that the tax reduction would grant an economic advantage only to ‘certain undertakings, namely agricultural undertakings that possess sustainable stable(s) or sustainable greenhouse(s)’. See also previous Cases NN 162/A/2003 and N 317/A/2006 of 4 July 2006—Austria Support of electricity production from renewable sources under the Austrian Green Electricity Act (feed-in tariffs) and Commission Decision of 4 July 2006, in which the Commission states that (para 43): ‘The beneficiaries constitute a selective group of undertakings in the sense of Art 87(1) EC as the measure solely favours certain green

State Aid and the EEA 541 similar to the existing Swedish rules on the depreciation of wind power plant investments, which has been in place since the 1990s (existing aid). The ESA concluded that the aid was compatible under Article 61(3)(c) of the EEA Agreement on the basis of the EEAG. In particular, the ESA noted that it will not be profitable to invest in and develop new wind power projects with the current electricity prices and that the low pricing was expected to continue (the combined expected price of electricity and certificates for new wind producers in the forward markets is below 40 €/MWh and hence below the marginal cost of wind power. The ESA further noted that tax depreciation rules are of general application (and similar to the rules in Sweden) and presumed less distortive than direct grants. As regards the incentive effect, the ESA referred to GBER Article 6(4)(a) and held that it was not necessary to show such effects in this context of tax advantages. In Article 6(4)(a) a derogation is made for aid measures consisting of tax advantages that establish a right to aid in accordance with objective criteria, and without further exercise of discretion. Furthermore, the aid intensity was low.113 In assessing the proportionality of the aid, the Authority found that the measure had some negative effects on hydropower investments, in particular marginal hydro projects, which are not realised without a depreciation incentive. On balance, such negative effects were, however, limited and outweighed by the positive effects of the measure in terms of contributing to the objective of common interest (see point 83 of the EEAG). iii. Aid in the Hydrocarbon Sector: Oil and Gas In Decision 411/06/COL, the ESA approved special depreciation rates for largescale LNG facilities in specific regions of northern Norway. The Norwegian Tax Act fixes the corporate profits tax at 25 per cent (then 28 per cent). Tangible operating assets are depreciated in accordance with the declining balance method. Operating assets are divided into depreciation groups. The rate of depreciation applicable for 2002 varied from 30 per cent of the remaining balance for office equipment down to 2 per cent for commercial buildings. The rate of depreciation applicable to the plant and equipment is 4 per cent of the remaining balance. The Petroleum Tax Act (PTA) sets out a tax regime that is specially adapted to income from extraction and transportation by pipeline of oil and gas on the Norwegian continental shelf. Income is subject to corporate tax on ordinary income

electricity producers’. Conversely, the ESA guidelines for ‘Application of State aid rules to measures relating to direct business taxation’ (s 3.1, paras 1 and 2) makes clear that the measure may not be selective if there are objectifiable differences between renewable and non-renewable energy sources: ‘The fact that some firms or some sectors benefit more than others from some of these tax measures does not necessarily mean that they are caught by the competition rules governing State aid. Thus, measures designed to reduce the taxation of labour for all firms have a relatively greater effect on labour-intensive industries than on capital-intensive industries, without necessarily constituting State aid. Similarly, tax incentives for environmental, R&D or training investment favour only the firms which undertake such investment, but again do not necessarily constitute State aid’. 113 With an aid intensity of 3.7%, the average individual aid amount under the notified measure would range from NOK 19 to 22 million.

542 Svein Terje Tveit and Jens Naas-Bibow at a rate of 25 per cent and special tax at a rate of 53 per cent on net income adjusted for uplift, ie, a total rate of 78 per cent. Uplift is shielding part of the income from the special tax. The PTA also contains special provisions on the calculation of income, including the use of norm prices for oil and gas for purposes of tax assessment. Expenses incurred in acquiring tangible operating assets relating to extraction and pipeline activities may be depreciated at a maximum rate of 16⅔ per cent per year. On 22 November 2001, the Norwegian Parliament approved an amendment to the PTA. Under the proposal, expenses for the acquisition of pipelines and production equipment would fall under the PTA and may be depreciated for tax purposes by up to 33⅓ per cent per year where the object, according to an approved plan for development and operation and a plan for installation and operation in accordance with the PTA, is the production and pipeline transport of gas which is to be cooled to liquid form in a new large-scale cooling plant. Only facilities where a plan for development and operation and a plan for installation and operation, in accordance with the PTA, has been approved after 1 January 2001 were defined as ‘new’. ‘Large-scale’ was defined as a minimum production capacity of four billion standard cubic metres per year. According to the general rules in the PTA, expenses for the acquisition of pipelines and production equipment may be depreciated for tax purposes by up to 16⅔ per cent per year. The amendment to the PTA implied that investment in conjunction with the production and pipeline transportation of gas that is to be cooled to liquid form in a new large-scale cooling plant could be depreciated over three years instead of over six years as in the ordinary rules in the PTA. LNG installations were deemed to constitute part of the pipeline transportation system, and were thus subject to the depreciation rule. Changing the depreciation period would provide lower tax income for the Norwegian State in the first three years and give a correspondingly higher income in the next three years. This means a loss for the State at present values and a corresponding gain for the companies.114 On 27 May 2002, the Norwegian government put forth a proposal to the Parliament to amend the geographical scope of the PTA Section 3(b), third sentence, to the effect that the rules on depreciation contained therein would be applicable only in cases where the large-scale LNG facility was located within the geographical areas of Finnmark County or the municipalities of Kåfjord, Skjervøy, Nordreisa or Kvænangen in Troms County. Otherwise, the increased depreciation rates and the other provisions of the PTA were the same as in Ot.prp. No 16 (2001–02), as described above. ESA assessed the compatibility of the proposed scheme with Article 61(3)(c) of the EEA Agreement and in light of the new Regional Aid Guidelines. The ESA observed that the area covered by the notified aid scheme, are all covered by the regional aid

114 With an approximate risk-free discount rate, the present value loss due to the amended depreciation rules seen in isolation would be over 4% of total investment (measured at present values), according to Ot.prp. No 16 (2001–02).

State Aid and the EEA 543 map for Norway for the period from 2007 to 2013 as approved by the ESA.115 The economic activity in this northern region of Norway has traditionally been based on natural resources, and fishing and agriculture still dominate the region. In addition, a considerable part of the workforce is employed in the public sector. The industry in the region is characterised by a low degree of diversification, and internal distances in the region are considerable.116 The ESA further observed that the Snøhvit project, the only project that so far has been granted aid under the scheme, is expected to create 350–400 new jobs in the region. Against this background, the ESA concluded that the aid scheme contributes to a coherent long-term regional development strategy, ie, by contributing to more diversified economic activity and thus to a more developed labour market in the region. The upper limit of the level of aid under the notified scheme would be 8 per cent and below the maximum aid ceiling set out in Decision 226/06/COL on the map of assisted areas and levels of aid in Norway and in the new Regional Aid Guidelines.117 On a more general level, there is a question as to whether the gassled tariffs as operated by the Gassled joint venture owning the Norwegian gas transportation system could entail State aid in favour of Norwegian gas distributors.118 This question raises difficult issues, among others, such as concerns as to whether State resources are involved and whether there is an economic advantage granted to the Norwegian gas distributors.119 iv. The EFTA Countries and the PCIs and Energy Infrastructure Projects The EEA EFTA Countries have three projects on the Commission’s list of Projects of Common Interest (PCIs):120 the interconnection between Wilster (DE) and Tonstad (NO) (NordLink); the Norway–United Kingdom interconnection; and the interconnection between Iceland and the United Kingdom (Ice Link). All three

115

Decision 226/06/COL on the regional aid map for Norway. The distance from the eastern to the western extremity of the zone is about 1000 kilometres. 117 The aid intensity applicable to regional aid in the area covered by the scheme is 15% gross grant equivalent for large enterprises. 118 The Gassled joint venture agreement was entered into force on 20 December 2002 between Petoro AS, Solveig Gas Norway AS and eight other Norwegian gas distributors, taking effect from 1 January 2003. Prior to this, the nine transportation systems included in Gassled were owned through separate joint ventures by 13 different companies. The gas transportation systems are all operated by Gassco, a company wholly owned by the Norwegian State. Thus, the establishment of Gassled (and Gassco) must be seen in connection with the phasing out of the Norwegian ‘GFU’ arrangement (where all gas sales were organised through a gas sales committee), the transition to a company-based sale of gas and the implementation of the EU gas directive. The reorganisation of the ownership also allows for efficient utilisation of the gas transportation system, including third-party access. 119 cf Case C-262/12 Vent de Colére, EU:C:2013:851, presented above where the Court confirms that measures not involving a transfer of State resources may also qualifiy as State aid if there is an intervention through State resources (as oppose to the judgment in Case C-379/98 Preussen Elektra), the Vent de Colére case concerned private undertakings appointed by the State to manage State resources. 120 See European Commission of 18 November 2015, Annex to Commission delegated regulation amending Regulation No 347/2013 of the European Parliament and of the Council as regards the Union list of projects of common interest. 116

544 Svein Terje Tveit and Jens Naas-Bibow interconnections are electricity interconnections. Currently, there is no State involved in the funding of these projects.121 v. State Support to Innovative Carbon Capture and Storage Projects (Test Centres at Mongstad and Kårstø) Norway has been a front-runner in the technological race to develop carbon capture and storage (CCS) technology which can help reducing CO2 emissions while continuing burning fossil fuels such as gas or coal.122 This has led to a body of case law from the ESA similar to that of the Commission.123 Despite the testing of CCS starting over a decade ago, the technology is still in its infancy and has not been applied on a large commercial scale. The first decision came in 2005, when the ESA decided that a research and development (R&D) programme run by Gassnova124 funding the development of prototype and demonstration projects for gas-fired power stations with CCS technology, constituted compatible aid under EEA Article 61(3)(c) in conjunction with ESA’s R&D Guidelines.125 Through Gassnova, the Norwegian State would later invest in the construction of a CCS test facility: the Test Centre Mongstad in western Norway. Hence, on 12 October 2006 the Norwegian Ministry of Petroleum and Energy signed an ‘Implementation Agreement’ with Statoil ASA, according to which the State would support the construction of the Mongstad Test Centre (Step 1). Pursuant to the Implementation Agreement, Statoil should perform a study and outline the requirements, risks and possible solutions towards a full-scale post-combustion CO2 capture facility and the State would then support the construction and operation of a full-scale CO2 capture project (Step 2). Step 2 consisted of two phases, ‘Mongstad Full-Scale CCS Development’ (Phase 1) and ‘Mongstad Full-Scale CCS Project Construction & Operation’ (Phase 2). The Development Agreement entered into by Statoil and the State on 5 April 2011 stated that the parties are ‘jointly responsible for realising the Capture Facility’, however, so that Statoil was responsible for the daily management of the project and in charge of the development work.

121 The Commission is concerned about national capacity markets and has previously approved intensive State aid in interconnection projects, cf Case SA.39050 of 15 July 2015 where the Commission approved €750 million aid for gas pipelines in Poland under the EEAG, connecting Europe via Poland to the Baltic, Adriatic and Black seas (as part of the ‘North–South gas interconnection priority corridor’). The aid from the European Regional Development Fund covered 64% of the total investment costs and the remainder was funded by the Polish TSO. The project was not sustainable without aid. In addition, there are other non-TSO interconnector projects such as the North Connect (Norway–UK). 122 See EU Directive 2009/EC/EC on geological storage of carbon dioxide [2009] OJ L140/114. The Directive is implemented in the EEA. 123 See CCS Demonstration Competition—Feed (Case N 74/2009) [2009] OJ C203/1; CCS Project in the Rotterdam Harbour Area (Case N 381/2010) [2011] OJ C149/3; CO2 Catch-up pilot project at Nuon Buggenum plant (Case N 190/2009) [2010] OJ C238/1. 124 Gassnova is a State company (Norw: ‘statsforetak’) established in 2007 with the purpose to, among others, administer the State’s interests in the CCM project. 125 Decision No 302/05/COL of 30 November 2005. Later, the ESA cleared certain amendments to the Gassnova scheme, see Decisions No 768/08/COL of 17 December 2008 and Decision 348/10/COL of 15 September 2010.

State Aid and the EEA 545 Mongstad was a showcase project for the viability of CCS on a larger scale and the Test Centre was envisaged to operate for five years without generating revenue or cost coverage in the period. The project was financed by the State. The Norwegian government held that they had a commercial interest in the development of the Centre as a leading nation in oil and gas production, and lower carbon costs affecting the gas price in the future. The ESA declined, however, to compare the investments with that of a hypothetical private investor and assessed the compatibility of the aid under EEA Article 61(3). In a series of decisions, the ESA declared the funding of the Test Centre Mongstad in the various stages as compatible with EEA Article 61(3)(c). First, the ESA concluded that the investment to own and operate TCM for a period of five years (Step 1) constituted compatible aid.126 In this decision, the ESA referred to the Commission’s policy (‘An energy policy for Europe’)127 and the target to ‘design a mechanism to stimulate the construction and operation by 2015 of up to 12 large-scale demonstrations of sustainable fossil fuels technologies in commercial power generation in the EU 25’ and ‘provide a clear perspective when coal- and gas-fired plants will need to install CO2 capture and storage’. The decision is in line with the supportive EU policy towards CCS. Second, the ESA concluded not to raise any objections to the financing of the Step 2 Development Agreement, the first phase of the Mongstad CCS Facility.128 In this decision, the ESA differentiated between three beneficiaries of the aid: Mongstad’s main shareholder Statoil (on which the aid confers an advantage, especially since it aims at exploiting its CCS-related know-how also commercially); Gassnova (which is not engaged in an economic activity); and technology suppliers (no advantage). Assessing the costs and benefits of the entire project, the ESA considered the project to be compatible with EEA Article 61(3)(c). Curiously, however, the Mongstad project was closed by the government with effect from 1 January 2014.129 In 2009, the ESA approved State funding to Gassnova SF in order to cover the costs of establishing a second CCS facility at Kårstø, Norway.130 The CCS will capture and store CO2 emitted by the gas power plant which is jointly owned by StatoilHydro ASA and Statkraft SF. The CCS was to be owned and managed by the 100 per cent State-owned company, Gassnova SF, which will also be the direct recipient of the State funding. The State funding covered 100 per cent of the investment cost and operating costs of the CCS over a period of 10 years. In the MEIP test under EEA Article 61(1), the ESA considered the price of ETS allowances (saved) and concluded that no private market investor would have undertaken a similar investment at the current price level. The ESA further noted that the price (costs 126

See Decision No 503/08/COL of 16 July 2008. ‘An energy policy for Europe’ (n 64). 128 Decision 91/12/COL of 15 March 2012. 129 The Office of the Auditor General of Norway published a report (Doc 3:14—Investigation into the State’s work with CO2 capture and storage (2012–2013)) in which the Ministry was criticised for the lack of efficient cost-control. The total expenditures of the Norwegian State for the period from 2007 to 2012 on the CCS facilitites on Mongstad and Kårstø amounted to NOK 7.4 billion. 130 Decision No 27/09/COL of 29 January 2008. 127

546 Svein Terje Tveit and Jens Naas-Bibow of allowances) to be paid by the operator of the CCS installation to Gassnova far from covered the real costs of a ton of CO2 stored underground. Building the installation with State funds thus conferred an economic advantage on the undertaking. In the balancing test under EEA Article 61(3), the ESA again concluded that the aid was compatible in view of the EEA objective to reduce CO2 emissions and combat climate change. The 2013 Brevik case concerning CCS in the cement industry was the first case outside the energy production sector. In brief, the ESA approved aid of approximately NOK 70 million from Norway to Norcem for the construction of CCS research facilities at Norcem’s cement plant in Brevik. Cement production involves the calcinations of limestone, which is responsible for high direct CO2 emissions. The ESA made reference to the previous decisions in Kårstø and Mongstad, and noted that the know-how developed would benefit the cement industry in Europe. vi. Emission Trading System In September 2013, the Authority approved a Norwegian aid scheme intended to compensate certain energy-intensive undertakings for increases in electricity prices resulting from the inclusion of the costs of greenhouse gas emissions. The aid scheme was mainly directed at certain predefined manufacturing industries (aluminium, chemicals, plastic, iron ore production etc). The scheme was intended to counter carbon leakage due to increased electricity prices resulting from the EU ETS and therefore approved under the ETS guidelines as aid for indirect emission costs.

C. Downstream Aid (Energy User) i. Overview The high level of energy generation in Norway and Iceland and the availability of power at comparably low prices have attracted a power-intensive industry made up mainly of manufacturing industries that consume around two-thirds of the total energy use in Norway and 80 per cent in Iceland. The power-intensive industry in Norway and Iceland has covered most of its need for power through long-term contracts under terms set by the governments, and the ESA has, on several occasions, dealt with State aid questions relating to Power Purchase Agreements (PPAs). This has led to several decisions from the EFTA Court and the ESA dealing with aid in the form of preferential rates, tax exemptions or other benefits for energy-intensive industries, which will be the focus of this chapter. Downstream aid has, however, also been granted to support the production and sale of environmentally friendly electric cars and to support alternative, renewable heating and electricity savings in private households (Norway).131 We will briefly comment on these cases below. 131 See also the recent ESA Decision 233/16/COL regarding the approval of an Enova aid scheme in the period from 2017 to 2022 intended to incentivise eco innovation. Note that the scheme was technology neutral, not sector-specific and directed at the end users adopting environmentally friendly innovations.

State Aid and the EEA 547 ii. Norwegian Long-Term Power Contracts: Aid to the Power-Intensive Industry Historically, the power-intensive industry in Norway has covered most of its need for power through long-term contracts under terms set by the government. This led to several State aid cases. Norway has since the early nineteenth century actively used the hydropower resources to develop power-intensive industry such as the aluminium industry, inter alia, by providing electricity at lower than market prices. Prior to the EEA Agreement, the EU (EC) several times threatened anti-dumping measures. After the entry into force of the EEA Agreement, anti-dumping concerns have been replaced by State aid concerns relating in particular to three issues: (i) derogations from electricity tax for certain production processes and reduced rates for certain industries (see section IV.C.iv below); (ii) the resale of concession power to below market prices (see section IV.C.v below); and the long-term power contracts (at below market prices) (this section). Long-term power contracts entered into between the State electricity company (Statkraft) and the power-intensive industry on terms fixed by the Norwegian Parliament which generally have been beneficial and below market price has a long tradition in Norway. Upon the entry into force of the EEA Agreement, several such PPAs existed. These agreements were not notified to the ESA as existing aid in 1994, and even if the ESA on several occasions considered the matter, at no point was there any requirement to abolish the contracts. There may be several reasons for this. First, there was an inherent difficulty in deciding what the market price under these contracts would have been, in particular prior to the liberalisation of the energy market. Second, most contracts expired between 2004 and 2011 and the Norwegian government (after a complaint by Bellona) agreed to let the existing agreements expire without further renewals. In order to secure the industry’s access to power on terms that were sustainable for the industry, the government later introduced a State guarantee scheme. Thus, by letter of 22 September 2010 (Event No 570557), the Norwegian authorities notified a guarantee scheme for the purchase of electric power on long-term contracts, pursuant to Article 1(3) of Part I of Protocol 3. The scheme was assessed under the Authority’s State Aid Guidelines on State Guarantees. The scheme was closed to borrowers in financial difficulty; each guarantee would be linked to a specific contract for the long-term supply of electricity with a fixed maximum amount and limited in time and a maximum of 80 per cent of the upfront payment or of the contract value would be guaranteed under the scheme. The assessment and calculation of the risks associated with the guarantee and the determination of appropriate premiums aimed to ensure that the guarantee scheme would be self-financing and operated on commercial terms.132 Specifically, although the guarantors were not subject to capital requirement rules as usual guarantors not under State guarantees would be, the pricing model took account of this disparity

132 The model reflected the following three factors of guarantee premiums: (i) the cost of capital associated with the relevant capital requirement; (ii) administrative costs; and (iii) general credit risk.

548 Svein Terje Tveit and Jens Naas-Bibow and required that the guarantee premiums covered remuneration for an adequate amount of capital even though this is not constituted. In the ESA’s view this ensured a level playing field for usual guarantors and an adequate remuneration to the State for the risk it would be taking. More generally, it should be noted that the testing of the private market investor principle in State aid cases concerning the power-intensive industry in the Nordic countries is difficult. Indeed, the power-intensive industry operating in the Nordic countries is dominated by a few large companies operating within different industries, such as aluminium, ferrosilicon and wood processing, where there is limited available data that could be useful in exploring the risks specific to each industry. Second, an assessment of the risk and the precise value of the premium may be difficult to determine in advance since the provision of guarantees for the purchase of electric power on long-term contracts is subject to a number of different risk factors, such as the solidity of the buyers, the future development of power markets etc. This was also the case in the Landsvirkjun case presented in the next section where the assessment rested mainly on the absence of a relevant market price to be relied on in the context of the private investor test, which in turn resulted from Iceland’s isolated status. iii. Long-Term Power Purchase Agreements/Stranded Costs: Case Study Iceland In 2015, the Authority reviewed two power contracts signed between Landsvirkjun (a fully State-owned electricity company) and energy-intensive users United Silicon and PCC (Decisions 67/15/COL and 207/15/COL, respectively).133 Ultimately, the ESA concluded that there was no State aid involved since the contracts reflected market terms. The ESA undertook a regular private market investor test and a fact-finding mission. As a general observation, it is worth noting that since the Icelandic market is isolated (no interconnectors) and the large majority of all electricity is sold to a few customers, no market price is available. The ESA has therefore not based the market economic operator test on benchmarking. Instead, the ESA assessed the power contracts individually and defined an acceptable level of return based on profitability (WACC), the expected NPV of the power contract and the related investment decision (in new power plants).134 In order to analyse the intensity and method of the ESA’s market investor principle testing, it is, however, useful to recap the facts and analysis from the decision to open a formal investigation.135 133 In the decision to open a formal investigation into the power purchase contract, the Authority could not exclude that State resources were involved ‘in light of the legal status of Landsvirkjun, the possible use of Landsvirkjun as a tool to attract foreign investment as a vehicle for job creation and economic development … and the public ownership of Landsvirkjun’. This appears to be in line with the Hungarian cases concerning certain PPAs. The Commission opinioned that the advantage stemmed from the use of State resources, because the decision to sign the PPA was a consequence of State policy implemented via the State-owned public utility wholesaler (State aid in energy sector, 7). 134 The latest example is ESA’s Decision 130/16/COL of 17 June 2016 on the sale of electricity to Thorsil (Iceland) in which ESA concluded that no State aid was involved. 135 See ESA’s invitation to submit comments pursuant to Art 1(2) in Pt I of Protocol 3 to the Surveillance and Court Agreement on State aid issues concerning a power contract between the Icelandic National

State Aid and the EEA 549 The PCC contract was concluded in March 2015, following a decision by the Authority to open an in-depth investigation into a similar contract signed in 2014.136 As the 2014 contract was terminated by the parties without entering into force, that investigation was closed by Decision 238/15/COL. The power contract with United Silicon was signed on 19 March 2014. The ESA looked into the profitability calculations and concluded that the contracts generate a reasonable rate of return for Landsvirkjun, hence a private market player would have accepted the terms of the PCC agreements. In doing so, the Authority made a number of observations. First, since a market price was not readily available, the Authority carried out an assessment of the profitability of the investment needed to provide PCC with the contract power in order to establish whether a private market economy operator would have concluded the contract on the same terms. The Authority sought reference to determine the return on the investment in the power station by calculating the NPV and/or internal rate of return (IRR) on the project. Second, the ESA noted that if Landsvirkjun needed to build new power facilities to provide the required electricity, the expected revenues from the power contracts should recoup those investments.137 Third, the ESA questioned Landsvirkjun’s submission that exploration and drilling costs should not be taken into account when calculating the profitability of the power contract. The Authority observed that the drilling of the wells was a considerable share of the capital cost of a geothermal power plant and that the wells were envisaged as production wells from the beginning, at the time when they were drilled as exploratory wells. Fourth, the ESA noted that Landsvirkjun appeared not to have included any required return on the accrued costs or any cost related to connecting the power station to the transmission system, which could be an unrealistic assumption given the need for a transmission agreement. Also, the PCC contract did not contain any conditions related to the Þeistareykir power station. This meant that Landsvirkjun bore the risk if the costs of constructing the power station increased, or if its construction was delayed. The ESA further questioned whether PCC would receive an economic advantage in the form of preferential transmission charges via the transmission agreement in the form of exemptions from costs that normally would be borne by the company in its normal course of business.138 The Icelandic system provided for a ‘Step-down Surcharge’ established to compensate for the actual extra costs related to the stepping-down of the electricity from 220 kV or 132 kV (the voltage of the grid) to the lower delivery voltage requested by the DSOs or energy-intensive user. The transmission agreement did not make an exemption for these costs for PCC Bakki-Silicon, but the calculation of the surcharge compensated only 80 per cent of the actual costs.

Power Company, Landsvirkjun, and PCC Bakki Silicon hf and an electricity transmission agreement between Landsvirkjun’s subsidiary, the transmission system operator, Landsnet, and PCC Bakki Silicon hf for a planned silicon metal plant with a capacity of 33,000 tons per annum to be built in Iceland. Invitation to submit comments of 19 March 2015 [2015] OJ C92/3. 136

See ESA Decision 543/14/COL. This was the case for PCC’s agreement, while the electricity for United Silicon’s new plant will be provided from existing facilities. 138 See ESA Decision 206/15/COL of 20 May 2015, para 60. 137

550 Svein Terje Tveit and Jens Naas-Bibow This was, however, ultimately accepted due to the existence of synergies such as the joint utilisation of facilities, equipment and reserve components. Second, the Authority noted that PCC did not pay the regular System Contribution Surcharge imposed on intensive users if the connection of the new plant caused additional costs for the existing users. However, the Surcharge was not required if the costs of the PCC project were not customer specific but related to the general investments that the local TSO (Landsnet) had an obligation to undertake. In effect, the decision was based on an application of the market investor test. For these reasons, the Authority initially concluded that it had doubts as to whether the power contract was concluded on market terms. The Icelandic authorities argued that the power contract yielded an acceptable return. In particular, the authorities stressed the business risks for the buyer, the duration (possibility to adjust to market developments), the profitability of the investments made by Landsvirkjun, and the comparability to other contracts with energy-intensive users. The Icelandic authorities submitted that ESA had to take into account: (i) that the power price was high compared with existing power contracts with energy-intensive users; (ii) the duration of the power contract was shorter than in existing power contracts with energy-intensive users; and (iii) there was the possibility of getting higher prices from the plant and its extension in the future, and to get higher prices from other energy-intensive users.49 The ESA turned in the final decision and concluded that it was satisfied with the evidence provided by the Icelandic government supporting the argument that the power contract would be profitable and entered into on market terms.139 The Authority started out with reiterating the basic premise in the ESA State aid guidelines that only where there are no objective grounds to reasonably expect that an investment will give an adequate rate of return that would be acceptable to a private investor in a comparable private undertaking operating under normal market conditions, is State aid involved.

The Authority undertook the assessment by reference to the NPV of the project. While the Authority and Iceland did not share the same opinion on whether the accrued costs should be taken into account, the 2015 power contract proved to be profitable with or without already accrued costs. The ESA in particular noted the short duration of the contracts allowing Landsvirkjun to adjust its prices to market developments elsewhere better than what was possible in past contracts, the Take or Pay obligation that ensures a constant stream of revenue regardless of the business success of PCC, the flexible curtailment options according to the power contract implying that Landsvirkjun could curtail more power from PCC than from other power-intensive industries if there is an energy shortage in Iceland without a need to 139 The ESA’s decision of 20 May 2015 on the sale of electricity to the PCC Silicon Metal Plant at Bakki under the 2015 Power Contract (Iceland), Decision 207/15/COL. The Authority initially opened an investigation and issued a statement of objections in relation to the power contract and the transmission contract (viewed as inseparable), but PCC and the Icelandic TSO (Landsnet) entered into a new transmission agreement in 2015 replacing the old 2014 agreement. The 2015 Transmission Agreement was also notified for legal certainty to the Authority and has been assessed separately (Decision 206/15/COL on the transmission of electricity to the PCC Silicon Metal Plant at Bakki (2015 Transmission Agreement)).

State Aid and the EEA 551 pay compensations for this, and finally noted that there are other likely additional revenue drivers that could stem directly or indirectly from the power contract.140 iv. Derogations from Electricity Tax for Certain Production Processes and Reduced Rates for Certain Industries Ever since the Norwegian tax on electricity consumption was introduced in 1971, certain power-intensive industries have benefited from either reduced rates or full tax exemptions. In 2002, this aid was estimated to constitute around NOK 4.6 billion per year. In principle, this exemption constitutes unlawful State aid under EEA Article 61(1). However, the EU has introduced harmonised rules on business taxation which allows substantially reduced rates if beneficial to the environment. The rationale being that a high levy on electricity consumption is generally desirable, but not obtainable without granting certain exemptions to the industry—thus the EU has chosen to harmonise the access to grant such exemptions. The Norwegian scheme was, however, not harmonised with the EU scheme. Following the Norwegian government’s procrastination on harmonising and notifying the measure, the ESA in 2004, inter alia, declared the sectoral exemption and the regional derogations from the electricity tax State aid within the meaning of Article 61(1) of the EEA Agreement and ordered the recovery of a smaller amount of the aid received from 6 February 2003 onwards.141 At the same time, the ESA made another decision in which they approved of a new (and notified) scheme for continued significant exemptions for the industry (chemical reduction, electrolytic processes as well as in mineralogical and metallurgical processes) based on the aid level as accepted in the EU.142 It is worth noting that the EU energy tax directive which justified the acceptance of the aid as compatible with the EEA Article 61(3)(c) was not a part of the EEA Agreement. Curiously, however, the ESA accepted that a similar arrangement could be made under the EEA law. Ultimately, the decisions meant that the traditional Norwegian electricity tax exemption scheme could be prolonged with some adjustments. For 2012, this aid was estimated to amount to around NOK 5.2 billion. v. Resale of Concession Power As stated above, the resale of concession power from the Norwegian municipalities to local enterprises has been a feature in the overall Norwegian policy to support the building of the power-intensive industry. Several local municipalities in Norway have access to cheap power, either by generating the power themselves or because, as a host municipality for power generating undertakings, they are entitled to receive a share (so-called concession power) at cost price.

140 eg, additional revenues through the sale of green certificates, by-products coming from geothermal power plants such as hot water, low-pressure steam, CO2 and H2S that can be developed into valuable products for sale, etc. 141 Decision No 148/04/COL of 30 June 2004. 142 ibid.

552 Svein Terje Tveit and Jens Naas-Bibow It is fair to assume that there are multiple instances of Norwegian municipalities granting State aid through the resale of concession power, without these cases being notified or approved by the ESA. To date, the ESA has not made any negative decisions in any single case. Instead, the ESA has had an ongoing dialogue with the Oil and Energy Ministry, whereby the Ministry has informed the municipalities of the responsibilities under State aid law. Over the last few years, there seems to be a shifting view among the municipalities and an increased focus on obtaining a market price for the power. In 2009, the ESA received a complaint filed against Narvik municipality (Norway) regarding the sale of concession power to Narvik Energy AS (NEAS). According to the complaint, on 16 October 2000 the municipality entered into a contract with NEAS for the sale of 128 GWh of annual concession power for a period of 50.5 years. For this, NEAS paid the municipality one upfront lump sum of NOK 126 million.143 The contract, that was entered into after negotiation between the contracting parties and without a competitive tender procedure, contained no index adjustment or other price adjustment clauses. The ESA ultimately decided that the contract with NEAS was concluded on market terms and closed the formal investigation.144 The ESA took note of the particular circumstances of the case including the fact that the municipalities’ finances were strained and it was in need of liquid capital and the concession power had previously been sold at a loss. More specifically, the ESA accepted that the prices for previously sold hydropower plants represented an adequate proxy to determine the lump sum market price to be paid for the long-term sale of concession power entitlements. vi. Individual Aid to the Industry under the Enova NETP Programme In 2015, the Authority approved aid by Enova to two demonstration plant projects through the New Energy Technology Programme (NETP), part of the Energy Fund scheme approved by the Authority in Decision 248/11/COL. The NETP provides support to demonstration projects for innovative technologies in order to foster their market diffusion. In both cases described below, an individual notification for a detailed assessment was required due to the high aid amounts. Later, in Decision 234/16/COL, the ESA also approved support to environmentally friendly demonstration projects and feasibility studies directly linked to the demonstration projects for the period from 2017 to 2022. The aid scheme is in particular meant to incentivise new technologies, products and processes within the fields of renewable energy production, energy efficiency and reduced greenhouse gas emissions, and the decision therefore comes as no surprise. In November 2015, the Authority approved a non-reimbursable grant to Tizir Titanium & Iron (TTI) of around NOK 122.7 million (approximately €12.9 million),

143 The Norwegian system of concession power is laid down in s 2(12) of the Industrial Licensing Act and s 12(15) of the Waterfalls Regulation Act and in effect entails any municipality which has hydropower production within its borders is entitled to receive an annual amount of concession power from concessionaires for waterfall exploitation. 144 See ESA Decision 258/13/COL.

State Aid and the EEA 553 for a full-scale demonstration project testing a new environmentally friendly technology (Decision 476/15/COL). TTI produces high purity pig iron and titanium dioxide slag from an ilmenite ore in a two-step process—pre-reduction and smelting. The purpose of the demonstration project is to verify, at full-scale, the use of a new water-cooled roof technology introduced in the smelting furnace. The technology is expected to result in both reduced energy consumption and the reduced emissions of CO2 during the production process. If this new technology proves to be successful, it may be applied in the whole metallurgical industry and may contribute to important reductions in energy consumption and CO2 emissions. The new technology will be available to third parties. In February 2015, the Authority approved a non-reimbursable grant of up to NOK 1.486 million (approximately €165 million) to Hydro Aluminium AS for the construction of a demonstration plant in the municipality of Karmøy, Norway (Decision 37/15/COL). The purpose of the aid is to enable the construction of a demonstration plant for the verification of a new smelting cell technology. Based on the research so far, this new technology will lead to an important reduction in both electricity consumption and direct greenhouse gas emissions. Due to the technological risks involved in the design of the new cells and the high investment costs, a verification of the technology at industrial scale is required before it can be put to commercial use. In its assessment, the Authority found that the project suffers from a funding gap and would not be pursued without aid. Furthermore, the maximum aid intensity of 50 per cent under the Energy Fund scheme was respected. There were also mechanisms to prevent overcompensation. Finally, the new technology will be available to third parties on commercial terms for investments in the EEA once it has been successfully verified in the demonstration plant. vii. State Aid for Electric Cars In Decision 150/15/COL, the Authority deemed several State aid indirectly benefiting manufacturers and dealers of electric vehicles and batteries as compatible aid under Article 61(83)(c) of the EEA Agreement. The comprehensive support programme intended to run to 31 December 2017 includes zero VAT rating for the supply, import and leasing of electric vehicles and the supply and import of batteries for electric vehicles, reduced annual vehicle tax, exemptions from road tolls and road ferries boarding charge and a favourable income tax calculation for employees benefiting from the private use of corporate electric cars. The overall aim is to reduce the CO2 emissions of the transport sector and support Norway’s overall goal to lower the greenhouse gas emissions by 2030 by at least 40 per cent compared with the 1990 level. The Authority held that the permission for electric vehicles to drive in bus lanes did not involve any commitment of State resources and therefore did not constitute State aid.145 Further, and unsurprisingly, the Authority refused to accept 145 This is in accordance with the decision of the Court of Justice in Case C-518/13 Eventech Ltd v The Parking Adjudicator, EU:C:2015:9, allowing London taxis to use bus lanes.

554 Svein Terje Tveit and Jens Naas-Bibow the zero VAT rating rules as justified by the logic and nature of the Norwegian VAT system and were therefore non-selective. The Authority noted that albeit exceptions to the basic (Norwegian) principle that the consumption of goods or services should be charged with a tax on consumption could be justified by an objective of common interest, it did not form part of the logic and general nature of a consumption tax system. The justification to promote the use of electric cars was considered external to the Norwegian fiscal system.146 In the compatibility assessment, the Authority agreed with the argument of the Norwegian State that the higher price of electric cars compared with conventional fuel cars suggests a market failure which the aid measures responded to. In the proportionality assessment, the Authority accepts that the measures are appropriate to achieve the general emission targets and focused on whether there in fact was a difference in price to conventional cars and whether this final price could be sufficiently accepted. The Authority did not, however, commence on a more general discussion as to whether other measures are better suited at targeting CO2 emission reductions. Recently, in Decision 232/16/COL, the Authority recently accepted ancillary aid to electric car owners (and manufacturers and suppliers) when deciding not to raise any objections to the prolongation and budget increase of a programme for alternative fuels infrastructure. The scheme consisted of five separate aid schemes with an annual budget of NOK 500 million run by Enova in the period from 2017 to 2022 directed at supporting electric charging infrastructure for vehicles, shoreside electricity supply for ships, hydrogen and biofuels refuelling infrastructure for vehiceles as well as LNG refuelling infrastructure for ships. The granting of the aid would be based on a bidding process or (if not suitable), the methodology for identifying eligible costs as set out in the EEAG (funding gap analysis and maxium aid intensities). The Authority, similarly as in Decision 150/15/COL, noted that the aid addressed a market failure evident in the lack of refuelling/recharging infrastructure—in particular in more rural areas of Norway.

146

See paras 105–08 of the decision.

23 State Aid Enforcement in the Energy Community MARCO BOTTA* AND ROZETA KAROVA*

I. INTRODUCTION

B

Y SIGNING THE Energy Community Treaty (EnCT) in October 2005,1 the Energy Community (EnC) Contracting Parties committed to implement the EU acquis in the field of energy in order to liberalise their energy markets. The initial signatories of the EnCT are potential and candidate EU Member States, which under the EnCT accepted to be bound by the relevant EU acquis even before the beginning of their formal EU accession process.2 Since subsidies in the energy sector might distort competition and jeopardise the process of market liberalisation, the EnCT requires the Contracting Parties to introduce a system of State aid control that mirrors the EU State aid rules and to apply it to the aid granted in their energy sectors. This chapter analyses the enforcement of State aid rules in the Energy Community. In particular, it first outlines the legal obligation of the Contracting Parties to introduce a State aid control regime under the Energy Community Treaty and the Association Agreements concluded with the EU. The focus then turns to the main types of subsidy granted by the Contracting Parties in the energy sectors. Finally, the system of enforcement of State aid rules in the Energy Community is analysed.

* The views expressed in the chapter are personal and they do not necessarily reflect the official position of the Energy Community Secretariat. 1 Treaty Establishing the Energy Community [2006] OJ L98/18. 2 The Energy Community Treaty (EnCT) was concluded between the EU and a number of Contracting Parties. At the moment, the Energy Community includes seven Contracting Parties: Bosnia and Herzegovina, Kosovo, Serbia, Macedonia, Montenegro, Albania, Serbia, Moldova and Ukraine. Before their accession to the EU, Croatia, Romania and Bulgaria were also Contracting Parties. Armenia, Norway and Turkey have Observer status, and Georgia became a fully-fledged member as of 1 July 2017. Further information concerning the members of the Energy Community is available at: www.energycommunity.org/aboutus/whoweare.html.

556 Marco Botta and Rozeta Karova II. THE LEGAL OBLIGATION OF THE ENERGY COMMUNITY CONTRACTING PARTIES TO INTRODUCE A STATE AID CONTROL REGIME

In accordance with the Stabilisation and Association Agreements (SAAs)3 concluded by the South-East European (SEE) countries with the EU, the SEE countries have a legal obligation to adopt a State aid law. The establishment of a system of State aid control is also one of the main EU accession criteria emphasised by the EU Commission in its Progress Reports published every year for each EU candidate country.4 Each SAA includes a general prohibition on any State aid that could distort the competition in the market by favouring certain undertakings.5 In addition, the SAAs include an obligation for the SEE countries to establish an ‘operationally independent authority’ ‘entrusted with the powers necessary for the full application’ of the State aid prohibition.6 A general prohibition vis-a-vis incompatible State aid was also included in the Partnership and Cooperation Agreements concluded by the EU in the 1990s with Moldova and Ukraine.7 The Association Agreements concluded by the EU with Moldova and Ukraine in June 2014 include a ‘stronger’ obligation to introduce a

3 The SAA represents the first step of the integration of the South-East European countries into the EU. The EU has concluded SAA with all the SEE countries. Finally, Montenegro, Serbia and Macedonia are candidate countries, while Bosnia and Herzegovina and Kosovo are potential candidates. For further information concerning the enlargement process see: www.ec.europa.eu/enlargement/countries/ index_en.htm. 4 Every year, DG Enlargement publishes a Progress Report for each SEE country, indicating the improvements made in relation to the adoption and enforcement of the EU acquis, as well as the lack of implementation of certain aspects of the acquis. The Progress Reports thus represent a benchmark whereby the EU Commission assesses every year the readiness of SEE to join the EU. The text of the Progress Reports for each SEE country is available at: www.ec.europa.eu/enlargement/press_corner/ key-documents/reports_oct_2011_en.htm. 5 The following provisions of the SAAs require the adoption of a State aid law:

SAA between the European Communities and their Member States and the Former Yugoslav Republic of Macedonia [2004] L84/13, Art 69(1)(c). SAA between the European Communities and their Member States and the Republic of Montenegro [2010] L108/3, Art 73(1)(iii). SAA between the European Communities and their Member States and the Republic of Albania [2009] L107/166, Art 71(1)(iii). SAA between the European Communities and their Member States and the Republic of Serbia [2013] L278/16, Art 73(1)(c). SAA between the European Communities and their Member States and the Republic of Bosnia and Herzegovina [2015] OJ L164/2, Art 71. 6 Only the SAA concluded with Macedonia does not explicitly mention in Art 69 that the general prohibition against State aids will have to be enforced by an independent State aid authority. 7 The Partnership and Cooperation Agreements concluded by the EU with Ukraine and Moldova included the following provision: ‘The Parties shall refrain from granting State aids favouring certain undertakings or the production of goods other than primary products as defined in the GATT, or the provision of services, which distort or threaten to distort competition insofar as they affect trade between the Community and the Republic of Moldova’. Partnership and Cooperation Agreement between the European Communities and their Member States and the Republic of Moldova [2008] OJ L181/3, Art 48(2)(2); Partnership and Cooperation Agreement between the European Communities and their Member States and Ukraine [1998] OJ L49/3, Art 49(2)(2).

State Aid Enforcement in the Energy Community 557 system of State aid control.8 Similar to the SAAs, in fact, the Association Agreements require Moldova and Ukraine to establish an ‘operationally independent authority’, in charge of enforcing State aid rules.9 The latter authority will enforce such rules in light of the interpretation of the Court of Justice of the European Union (CJEU) under Article 107 TFEU.10 The incentive to adopt a State aid law for these two groups of countries is thus related to the EU external conditionality: while for the SEE countries such conditionality is linked to the EU accession prospect, for Moldova and Ukraine the conditionality is related to the possibility to get access to the EU internal market under the conditions of the Association Agreements.11 Besides the SAAs and the Association Agreements, a further obligation to introduce a system of State aid control in the energy sector derives from the Energy Community Treaty. Similar to the Treaty of Rome, the EnCT contains a general prohibition on ‘any aid which distorts or threatens to distort competition among the undertakings active in the energy sector’.12 The Energy Community founding fathers were well aware that the EnCT should include a system of State aid control, in order to avoid a subsidies race between the Contracting Parties while establishing a regional energy market that should later be included in the EU internal market. In particular, Annex III of the EnC Treaty replicates the text of Article 107 TFEU.

A. Decentralised Enforcement System In spite of the influence of the EU State aid rules in the EnCT, some major differences exist between the two regimes, related to the institutional set-up of the Energy Community, as well as its aim to achieve strictly sectoral integration between its Parties. For instance, the Energy Community law covers mostly selected pieces of the relevant EU acquis. Furthermore, the horizontal policies such as competition and State aid are adopted as Energy Community law by including only substantive rules in the Energy Community Treaty. Procedural competition and State aid rules are not part of the obligations undertaken by the Contracting Parties. Even more important is the difference related to the enforcement system: the EnCT introduced a ‘decentralised’ system of enforcement of State aid rules whereby each Contracting Party should adopt a national State aid law and establish a national State aid enforcement authority. Unlike in the EU, only the Ministerial Council in the Energy Community has a decision-making power.13 The power of the Secretariat, 8 Association Agreement between the European Union and the European Atomic Energy Community and their Member States, on the one part, and the Republic of Moldova, on the other part [2014] OJ L260/4; Association Agreement between the European Union and its Member States, on the one part, and Ukraine, on the other part [2014] OJ L161/3. 9 EU–Moldova Association Agreement, Art 341; EU–Ukraine Association Agreement, Art 267(1). 10 EU–Moldova Association Agreement (n 9) Art 340. 11 M Botta, ‘State aid Control in South-East Europe, the Endless Transition’ (2013) 11(1) European State Aid Law Quarterly 85. 12 Art 18(c) EnCT. 13 Even though the EnCT gives the possibility for the ECRB and PHLG to take decisions, if empowered to do so by the Ministerial Council, that has never happened so far and the activities of these two institutions are mostly advisory.

558 Marco Botta and Rozeta Karova which is the only independent and centrally coordinating institution of the Energy Community, to take decisions is explicitly excluded by Article 67(d) of the EnCT.14 Therefore, it is up to each Contracting Party to establish and entrust a national institution with State aid enforcement—ie, with review of new aid schemes notified by national or local granting authorities and deciding on blocking or approval subject to the conditions of the new subsidy schemes. Every State aid enforcement authority would assess the notified aid scheme in accordance with the national provision corresponding to Article 107 TFEU and in the light of the CJEU case law. On the other hand, the Energy Community Secretariat supervises the compliance with the EnCT by the Contracting Parties,15 but it does not exercise an ex ante control of the Contracting Parties’ national aid schemes similar to that performed by the EU Commission vis-a-vis the aid schemes notified by the EU Member States.

B. National State Aid Rules In recent years, all Contracting Parties have adopted State aid law in accordance with their international legal obligations.16 The last countries that adopted such legislation were Moldova,17 Kosovo18 and Bosnia and Herzegovina.19 Ukraine adopted national State aid law in July 2014, but its entry into force took place only in August 2017.20 Most of these countries have also adopted national procedural rules applicable in State aid cases and have transposed the secondary legislation including the Commission’s Guidelines and Regulations.21 All Contracting Parties have established enforcement authorities in accordance with two institutional set-ups:22 either by establishing an inter-governmental State aid Commission assisted by an administrative State aid Office (eg, Serbia,23 14 According to Art 70 EnCT, the Director and the staff are not allowed to ‘seek or receive instructions from any Party to the EnCT and shall act impartially and promote the interests of the Energy Community’. 15 Art 67(b) EnCT. 16 Albania: Law No 9374, adopted by the Albania Parliamentary Assembly on 21 April 2005. Law 9374/2005 was amended by Law 10183, adopted on 29 October 2009; Macedonia: Law on State aid Control, Official Gazette 145/2010; Montenegro: Law of Control of State aid Support and Aid, adopted by the Montenegro Parliament on 9 November 2009; Serbia: Law on State aid Control, Official Gazette 51/09. State aid Act of the Republic of Croatia, entered into force on 2 April 2003, Official Gazette 47/03. 17 Law on State aid of the Republic of Moldova, Law No 139 (adopted on 15 June 2012, entered into force August 2013). 18 Law 2011/04-L-024, Law on State aid adopted by the Assembly of the Republic of Kosovo on 29 July 2011 (entered into force on 1 January 2012). 19 Bosnia and Herzegovina adopted the State aid law in February 2012. Energy Community Secretariat, Annual Report on the Implementation of the Acquis under the Treaty Establishing the Energy Community (2012) 135. 20 Law of Ukraine of State aid to Business Entities, No 34, 1 July 2014, No 1555-VII. 21 The newest Contracting Party, Georgia, also has competition and State aid rules at national level governed by the Law on competition, No 6148-IS (8 May 2012). 22 L Biegunski, ‘Forms of State aid Authorities in Associated Countries of Central and Eastern Europe’. (2012) 11 (3) European State Aid Law Quarterly 567. 23 The State aid Commission includes five members, representatives of the Ministry of Finance (chairman), Competition Authority (deputy chairman), Ministry of the Economy and Regional Development, Ministry of Infrastructure, Ministry of Environment. Serbia State aid law (n 16) Art 6.

State Aid Enforcement in the Energy Community 559 Montenegro,24 Albania25 and Kosovo),26 or by empowering the National Competition Authority (NCA) to review the notified aid schemes (eg, Macedonia,27 Moldova,28 Ukraine29 and Georgia).30 To sum up, over the last decade, most of the Contracting Parties of the Energy Community have introduced systems of State aid control in accordance with the legal obligation stemming from the SAAs and the Association Agreements, as well as from the obligation deriving from the Energy Community Treaty. Although the EnCT introduced a ‘sectoral’ system of State aid control (ie, covering only the subsidies granted to energy undertakings, or affecting the energy sectors), the latter are of great relevance for the Contracting Parties. As the following section will demonstrate, due to the high degree of subsidisation of the energy sectors, the State aid prohibition included in Article 18 EnCT has important consequences for these countries.

III. ENERGY SUBSIDIES GRANTED BY THE ENERGY COMMUNITY CONTRACTING PARTIES AND NATIONAL ENFORCEMENT OF STATE AID RULES

According to a study published in 2011 by the United Nations Development Programme (UNDP), countries of the Western Balkans heavily subsidise their energy sectors; the average level of subsidy is estimated at around 9 per cent of the GDP of these countries.31 In spite of the relevance of the energy subsidies in the region

24 The State aid Commission includes nine members, representatives of different ministries, as well as one representative of the municipalities and one representative of the association of the employers. Montenegro State aid law (n 16) Art 11. 25 The State aid Commission includes five members appointed by the Council of Ministers for a period of four years. The Commission is chaired by the Minister of Economic Affairs. Albania State aid law (n 16) Art 16. 26 The State aid Commission includes five members: Minister of Finance (chairman), Minister for European Integration, Minister of Trade, representative of the civil society, and chairman of the municipalities association. Under Art 7 of the State aid law, the State aid Office should be placed within the Competition Authority. However, due to the fact that the State aid law entered into force only on 1 January 2012, the State aid Office has not been established yet. Kosovo State aid law (n 18) Arts 7–8. 27 The enforcement of the 2003 Macedonian State aid law was assigned to an intergovernmental State aid Commission supported by the Ministry of Finance. This institutional setting, which currently characterises the majority of the SEE countries, proved not to be successful in ensuring an autonomous enforcement of the legislation. Consequently, following the example of Croatia, in 2006 the functions of State aid control were transferred to the Комисијата за заштита на конкуренцијата (Macedonian Competition Commission, KZK). Law on Amending and Supplementing the Law on State aid (adopted on 6 June 2006). Published on the Official Gazette 70/06. An English translation of the text of the legislation is available at: www.kzk.gov.mk/images/Vestiimages/460/DOWNLOAD.PDF. 28 The new Moldova State aid law grants the task of State aid enforcement authority to the NCA. Moldova State aid law (n 17) Art 8. 29 Under the 2014 State aid law, the Anti-Monopoly Committee of Ukraine is tasked to enforce the State aid law. It shall however start actively enforcing the provisions of the Law after its entry into force, in July 2017. 30 Article 4 of the Law on Competition of 2012 designates the Competition Agency as an authorised body to ensure compliance with the provisions of the law. 31 UNDP, Fossil Fuel Subsidies in the Western Balkans (2011) 18. The text of the report is available at: www.undp.org.tr/publicationsDocuments/Fossil_Fuel_Subsidies_F.pdf.

560 Marco Botta and Rozeta Karova and though most of the Contracting Parties have already adopted a State aid law, so far the State aid enforcement authorities of these countries have played a marginal role in reviewing the subsidies granted. In particular, only the State aid authorities of a few Contracting Parties have analysed a limited number of subsidy schemes granted in the energy sector.32 A study carried out in 2011 by the law firm Hunton & Williams on behalf of the EnC Secretariat has also pointed out that most of the aid schemes granted in the energy sector by the Contracting Parties have not been notified to the State aid enforcement authorities. The report established a ‘State aid inventory’ of the energy schemes present in the Contracting Parties.33 The report underlined the existence of a great discrepancy between the large number of energy aid schemes present in these countries and the limited number of decisions adopted by the State aid enforcement authorities in this field. Finally, even when notified, the State aid enforcement authorities have generally approved without conditions the notified aid schemes; as recognised by the EnC Secretariat, ‘no Contracting Party has ever taken a decision to prohibit the granting of aid to an energy undertaking or ordered its recovery in the energy sector’.34 The types of State aid granted in the energy sectors by the Contracting Parties are quite similar to the energy subsidies granted by the EU Member States.

A. Long-Term Agreements During the past decade, long-term Power Purchase Agreements (PPAs) were concluded in several Contracting Parties. In Albania KESH, the Albanian State-owned electricity wholesale supplier, signed PPAs with a number of hydropower plants.35 Under Law 9470/2006 on the privatisation of local hydropower plants, KESH was legally required to enter into power purchasing agreements with the privatised hydropower plants according to a fixed price determined by the Albanian Energy Regulatory Authority. The measure aimed at incentivising the purchase of hydropower plants by foreign investors during their privatisation process. These PPAs were similar to the Hungarian36 and Polish cases investigated by the EU Commission.37

32 See 2015 EnC Secretariat’s annual report. Energy Community Secretariat, 2015 Annual Report on the Implementation of the Acquis under the Treaty Establishing the Energy Community (2015). Published on 8 October 2015. The annual Implementation Reports are available at: www.energy-community. org/implementation/reports.html. 33 In April 2011, the law firms Hunton & Williams, and Eisenberg Herzog, published a report concerning the enforcement of State aid rules in the electricity sector by the Contracting Parties. The study was commissioned by the EnC Secretariat. The text of the study is available at: www.energy-community.org/ pls/portal/docs/948179.PDF. 34 Energy Community Secretariat, Annual Report on the Implementation of the Acquis under the Treaty Establishing the Energy Community (2012) 130. The annual Implementation Reports are available at: www.energy-community.org/implementation/reports.html. 35 Hunton & Williams (n 33) 75. 36 Hungarian Stranded Costs (Case SA.17365) Commission Decision 2009/609/EC [2009] OJ L225/53. 37 Stranded Costs Compesations in Poland (SA. 18989) Commission Decision 2009/287/EC [2009] OJ 83/1.

State Aid Enforcement in the Energy Community 561 Nevertheless, these long-term agreements have never been notified to the Albanian State Aid Commission.38 The Contracting Parties have also included PPAs in the privatisation agreements of the distribution electricity companies. A good example in this regard concerns the contract of privatisation of KEDS (ie, the Kosovo distribution and supply electricity company).39 After having been separated from KEK, which remained a State-owned generation company, KEDS has been sold to foreign investors. Under the terms of the privatisation agreement, KEDS will continue to benefit from the PPAs previously concluded with KEK: KEDS will be able to purchase electricity from KEK at a guaranteed price over a long period of time.40 Similar long-term arrangements have also been included in the privatisation of the Macedonian distribution and supply company, sold to the Austrian EVN, as well as the Albanian privatisation when CEZ bought the Albanian distribution and supply. These examples show that the PPAs are not only concluded in the context of the privatisation of State-owned generators, as in Poland and Hungary, but also in the context of the privatisation of supply and distribution companies, as well as in the tenders to build new power plants. The examples of the PPAs concluded in Kosovo and Albania show that the Contracting Parties are willing to rely on these long-term arrangements in order to attract foreign investors. Nevertheless, in none of these examples does the State aid enforcement authority seem to have been notified of these agreements, in order to assess their compatibility with the State aid rules.

B. Price Regulation Price regulation might involve State aid with the potential to affect the trading of electricity between the Parties of the Energy Community. If an undertaking is supplied with electricity at low regulated prices (ie, below market prices), an advantage would be conferred on that recipient of the electricity as compared with other market participants in the Energy Community. If the public supplier and/or generator are State-owned companies (as is the case in most Contracting Parties), the aid is provided from State resources. The loss for supplying at low regulated prices incurred by the supplier and/or the State-owned generator may further need to be covered from the State budget in the form of additional State aid to the electricity supplier and/or generator in question. In the EU, the Commission has opened a number of investigations into whether the low regulated prices for large undertakings constitute State aid and, if so, whether they could be justified on the basis of, inter alia, Article 3(4) of Directive 2003/54/EC. The latter provision requires that ‘any financial compensation or exclusive right granted to the company in charge of 38

Hunton & Williams (n 33) 75. See: mzhe-ks.net. 40 Republic of Kosovo, Implementation Agreement. Kosovo Electricity Distribution and Supply Privatisation. Published in October 2012, Art 4.3. The text of the agreement is available at: mzhe-ks.net. 39

562 Marco Botta and Rozeta Karova providing the Public Service Obligation has to be assigned in a non-discriminatory and transparent manner’. The French case41 and the Spanish case42 are particularly relevant and instructive in this respect, as they represent examples of what types of conflict with State aid rules might be faced when maintaining in force low regulated electricity prices for all customers’ categories. Most of the Contracting Parties provide either regulated tariffs or subsidies for vulnerable customers. For instance, in accordance with its Social Action Plan, in 2010 the Macedonian government adopted a regime of electricity subsidies for vulnerable households. The latter could receive a partial reimbursement of the paid energy bills from the State’s budget.43 On the other hand, other Contracting Parties (eg, Bosnia and Herzegovina,44 Montenegro45 and Serbia)46 opted for a system of regulated tariffs for all households. Regulated tariffs may jeopardise the liberalisation of the electricity industry if they are not cost-reflective and if they involve most of the customers. From the point of view of State aid rules, regulated tariffs for households would not qualify as aids under Article 107(1) TFEU, since they would not affect undertakings. On the other hand, regulated electricity tariffs for industrial customers might be considered State aids when funded by State resources. A study conducted in 2012 by the EnC Secretariat pointed out that in most of the Contracting Parties, industrial customers still benefit from regulated energy tariffs,47 which have neither been notified nor assessed by the national State aid authorities.

C. Stimulating Renewables The Contracting Parties have also adopted a number of subsidy schemes to incentivise the production of renewable energy and to support projects for incentivising energy efficiency. In particular, a number of Contracting Parties have introduced a legal obligation for the electricity wholesale supplier to conclude long-term PPAs with electricity generators relying on renewable energy sources at fixed prices (eg, Kosovo,48 Macedonia,49 Montenegro,50 Serbia,51 and Bosnia and Herzegovina).52 In addition, some Contracting Parties have also provided tax reliefs (eg, Albania,53

41

Regulated electricity tariffs in France, European Commission, State Aid C 17/07 (ex NN 19/07). Regulated electricity tariffs in Spain, European Commission, Decision: State Aid C 3/07 (ex NN 66/06). 43 Hunton & Williams (n 33) 88. 44 ibid, 79. 45 ibid, 93. 46 ibid, 100. 47 Energy Community Secretariat, Regulated Energy Prices in the Energy Community—State of Play and Recommendations for Reform (2012). The text of the report is available at: www.energycommunity.org/pls/portal/docs/1660178.PDF. See also R Karova, ‘Public Service Obligation Provision as a Regulatory Tool: End-users’ Electricity Price Regulation in the Energy Community as a Study’ (2014) 2 European Networks Law and Regulation Quarterly 2. 48 Hunton & Williams (n 33) 104. 49 ibid, 89. 50 ibid, 93. 51 ibid, 101. 52 ibid. 53 For instance, Law 8987/2002 provided for a customs duties exemption for imported goods necessary to build new generators relying on renewable energy sources. 42

State Aid Enforcement in the Energy Community 563 Serbia),54 direct grants (Montenegro)55 and introduced ‘feed-in tariffs’ (FITs)56 to support green energy production. In this regard, the EnC Secretariat has recently published Policy Guidelines, endorsing the EU Commission’s 2014 Energy and Environmental Guidelines.57 This means that Energy and Environmental Guidelines are to be followed by the State aid authorities of the Contracting Parties, and the Secretariat will make them a point of reference in its review of the Contracting Parties’ assessment, in order to ensure their uniform and homogeneous application in the entire Energy Community. However, to date, only the Moldovan enforcement authority reviewed a State aid notification from the Energy Efficiency Fund aimed at introducing a State aid scheme to support renewable energy and stimulate energy efficiency.58 The Moldovan Competition Council approved the aid as the aid intensity was below the 30 per cent threshold and the aid did not aim at financing operational activities.59 However, the Competition Council failed to examine the limitations concerning the eligible costs of the aid.60

D. Operating Aid to State-Owned Energy Undertakings The last issue that characterises the Contracting Parties concerns the widespread ‘operating’ aids granted to State-owned energy companies that face difficulties in profitably remaining in the market. A number of Contracting Parties have covered the losses cumulated by their energy companies through loans and financial guarantees (eg, Serbia),61 measures of debts cancellation (eg, Albania)62 and tax exemptions (eg, Albania).63 A good example of this tendency is Kosovo, where the formerly

54

Hunton & Williams (n 33) 101. In May 2011, Montenegro established an Energy Efficiency Fund to support energy efficiency projects. See: www.energetska-efikasnost.me. 56 The key principle of FiTs is to offer guaranteed prices to renewable energy generators for a certain period of time. FiTs may either include a ‘minimum fixed price’ paid to the generators independently by the electricity price fluctuations in the wholesale market, or a ‘premium’ price which is added onto the electricity market price. While ‘minimum’ FiTs aim at reducing the uncertainty faced by RES generators which would not get a return of their investment in the construction of a new RES power plant if the electricity prices declined, ‘premium’ FiTs directly compensate the RES generators for the additional operating and investments costs faced in comparison to fossil fuel generators. T Couture and Y Gagnon, ‘An Analysis of Feed-in Tariff Remuneration Models: Implications for Renewable Energy Investment’ (2010) 38 Energy Policy 955. 57 European Commission, ‘Guidelines on State aid for environmental protection and energy 2014–2020’ [2014] OJ C200/1. 58 EnC Implementation Report 2015 (n 32) 148. 59 ibid. 60 ibid. 61 In recent years, Serbia has provided extensive loans and guarantees to Elektroprivreda Srbije, the main State-owned generator active in the country. A table summarising the loans and financial guarantees granted by Serbian government in favour of Elektroprivreda Srbije is available at Hunton & Williams (n 33) 97–99. 62 Over the last few decades, the Albanian governments adopted a number of decisions to cancel the debts cumulated by KESH based on unpaid taxes and contributions. Hunton & Williams (n 33) 76. 63 For instance, in July 2007 the Albanian State aid Commission approved an aid scheme deferring the VAT payment for imported equipment by KESH to be used in electrical power plants. See Decision of the Albanian State aid Commission (n 16) (adopted on 16 July 2007). 55

564 Marco Botta and Rozeta Karova State-owned electricity company KEK cumulated a debt of €157 million vis-a-vis the Kosovo government over the period from 2005 to 2009.64 Due to the low electricity bills collection rate in the country (ie, which varied between 51 per cent and 64 per cent over this period of time), the government of Kosovo had to directly intervene via loans and financial guarantees to ensure that KEK had sufficient liquidity to purchase electricity.65 The cumulated debt was periodically reduced by the Kosovo government via measures of debt cancellation. Such measures could be considered as operational State aid schemes, necessary to allow KEK to stay in the market. In view of their potential anticompetitive effect, it is well known that operating State aid should be allowed only for a limited period of time, in the context of a restructuring programme of the beneficiary undertaking. However, in some Contracting Parties, operating State aids to State-owned energy companies have been granted over a long period of time; such aid schemes have generally not been notified to the State aid enforcement authorities at all.

E. National State Aid Law Enforcement in the Energy Sector The main issue of concern of the Contracting Parties in relation to the enforcement of State aid rules vis-a-vis long-term PPAs, regulated retail electricity prices and green energy subsidies, is the lack of notification of the new aid schemes to the State aid enforcement authority. Even when notified, however, the State aid enforcement authorities have often incorrectly applied the correctly transposed EU State aid rules when analysing aid for SGEI in the energy sectors. There is no example of an ex officio investigation and assessment of State aid in the energy sector in any Contracting Party so far. Good examples of this are two decisions adopted in December 2011 by the Macedonian Competition Commission (KZK), acting as the State aid enforcement authority in the country.66 KZK approved two aid schemes provided by the Stateowned companies MEPSO (the Macedonian Transmission System Operator, TSO) and ELEM (the Macedonian power generation company) to Toplifikacija (ie, the private company providing district heating in the city of Skopje). MEPSO and ELEM granted a loan to Toplifikacija at an interest rate lower than the ordinary interest rate applied by commercial banks. The loan satisfied the conditions to be considered as a State aid under Article 107(1) TFEU: the loan implied disbursement of State resources by State-owned companies, it created a selective advantage for Toplifikacija, and it breached the market investor principle (ie, the interest rate was lower than the commercial interest rate available on the market). The Macedonian NCA, however, approved the two aid schemes by arguing that ‘companies performing activities of

64

UNDP, Fossil Fuel Subsidies (n 31) 41. ibid. Decision Nos 10-84 and 10-86 adopted by the Macedonian NCA on 21 December 2011. The text of the two decisions of the Macedonian NCA is available at: www.kzk.gov.mk/mak/zapis1. asp?id=1001&kategorija=11. 65 66

State Aid Enforcement in the Energy Community 565 public interest may receive State aid to cover any arising costs, including a reasonable profit from performing activities of public interest.67 KZK thus argued that the aid was granted in view of the ‘public interest’ function performed by Toplifikacija (ie, the provision of district heating in Skopje), and the aid did not ‘overcompensate’ Toplifikacija for the public service performed. Prohibition of overcompensation is one of the four Altmark conditions; nevertheless, the Macedonian NCA omitted to assess whether the other three conditions were complied with as well. In particular, in its decision, the KZK did not assess whether Toplifikacija had been entrusted by an act of the State to perform the SGEI; whether the amount of compensation had been defined in advance when Toplifikacija had been selected as the SGEI provider; and whether Toplifikacija had been selected through a public tender. In this case, KZK eventually combined the conditions for assessing if a subsidy could be considered State aid with the conditions for verifying compatibility of the aid. A similar line of reasoning was followed by the Serbian State aid authority in the TPP Kolubara case. The latter is the largest coal mine in Serbia; the coal produced by Kolubara was mostly used by EPS’s thermo power plants to produce electricity, thus representing an important energy source for the country. The State aid was granted via State guarantees and the transfer of public property in order to increase the mine efficiency. In April 2015, the Serbian State aid Commission approved the aid considering that EPS was in charge of ensuring a safe and regular power supply to tariff customers, but on the one hand it failed to assess whether four of the measures provided by the Republic of Serbia were compatible with the Energy Community internal market, and on the other hand, the single compatibility assessment undertaken by the Commission for State Aid Control was not in line with the State aid acquis. The Toplifikacija and the Kolubara decisions show the tendency by the State aid enforcement authorities of the Contracting Parties to approve any aid scheme in the energy field which aims at satisfying general ‘public interest’ objectives, without analysing in detail whether the four Altmark conditions are cumulatively satisfied. An issue closely connected to the ‘misapplication’ of the Altmark conditions is the tendency by the State aid enforcement authorities of the Contracting Parties to approve any aid scheme granted to the State-owned companies operating in the energy industry. In June 2011, the KZK approved a State guarantee in favour of MEPSO at more favourable conditions than ordinary market conditions.68 MEPSO needed a financial guarantee in order to receive a loan from international financial institutions (eg, the World Bank and the European Bank for Reconstruction and Development); the loan would finance the construction of new interconnectors with Serbia and Bulgaria. In its decision, the KZK recognised that the guarantee created an advantage for MEPSO. However, since MEPSO was the only energy undertaking in the country licensed for electricity transmission, the Macedonian NCA concluded that the subsidy did not distort the competition in the market (ie, the subsidy did not qualify as State aid under Article 107(1) TFEU).69 67

ibid. Decision No 10-64 adopted by the Macedonian NCA on 28 June 2011. The text of the decision is available at: www.kzk.gov.mk/mak/zapis1.asp?id=977&kategorija=11. 69 ibid. 68

566 Marco Botta and Rozeta Karova A similar case was decided at the beginning of 2013 by the Montenegrin State aid Commission, which authorised a loan granted by the central government to CGES (ie, the Montenegrin TSO), in order to build a new under-sea electricity transmission line connecting Montenegro with Italy.70 Similar to the KZK decision in the MEPSO case, the Montenegrin State aid Commission authorised the aid without a serious analysis of the effects on competition caused by the aid.71 The MEPSO and CGES decisions are not isolated cases of this type of reasoning: in September 2012, the Albanian State aid Commission authorised a State guarantee to allow KESH (ie, the State-owned wholesale supplier) to import electricity from abroad.72 In its decision, the Albanian State aid Commission concluded that the State guarantee would not distort competition since KESH was the only licensed entity in the country which had the obligation to purchase electricity either from generators present in the country, or to import it from abroad.73 The Albanian State aid Commission concluded that the State guarantee was not a State aid since KESH was a State-owned company and it acted as a monopolist in the wholesale electricity supply market, rather than exempting the aid under the national provision equivalent to Article 107(3)(c)TFEU.

IV. THE ENFORCEMENT OF STATE AID POLICY BY THE ENERGY COMMUNITY SECRETARIAT

A. Infringement Actions for Lack of Transposition and Implementation of State Aid Rules The EnC lacks a central authority for the enforcement of the competition and State aid rules. Due to its lack of decision-making powers, the Secretariat cannot enforce directly Article 18(1)(c) of the Energy Community Treaty by assessing nationally designed aid schemes in the energy sector. On the other hand, the Secretariat can initiate infringement proceedings against the Contracting Parties for the lack or improper implementation of competition and State aid rules in the energy sector. The Secretariat monitors the implementation of the Treaty obligations by the Contracting Parties and initiates dispute settlement procedures governed by Articles 90–93 of the EnCT and the Dispute Settlement Rules (DSR).74 The procedure has been modelled upon the EU infringement procedure; it covers only infringement actions against the Parties to the Treaty and not disputes between different countries

70 Energy Community Secretariat, Annual Report on the Implementation of the Acquis under the Treaty Establishing the Energy Community (2013) 84. The text of the report is available at: www.energycommunity.org/portal/page/portal/ENC_HOME/AREAS_OF_WORK/Implementation/Report. 71 ibid, 84. 72 Decision No 42 of the Albanian State aid Commission (adopted on 11 September 2012). The text of the decisions of the Albanian State aid Commission are available at: www.mete.gov.al. 73 ibid. 74 Ministerial Council, Procedural Act on the Rules of Procedure for Dispute Settlement under the Treaty, No 2008/01/MC-EnC, 27 June 2008. Revised by the Ministerial Council Decision on 16 October 2015.

State Aid Enforcement in the Energy Community 567 or annulment and preliminary ruling proceedings. The procedure also does not cover disputes between private parties.75 The dispute settlement is initiated by the Secretariat against Parties to the Treaty for either a failure to comply with Treaty obligations or to implement a Decision addressed to them within the required period of time.76 The DSR define very broadly the scope of the acts of failure: they cover any of the Parties’ measures (actions or omissions)77 and ‘any measure by the public authorities of the Party (central, regional or local as well as legislative, administrative or judicative)’. This expression thus covers acts of competition and State aid authorities, as well as public undertakings and undertakings to which special or exclusive rights have been granted.78 The Secretariat may initiate a dispute settlement procedure either upon a complaint (ie, by a private body, another Party or by the European Community Regulatory Board)79 or by its own initiative.80 The right to submit a complaint to the Secretariat is also broadly defined; ‘all natural and legal persons as well as companies, firms or associations having no legal personality’ can submit a complaint.81 Such broad legal standing is important in State aid cases, since the Secretariat often learns about un-notified aid measures or cases investigated by the national State aid authorities through the complaint submitted either by competitors, or any legal or natural person interested in the case.82 The monitoring activities carried out by the Secretariat have resulted in several infringement actions being initiated so far. Three cases have been initiated by the Secretariat on the lack of transposition and implementation of State aid rules: Case ECS-1/10 against Bosnia and Herzegovina opened in September 2010;83 Case ECS-7/11 against Kosovo opened in February 2011;84 and Case ECS-8/14 against Ukraine opened in April 2014.85 Bosnia and Herzegovina as well as Kosovo have been under the obligation to enforce State aid rules since the entry into force of the EnC Treaty, whereas the obligation for Ukraine is stipulated in Article 1(2) of its Accession Protocol to the Energy Community. By not adopting State aid legislation

75

Art 4 DSR. Art 90(1) EnCT. 77 Art 2(1) DSR. 78 ibid, Art 2(2). 79 The European Community Regulatory Board (ECRB) includes the representatives of the energy regulatory authorities of the EnC Contracting Parties. 80 Art 11 DSR. 81 ibid, Art 19(2). 82 Such notifications, for instance, awake the interests of the environmental non-governmental organisations which, even though not directly affected by the aid measures granted for construction or other support to thermal power plants, have interests in the cases. 83 Case ECS 1/2010 Dispute settlement procedure initiated against Bosnia and Herzegovina for the failure to adopt State aid legislation. Opening Letter sent on 22 September. See: www.energy-community. org/portal/page/portal/ENC_HOME/AREAS_OF_WORK/Dispute_Settlement/2010. 84 Case ECS 7/2011 Dispute settlement procedure initiated against Kosovo/UNMIK1 for the failure to adopt State aid legislation. Opening Letter sent on 9 February 2011. See: www.energy-community.org/ portal/page/portal/ENC_HOME/AREAS_OF_WORK/Dispute_Settlement/2011/07_11. 85 Case ECS 8/14 Dispute settlement procedure initiated against Ukraine for the failure to adopt State aid legislation. Opening Letter sent on 22 April 2014. See: www.energy-community.org/portal/page/ portal/ENC_HOME/AREAS_OF_WORK/Dispute_Settlement/2014/8. 76

568 Marco Botta and Rozeta Karova within the time limit required, these three Contracting Parties have clearly breached their obligations under Article 18 EnCT. Following the Opening Letters sent by the ECS, these three Contracting Parties adopted State aid laws. However, besides mere transposition, the closure of the cases depends on their implementation in practice— ie, establishing enforcement authorities and making them operational. While the State aid legislation in Kosovo86 and Bosnia and Herzegovina have established national enforcement authorities, these institutions have not started applying the State aid rules to the energy sectors. In fact, the 2017 Law on State aid of Kosovo has changed the State aid enforcement system,87 but neither the previous nor the new State aid Commission has ever taken a decision applying State aid to the energy sector. In Ukraine, on the other hand, even though the Antimonopoly Committee has been entrusted with State aid law enforcement, the entry into force of the State aid law has been postponed to 2017.88 Therefore, the Energy Community Secretariat (ECS) has not closed these cases upon the adoption of primary legislation.89 By doing so, the ECS aimed at ensuring not only the mere legislative transposition of the EnC acquis into the legal systems of the Contracting Parties, but also its proper and effective implementation and enforcement.90 Besides the three cases concerning lack of transposition of State aid rules, the Secretariat has also initiated infringement actions for the lack of proper implementation of those rules in the energy sector. On 14 July 2016, the Secretariat initiated Case ECS-11/14 against Serbia for its failure to comply with its obligations under Articles 18 and 19 of the Energy Community Treaty in the Kolubara case. As mentioned in the previous section, the Serbian Commission for State Aid Control incorrectly assessed the compatibility of State aid granted to Elektroprivreda Srbije (EPS) for the Kolubara power plant project. The infringement of EnC acquis concerns four State guarantees issued by the Serbian State in relation to loans received by EPS from international financial institutions for the Kolubara B power plant project, as well as the transfer of property and land from the Republic of Serbia to EPS for the same project. In its preliminary assessment, the Secretariat found that these measures constitute State aid and that Serbia infringed the EnC Treaty. First, the Commission for State Aid Control failed to assess whether four of the measures provided by the Republic of Serbia were compatible with the Energy Community internal market.

86 The new Law on State aid was adopted in December 2016 and entered into force in January 2017, Law No 05/L-100. 87 The State Aid Office was transferred from the Kosovo Competition Authority to the Ministry of Finance and was renamed as the State Aid Department. When the term of the current State Aid Commission expires in January 2018, its new members shall be selected via a merit-based recruitment process and shall be appointed by the Parliament of Kosovo*, following a selection committee’s proposal (Article 9 read in conjunction with Article 24 of the Law). 88 The Antimonopoly Committee of Ukraine with assistance from an EU-funded project and with close involvement of the Secretariat initiated early implementation of the State aid rules by preparing a case inventory in the energy sector, which was a reason for the Secretariat not to proceed with the infringement procedure. 89 Energy Community Secretariat, Annual Implementation Report (2013) (n 70) 84, 206. 90 Botta (n 11).

State Aid Enforcement in the Energy Community 569 Secondly, the compatibility assessment undertaken by the Commission for State Aid Control of one of the measures was not in line with the State aid acquis.91 It is worth noting that a number of complaints have been submitted to the Secretariat in several other State aid cases. Most of these complaints concern either guarantees for loans to State-owned undertakings (ie, usually loans granted for the construction of thermal power plants), or other issues where providers of services of general economic interest in the energy sector have benefited from State support. This shows that the awareness in the Contracting Parties about the applicability of the State aid rules to the energy sectors in the Contracting Parties is growing.

B. A New Tool to Ensure Compliance: Article 2 of the Dispute Settlement Rules The Energy Community Dispute Settlement rules were amended by the Ministerial Council in 2015. The newly introduced Article 2 has formalised the cooperation between the national authorities of the Contracting Parties and the Secretariat. In particular, this Article introduces a new legal obligation for the national authorities of the Contracting Parties (ie, including the State aid and competition authorities): [W]here a question concerning the interpretation or application of Energy Community law is raised in proceedings before a national authority of a Contracting Party, such authority, upon request of a party to the procedure before it or on its own motion shall notify the question to the Secretariat in writing at the earliest stage possible in the procedure. At the same time, Article 2 introduces an obligation for the Contracting Parties to ensure that the national courts have a right to notify the Secretariat about questions concerning the interpretation and application of Energy Community law. While State aid authorities have the obligation to ask for an opinion from the Secretariat, national courts have to have a right to do so.

The new provision governs the cooperation between the Secretariat and the national authorities and courts of the Contracting Parties. The reason for such cooperation is the need to ensure a uniform interpretation of the Energy Community rules.92 Under Article 94 of the EnCT, the obligation to interpret the Energy Community law in conformity with the case law of the CJEU and the General Court exists not only for the Energy Community institutions, but also for the Contracting Parties. Therefore, the cooperation between the Secretariat and the national State aid authorities under Article 2 would ensure that such harmonious interpretation takes place throughout the Energy Community. The cooperation introduced by Article 2 would follow the example of cooperation between the EU Commission and the national authorities and courts of the

91 After conducting a preliminary procedure, the Secretariat submitted a Reasoned Request in Case ECS-11/14 to the Ministerial Council on 19 May 2017, requesting a decision on Serbia’s failure to comply with the Energy Community State aid acquis. 92 On the homogeneity of interpretation of Energy Community law, see R Karova, ‘The Dispute Settlement System of the Energy Community: Testing its Effectiveness’ in D Buschle and K Talus (eds), The Energy Community: A New Energy Governance System (Cambridge, Intersentia, 2015).

570 Marco Botta and Rozeta Karova EU Member States.93 The proposal builds upon the existing cooperation between the Secretariat and the NCAs, State aid authorities and National Regulatory Authorities (NRAs). In the area of competition, such cooperation exists through the Energy Community Competition Network, established by a Memorandum of Understanding signed between the Secretariat and the National Competition Authorities of the Contracting Parties. The Network aims at facilitating the cooperation between the competition and State aid authorities, and helps them to exchange experience and best practices.94 Moreover, several Memoranda of Understanding have been signed between the Secretariat and the National Regulatory Authorities of the Contracting Parties, establishing an Implementation Partnership with the Secretariat. Finally, under Article 2, the Secretariat has to prepare an annual report to the Ministerial Council on the application and interpretation of the Energy Community law by national authorities and courts. Such reports shall keep all Parties and institutions of the Energy Community informed about national developments, and thus would contribute to the harmonious application of Energy Community law.

V. CONCLUSIONS

The chapter has analysed the enforcement of State aid rules in the Energy Community. Every EnC Contracting Party has adopted a State aid law in accordance with its international obligations under the EnCT and the Association Agreement concluded with the EU. Nevertheless, the chapter has pointed out a general lack of enforcement of State aid rules in relation to subsidies granted by the Contracting Parties to their energy operators. In particular, the State aid monitoring authorities have generally not been notified about long-term PPAs, regulated energy tariffs to industrial customers, subsidies to stimulate renewable energy sources and operating aids to Stateowned energy operators. Even when informed of aids granted to the provider of a service of general economic interest, the State aid monitoring authorities have failed to correctly apply the Altmark criteria, approving any aid scheme that was ‘in the public interest’. During the past few years, the EnC Secretariat has tried to strengthen the enforcement of State aid rules in the EnC by opening investigations concerning the Contracting Parties which have failed to establish a full functional State aid authority. The recently initiated infringement action in the Kolubara case represent a step forward: for the first time, the Secretariat contends the compatibility assessment previously carried out by a national State Aid Authority. Besides the infringement proceedings, the Secretariat has also tried to strengthen the cooperation with the national authorities by establishing the Energy Community Competition Network.

93 For instance, in the area of competition, Arts 11–13 of Regulation (EC) 1/2003 govern the cooperation with the NCAs, and Art 15 of Regulation (EC) 1/2003 and the EU Notice on the cooperation between the Commission and the courts of the EU Member States. 94 For information about the Network, see: www.energy-community.org/portal/page/portal/ ENC_HOME/AREAS_OF_WORK/Instruments/Competition/Network.

State Aid Enforcement in the Energy Community 571 The main limit of the system of State aid enforcement in the Energy Community concerns the lack of a supranational enforcement authority. Without an obligation for the Contracting Parties to notify the Secretariat about new aid schemes in the energy sector, it is difficult for the Secretariat to monitor whether the case law in the Contracting Parties complies with EU case law. Secondly, in the EnC there is no supranational court to which the national courts could address questions regarding the proper application of Energy Community law. The national courts of the Contracting Parties to the European Common Aviation Area95 can directly ask for preliminary rulings from the CJEU; this is not the case under the EnCT, although the Contracting Parties to the two agreements are mostly identical. Under Article 2 of the revised Disputes Settlement Rules, national State aid authorities shall ask for an opinion from the Secretariat, while national courts may decide to do so. The requirement for obtaining an opinion from the Secretariat is not new; such an obligation for the national administrative authorities of the Contracting Parties had already been introduced in relation to the process of implementation of the Third Energy Package. In particular, the Contracting Parties have to obtain from the Secretariat an opinion concerning Public Service Obligations, as well as a decision within the certification procedure for TSOs, and exemptions for new interconnectors. In those cases, the Third Energy Package imposes an obligation on the Contracting Parties to obtain an opinion of the Secretariat. In case the national authority’s decision diverges from ECS opinion, reasons for such divergence should be published together with the opinion. Compared with the Third Package required notifications, Article 2 of the Dispute Settlement Rules does not oblige a national authority/court to defend its position at the Ministerial Council.96 There have not yet been cases of formal application under Article 2 of the Dispute Settlement Rules. Nevertheless, some Contracting Parties had already communicated and coordinated their national assessments with the Secretariat in energy cases97 before the introduction of Article 2 in the Energy Community procedural rules on dispute settlement. However, in the long term, it will hopefully increase the involvement of the Secretariat in national State aid proceedings by strengthening the degree of enforcement of State aid rules in the Energy Community and assisting the national authorities in applying correctly the EU State aid rules to the energy sectors.

95 Art 16 ECAA Agreement. The ECAA was established in 2006. Multilateral Agreement between the European Community and its Member States, the Republic of Albania, Bosnia and Herzegovina, the Republic of Bulgaria, the Republic of Croatia, the former Yugoslav Republic of Macedonia, the Republic of Iceland, the Republic of Montenegro, the Kingdom of Norway, Romania, the Republic of Serbia and the United Nations Interim Administration Mission in Kosovo on the establishment of a European Common Aviation Area [2006] OJ L285/3 (ECAA Agreement). 96 Art 36 of Directive 2009/73/EC and Art 3 of Regulation 715/2009, as adapted for the Energy Community by Ministerial Council Decision. 97 A good example is the practice of the Albanian competition authority, as well as the State aid authority. Both maintain a practice of informing the Secretariat, and discussing cases in which they are engaged in the energy sectors.

Index Introductory Note References such as ‘128–29’ 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 ‘State aid’ and ‘energy’, the use of these terms (and certain others which occur constantly throughout the book) as entry points has been restricted. Information will be found under the corresponding detailed topics. absolute impossibility 306–7, 317–20 accessions 220, 225, 327, 337, 340–41, 347 ACER (Agency for Cooperation of Energy Regulators) 156, 159–60, 175–76, 240–43, 247, 527 acquis 515, 527, 555, 557 acte clair principle 478 actions for damages 364, 367, 369, 373–75, 473, 484 additional costs 19–20, 58–59, 61, 72, 74, 84–85, 89–90, 413–14 adequacy assessment 147, 154–58, 162, 176 administrative courts 324, 392, 411, 413–14, 418, 463, 469, 472 advantages imputable to the State 22–23 Agency for Cooperation of Energy Regulators, see ACER aid, see also Introductory Note amounts 14–15, 28, 122, 218, 267, 401, 430, 536 character 362, 370, 378–82 compatibility, see compatibility definition of State aid 3–29, 143, 457–58, 533 Art 107(1) elements 7–9 energy sector focus areas 29 general 5–7 no State functions/action 9 for energy infrastructure 258–59 incompatible 54, 183, 297, 359, 408, 483, 523, 556 intensity 84, 259–60, 398–401, 536, 541, 563 maximum 84, 259, 268, 553 measures 28, 119–27, 256–57, 359, 367–68, 399–401, 445–47, 449 non-efficient 120–21 to nuclear and coal 201–33 recipients 215, 314, 325, 331, 369, 375, 462, 468 recovery 301–25, 362, 371–72, 383–84, 461, 470, 476, 481–84 rescue and restructuring 206, 214, 218

schemes 54, 56, 141, 224, 443–44, 449–50, 558, 564–66 notified 542, 558–60 and SGEIs (services of general economic interest) 283–90 unlawful 308–9, 315–17, 321–24, 372–73, 376–77, 460–61, 470–71, 524 Albania 559–63, 566 Alcoa 39–42, 47, 464–67, 472–73, 536 allowances 69, 88–89, 223, 392, 525, 546 emission 6, 21, 288, 392–93 Alouminion 479–80 Altmark 54, 273, 283–84, 286, 289, 297–98 conditions 211, 285, 287, 289–90, 565 criteria 169–70, 211, 288–89, 293, 422, 570 aluminium 39, 41, 45, 48, 479, 531, 546, 548 ammonia producers 33, 36 analysis cost–benefit 241–42, 247, 250 economic 113–14, 117–21, 123, 125, 133–34, 141, 143, 361 ancillary markets 527, 530 annulment of Commission decisions 307–8, 323–25, 348, 382, 390–91, 411–14, 438–39, 501 appeals 67–68, 77–78, 275, 323–24, 438–39, 444, 447–49, 468–69 appropriateness 122, 147, 165, 216, 257–58, 261, 267, 269 arbitral awards 327–28, 342–43, 346, 349–52, 354, 430, 496, 506–9 arbitral tribunals 327–28, 336, 340, 346 arbitration 327–55 enforcement of awards in conflict with EU law 342–53 investment treaty 327, 336–42, 353 and legitimate expectations as key criterion under FET standard 333–36 and protection of legitimate expectations in State aid proceedings 328–33 arbitrators 336, 355, 506, 508

574 Index ARENH mechanism 408–9 assessment adequacy 147, 154–58, 162, 176 common assessment principles 121, 164, 206, 255 compatibility 113–14, 120–21, 253, 257–58, 260, 266, 362, 569–70 ex ante 26, 123, 126, 128, 189, 297, 401–2, 417 ex post 288, 296–97, 417 manifest errors of 11, 314 preliminary 378, 390–91, 422, 568 assets 16, 18, 210–11, 215, 306, 312–14, 318–19, 540 operating 541–42 State, see State assets assistance 108, 155, 219, 357–58, 366–67 financial 236, 248–50 Association Agreements 555–57, 559, 570 asymmetry of information 128–29 auctions 20–21, 88, 113–14, 129–31, 133, 141, 145–77, 429–30 organised 170–71, 420 Austria 26, 73–74, 174–75, 186, 217, 274, 441–51, 531 EAVG 441–44, 446–51 standstill 443–46 Wellnesshotel 446–47, 449–50 authorities competent 246–47, 322, 325, 345, 355 competition 132, 529, 533, 559, 569–70 autonomy, procedural 362–63, 367–68, 371, 481 availability obligations 191, 193–94 balance 64, 136, 139, 172–73, 260, 533, 535, 541 balancing markets 153, 158, 162, 165, 173, 175–76, 191, 193 balancing services 11, 395 balancing test 120–21, 123, 125–26, 132, 165, 167, 213, 546 BalticConnector 245 basis amount 432–35 Belgium 35, 149, 191, 207, 216, 220, 485–90, 526 capacity mechanism 489–90 electricity market liberalisation 489 offshore wind subsidy scheme 485–87 Belgium, nuclear phase-out support 487–89 benchmarks 15, 24, 96, 113–14, 117, 284, 297 beneficiaries 80–84, 264–69, 310–13, 318–23, 330–32, 367–70, 373–78, 466–70 beneficiary undertakings 9, 19, 28, 406, 438, 564 benefits 66–67, 94–96, 128–30, 132–33, 152–53, 198–99, 367–68, 406 economic 9, 259 potential 180, 357 bidders 17, 131, 196, 287

bidding competitive, see competitive bidding zones 158, 174–75, 182 bids 129–31, 139, 141, 167, 169, 191, 193, 196 bio fuels 26, 399–401 biomass 230, 437, 456, 534 black clauses 114, 118, 120 block exemptions 3, 118–20, 123, 126, 143, 254, 397–98, 447 Bosnia and Herzegovina 558, 562, 567–68 Brittany 171, 289, 403, 421–23 budgets 6, 130–31, 209, 250, 253, 434, 468, 554 State 232, 416, 538, 561 Bulgaria 221, 224, 251, 565 business risks 258, 267, 334, 550 business tax 309–10, 321 buyers 12, 211, 288, 312, 319, 548, 550 Canada 95, 100–1, 104, 106–7 capacities cross-border 174, 179, 181–82, 188, 195 domestic 186, 192, 196, 199 excess 39, 48, 188 existing 148, 170–71, 419 foreign 141–42, 170, 174, 176, 185–86, 188–90, 192–96, 419–21 generation 53, 151, 159–60, 173, 180, 211, 280, 289 installed 124, 132, 400, 405, 434, 537 interconnection 164–65, 174, 186, 188, 192–96, 198–99, 420 maximum 130–31, 192, 400 nuclear 201, 276, 279 overcapacity 16, 38–40, 46–47, 137–38, 186, 268 production 214, 225, 289, 433, 526 capacity auctions, see auctions capacity certificates 21, 169–70, 185, 416–17 capacity contracts 149, 180 capacity market 139, 171, 196, 422 capacity mechanisms (CMs) 145–77, 179–82, 184–86, 188–91, 195–99, 403, 419, 528 Belgium 489–90 cross-border participation 179–99 evolution of EU policy 147–49 France 21, 170, 186, 191, 289, 403, 416–24 market design initiative (MDI) 146–47, 150, 153–61 market-wide 162, 166, 177 material issues 150–53 planned 158, 161, 418 recent debate 145–46 sector inquiry on capacity remuneration mechanisms (CRMs) 161–63 and State aid guidelines 163–75 capacity payments 145, 148–49, 159, 163, 168, 280, 285, 494 capacity prices 171, 195–97

Index capacity providers 160–61, 167–68, 170, 188–91, 193, 196, 289, 417 foreign 188, 192, 196, 419 capacity remuneration mechanisms (CRMs) 145–47, 149–60, 162–64, 176, 180–83, 185, 187, 189–91; see also capacity mechanisms sector inquiry 161–63 capacity shortages 151, 173–74, 231 capital 39, 91, 128–29, 209, 284, 288, 535, 548 contributions 13, 315–16 costs 54, 129, 549 carbon capture and storage, see CCS carbon dioxide 238, 241, 250, 261 emissions 69–70, 81, 88, 230, 232, 239, 546, 553 transport 237, 239–41 carbon leakage 64, 88–90 carbon taxes 83, 257 case-by-case basis 102, 246, 260, 264, 462 casinos 482–83 causal links 374, 376, 460–62, 474 CCS (carbon capture and storage) 222, 256, 259, 544–46 CEF (Connecting Europe Facility) 248–50, 264, 269 cement industry 43, 546 certificates 21, 170–71, 288, 418–21, 485, 487, 537–39, 541 capacity 21, 169–70, 185, 416–17 electricity 529, 537–39 certificates markets 170, 288, 487, 529, 537, 539 green 537, 539–40 regional 485 CfD, see contracts for difference charges 6, 19–20, 49–50, 73, 77–78, 86–88, 471, 511–13 compulsory 74, 78, 87 electricity 73–74, 396 network 49–50, 75, 87, 396–97 parafiscal 18, 20, 78, 389 claimants 338–39, 341–43, 353, 357–58, 363–64, 371–74, 376, 525 clawback mechanisms 290, 401–2 clean energy 92, 98, 103–6, 108, 110, 153, 175–76, 226 Clean Energy Package 153, 175–76 clearing prices 73, 196 clients 40, 417, 420 closures 39–40, 177, 226–29, 232, 488, 568 aid 227–28, 231 CO2 emissions, see carbon dioxide, emissions coal 201, 203, 205, 207, 221–33, 278, 439–40, 544–45 accession States 225 current State aid regime and practice 226–33 ECSC (European Coal and Steel Community) 201, 203–4, 221–25, 443 hard 227, 231, 287 imported 221–22, 233, 291

575

indigenous 224, 229, 287, 291, 494 Netherlands 439–40 uncompetitive mines 225–29 coal-fired power plants 229, 287, 387, 392, 402, 439 co-financing 264–65 cogeneration 88, 251, 259, 274, 535 coincident scarcity 188, 193–94, 199 commercial terms 16, 210, 547, 553 Commission 33–46, 71–87, 202–33, 250–69, 285–98, 308–24, 371–84, 414–25 DG Energy 146–47, 150, 153–54, 156–58, 176 Commission investigations 418, 421 Commission practice 228, 272–73 commitments 55, 175, 193–94, 229–30, 311, 333–35, 407, 431 common assessment principles 121, 164, 206, 255 common interest 115, 123–25, 157, 215–16, 253–56, 259–61, 265, 267–69 objectives of 122, 164, 166, 216–19, 256–57, 259, 261, 436–37 common market 79–80, 304–5, 316–17, 359–60, 382, 435–36, 444–45, 465 principles 181–82 comparable situations 23–25 compatibility 50–53, 119–23, 183, 205–6, 290, 292–93, 302, 476–78 assessment 113–14, 120–21, 253, 257–58, 260, 266, 362, 569–70 criteria 143, 164, 255, 436 and economic analysis 114–23 environmental protection and energy 78–90 levies or surcharges to finance renewable energy 84–87 renewable energy sources 113–43 SGEIs 290–98 TEN projects 252–68 compensation 220, 232–33, 283–88, 296–98, 344–49, 428–31, 467–68, 504–5 financial 207, 229, 358, 374, 488, 561 payment 342, 344–47, 349, 351, 396, 502 competence 56, 460, 463, 473, 478, 520, 522, 527 exclusive 116, 253, 323, 359–60, 464, 477–78 competent authorities 246–47, 322, 325, 345, 355 competition 26–28, 115, 117–18, 215–19, 238–39, 260, 473–74, 565–67 distortion(s) 5, 7, 26–27, 122, 124, 141, 458, 460 effective 306, 529 law 7, 56, 146, 346, 404, 408, 456, 474 and trade 26, 28, 122, 126, 407, 419, 421, 423 competitive advantage 268, 307–8 competitive bidding 129, 132, 165 processes 86, 256, 258–59, 436–37, 554 competitive conditions 27, 283, 312, 392

576 Index competitive disadvantage 27, 42, 173 competitive markets 48, 93, 252, 281, 407, 457, 465 competitive positions 27, 44, 52, 220, 284, 383 competitive tenders 127, 129–31, 141, 143 competitiveness 43, 51–52, 124, 126, 135, 454–55, 457, 523 competitors 137, 304, 354, 367–70, 376, 456–57, 459–63, 472–74 potential 421, 462, 483 complainants 303, 397, 488, 507, 509 complaints 75–76, 479, 482–83, 506–10, 521, 552, 567, 569 complexity 33, 159–60, 199, 279, 358 compliance 263, 266, 282, 284, 293, 304–5, 328, 529 compulsory charges 74, 78, 87 compulsory surcharges 75–76, 78 concession power 547, 551–52 concessions 284, 466, 530 public 524, 530 concurrent proceedings 358, 361, 365, 373, 376–84 conflicts of interest 155, 189 Connecting Europe Facility, see CEF consolidation model 530, 540 constitutional courts 389, 442–43, 446, 448, 450, 489, 499–503, 513 consultations, public 148, 246–47 consumer subsidies 97–99, 103 consumers 58–61, 98–99, 102–3, 140–41, 150–54, 404–5, 492–93, 509–13 disadvantaged 491, 510 domestic 98, 282 electricity 40, 50, 59, 75, 85, 87, 389, 504–5 final 14, 19–20, 72–74, 76–78, 96–97, 129, 396–97, 405–6 intermediate 99 large 34, 75, 414, 428 consumption 51, 81, 92–93, 98, 230, 415, 420, 554 electricity 41, 169, 406, 495, 520, 551, 553 energy 39, 50, 75, 80, 89, 236, 399, 553 profiles 34–35, 46–47, 50 continuity, economic 311–14 contracts 53–54, 206–8, 276, 364–65, 479–80, 532, 547–50, 552 capacity 149, 180 import 425, 427–28 long-term 11, 53, 105, 124, 422, 425, 428, 546–48 control 61, 63, 202, 207, 209, 298, 497, 499 mechanisms 75, 77, 87 State 18, 59–61, 75, 87, 232, 389 conversions 4, 38, 204, 230, 293 cooperation 142, 155, 319, 350, 529, 569–70 mechanisms 142–43, 538 sincere 349–50, 360–62, 365–66, 480 coordination 139, 152, 155, 246, 257 corporate groups 493, 507–8, 511 cost price 432, 434, 551

cost–benefit analysis 241–42, 247, 250 costs 72–74, 76–79, 84–85, 127–30, 282–86, 288, 425–31, 549–50 additional 19–20, 58–59, 61, 72, 74, 84–85, 89–90, 413–14 allocation 247, 296 capital 54, 129, 549 eligible 120, 249, 259–60, 267, 398–402, 535–36, 554, 563 emission 67–89 exceptional 227–28 extra 60–63, 231, 259, 401, 426, 549 fixed 35, 38, 40, 46–47, 140, 168, 207, 288 indirect emissions 87–90, 546 investment 151, 287–88, 292, 398–99, 401, 534–36, 545, 553 least 17, 121, 126, 285, 287, 291 levelised 134, 427, 434, 487 marginal 40–41, 47, 194, 396, 541 net 121, 283–84, 296–97 operating 217, 297, 545 production 40, 42, 44, 82, 85, 89, 435–36, 538–39 regulated 504, 506, 511 restructuring 219, 504 social 228–29 stranded 45, 48, 220, 426–31, 456–57, 489, 493, 495 total 215, 244, 297, 411, 511 variable 38, 47–48, 207, 288, 488 Council of State 59, 61, 477–78 counterfactual scenario 258–59, 267 countervailing duties (CVDs) 94 courts of appeal 322, 350, 388, 390, 395, 440 constitutional 389, 442–43, 446, 448, 450, 489, 499–503, 513 English 349–51 France 411 Germany 364–65, 378–79, 388–89, 395, 402 Greece 45, 475, 477–78, 484 national, see national courts referring 205, 282, 311, 324, 380 supreme 352, 387–88, 440, 463–64, 470, 473, 501–5, 509–11 criteria eligibility 243, 248–50, 263, 265 non-discriminatory 165, 400, 415 objective 24, 81, 285, 292, 297, 541 transparent 16, 80, 82, 85 CRMs, see capacity remuneration mechanisms cross-border capacities 174, 179, 181–82, 188, 195 cross-border participation 155–56, 159–60, 163, 179–83, 185–87, 189, 191–93, 195–99 direct 185–86, 189 explicit 159, 180, 199 indirect 185 key issues for implementation 185–98 legal grounds for imposition 181–85 necessity 198–99

Index cross-subsidisation 27, 409, 513, 526–27 cumulative conditions/criteria 205–6, 263–64, 284 customers 19, 34–35, 64, 405–6, 426, 445, 493–94, 562 household 64, 533 industrial 34, 533, 562, 570 non-residential 49, 409–10 residential 32, 54, 141 vulnerable 455, 562 CVDs (countervailing duties) 94 Czech Republic 221, 226, 229, 231 damages 232, 345, 347, 364–65, 373–76, 460–64, 472–74, 524–25 actions for 364, 367, 369, 373–75, 473, 484 non-contractual 461 reimbursement 462, 473 de minimis exemption 4, 28–29, 254, 272 debts 14, 213, 306, 317, 494, 563–64 decentralised enforcement 475, 521, 557 decision-making powers 183, 527, 557, 566 decommissioning 29, 201, 208–9, 212–14, 216, 228–29, 231, 395 deficits 477, 493, 495–97, 506, 512 tariff 64, 494–98, 500, 503 Deggendorf principle 306 delay 318, 320–21, 323, 378, 381, 459, 461, 466 delivery obligation 191, 193–94 demand 39–40, 89, 151, 163, 280, 329, 331, 538–39 demand-side response, see DSR Demkolec 427–30 demonstration projects 127, 129, 132, 137, 536, 544, 552–53 Denmark 69, 81–83, 140, 149, 244, 266, 530, 532 depreciation 35, 38, 288, 292, 540–42 derogations 65, 219, 277, 482, 489, 541, 547, 551 from levies or surcharges 70–78 tax 66–70 design 107, 122–23, 125, 134, 155–56, 160–61, 164–65, 191 features 126, 131–32 development 113–15, 125–26, 136, 235–37, 251–52, 265–67, 426–27, 544–45 development bank loans 252 DG Energy 146–47, 150, 153–54, 156–58, 176 difference, contracts for 54, 166, 208–12, 217, 230, 277, 279 differentiation 25, 137, 392–93, 421, 442 diligent business man 321, 331–32, 372 direct effect 305, 357, 368, 459–60, 524 disadvantage, competitive 27, 42, 173 discretion 254, 281, 285, 287, 298, 374, 478, 481

577

discrimination 85–86, 107, 109, 163, 296, 334, 423, 525 disputes 102, 106, 110, 337–38, 340–41, 428–29, 445, 566–67 distortions 27–28, 125–26, 132–33, 140, 216–19, 368, 473–74, 533 potential 126, 230 regulatory 157, 176 structural 134, 477 undue 216, 218, 260, 268–69, 488 distribution 33, 43, 49, 73–74, 98, 251, 492, 510–11 networks 401, 493, 495 distribution system operators (DSOs) 18, 72–77, 85, 154, 240, 396, 429, 549 district heating 88, 398, 401, 425, 428, 430–31, 535, 564–65 divergence 117, 353–54, 526, 571 diversification 239, 249, 281, 437, 543 domestic capacities 186, 192, 196, 199 double infringement 325 downstream markets 64, 141, 268, 536 DSOs, see distribution system operators DSR (demand-side response) 157–58, 162, 164–66, 168, 185, 188–89, 193, 566–67 dual pricing 99–100, 102 duration 170–71, 292–93, 297, 419, 421, 424, 480, 550 EA, see Europe Agreements economic advantage 7, 9–10, 206–7, 283–84, 288, 406, 477, 479 economic analysis 113, 117–21, 181, 361, 366 and renewable energy support schemes 123–33 role in compatibility assessment 114–23 unintended consequences 133–43 economic continuity 311–14 economic interest, general 275, 279, 282, 289, 291, 293, 295, 411 economic operators 6, 64, 122, 190, 295, 375, 516, 519 economic theory 129–30, 140, 188 economics 117, 120, 125, 134, 139 ECS, see Energy Community Secretariat ECSC (European Coal and Steel Community) 201, 203–4, 221–25, 443 Ecuador 335 EDF 38–39, 55, 170–71, 207–8, 406–9, 423, 456, 458 EEA (European Economic Area) 237–38, 385, 515–54 downstream aid 546–54 energy market 526–33 enforcement 533–54 ESA (EFTA Surveillance Authority) 519–27, 529, 534–36, 539–52 scope of Agreement and application of State aid rules 515–20 State aid procedure 520–26 upstream aid 539–46

578 Index EEAG (Energy and Environmental Aid Guidelines) 78–86, 125–26, 147–48, 183–84, 254–62, 414–16, 423, 436–37 EEG-surcharge 77, 85–87, 389–90 effect on trade 27–28, 260, 519, 533, 536 effective enforcement 359, 381, 482–83 effective implementation 298, 306, 323, 354, 359, 479, 482, 568 effective judicial protection 307, 324, 369, 478 effectiveness 305–7, 321–22, 324, 362–65, 367, 369, 481, 484 efficiency 121, 125, 127, 132, 137, 195–96, 238–39, 296–97 energy 81, 83, 157–58, 236, 251, 398, 552, 562 incentives 296–97 targets 296–97 EFSI (European Fund for Strategic Investments) 248, 251–52 EFTA Court 519–20, 522–23, 527, 540, 546 EFTA Surveillance Authority, see ESA EIUs (energy-intensive undertakings) 19, 44, 69, 86, 389–90, 441, 444, 446 Elcogás 495, 504–5, 509–10 Electrabel 214, 336–39, 426, 489 electricity 208, 277–79, 393–95, 404–6, 423–28, 439–40, 492–96, 531–33 certificates 529, 537–39 charges 73–74, 396 consumption 41, 169, 406, 495, 520, 551, 553 costs 84, 89, 389 green 18, 25–26, 73, 77–78, 445, 538 low-carbon 165–67, 211, 279 markets 42, 158, 216, 224–25, 465, 488–89, 526, 531 regulated 216, 488 production 20, 83, 216, 221, 279, 388–89, 394, 426 renewable 73, 77, 85, 256, 426, 431–32, 434–35, 538–39; see also renewable energy sale of 165, 184, 407, 531 suppliers 64, 72–73, 75, 77–78, 85, 169, 405, 537–38 tariffs 42–43, 47, 404, 465–67, 473, 480, 510, 562 tax 83, 547, 551 electricity-intensive users 49–52, 396–97 electro-intensity 52, 85–86 Elia 485–87 eligibility, criteria 243, 248–50, 263, 265 eligible costs 120, 249, 259–60, 267, 398–402, 535–36, 554, 563 emission allowances 6, 21, 288, 392–93 Emission Trading Scheme, see ETS emissions 21, 69, 83, 88, 90, 103, 163, 525 costs 67–89 indirect 87–90, 546 targets 230, 232 EnC, see Energy Community

end users 31, 51, 58, 60–61, 64, 152, 526, 537–38 ENEL 40–41, 43, 209, 454, 465, 467 energy, see also Introductory Note dual pricing, see dual pricing efficiency 81, 83, 157–58, 236, 251, 398, 552, 562 Greek case 45–46 green 494, 521 Hinkley Point 53–54, 206, 208–9, 211–12, 216–17, 273–74, 276–77, 279 Hungarian long-term contracts 53–54 infrastructure, trans-European 236–37, 239, 252, 255, 268–69 Italian cases 39–45 law 85, 409, 453, 492 long-term contracts and power-purchase agreements 53–54 market prices 134, 437 markets 44, 46, 123–26, 128, 132–33, 162, 194–95, 529–30 internal 146, 162–63, 256, 260, 265, 268, 515, 517 nuclear 7, 201–2, 204, 212, 216–18, 273–74, 279, 488 photovoltaic 495, 512–13 policy 64, 136, 277, 279, 453, 516, 534, 545 price regulation 56–57 production 84, 91, 230, 393, 497, 510, 512, 539 products 25, 81, 83, 208, 439–40, 515, 523, 533 renewable, see renewable energy residential customers 54–56 resources 134, 181, 233, 278, 530, 539 solar 106, 130, 132, 151, 534 State aid and preferential tariffs for energy-intensive industries 32–52 State resources 57–64 subsidies 92, 97, 103–5, 110 Energy Community (EnC) 559–66 supply, security of 32, 181, 204, 223, 264, 277 sustainable 103, 165, 237, 251, 389 taxes 79, 441–42, 523, 551 rebates 441–42, 446, 449–50 and WTO subsidy rules 91–110 Energy and Environmental Aid Guidelines, see EEAG Energy Charter 506–8 Energy Community (EnC) 385, 555–71 energy subsidies granted by Contracting Partners 559–66 legal obligation of Contracting Parties to introduce State aid control regime 556–59 Secretariat enforcement 566–70 Energy Community Secretariat (ECS) 555, 558, 566–70 Energy Fund 534–36, 552–53 energy infrastructure 5, 13–14, 204–5, 235, 237, 251–62, 398, 401

Index Energy Union 4, 143, 161, 179, 181, 265 energy-intensive companies 44, 66, 69, 73, 77, 81, 84–87, 89–90 energy-intensive industries 76, 85, 97, 100, 546 State aid and preferential tariffs for 32–52 energy-intensive undertakings, see EIUs energy-intensive users 42, 64, 77, 172, 388, 415, 531, 548–50 enforcement 298, 302, 308, 357–58, 360, 456, 460–61, 528–29 of arbitration awards in conflict with EU law 342–53 decentralised 475, 521, 557 effective 359, 381, 482–83 Energy Community (EnC) 385, 555–71 notices 303, 308, 359, 366, 373, 376, 378 private 357–59, 361–63, 365, 367–69, 371, 373–75, 383, 473–74 role of Commission and national courts 358–67 Engie-Electrabel 207, 216, 488–89 Enova 534–36, 539, 552, 554 entrants, new 88, 131–32, 404, 457, 462, 471–72, 489 entrustment 211, 275–76, 284, 290, 292, 294, 296–97 environment 63, 65, 68, 78–79, 84, 99–100, 229, 233 environmental protection 3, 78–79, 255–57, 261–62, 266, 269, 389, 397–99 guidelines on State aid for 14, 65, 122, 125, 183, 435–36, 534–35, 539 projects 261–62, 269 environmental taxes 79–80, 416, 439, 447 harmonised 80–84 non-harmonised 82 equal treatment 442–43, 448, 450 equitable treatment 328, 333, 336–39, 341 equivalence 13, 305, 362–64 ERDF (European Regional Development Fund) 244, 248, 250–51, 254, 269, 278 errors, manifest 35, 233, 271, 276, 278, 379, 469 ESA (EFTA Surveillance Authority) 519–27, 529, 534–36, 539–52 ESIF (European Structural and Investment Funds) 250, 254 Essent 57, 59, 61, 72–73, 77–78, 185, 429 ETS (Emission Trading Scheme) 21, 69–70, 81–84, 87–90, 222, 288, 546 Euratom 201–6, 212, 214–17, 219 and State aid rules 203–5 Europe Agreements (EA) 202–6, 218–19, 225, 342 European Coal and Steel Community, see ECSC European Commission, see Commission European Economic Area 237–38, 515 European Fund for Strategic Investments, see EFSI European Regional Development Fund, see ERDF European Union, see Introductory Note

579

ex ante assessments 26, 123, 126, 128, 189, 297, 401–2, 417 ex post assessment 288, 296–97, 417 exceptional circumstances 297, 304, 320–21, 332, 354, 367, 372, 412 excess capacity 39, 48, 188 exchanges 192, 207, 319–20, 416–17, 421, 495 exclusive competence 116, 253, 323, 359–60, 464, 477–78 exemptions 46–47, 49–52, 66–69, 74–75, 116–20, 294–95, 393–94, 439–40 tax 22, 25, 65–66, 68–70, 82–83, 425–26, 439–40, 469–70 existing installations 399, 499, 502–3 expectations 136, 328, 331, 333, 335–36, 342 fair 334 justifiable 333 legitimate, see legitimate expectations expected profits 499–500, 502, 507 expenses 27, 49, 51–52, 96, 493, 495–96, 521, 542 regulated 493, 495–97, 504, 512 explicit participation 159, 180, 188–90, 199 export taxes 99, 102 expropriation 466–67, 500, 503, 508 regulatory 337, 339 externalities 102, 124, 126, 135 negative 93–94, 100, 107 positive 250, 536 extra-cost approach 534–35 fair and equitable treatment (FET) 328, 333–34, 336–39, 341, 355 FCA (French Competition Authority) 184, 191, 404, 408–10, 417 feed-in premiums 114, 126–30, 132–33, 141, 143 feed-in tariffs 86, 91, 97–98, 113, 124, 126–30, 253, 478 guaranteed 477–78 fees, risk 206–7, 209, 218 FET, see fair and equitable treatment final consumers 14, 19–20, 72–74, 76–78, 96–97, 129, 396–97, 405–6 financial advantage 27, 36, 283–84, 305 financial burden 17, 61, 64, 75–77, 82, 108, 388, 393 financial compensation 207, 229, 358, 374, 488, 561 financial flows 75, 77, 87, 487 financial institutions 14, 405, 413, 565, 568 financial markets 60, 151, 274 financial resources 19, 57, 60, 73, 261, 407 financial risk 151, 193, 264 financial support 210, 229, 244, 248, 264, 268–69, 457, 461 financial transparency 293 financing mechanisms 61, 75, 77, 104, 403, 414, 428, 510–11 Finland 148–49, 209, 211, 215, 245, 398, 530, 532

580 Index FIT schemes 105–8, 413–14 fixed costs 35, 38, 40, 46–47, 140, 168, 207, 288 fixed prices 11, 560, 562 flexibility 130, 133, 135, 168, 294, 381 flows 160, 190, 192–93, 195, 365, 389, 533 financial 75, 77, 87 force majeure 193 foreign capacities 141–42, 170, 174, 176, 185–86, 188–90, 192–96, 419–21 foreign entities 180, 188, 191–92, 194, 199 foreign generators 159, 174, 186, 194–96, 199 foreign investors 327, 333, 335, 500, 560–61 formal investigations 169, 171, 209, 372, 377–79, 381–82, 548, 552 procedure 21, 75–76, 87, 309, 390–91, 396, 415, 482 fossil fuels 84, 89, 91–92, 97–100, 222, 406, 489 subsidies 97–103, 105, 110 France 12–13, 58–59, 133, 149, 170, 243–44, 318–20, 403–24 capacity mechanisms 21, 170, 186, 191, 289, 403, 416–24 Competition Authority, see FCA Council of State 59, 61, 477–78 courts 58–59, 411 Energy Code 410–11, 416–17 government 55, 60, 209, 215, 219, 403, 416, 418 market 169, 171, 407–8, 424 regulated prices 403–12 renewable energy 412–16 free market 53, 64, 271, 533 free movement of goods 6, 86, 183–84 free zone benefits 336, 341–42, 353 funding 58–59, 62, 71–76, 236, 248, 250, 269, 544–45 public 13–14, 29, 283, 315 gas 33–35, 37, 236–38, 240–41, 403–4, 409, 515–17, 541–42 LNG (liquefied natural gas) 235, 252, 281, 292–94, 296–97, 523, 542 prices 410–11, 545 projects 239, 241–42, 523 supply 33–34, 36, 281, 404, 411 gas-fired power plants 168, 289, 291, 440, 488, 544–45 Gassnova 544–46 general economic interest 275, 279, 282, 289, 291, 293, 295, 411 general measures 24, 66–67 generation adequacy 146, 148, 153–54, 163–64, 174, 180, 182–83, 280–81 assessment 156, 174 generation capacity 53, 151, 159–60, 173, 180, 211, 280, 289

generators 86, 126–29, 137, 139–40, 159–60, 165–67, 285–88, 439–40 geothermal power plants 471, 549 Germany 18–19, 76–78, 85–87, 173–76, 221–22, 226–27, 230–32, 381–82, 387–402 coal and nuclear 392–94 courts 364–65, 378–79, 388–89, 395, 402 grid-related issues 394–97 law 51, 72–73, 232, 311, 370, 392 market 133, 392 renewable energy 388–91 State aid measures in energy area 397–402 good faith 60, 306, 319, 330, 341 goods 5–8, 16, 20, 23, 91–92, 115, 183, 185 free movement of 6, 86, 183–84 governance 61, 63, 455 Greece 45, 48, 133, 221, 226, 475–84 courts 45, 475, 477–78, 484 recovery 481–84 violations of notification and standstill obligation 475–81 green certificates 104, 127–28, 142, 418, 485, 537–39 market 537, 539–40 green electricity 18, 25–26, 73, 77–78, 445, 538 green return tariffs 54–55, 405–7 grids 4, 106, 130, 175, 237–38, 395, 429, 512–13 Germany 394–97 smart 235, 238–39, 241–43, 250–51 guarantee premiums 14–15, 548 guaranteed feed-in tariffs 477–78 guarantees 14–16, 192–95, 209–10, 219–20, 251–53, 307–8, 311, 547–48 financial 563–65 investment 207, 216 of operation 431, 456 of origin 185, 432 State 14–15, 22, 209, 217, 220, 547, 565–66, 568 guarantors 15, 61–62, 210, 547–48 guidelines on State aid for environmental protection 14, 65, 122, 183, 254, 435–36, 539 hard coal 227, 231, 287 harmonisation 82, 135, 142–43, 146, 155, 443, 520 harmonised environmental taxes 80–84 harmonised taxes 79–84 heating, district 88, 398, 401, 425, 428, 430–31, 535, 564–65 host States 333–35, 353–54 households 64, 96–99, 405, 410, 428, 510, 533, 562 Hungary 11, 53, 210, 218, 224, 226, 336–40, 352

Index hydroelectric plants 399, 466, 468, 537, 552, 560 hydropower 528, 531, 539, 541 Iceland 515, 522, 531, 540, 543, 546, 548–50 market 531, 548 ICSID (International Centre for Settlement of Investment Disputes) 339, 343, 345, 348–49, 351–52, 506 IEA (International Energy Agency) 454–55 illegal State aid 305, 328–29, 341–51, 396–97, 440, 459, 461–63, 476–77 illegality 308, 414, 444–45, 476 interest 373, 414 implementation 55–56, 222–24, 345, 379–80, 458–60, 480, 527, 566–68 effective 298, 306, 323, 354, 359, 479, 482, 568 unlawful 359–60, 378 implicit participation 192, 199 import contracts 425, 427–28 important projects of common European interest, see IPCEIs imports 86, 186–87, 222, 224, 227, 528, 531, 553 impossibility, absolute 306–7, 317–20 imputability 23, 55, 208, 345–46, 406 incentive effects 216, 218, 258, 261, 267, 269, 420, 423 incentives 90, 105–6, 188–89, 199, 279, 342, 347, 494 efficiency 296–97 inefficient 138 income 9, 54, 101, 168, 493, 495–96, 534, 541–42 sources 493–94, 496–97 incompatibility 190, 311, 411, 481 incompatible State aids 54, 183, 297, 359, 408, 483, 523, 556 incumbents 168, 170, 409, 411, 454, 456–57, 462, 471 indigenous coal 224, 229, 287, 291, 494 indirect emissions costs 87–90, 546 indirect transfer of State resources 17–18, 72, 388–89 individual rights 360, 363–64, 377, 464, 484, 500, 502–3 industrial customers 34, 533, 562, 570 information 128–29, 204–5, 246–47, 315–17, 322–23, 365–66, 447, 449–50 asymmetry 128–29 infrastructure 4, 13–14, 229, 236–37, 251, 256–57, 260, 267–68 energy 5, 13–14, 204–5, 235, 237, 251–62, 398, 401 trans-European 236–37, 239, 252, 255, 268–69 projects 236, 248, 254, 264 regulated 189–90 in-house provision of services principle 295

581

injury 93, 106, 108, 525 innovation 9, 125, 132, 135, 137, 250–51, 267, 456–57 installations 90, 399–400, 412, 494–97, 502–4, 508, 512–13, 542 existing 399, 499, 502–3 installed capacity 124, 132, 400, 405, 434, 537 institutions 271, 328, 330, 353, 355, 366, 517–18, 568 financial 14, 405, 413, 565, 568 integration 125–28, 134–35, 137, 143, 146, 148, 238–39, 243–45 of RES 126, 134–35, 137–38, 140, 142 intensity, aid 84, 259–60, 398–401, 536, 541, 563 interconnections 46–47, 180–81, 185–90, 192–96, 198–99, 252, 255, 543–44 interconnectors 159–60, 165–67, 174, 185–86, 188–92, 199, 527–28, 530–31 interest rates 14–15, 66, 212, 253, 564 interim measures 225, 371, 373, 479–80 intermediaries 19, 60, 71–78, 84, 87, 249, 413 internal energy market 146, 162–63, 256, 260, 265, 268, 515, 517 internal market 54–55, 115, 182–84, 253–56, 260–62, 272, 302, 476–81 International Centre for Settlement of Investment Disputes, see ICSID International Energy Agency, see IEA international investors 494, 496, 499, 502, 506–7 international law 338, 348–49, 352 international obligations 307, 350, 570 interoperability 236, 238–39, 252, 262 interruptibility schemes 149, 163, 172–75 investigations 377, 379, 416, 425, 465, 467, 470, 491 Commission 418, 421 formal, see formal investigations investment 10–13, 151–52, 333–37, 498–99, 506–9, 536–37, 540–42, 549–50 aid 54, 226–27, 256, 258–59, 398–400, 534–35 arbitration 328, 334, 342, 344, 353–55 decisions 11, 96, 151, 218, 536, 548 efficient 153, 538 guarantees 207, 216 network 139, 238 new 137, 166–67, 170, 199, 218, 251, 498 investment treaty arbitration 327, 336–42, 353 investors 327–28, 333–36, 340–44, 353–55, 495, 499–500, 502–3, 507–9 foreign 327, 333, 335, 500, 560–61 international 494, 496, 499, 502, 506–7 market economy 10–11 private 10, 12–13, 46, 96, 209–10, 545, 548, 550 IPCEIs (important projects of common European interest) 115, 254, 261–69 Ireland 145, 149, 280, 285–86

582 Index Isolux 507–8 Italy 26, 39–40, 42–44, 133, 145, 149, 243–44, 453–74 case law 461–69 government 39, 453, 455, 465–66 State aid and enforcement notions 456–61 joint ventures 156, 525 judicial protection, effective 307, 324, 369, 478 jurisdiction 305, 337–38, 342, 344, 352, 354, 463, 522 justifications 25–26, 43, 45, 48, 66–69, 275, 278, 282 Kolubara 565, 568, 570 Kosovo 559, 561–64, 567–68 Landsvirkjun 531, 548–50 large electricity consumers (LEC) 50, 75, 87, 415, 429, 464 lawfulness 311, 372, 378, 443–44, 457 LCRs, see local content requirements legal actions 364, 456, 458–62, 464, 472–73, 490, 522 legal bases 204, 221, 310, 375, 384, 516, 518, 524 legal obligations 339, 341, 353, 555–56, 558–59, 562, 569 legal orders 322, 328, 352–53, 355, 484 legal standing 363, 367–69, 501, 567 legality 195, 215, 253, 323–24, 329, 396, 458 legislative reforms 496, 502–3, 506–8 legislators 64, 129–30, 404, 442, 446, 477–78, 498 legitimate expectations 307, 320–21, 328–36, 338–39, 341–42, 353–55, 468–69, 499–500 as key criterion under FET standard 333–36 protection 307, 309, 320–21, 328–33, 338, 468–69 level playing field 127, 143, 155, 259, 283, 423, 464, 474 levelised cost of energy (LCOE) 130, 134, 427, 434, 487 levies 68, 70, 83–85, 208, 213–14, 393–94, 428–29, 482–83 parafiscal 40, 70–72, 78, 84, 388, 428, 467 lex generalis 205 lex specialis 202, 205 liability 15, 207, 213–14, 318, 375, 461 contractual 463 non-contractual 370, 375 State 461, 523 liberalisation 33, 407, 426, 457, 471, 492, 517, 526 liberalised markets 42, 271, 285, 287 Liechtenstein 515, 531 lignite 222, 226–27, 230–31 limitation periods 307, 309–10, 524 Lithuania 133, 255, 281, 293–95, 532 LME (London Metals Exchange) 89

LNG (liquefied natural gas) 235, 252, 281, 292–94, 296–97, 523, 542 loads 154, 172–73, 187, 244 balancing 50 interruptible 172, 175 loans 10, 14–15, 209, 252–53, 307, 311, 563–66, 568–69 local content requirements (LCRs) 94, 96, 106–8 London Metals Exchange (LME) 89 long-term contracts 11, 53, 105, 124, 422, 425, 428, 546–48 long-term power purchase agreements 11, 524, 531, 547–48, 560, 562, 564, 570 losses 347, 374, 376, 483, 485, 508, 561, 563 low-carbon electricity 165–67, 211, 279 Lufthansa 379–82, 397 Luxembourg 113, 507 Macedonia 559, 561–62, 564–65 main proceedings 59–60, 74, 391, 414 manifest errors 35, 233, 271, 276, 278, 379, 469 marginal costs 40–41, 47, 194, 396, 541 market conditions 13, 42, 100, 105, 143, 167, 314, 319 normal 9–10, 16, 42, 65, 273, 279, 307, 397 market design initiative (MDI) 146–47, 150, 153–61 market designs 133–34, 138–39, 143, 145–47, 152–53, 161, 163, 176 market dominance 165, 167, 423 market economy investors 10–11 market economy operator principle 10–13, 209, 211 market equilibrium 140–41 market failures 42–43, 94, 124–27, 161–62, 257–58, 273–75, 279, 536 residual 176, 257 well-defined 122, 218 market forces 45, 174, 275, 279, 289, 407, 411 market integration 125, 128, 134–35, 143, 146, 148, 163, 238–39 market investor test 457, 537, 548, 550, 564 market mechanisms 158, 280, 533, 538–39 market operators 11, 53, 286, 312, 383, 477–78 market participants 58, 159, 177, 195, 255, 454, 536, 538 market players 146, 158, 170, 418, 421, 478, 492, 527 market power 42, 128–29, 131–32, 146, 176 market prices 54–55, 72–74, 208, 312–13, 404–6, 432–33, 435, 547–49 market reforms 147, 162, 173, 177 market risk 128–29, 187–88 market-based mechanisms 126, 128–29, 170, 185 market-based remuneration 14, 397, 540 markets balancing 153, 158, 162, 165, 173, 175–76, 191, 193

Index competitive 48, 93, 252, 281, 407, 457, 465 downstream 64, 141, 268, 536 electricity, see electricity, markets energy 44, 46, 123–26, 128, 132–33, 162, 194–95, 529–30 France 169, 171, 407–8, 424 Germany 133, 392 green certificates 537, 539–40 Iceland 531, 548 liberalised 42, 271, 285, 287 relevant 274, 298, 332, 481–82 retail 154–55, 424–26, 455, 531, 533 short-term 151, 153, 155, 157, 176 single market 199, 271, 283, 353, 355, 415, 466–67, 515–16 spot 77, 85, 289 market-wide capacity mechanisms 162, 166, 177 mature technologies 125, 129, 131–33, 137, 275, 279 maturity 125–26, 135–36, 236, 247, 249–50 maximum aid intensities 84, 259, 268, 553 maximum capacity 130–31, 192, 400 MDI, see market design initiative measures justified by the nature of the system 25–26 MEP-subsidy 426–27, 431, 435 methodology 156, 158–60, 220, 296, 428, 431, 489, 498 reference framework 66–67, 69 Micula 328, 336, 340, 342, 344, 346–54, 502, 509–10 mines 221, 227–29 minimum prices 17, 72, 108, 388, 485, 487 missing money problem 151–52, 171, 176, 289 modernisation 53, 133, 143, 223, 252, 256, 456 Moldova 556–59, 563 Mongstad 544–46 monopolies 33, 43, 99, 119–20, 290, 407 legal 27, 491–92, 511 Montenegro 559, 562–63, 566 mothballing 137, 231–32 multilateral trading system 92–93, 102, 110 national authorities 250, 258, 312–13, 315, 318–20, 330–31, 358, 569–71 national competition authorities 533, 559, 570 national courts 301–4, 307–11, 321–24, 357–73, 375–84, 390–91, 459–61, 475–81 obligations in cases of concurrent proceedings 376–84 private enforcement before 367–76 National CRMs, see capacity remuneration mechanisms national judges 358, 373, 463, 472, 476–78, 484 national law 245–46, 306–7, 317–18, 321–23, 364, 375–76, 383, 481–82 national legislation 60, 208, 284, 297, 371, 393, 492, 499–501

583

national level 31, 64, 107, 141, 404, 410–11, 456, 459 national provisions 470, 482, 558, 566 national regulatory authorities, see NRAs natural gas, see gas natural resources 539–40, 543 NCAs, see national competition authorities necessity 43, 52, 80, 82, 161, 164–66, 215, 257 negative effects 53, 93, 100, 122, 128, 132, 268, 541 negative externalities 93–94, 100, 107 negative prices 86, 127–28, 400 neighbouring countries 152, 160, 187–89, 192, 194–96, 199 net costs 121, 283–84, 296–97 net present value, see NPV Netherlands 20, 33, 35–37, 133, 425–40, 507, 530 ammonia producers 33–34, 36 coal tax 439–40 development of projects 437–39 government 34, 425–28, 431–32, 439 SDE-subsidy 427, 431–37 stranded costs 427–31 Supreme Court 307, 310, 440 network charges 49–50, 75, 87, 396–97 network investments 139, 238 network operators 18, 72–73, 77–78, 108, 417, 493 networks 49–50, 154, 173, 182, 189, 199, 456, 464 new entrants 88, 131–32, 404, 457, 462, 471–72, 489 non-actionable subsidies 94–95, 108–10 non-compliance 163, 275–76, 449, 458, 483, 529 non-contractual liability 370, 375 non-discriminatory criteria 165, 400, 415 non-harmonised taxes 79–80, 82–83 non-ICSID awards 343, 346, 350–52 non-reimbursable grants 536, 552–53 non-residential customers 49, 409–10 Nord Pool 532–33 normal business risk 258, 267 normal market conditions 9–10, 16, 42, 65, 273, 279, 307, 397 Norway 142, 515, 521, 523, 525–29, 536–41, 543–47, 551–54 Energy Fund 534–36, 552–53 government 523, 525, 527, 529–30, 535, 538, 540, 544–45 notification 118–19, 202, 204–5, 446, 448, 475, 477, 502 (and standstill) obligation 357, 360–61, 363, 375–76, 441–46, 475–78, 480, 524 thresholds 256, 398–402 notified measures 4, 166, 183, 211, 289–90, 540, 542–43, 558–60 NPPs, see nuclear power plants NPV (net present value) 139, 267–68, 534–35, 549–50

584 Index NRAs (national regulatory authorities) 160, 240–42, 247, 296, 570 nuclear and coal, aid to 201–33 nuclear capacity 201, 276, 279 nuclear energy 7, 201–2, 204, 212, 216–18, 273–74, 279, 488 nuclear fuel 203, 205, 208, 212, 214, 393–94 nuclear industries 201–21, 230 application of Art 106(2) 221 application of Art 107 205–20 nuclear power plants (NPPs) 207–10, 212, 214, 216, 218–20, 274, 394–95, 487–88 nuclear tax 207, 488–89 objective criteria 24, 81, 285, 292, 297, 541 objectives of common interest 122, 164, 166, 216–19, 256–57, 259, 261, 436–37 policy 5, 11, 25, 103, 123, 125, 153, 286 obligation to recover 117, 302, 309, 325, 466, 524 offshore wind farms 143, 244, 434–37, 485 oil 238, 240–42, 261, 515–16, 528, 541–42, 545, 552 Olympic Airways Group 313–14 Ontario 106–7 operating aid 41, 47–49, 226–28, 256, 258–59, 400–1, 465–67, 563–64 operating assets 541–42 origin of resources, State 59–60, 63 overcapacity 16, 38–40, 46–47, 137–38, 186, 268 overcompensation 163, 165, 171, 173, 230, 297, 534–35, 539 risk 171, 217, 230, 437 ownership 454, 530, 539 parafiscal charges 18, 20, 78, 389 parafiscal levies 40, 70–72, 78, 84, 388, 428, 467 participation 159–60, 165, 179, 182–83, 185–86, 190–91, 194, 198–99 direct 125, 135, 170, 182, 189–90, 192, 194–95 explicit 159, 180, 188–90, 199 implicit 192, 199 indirect 194 pay-as-bid design 131 pay-as-clear design 131 PCC 548–50 PCIs, see Projects of Common Interest photovoltaic energy 495, 512–13 pipelines 4, 237, 245, 292–93, 408, 454, 541–42 Poland 133, 149, 221–22, 224, 226, 228–29, 255, 561 policy failures 150, 152, 154, 158 policy objectives 5, 11, 25, 103, 123, 125, 153, 286 polluter pays principle 203, 215–16, 218, 281 Portovesme 39–40, 42, 48, 465

Portugal 149, 251, 283, 493, 505 positive effects 104, 122, 125, 129, 167, 541 positive externalities 250, 257, 265, 536 postal services 271, 292 power plants 53–54, 137, 215–16, 218, 220, 231–33, 395–96, 537–38 coal-fired 229, 287, 387, 392, 402, 439 gas-fired 168, 289, 291, 440, 488, 544–45 nuclear 207–10, 212, 214, 216, 218–20, 274, 394–95, 487–88 thermal 569 virtual 41, 56 wind 540 power purchase agreements (PPAs) 11–12, 53–54, 275, 336–39, 352, 432–33, 546–48, 560–61 long-term 11, 524, 531, 547–48, 560, 562, 564, 570 PPAs, see power purchase agreements PPC 168, 293, 479–80 Praxair 413–14 precedents 43, 45, 217, 278, 303, 461 preferential tariffs 32–38, 40–47, 49, 51, 55, 465–66, 468, 479–80 preliminary assessment 378, 390–91, 422, 568 preliminary rulings 324, 379–81, 387, 390–91, 393–95, 411–14, 428–29, 505 premiums feed-in 114, 126–30, 132–33, 141, 143 guarantee 14–15, 548 PreussenElektra 17–21, 57–63, 72–78, 119, 231, 388–90, 486, 490 price caps 194, 339–40 price formation 5, 64, 151, 454, 464 price regulation 31, 33, 35, 37, 55–57, 63–64, 389, 561 price signals 127, 151–52, 172, 177, 195, 289 price spikes 151, 163 price volatility 212, 420 prices 38–42, 87–89, 98–99, 137, 196–97, 404–6, 408–11, 531–33 clearing 73, 196 competitive 32, 42, 64, 165 electricity 42, 44, 158, 163, 167, 433, 528, 546 energy 31–33, 44, 64, 99, 133–35, 137–38, 141, 192 fixed 11, 560, 562 market 54–55, 72–74, 208, 312–13, 404–6, 432–33, 435, 547–49 minimum 17, 72, 108, 388, 485, 487 negative 86, 127–28, 400 reference 54, 437, 532 regulated 31, 59, 403–5, 407, 409–11, 483, 561 retail 31–32, 155, 298, 422, 454–56 private creditor test 10, 14–16 private enforcement 357–84, 461, 473–74 before national courts 367–76 private investors 10, 12–13, 46, 96, 209–10, 545, 548, 550

Index private operators 16, 20, 23, 40, 53, 57, 63, 74 private parties 116, 276, 344, 357, 374, 567 private seller test 10, 211 privatisation 209, 213, 313, 315, 409, 462, 560–61 procedural autonomy 362–63, 367–68, 371, 481 producer subsidies 97–98, 110 producers, electricity 72–73, 77, 489, 493, 495, 497, 537, 540–41 production 7–8, 26, 40–41, 43, 115, 399–400, 426–27, 492–94 electricity 20, 83, 216, 221, 279, 388–89, 394, 426 profitability 220, 258, 267, 498, 537, 548–50 profits 8, 12, 36–37, 128–29, 165, 231–32, 285–86, 502 expected 499–500, 502, 507 reasonable 10, 17, 284–85, 287, 296–97, 565 project promoters 240–42, 244–48, 250, 265 Projects of Common Interest (PCIs) 235–69, 543; see also TEN-E consequences of status 245–52 examples 243–45 implementation and delisting 242–43 prolongations 43–44, 405, 467, 554 promoters, project 240–42, 244–48, 250, 265 proportionality 55–56, 80, 82, 121–22, 216, 259, 282, 410–11 test 215, 290–91 protection of legitimate expectations 307, 309, 320–21, 328–33, 338, 468–69 provisional measures 318, 324, 379, 384, 390 PSOs, see public service obligations public service compensation 233, 272, 286 public service obligations (PSOs) 16–17, 169, 275–78, 281–84, 286, 288–92, 296–98, 422–23 public services 27, 120–21, 271, 405, 416, 565 public support 29, 58, 217, 267, 274, 286, 508 public undertakings 9, 11, 22–23, 293, 530, 567 purchase obligations 17–18, 20, 72–74, 76–78, 108, 388, 412–13 quid pluris factor 275–76 radioactive waste 202–3, 212 rate of return, reasonable 165, 167, 259, 549 R&D 104, 251, 262, 264, 267, 541, 544 reasonable profits 10, 17, 284–85, 287, 296–97, 565 reasonable rate of return 165, 167, 259, 549 reasoned opinions 33, 202, 410 rebates 35, 441–43, 449 tax 66, 441–44, 447–50 recovery aid 301–25, 362, 371–72, 383–84, 461, 470, 476, 481–84 case law illustration of principles 307–17 damages 464, 473

585

decisions 301, 312–13, 358–60, 384, 481–82, 524 effective 314, 324, 483 forms 317–25 general principles 303–7 Greece 481–84 obligation 311–12, 314, 465, 483 orders 372–73, 382, 469–70 procedures 305, 319–20, 482 purpose 304, 467 RECs (renewable energy certificates) 485–87 reference framework 66–67, 69 reference geographic area 25 reference prices 54, 437, 532 reference systems 24–25, 69 referring courts 205, 282, 311, 324, 380 reforms 125, 162, 176, 446–48, 491, 494–501, 503, 506–9 legislative 496, 502–3, 506–8 market 147, 162, 173, 177 regional governments 107, 499–501, 513 regional lists 241–42 regulated costs 504, 506, 511 regulated electricity tariffs 42, 404, 510, 562 regulated expenses 493, 495–97, 504, 512 regulated infrastructure 189–90 regulated interconnectors 189–90 regulated prices 31, 59, 483, 561 France 403–12 regulated tariffs 54–55, 281, 404–8, 411, 492–95, 509, 562 regulators 56, 128–31, 179, 241, 420 domestic 56–57, 240, 570 regulatory distortions 157, 176 regulatory expropriation 337, 339 regulatory failures 154, 162, 168, 175–76 regulatory frameworks 235–36, 248, 262, 273, 333, 453, 529 reimbursement 86, 304, 317, 413–14, 440, 442, 460, 462 relevant markets 274, 298, 332, 481–82 reliability 146, 153, 157–58, 161–62, 166, 180, 192 remuneration 184, 186, 188, 231–32, 492–93, 497–98, 502, 504–5 market-based 14, 397, 540 systems 280, 495, 498–99, 501, 503 renewable electricity 73, 77, 85, 256, 426, 431–32, 434–35, 538–39 renewable energy 18–19, 51–52, 70–71, 97, 103–5, 494–99, 505–6, 537–39 certificates, see RECs compatibility 113–43 France 412–16 Germany 388–91 levies or surcharges to finance 84–87 penetration 126, 129, 133, 135 producers 72, 74, 84, 86, 498–99, 504, 537, 539 projects 70, 257, 387, 427, 534

586 Index sources (RES) 17, 85–86, 113–43, 243–44, 278–79, 399–400, 436–37, 494–95 Spain 494–510 subsidies 92, 103–5, 110, 505 support 129, 134–35, 140–41, 387–88, 398–400 renewable surcharges 85, 415 renewables, see renewable energy repayment 14, 304, 311, 313–14, 317–18, 323, 362, 378 RES, see renewable energy, sources resale of concession power 547, 551–52 rescue and restructuring aid 206, 214, 218 reserves 53, 139–40, 151, 162, 167, 174, 212, 275 strategic 145, 149, 162, 173–74, 177, 191, 490 residential customers 32, 54–56, 141 Residex 310–11 residual market failures 176, 257 resources 18–22, 57–63, 209, 239, 251, 253–54, 322, 464–65 natural 539–40, 543 public 17, 21, 418 restructuring, plans 210, 214, 219 retail markets 154–55, 424–26, 455, 531, 533 retail prices 31–32, 155, 298, 422, 454–56 retroactive effects 412, 432, 500, 502–3, 506–7 retroactivity 501, 503, 508 return tariffs green 54–55, 405–7 yellow 406 revenues 20–21, 69–70, 101–2, 139, 195–96, 199, 405–6, 493–95 revocation 342, 347, 350–52, 529 rights effective 360, 367, 377 individual 360, 363–64, 377, 464, 484, 500, 502–3 subjective 302, 464, 473 transmission 195–96 risk 125–26, 128–29, 131–33, 167–68, 199, 209–10, 212–13, 547–49 business 258, 267, 334, 550 of distortion 142, 173, 533 fees 206–7, 209, 218 financial 151, 193, 264 overcapacity 137, 268 overcompensation 171, 217, 230, 437 political 336 premiums 129, 207, 212, 535 relocation 43, 46–48 simultaneous scarcity 187, 193–94 Romania 221, 229, 336, 340–42, 346–50, 353, 509 SAAP, see State Aid Action Plan SAAs (Stabilisation and Association Agreements) 556–57, 559 safe management of nuclear liabilities 202–3, 214, 218–19

safeguards 129, 132, 359–60, 363, 370, 373, 377–79, 475 safety 170, 203, 214, 219, 228–29 sale or supply of assets or services 16 SAM (State aid modernisation) 113, 120, 122, 206, 213, 521 sanctions 60, 62, 116, 275–76, 304, 325, 349 administrative 58, 60, 62 Sardinia 39–40, 42, 47, 464–66 savings 34–35, 46, 50, 455 scarcity 187–88, 191–94, 196 coincident 188, 193–94, 199 simultaneous 187–88, 193–94 SDE-subsidy 427, 431–37 SDE-subsidy scheme 427, 431–33, 435–37 sector inquiries 146–47, 149–50, 152, 155–57, 159, 161, 163–64, 172–73 security 176, 179, 239, 277, 281–82, 289, 294, 417 interests 294–95, 454 of supply 146–47, 179–82, 185–88, 191–95, 198–99, 238–39, 277–82, 289 selection criteria 236–38, 290 selective advantage 21, 25, 67–68, 172, 211–13, 415, 419, 422 selective measures 25, 66, 68, 394, 489 selectivity 23–25, 56, 65–66, 68–69, 211–12, 392, 406, 505 criterion 7, 23–26, 65–72, 483 self-consumption 491, 512–13 Serbia 558, 562–63, 565, 568 service providers 27, 272, 442–44, 446–48, 450 services postal 271 public 27, 120–21, 271, 405, 416, 565 services of general economic interest, see SGEIs SGEIs (services of general economic interest) 15–16, 29, 120–21, 211, 221, 232–33, 271–98, 564–65 and aid 283–90 and compatibility with internal market 290–98 definition 273–82 shareholders 12–13, 209, 312–13, 427–29, 462 shortages, capacity 151, 173–74, 231 short-term markets 151, 153, 155, 157, 176 simultaneous scarcity 187–88, 193–94 sincere cooperation 349–50, 360–62, 365–66, 480 single market 199, 271, 283, 353, 355, 415, 466–67, 515–16 Slovakia 209, 213, 216, 224, 226, 322–23 Slovenia 220, 228, 251 small- and medium-sized enterprises, see SMEs smart grids 235, 238–39, 241–43, 250–51 SMEs (small- and medium-sized enterprises) 26, 66, 251, 259, 405 social bond 491, 510–11 social costs 228–29 social welfare benefits 199, 228 solar energy 106, 130, 132, 151, 534

Index SoS, see security, of supply Spain 139, 145, 149, 221, 224, 226, 287–88, 491–513 Constitution 499–500, 502–3 courts 491, 499–501, 504, 509–11 government 491, 499, 506–7, 511 legislation 492, 495, 510–11 national legal framework 491–94 public debt 498, 500 regions 501, 504, 513 renewable energy 494–510 special subsidised regime 501 specificity 94, 96–98, 101–2, 105, 110 spot market 77, 85, 289 Stabilisation and Association Agreements, see SAAs stakeholders 145, 155, 185, 226–27, 240, 248, 479 standard tariffs 40, 54, 406 standing, legal 363, 367–69, 501, 567 standstill, notification (and standstill) obligation 357, 360–61, 363, 375–76, 441–46, 475–78, 480, 524 State aid, see aid and Introductory Note State Aid Action Plan (SAAP) 120, 122, 213 State aid modernisation, see SAM State assets 292, 296 permanent 22 State budgets 232, 416, 538, 561 State control 18, 59–61, 75, 87, 232, 389 State guarantees 14–16, 22, 209, 217, 220, 547, 565–66, 568 State liability 461, 523 State origin of resources 59–60, 63 State resources 17–22, 57–65, 70–78, 231–32, 252–54, 406, 504–5, 561–62 directly or indirectly granted through 18–22 energy sector 57–64 transfer 20, 57, 65, 72, 78, 470, 485 indirect 17–18, 72, 388–89 State-owned companies 12, 43, 45, 545, 548, 561, 563–66, 569–70 Statnett SF 527, 529–30, 533, 537 storage 208, 214, 216, 235, 237, 256, 261, 544–45 projects 238, 242, 544 stranded costs 45, 48, 220, 426–31, 456–57, 489, 493, 495 strategic reserves 145, 149, 162, 173–74, 177, 191, 490 stress situations 151, 158, 160, 191 StromNEV 75–76, 78, 87 subjective rights 302, 464, 473 subsidiaries 12, 39, 72, 506, 549 subsidies 6, 91–96, 98–110, 433–35, 437–38, 493–95, 499, 504–5 actionable 94, 106–8 consumer 97–99, 103 energy 92, 97, 103–5, 110, 559–60, 564 fossil fuel, see fossil fuels, subsidies non-actionable 94–95, 108–10

587

producer 97–98, 110 renewable energy 92, 103–5, 110, 505 substitutable technologies 165, 171 successor schemes 18, 388 sui generis decisions 313 supervision 64, 358, 527, 529 suppliers 72, 77, 404–5, 408–9, 417, 420, 492–93, 560–62 electricity 64, 72–73, 75, 77–78, 85, 169, 405, 537–38 supply electricity 45, 51, 161, 171, 174, 179, 417, 422 gas 33–34, 36, 281, 404, 411 security of 158, 179–82, 185–86, 191–95, 198–99, 238–39, 277–78, 280–82 support 124–25, 127–29, 131–36, 166, 250–51, 253–55, 398–400, 504–5 efficient 135–36 financial 210, 229, 244, 248, 264, 268–69, 457, 461 public 29, 217, 267, 274, 286, 508 supreme courts 352, 387–88, 440, 463–64, 470, 473, 501–5, 509–11 surcharges 65, 70, 72–78, 84–87, 415, 445–46, 485–86, 549–50 compulsory 75–76, 78 renewable 85, 415 special 75, 87 suspension 324, 371, 378, 480 sustainability 238–39 financial 496 sustainable energy 103, 165, 237, 251, 389 Sweden 81, 83, 148–49, 265–66, 509, 529–31, 537–39, 541 Switzerland 352, 531 system failure 117–18 system flexibility 238–39 systemic failures 261, 267 targets efficiency 296–97 emissions 230, 232 tariff deficits 64, 494–98, 500, 503 tariff structures 33–34, 36, 455 tariffs 33–38, 40–48, 54–56, 104, 406–7, 410–11, 465–67, 472 electricity 42–43, 47, 404, 465–67, 473, 480, 510, 562 feed-in 86, 91, 97–98, 113, 124, 126–30, 253, 478 preferential 32–38, 40–47, 49, 51, 55, 465–66, 468, 479–80 regulated 54–55, 281, 404–8, 411, 492–95, 509, 562 standard 40, 54, 406 tax derogations 66–70 tax exemptions 22, 25, 65–66, 68–70, 82–83, 425–26, 439–40, 469–70 tax rebates 66, 441–44, 447–50 tax reductions 79–83, 258, 414, 416

588 Index taxes 49, 64–90, 368, 439–40, 497, 507, 509, 520 corporate 12, 67, 541 electricity 83, 547, 551 environmental 79–80, 82–83, 416, 439, 447 harmonised 79–84 non-harmonised 79–80, 82–83 nuclear 207, 488–89 technological neutrality 125, 133, 165 technologies innovative 127, 136, 552 mature 125, 129, 131–33, 137, 275, 279 new 124, 135, 275, 536, 552–53 substitutable 165, 171 telecommunications 251, 271, 395, 453 tender amounts 434–35 tenderers 285, 289–90, 422, 434 tenders 129, 133, 137, 276, 312, 422, 434, 561 competitive 127, 129–31, 141, 143 TEN-E 236–43, 245–50, 252, 255, 257, 261–63, 268–69; see also Projects of Common Interest termination 211, 276, 332, 338, 341, 479–80 Terni 43–44, 48, 466 territorial restrictions 180–81 tests 10–11, 96, 104, 110, 126, 282, 286, 298 balancing 120–21, 123, 125–26, 132, 165, 167, 213, 546 market economy 10, 53 market investor 457, 537, 548, 550, 564 private creditor 10, 14–16 private seller 10, 211 proportionality 215, 290–91 third parties 302, 313, 331–32, 354–55, 369, 375, 445, 475–76 ThyssenKrupp 43, 321, 466–67, 472 tickets 195–96, 198 trade 26, 28, 92, 122, 290–91, 407, 418–19, 423 and competition 26, 28, 122, 126, 407, 419, 421, 423 effect on 27–28, 260, 519, 533, 536 Trade-Related Investment Measures, see TRIMS trading conditions 115, 215, 255 trans-European energy infrastructure 236–37, 239, 252, 255, 268–69 transfer 43, 59–60, 102, 209, 215–16, 312–14, 319, 467 of State resources 20, 57, 65, 72, 78, 470, 485 transit 61–62, 64 transmission 29, 33, 49, 159–60, 244, 483, 492–93, 511 transmission costs 49 transmission rights 195–96 transmission systems operators (TSOs) 18–19, 64, 75–78, 85, 172, 174, 189–93, 395–96 transparency 127, 154–55, 246–47, 293, 420 financial 293 obligations 260, 268, 437 transparent criteria 16, 80, 82, 85

transport 98, 123, 251, 261, 271, 492–93 carbon dioxide 237, 239–41 transposition 23, 566–68 treatment equal 442–43, 448, 450 equitable 328, 333, 336–39, 341 preferential 89, 466 regulatory 245, 247 treaty obligations 566–67 TRIMS (Trade-Related Investment Measures) 106–7 TSOs, see transmission systems operators TVO 209, 211, 215 UK, see United Kingdom Ukraine 556–59, 567–68 unbundling 160, 189–90 uncompetitive coal mines 225–29 under-investment 124, 150 undertakings, beneficiary 9, 19, 28, 406, 438, 564 undistorted competition 190, 482–83 undue negative effects on competition and trade 122, 183, 216, 421, 423 United Kingdom (UK) 83–84, 133, 148–49, 166, 168, 206–7, 211–14, 217–18 authorities 84, 209–11, 213, 216–17, 276 courts 349–51 government 166, 206–8, 210–11, 214–15, 218–19, 221, 230, 279 United States 95–96, 99–100, 104, 145, 222, 230, 352, 355 unlawful aid 308–9, 315–17, 321–24, 372–73, 376–77, 460–61, 470–71, 524 unlawful implementation 359–60, 378 unlawful retroactive effects 500, 502–3 unlawfulness 302, 307, 316–17, 373, 383, 445, 459 upfront payments 296–97, 547 US, see United States use of State resources 46, 57–59, 418, 483 validity 308, 317, 353, 362, 380, 390, 461 variable costs 38, 47–48, 207, 288, 488 VAT 81, 225, 520, 553–54 Vent de Colère 57–59, 61–63, 74, 78, 403, 412–14, 487, 504 virtual power plants 41, 56 vulnerable customers 455, 562 waste 206–7, 212, 214, 216, 218, 497 radioactive 202–3, 212 white clauses 114, 118, 120, 123, 132 wind energy/power 58, 61, 106, 132, 244, 412–13, 495, 539–41 wind farms 244, 432, 434–35, 437 offshore 143, 244, 434–37, 485 World Trade Organization, see WTO WTO (World Trade Organization) 91–110 fossil fuels and renewables 97–109 regulation of subsidies 92–97