European Competition Law Annual 2011: Integrating Public and Private Enforcement of Competition Law - Implications for Courts and Agencies 9781474201797, 9781849463515

This volume contains the written contributions and oral proceedings of the “16th Annual EU Competition Law and Policy Wo

221 92 9MB

English Pages [623] Year 2014

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

European Competition Law Annual 2011: Integrating Public and Private Enforcement of Competition Law - Implications for Courts and Agencies
 9781474201797, 9781849463515

Citation preview

SPONSORS OF THE E U C O M P E T I T I O N LAW AN D POLICY WORKSHOP Baker Botts LLP Contact: James Rill, Esq. 1299 Pennsylvania Ave., NW Washington, DC 20004 USA Tel.: +1 202 639 78 83 Fax: +1 202 639 78 90 E-mail: [email protected] Blake, Cassels & Graydon LLP Contact: Calvin S. Goldman Commerce Court West 199 Bay Street Toronto, Ontario Canada M5L 1A9 Tel.: +1 416 863 22 80 Fax: +1 416 863 26 53 E-mail: [email protected] Cleary, Gottlieb, Steen & Hamilton Contact: Mario Siragusa Rome Office Piazza di Spagna, 15 I-00187 Rome Italy Tel.: +39 06 695 221 Fax: +39 06 692 00 665 E-mail: [email protected] Compass Lexecon Contact: Atilano Jorge Padilla Davidson Building 5 South Hampton Street B-1000 London WC2E 7HA Tel. + 44 20 7632 5005 Fax: + 44 20 7632 5155 E-mail: [email protected]

vi  Sponsors CRA International Contact: Cristina Caffarri 99 Bishopsgate London EC2M 3XD UK Tel.: +44 20 7664 3700 Fax: +44 20 7664 3998 E-mail: [email protected] Hengeler Müller Contact: Jochen Burrichter Benrater Strasse 18-20 D-40213 Düsseldorf Germany Fax: +49 211 83 04 222 E-mail: [email protected] Martínez Lage, Allendesalazar & Brokelmann Contact: Santiago Martínez Lage Claudio Coello 37 ES-28001 Madrid Spain Tel.: +34 91 426 44 70 Fax: +34 91 577 37 74 E-mail: [email protected] RBB Economics Contact: Andrea Lofaro RBB Economics London The Connection 198 High Holborn London, WC1V 7BD UK Tel.: +44 20 7421 2414 Fax: +44 20 7421 2411 E-mail: [email protected] Skadden, Arps, Slate, Meagher & Flom LLP Contact: James Venit Brussels Office 523 Avenue Louise B-1050 Brussels Belgium Tel.: +32 2 639 03 00 Fax: +32 2 639 03 39 E-mail: [email protected]

Sponsors  White & Case LLP Contact: Ian S Forrester Brussels Office 62, rue de la Loi B-1040 Brussels Belgium Tel.: +32 2 219 16 20 Fax: +32 2 219 16 26 E-mail: [email protected] WilmerHale LLP Contact: John Ratliff Bastion Tower Place du Champ de Mars/ Marsveldplein 5 B-1050 Brussels Belgium Tel.: +32 2 285 49 08 Fax: +32 2 285 49 49 E-mail: [email protected]

vii  

16th ANNUAL EU C OMP E T IT ION L AW AND P OLIC Y W OR K S HOP : INTEGRATING PUBLIC AND PRIVATE ENFORCEMENT IMPLICATIONS FOR COURTS AND AGENCIES

Organizers Philip Lowe and Mel Marquis Contributing authors Enno Ahlenstiel Donald Baker Jochen Burrichter Horst Butz

Hengeler Müller, Düsseldorf Baker & Miller PLLC, Washington, D.C. Hengeler Müller, Düsseldorf Landgericht Düsseldorf and Heinrich Heine University, Düsseldorf Scott Campbell Stewarts Law LLP, London Brian Facey Blake, Cassels & Graydon LLP, Toronto Tristan Feunteun Stewarts Law LLP, London Ian Forrester White and Case, Brussels; Glasgow University Andrew Foster Skadden Arps, Brussels Andrew Gavil Howard University, Washington, D.C. (Now U.S. Federal Trade Commission) Barry Hawk Fordham University School of Law, New York James Keyte Skadden Arps, New York Assimakis Komninos Competition Commission, Athens; University College London (now White and Case, Brussels) Bruno Lasserre Competition Authority, Paris Frédéric Louis WilmerHale, Brussels Mel Marquis European University Institute, Florence Veljko Milutinovič European University Institute, Florence Luis Silva Morais Paz Ferreira, Lisbon; University of Lisbon Tom Ottervanger Allen & Overy, Amsterdam; University of Leiden Silvia Pietrini Université Lille 2, Droit et Santé, Centre Réné DemogueCRDP Mark Powell White and Case, Brussels John Ratliff WilmerHale, Brussels J Thomas Rosch U.S. Federal Trade Commission

xii  List of Participants David Rosner Mario Siragusa James Venit

Blake, Cassels & Graydon LLP, Toronto Cleary Gottlieb Steen Hamilton, Rome Skadden Arps, Brussels

Mel Marquis*

Perchance to Dream: Well Integrated Public and Private Antitrust Enforcement in the European Union

This volume presents contributions prepared for the 16th edition of the Annual EU Competition Law and Policy Workshop, held on 17–18 June 2011 at the European University Institute in Florence. The 2011 Workshop was devoted to issues pertaining to the development of private antitrust litigation in the EU, with an accent on the interplay between private antitrust litigation and public enforcement of the competition rules by European authorities. This was the second of the workshops I have organized with Philip Lowe, and it marks an organizational change in the sense that we conducted the event under the auspices of the EUI’s Law Department, where our activities now continue in collaboration (since 2012) with a third co-director, Professor Giorgio Monti. Aside from the usual lag between the time our sessions convene and the time the proceedings are published, I am also responsible for holding up the production of this volume until the European Commission came forward in June of 2013 with a package whose measures include plans for a Directive in the field of ‘certain rules governing actions for damages under national laws for infringements of the competition law provisions of the Member States and of the European Union’ (see Annex I). This draft Directive, an unprecedented legal instrument in this field of law, is currently being scrutinized, and its more sensitive provisions are quietly being amended, by the European Parliament and by the governments of the EU Member States. Nevertheless, the basic contours of the way forward can be seen, and we now proceed with the present publication, a collection of analyses which shed abundant light on our subject. The Commission’s draft legislative text and flanking measures – which include * Part-time Professor of Law, European University Institute, Florence; Professore a contratto, LUMSA University, Rome. Co-Director, EU Competition Law and Policy Workshop; Co-Director, Rome Antitrust Policy Forum. I have benefited from the kind support of numerous individuals who can’t all be named but Philip Lowe, Giorgio Monti, Claus Ehlermann and our illustrious program sponsors are certainly among them, as are the many gifted scholars and practitioners that joined in our discussions and contributed the chapters herein. On specific points I have benefited from comments kindly provided by Nicholas Khan. (As if it had to be added, neither he nor his institution is engaged by any of the views expressed in this chapter.) I have also benefited from the views exchanged in a workshop I co-organized on 5 October 2013 with Hans Micklitz, Giorgio Monti and Andreas Schwab, entitled ‘Shaping Private Antitrust Enforcement in the EU’. I would like to thank all the speakers who joined us on that occasion as well. A collection of essays based on the latter event is under construction and will be published.

xiv  Mel Marquis a Recommendation on collective redress and other non-binding instruments addressing the quantification of harm in private damages actions – have been taken into account to the extent possible given the subsequent scurrying to get the product to market. My own brief conclusions regarding the Directive-tobe and the Recommendation on collective redress appear at the end of this chapter. The Commission’s package represents an early milestone in the course of a longer discussion that will develop over many decades to come and which was launched by Claus Ehlermann and his Workshop guests in 2001 while anticipating the judgment of the Court of Justice in the seminal Courage case.1 Further reflections will undoubtedly follow in our forum once the empirics lend themselves to more analysis and lesson-learning.

Background The EU Competition Law and Policy Workshop is an ongoing program that explores topical policy and enforcement issues in the area of competition law and economics.2 Each year the Workshop brings together a group of top-level EU and international policy makers, judges, legal practitioners, economic experts and scholars to take part in intensive debates that explore specific competitionrelated issues in an informal and non-commercial environment. A chief aim is to stimulate critical reflection on the part of both the Workshop participants and the broader public.

Structure of the Book The present volume consists of five main parts. The first four correspond to the organization of the discussions at our meeting, while the fifth is something of a bonus feature that looks both backward and forward. The five parts are as follows: 1 See Claus-Dieter Ehlermann and Isabela Atanasiu (eds), European Competition Law Annual 2001: Effective Private Enforcement of EC Antitrust Law (Hart Publishing, 2003). See also ClausDieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2006: Enforcement of Prohibition of cartels, Hart Publishing, 2007, 425-485. 2 In this series, which is published by Hart of Oxford (see http://www.hartpub.co.uk/SeriesDetails. aspx?SeriesName=European+Competition+Law+Annual), we have covered (to name only a few topics) the abuse of dominance, the application of competition rules in intellectual property scenarios, the prohibition against cartels, cartel settlements and commitment decisions, judicial review by the EU Courts and the evaluation of evidence, and merger control from the perspective of various jurisdictions worldwide. Two forthcoming works will address: the relationship between competition policy and many other public policies; and the dilemma, figuratively speaking, of enforcing competition law in a way that is both effective and legitimate.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xv  

(1) Designing a balanced system: damages, deterrence, leniency and litigants’ rights; (2) Integrating public and private enforcement in Europe. Legal and jurisdictional issues; (3) Options for collective redress in the European Union; (4) Drawing lessons and conclusions; and (5) Private damages claims and the elusive future. The remainder of this introduction previews the written contributions in this book and provides related commentary to add context, draw connections between the various papers, or interject personal views or musings. As I’m always at pains to stress, the introduction is no substitute for the other contributions, not least because I occasionally pass over some of their contents due to the already exaggerated loquacity of the remarks that follow.

Part I

Designing a Balanced System: Damages, Deterrence, Leniency and Litigants’ Rights

Andy Gavil: ‘Designing Private Rights of Action for Competition Policy Systems: The Role of Interdependence and the Advantages of a Sequential Approach’. In this chapter, Gavil considers the relative merits of establishing a private enforcement system by way of comprehensive legislation versus a ‘sequential’, or incremental approach that introduces a few elements of the system but leaves space for organic growth. Gavil’s suggestion is that it is risky to usher in a seemingly complete apparatus all at once. The argument underlines the interdependence of the many threads woven into a mature system of private antitrust enforcement. The interplay of those threads – for example, in terms of litigation incentive structures, cost rules, locus standi criteria, evidentiary disclosure mechanisms, the role of expert economic evidence, manageable standards of proof, effective judicial systems, adequate remedies, etc. – is bound to produce results that are practically impossible to predict ex ante. This unpredictability arguably points to the superiority of a more humble approach in which foundations are laid for gradually building experience and making periodic adjustments after cycles of trial, error and observation. One concern is the perception in the United States (not always grounded in fact) that private enforcement has run amok. As others such as Stephen Calkins and Bill Kovacic have often noted, if some elements of the antitrust enforcement ecosystem are perceived to be imbalanced, other elements will be ‘re-equilibrated’ in various ways, including by way of procedural devices, or by re-interpretation of substantive law, or by the sweeping conclusion that antitrust is, in light of its social costs, simply an inferior institution best left on the shelf in certain economic regulatory contexts. Such instances of ‘backlash’ can

xvi  Mel Marquis then potentially spill over to other areas, not just making it more difficult to file private claims but also indirectly harming public enforcement efforts, despite the quite different objectives driving private enforcement on the one hand and public enforcement on the other.3 But ironically, the United States does not only provide a cautionary tale of what can happen if the excesses of private antitrust litigation provoke too much reequilibration; Gavil also emphasizes that the development of private enforcement in the United States epitomizes the sequential approach he advocates. In this regard a historical overview is presented to remind the reader that although the much-reviled treble damages provision in U.S. federal antitrust law (15 U.S.C. § 15, which also allows recovery of costs and reasonable attorney fees) dates back to 1890,4 it was not until the 1960s and 1970s that crucial pieces of the private litigation and class action framework were put in place.5 Gavil does not describe a developmental process of error and adjustment, and there may be doubt about whether the sequential experience in the U.S. produced an outcome superior to some counterfactual and hence unobservable big-bang approach. Nevertheless, if the U.S. experience has thus far been unsatisfactory, it does not follow that sub-optimal incrementalism in the U.S. must be echoed by sub-optimal incrementalism in Europe. Gavil’s advice on taking sequential steps appears to have been taken to heart by the European Commission, whose proposed Directive published in June 20136 is by no means an exhaustive, comprehensive prescription for national antitrust litigation. Several elements of the future private enforcement system seem to have ample space to develop over time, and the construction of the edifice will surely proceed in an iterative manner, not only because the proposed 3 Gavil points out, though, that the range of options considered in Europe to promote private litigation steer wide of system features commonly regarded as being (singly or in combination) pernicious in the U.S. system, such as contingency fees, automatic and burdensome discovery rules, the right of plaintiffs to recover legal fees, lay juries and so on. Consequently, Gavil considers that the risk of a backlash in Europe similar to that seen in the U.S. federal courts is attenuated. 4 See, eg, Donald Baker, ‘Rewriting History – What Have We Learned About Private Antitrust Enforcement That We Would Recommend To Others?’, 16 Loyola Consumer Law Review 379– 408 (2004), at 379–382; David Gerber, ‘Private enforcement of competition law: a comparative perspective’, in Thomas Möllers and Andreas Heinemann, eds., The Enforcement of Competition Law in Europe, Cambridge University Press, 2007, pp 431–451, at 434; Wouter Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for Damages’, 32 World Competition 3–26 (2009), at 14. 5 Neither the Sherman Act nor the Clayton Act provides for class action litigation. The non antitrustspecific possibility of aggregating claims in a class action was introduced, as a continuation of an earlier rule applied by equity courts, in a federal rule of civil procedure (FRCP 23) in 1938. See Harry Kalven, Jr. and Maurice Rosenfield, ‘The Contemporary Function of the Class Suit’, 8 University of Chicago Law Review 684–721 (1941). (As result of later amendments in 1966, which introduced the ‘opt out’ principle, Rule 23 came of age, as it were, assuming its principal modern features.) Aggregated claims in the U.S. are traced back to West v Randall, 29 F. Cas. 718 (R.I. 1820). However, far earlier antecedents from medieval England are well documented. See Stephen Yeazell, From Medieval Group Litigation to the Modern Class Action, Yale University Press, 1987. 6 The proposed Directive is included as Annex I to this volume. By the time the Council of Ministers and the European Parliament are done amending the proposal, the original proposal will of course be outdated. The Annex is nevertheless deliberately included for purposes of comparison and in order to provide historical context to inform future research.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xvii  

directive is an instrument of minimal harmonization but also obviously because, notwithstanding the Commission’s current soft law efforts (taking the form of a Recommendation addressed to the Member States), binding harmonization has not yet been attempted in the politically delicate and legally complex field of collective redress.7 Tom Ottervanger: ‘Designing a balanced system: Damages, Deterrence, Leniency and Litigants’ Rights’. This essay presents the view, held by many in Europe, that it is not appropriate to conceive of private antitrust litigation as furthering traditional objectives of public enforcement, such as, in particular, deterrence and deterrence-induced compliance.8 Ottervanger acknowledges that marginal gains in deterrence might flow as a by-product of more robust damages claims, but he stresses, among other things, that the interests of private claimants do not coincide with the public interest, which is more properly the lodestar of the public enforcer. Private claimants care little, for example, about the future consequences of the jurisprudence to which they contribute today, so long as they succeed in vindicating their (generally economic) rights. This line of reasoning seems to lead to the conclusion, espoused by Ottervanger, that it is not entirely appropriate for the EU legislator to ‘design’ a system which encourages antitrust damages claims, or which ‘integrates’ such litigation with public enforcement, since integration may imply cumulative application. Thus, recalling the premises guiding the adoption of Regulation 1/2003, it is suggested that the empowerment of national courts, in particular by making Article 101(3) directly applicable, should result in private litigation substituting for the use of public resources to investigate and punish unlawful conduct; but that subjecting defendants first to public exposure (ending in fines) and then to private exposure (ending in payment of damages or in a costly settlement) is both inconsistent with the original aims of the reform and apt to constitute ‘disproportionate double deterrence’.9 From 7 For discussion of collective redress in the European context, see the contributions to this volume by Mario Siragusa, by Silvia Pietrini and by Bruno Lasserre; and Roger Van den Bergh, ‘Private Enforcement of European Competition Law and the Persisting Collective Action Problem’, 20 Maastricht Journal of European and Comparative Law 12–34 (2013). For extended discussions, see Silvia Pietrini, L’action collective en droit des pratiques anticoncurrentielles. Perspectives nationale, européenne et internationale, Bruylant, 2012; Sonja Keske, Group Litigation in European Competition Law: A Law and Economics Perspective, Intersentia, 2010. See also Jurgen Backhaus, Alberto Cassone and Giovanni Ramello, eds., The Law and Economics of Class Actions in Europe. Lessons from America, Edward Elgar, 2012. 8 On this theme, see also, eg, Wouter Wils, ‘Should Private Antitrust Enforcement Be Encouraged in Europe?’, 26 World Competition 473–488 (2003). 9 For further discussion of the poor coordination (eg, given the lack of ‘equalization’ mechanisms) between public and private enforcement, see Michael Frese, ‘Fines and Damages under EU Competition Law – Implication of the Accumulation of Liability’, Amsterdam Center for Law & Economics Working Paper No. 2011-05, with references; Damien Geradin and Laurie-Anne Grelier, ‘Cartel Damages Claims in the European Union: Have We Only Seen the Tip of the Iceberg?’, http://ssrn.com/abstract=2362386 (2013). The idea that adding civil liability to fines imposed by the European Commission leads to disproportionate combined exposure seems to presuppose that the fines imposed by the Commission are sufficiently deterrent. However, even at their infamously high level, it seems that EU fines frequently fall short of their optimal levels. See Mario Mariniello, ‘Do

xviii  Mel Marquis this perspective, Ottervanger argues that follow-on claims for damages are ‘not some form of policy instrument for attaining general interest objectives through further deterrence. They are simply tort actions.’10 A logical implication of this remark would seem to be that, since tort litigation traditionally (though not exclusively) belongs to the domain of national private law and national civil procedure, legislative action in this field at the EU level should remain minimal. Indeed, even as a matter of policy the Commission is criticized for including, in its press releases, the standard coda that informs potential victims of antitrust infringements of the EU-law right to seek damages before national courts. One could not take issue with the proposition that private litigants engage in self-serving behaviour (which in the worst cases may consist of seeking and possibly obtaining anticompetitive judicial interventions), and that they pursue the public interest only when their individualized decision tree tells them they should, ie, when the public interest and their own interests converge. There is also some appeal in the prevalent argument that the instruments of public enforcement and private litigation should not be confused or conflated, and that the objectives of deterrence and compensation should be kept separate. However, private litigation is not solely private; it occurs in a highly developed judicial system in which the public interest is represented and guarded in various ways. As a matter of EU jurisprudence, moreover, the Court of Justice has not hesitated to state that private claims to enforce the directly effective antitrust rules of the Treaty before national courts contributes to the effectiveness of those rules; from the Court’s point of view, private litigation thus transcends purely private interest.11 It also European Union fines deter price-fixing?’, Bruegel Policy Brief 2013/04, May 2013. Having said that, the ‘optimal fine’ theory that animates EU fines seems unlikely to secure optimal compliance. To state the argument briefly, a key employee deciding whether to collude can rationally expect to face no adverse consequences at all (since few cartels are ever unearthed) or to face them only after years of delay, by which time she will probably be long gone or dead. (The period of delay, if the cartel is exposed, will be the lifetime of the cartel plus years of investigation time – see Mariniello, ibid, referring to 10–20 years of lag from the moment of decision to the moment of sanction, if any. One may add that even in a scenario where an undertaking is punished, shareholders may not sanction responsible individuals in a manner that promotes adequate compliance, making the link between public fines and compliance more tenuous still.) The weaknesses in the links between public fines and conformity with antitrust rules may also suggest that, at least with respect to certain kinds of cases (eg, follow-on cartel damages actions), there may be only a weak link between the prospect of exposure to civil damages and compliant behavior, and by the same token there may be only a weak link between the combined effect of public and private liability and compliance. 10 See page 21. (Note: cross-references made in this chapter to specific page numbers are generally accurate but a slight shift in pagination may have occurred due to typesetting.) 11 The latter two points are espoused, for example, in Assimakis Komninos, ‘The Relationship between Public and Private Enforcement: quod Dei Deo, quod Caesaris Ceasari’, this volume, pages 143–144. Cf. also Andreas Reindl, ‘The European Commission’s Package on Private Enforcement in Competition Cases’, CPI Antitrust Chronicle, August 2013 (1), for example at page 4 (‘According to Pfleiderer, deterrence consideration[s] should […] be of overriding importance when shaping the rules governing the private litigation regime.’); and Andreas Reindl, Secretariat Note in OECD Roundtable on Private Remedies 2007 (2006), DAF/COMP(2006)34, 8–20, at 10. On the other hand, while it seems clear enough that the ‘right to damages’ as it emerged in Courage v Crehan was mainly driven by the imperative of ‘effectiveness’, and while Pfleiderer pointedly seems to underline this, it has also been argued that Courage, when read ‘in the light of the wider case law of the Court of Justice’, is in

Well Integrated Public and Private Antitrust Enforcement in the EU 

xix  

seems a bit idealized to say (as some do, although Ottervanger stops short of saying it) that deterrence simply has no role in the context of private litigation. The matter is rather more nuanced than the mere observation that deterrence may be an unintended side-effect of private actions that lead to compensation, and that the relationship goes no further. The more realistic way to look at the issue is to ask why compensation mechanisms exist. Do they exist purely for the sake of compensation? Maybe, but if they somehow signal that a system of law will not rest until corrective justice is done (or until the statute of limitations runs), and if that signal alters the incentives of some marginal subset of a larger group of actors that could potentially have to pay compensation, such that some instances of tortious behaviour are not undertaken in the first place, it is legitimate to recognize that the system of private litigation has wider public benefits. By making collusion more expensive, a margin of such behaviour is avoided, which likewise avoids on some marginal basis the need to be compensated. A private litigation system does not only protect those wronged in the past; its very existence is in part supposed to be a prophylactic against its own use.12 It is better to recognize this property of damages recovery than to erect an artificially binary system of compensation versus deterrence as a way of keeping public enforcement and private litigation in separate spheres that do not touch, or that touch only accidentally.13 Ottervanger discusses other subjects as well including, for example, access to documents in the possession of a competition authority in light of judgments such as Pfleiderer and, with regard to the application of Regulation 1049/2001, EnBW Energie Baden-Wurttemburg and Editions Odile Jacob. I will just call attention here to his reference to the 2005 Dutch Act on the collective settlement of mass damages claims (ie, the ‘WCAM’).14 As the name suggests, the Act applies only where private litigation is settled out of court, the settlement being conditional on approval by the Amsterdam Court of Appeal. This legislation represents a hybrid model of dispute resolution in the sense that no mass claim can be forced upon a reality concerned more with the compensation of victims than it is with deterrence. For this argument, see Paolisa Nebbia, ‘Damages actions for the infringement of EC competition law: compensation or deterrence?’, 33 European Law Review 23–43 (2008); Wouter Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for Damages’, cited above note 4, quote at 14. 12 As already suggested, however, the effectiveness of the prophylactic should not be overstated. See above note 8 in fine. 13 Cf. Van den Bergh, ‘Persisting Collective Action Problem’, cited above note 7, at 32–33 (stating pithily that ‘public and private enforcement are not mutually exclusive but [rather] should reinforce each other’ and accepting as desirable the additional deterrence effects that could be produced by a functioning private enforcement system). Cf. also Commission, Impact Assessment Report accompanying the proposal for a Directive of the European Parliament and of the Council, SWD(2013) 203 final of 11 June 2013, included in this volume as Annex II, para 29 (‘All stakeholder groups apart from business are generally of the view that both instruments are in principle equally important and must hence be independent and complementary mechanisms.’). 14 See also, eg, Willem van Boom, ‘Collective Settlement of Mass Claims in the Netherlands’, in Matthias Casper, André Janssen, Petra Pohlmann, Reiner Schulze, eds., Auf dem Weg zu einer europäischen Sammelklage?, Sellier, 2009, pp 171–192 (assessing the Dutch Act positively in the main, subject to some criticisms); Erik Werlauff, ‘Class Action and Class Settlement in a European Perspective’, [2013] European Business Law Review 173–186, at 182–184.

xx  Mel Marquis defendant without his consent, which is by definition a part of the settlement; but at the same time, the claimants are bundled together on an opt-out basis. Failure to opt out means the claimant concerned will have to be content with the terms of the settlement. This inspired solution (which has led to some concrete results in the Netherlands15 and which the UK Government has decided to emulate as a complement to class litigation in its current initiative on private enforcement16) defuses or at least attenuates the perceived dangers of opt-out approaches while overcoming the collective action problem that makes opt-in approaches likely (in my view, and in the view of consumer organizations) to fail or to deliver poor results. Although some Member States have collective redress mechanisms based on the opt-out principle or on opt-in/opt-out combinations (accompanied by safeguards to minimize abuse),17 the Commission frankly seems to have bent to the will of industry (a sign of direct lobbying pressure and indirect pressure via the European Parliament), advising Member States in Recommendation 2013/396/EU on collective redress to avoid opt-out bundling of claims, and to specifically defend any exceptions, by way of law or court order, on grounds of ‘sound administration of justice’.18 (To put this in perspective, no justification 15 See, eg, Werlauff, cited previous footnote, at 182 and 183 (observing that ‘[i]n practice the model has proved to be of very great value’ to both claimants and tortfeasors; and citing a product liability dispute and several cases concerning misconduct in the financial sector). 16 See UK Department for Business, Information and Skills, ‘Private Actions in Competition Law: A consultation on options for reform – government response’, BIS/13/501 (29 January 2013), pages 50–51. The report is available at https://www.gov.uk/government/uploads/system/uploads/ attachment_data/file/70185/13-501-private-actions-in-competition-law-a-consultation-on-optionsfor-reform-government-response1.pdf. Where the UK collective settlement procedure is employed, the opt-out element will apply only to class members domiciled in the UK; those domiciled abroad will have the opportunity to opt in. See Renato Nazzini, ‘Competition Litigation in the United Kingdom: What Lies Ahead?’, CPI Antitrust Chronicle, April 2013(1), at pp 5–6. Limiting the opt-out principle to the UK border in this way is intended to lessen the incidence of forum-shopping. See Silvia Pietrini, ‘The Future of Collective Damages Actions in Europe’, this volume, page 264. 17 Apart from the Netherlands and now the UK, some (generally mixed or incomplete) version of an opt-out collective action regime has been established in Bulgaria, Denmark and Portugal; in general, the experience so far in these latter jurisdictions has been relatively limited. About twice as many Member States (Austria, Finland, France, Hungary, Italy, Poland, Spain, Sweden and others) have, as of this writing, opt-in rules of one form or another. The pending French bill is also constructed around an opt-in ‘action de groupe’. See projet de loi relatif à la consommation (EFIX1307316L), http://www.legifrance.gouv.fr/affichLoiPreparation.do?idDocument=JORFDOLE000027383756&t ype=general. 18 Commission Recommendation 2013/396/EU of 11 June 2013 on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law, 2013 OJ L201/60, para 28. The Commission contends that opt-in systems ‘respect the right of a person to decide whether to participate or not’, and adds that they make it easier to determine the combined value of the individual claims. Furthermore, tracking down beneficiaries of a successful recovery is difficult, so disbursements might not reach those entitled to them. See Commission, ‘Towards a European Horizontal Framework for Collective Redress’, COM(2013) 401/2, pages 11–12. The counter-arguments to mandatory opt-in procedures are relegated to footnotes (37 and 38) in the Communication. The presumption in favour of opt-in actions and against opt-out is curious given that the effectiveness of opt-in procedures has never to my knowledge been demonstrated (whereas notorious instances of its failure have been recorded – see below note 171). A reversal of that presumption seems more sensible to me. (For more details as to why Recommendation 2013/396 is apt to be of limited impact, see Sonja Keske, ‘Collective Redress – (Too) Great Expectations?’, presented at a conference I organized at LUMSA University in Rome on 8 November 2013, not yet published.)

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxi  

requirement seems to apply if a Member State simply chooses not to adopt or not to maintain any form of collective redress, contrary to what is recommended by the Commission.) In the Recommendation, it is stated that the Member States ‘should ensure that the parties to a dispute in a mass harm situation are encouraged to settle the dispute about compensation consensually or out-of-court, both at the pre-trial stage and during civil trial […]’.19 However, this reference to collective consensual dispute resolution (‘CDR’) does not seem to accommodate Dutch-style settlement solutions, because the envisaged pre-trial and midst-of-trial settlements relate to cases brought on an opt-in basis. In the future, more attention should be given to other models, including the Dutch settlement model and the two-track opt-out/opt-in model soon to take effect in the UK.20 As for the possibility to justify opt-out systems as being necessary for the sound administration of justice, although this may be some sort of green light in yellow camouflage, the nature of the signal may be questioned. The message seems to be that judicial economy and the avoidance of potentially conflicting solutions arising from fragmented litigation are legitimate goals, whereas effective access to justice for consumers is not. This is an unsustainable position. But the real problem with ‘opt-out’ is that it is tainted by its American heritage, and it is commonplace to point to the U.S. Supreme Court itself and its frosty attitude toward class actions in judgments such as (i) Wal-Mart v Dukes (564 U.S. __, 131 S.Ct. 2541 (2011) (a judgment whose relevance to antitrust cases was however rather limited, in part but not only because it was a Rule 23(b)(2) case, not a Rule 23(b)(3) case); see also Comcast Corp. v Behrend, 569 U.S. __ (2013)) and (ii) American Express v Italian Colors (570 U.S. __ (2013)), discussed below, notes 116–120 and accompanying text) as confirmation that opt-out class actions must be inherently pernicious. Sober accounts of the U.S. system show that it is not ‘opt-out’ as such that leads to abuse, but rather its undesirable dis-synergies with other features such as the asymmetric shifting of legal costs in favor of plaintiffs, the mandatory trebling of damages as applied to most defendants (ie, those not benefiting from a de-trebling under ACPERA), and other features discussed later in the main text. In light of the political sensitivity to ‘opt-out’ collective actions, the Commission would have been better advised to abstain from making a recommendation one way or another. It could have left the choice to the Member States, which would then have been in a position to accumulate experience and later to compare results. But in my view this is probably what the Commission seeks to achieve anyway, notwithstanding the tenor of the Recommendation. The Commission knows some Member States such as the UK will likely experiment in any case with the ‘opt-out’ approach (which may incorporate courtapproved cy près distribution of unclaimed damages), and that if positive experiences are forthcoming, they can be used as ammunition at the time of the Recommendation’s review in 2017 or thereafter to support a change of heart. 19 Recommendation 2013/396, cited previous footnote, para 25. 20 The commendable idea of eschewing the binary choice of opt-in and opt-out (and instead granting discretion to judges to adopt the procedures most appropriate for the cases before them) arises from the recognition that some claims accrue to medium and large businesses that do not require the special features of an opt-out to ensure they will enforce their rights, whereas other claims accrue to dispersed victims which individually tend to lack the resources and incentives to do so. Besides the Netherlands and the UK, other opt-out class action systems that could be explored as potential sources of inspiration include those in Australia (where the main ‘safeguard’ is the loser-pays rule, although, oddly, there is no certification stage) and some of the provincial systems in Canada. For discussion of the Australian system, see, eg, Stuart Clark and Christina Harris, ‘The Push to Reform Class Action Procedure in Australia: Evolution or Revolution’, 32 Melbourne University Law Review 775 (2008). Class litigation in Canada is discussed by Brian Facey and David Rosner, ‘Collective Redress for Cartel Damages in Canada’, this volume.

xxii  Mel Marquis Scott Campbell and Tristan Feunteun: ‘Designing a Balanced System: Damages, Deterrence, Leniency and Litigants’ Rights – A Claimant’s Perspective’. Campbell and Feunteun provide comments that recognize Europe’s instinctive aversion to US-style antitrust litigation (although they correctly observe that notions circulating in Europe about the US system are often caricaturized and incomplete) and at the same time call for an EU Directive that will reinforce the incipient trend, in some Member States, toward better means of obtaining compensation for anticompetitive harm. The chapter is generally focused on the efforts of the EU and the UK to encourage private enforcement, and in that regard the authors discuss, for example, the Joint Information Note issued by Commissioners Reding, Almunia and Dalli on 5 October 2010,21 and the consultation on options for reforming private antitrust litigation launched by the UK Department for Business, Information and Skills in April of 2012.22 Like other authors in this volume they also discuss the crucial subject of access to evidence for claimants in national litigation, especially with regard to materials in the case files of competition authorities. The judgments in Pfleiderer and CDC Hydrogene Peroxide are mentioned, and the authors make suggestions about how to develop a differentiated approach to the disclosure of evidence. Here it is emphasized that, while certain measures of protection are justified and proper, in particular corporate statements and settlement submissions, there is no need to protect preSee http://ec.europa.eu/transparency/regdoc/rep/2/2010/EN/2-2010-1192-EN-1-0.Pdf. See https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/31528/12742-private-actions-in-competition-law-consultation.pdf. The BIS followed up the consultation with its ‘Government Response’ in January of 2013 (cited above note 16). Under the UK’s envisaged system the CAT, which will have exclusive jurisdiction to hear and adjudicate collective actions, will decide whether to certify a case and whether it is amenable to opt-out or whether an opt-in procedure should be used instead. See page 31 of the Government Response. With regard to the new possibility of proceeding on an opt-out basis, at page 6 (see also page 40), the BIS summarizes the precautions taken in its proposal: Recognising the concerns raised that this could lead to frivolous or unmeritorious litigation, the Government is introducing a set of strong safeguards, including: – Strict judicial certification of cases so that only meritorious cases are taken forward. – No treble damages. – No contingency fees for lawyers. – Maintaining the ‘loser-pays’ rule so that those who bring unsuccessful cases pay the full price. Claims will only be allowed to be brought by claimants or by genuine representatives of the claimants, such as trade associations or consumer associations, not by law firms, third party funders or special purpose vehicles. Any unclaimed sums would be allocated to the Access to Justice Foundation (AtJF). [Furthermore, as mentioned above, the opt-out rule will apply only to class members domiciled in the UK, while non-UK claimants may opt in – see above note 16.] The UK proposals have been discussed in various commentaries. See, eg, Nazzini, ‘What Lies Ahead?’, cited above note 16; Gerald Barling, ‘The UK Competition Regime: Recent Developments and Further Proposals for Change’, in Philip Lowe, Giorgio Monti and Mel Marquis, eds., Effective and Legitimate Enforcement of Competition Law, Hart Publishing, forthcoming (praising the Government’s plans to strengthen collective actions, given the inadequacy of the over-restrictive Article 47B of the Competition Act 1998 (opt-in only), and Article 19(6) of the Rules of Civil Procedure; but also praising the stipulation of various safeguards to avoid the encouragement of a litigation culture in the UK). See also Silvia Pietrini, ‘The Future of Collective Damages Actions’, this volume (advocating the use of the proposed UK model, with its combination of territorial opt-out and safeguards against abuse, as a desirable model for common European rules). 21 22

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxiii  

existing documents.23 They do not refer to other types of documents prepared in connection with an investigation, such as replies to requests for information; but since documents such as these are not ‘pre-existing’, the authors may countenance some degree of protection. It is asserted that nothing should prevent a leniency applicant or party to a cartel settlement from disclosing corporate statements or settlement submissions, on a voluntary basis and after the Statement of Objections or similar proposed decision has been issued, if it so chooses in the context of settlement discussions with private claimants. According to Campbell and Feunteun, ‘voluntary evidence disclosure by leniency applicants and other defendants is often the key to ensuring that settlement negotiations can progress properly and be appropriately supervised by the courts’.24 Without specifying mandatory rules for Member States with regard to voluntary disclosure of such evidence, the Commission did come forward in June of 2013 with a relatively, but not perfectly clear category-based approach to evidence disclosure via court order.25 The Commission’s proposal is presented in schematic form below.

23 For a critical view of the distinction between those materials worth protecting and ‘pre-existing’ documents, see Sven Völcker’s case note on Case C-360/09, Pfleiderer AG v Bundeskartellamt, 49 Common Market Law Review 695-720 (2012), at 709–712. 24 Page 35. 25 In section 4.3.3 of his chapter, Luís Morais calls for further precision with regard to these rules. For example, he says ‘the Directive should specify that protection against disclosure should apply not only to the first oral statement made by a party within a leniency procedure but also to subsequent and complementary submissions’. See Morais, ‘Integrating Public and Private Enforcement of Competition Law in Europe: Legal Issues’, this volume, page 130; similarly, see Geradin and Grelier, ‘Tip of the Iceberg’, cited above note 9, at 11 and 19. Morais, Geradin and Grelier are also wary – and others will be too, surely – of the only temporary protection that would apply as regards the ‘grey list’ documents seen in the chart below.

xxiv  Mel Marquis Access to leniency and other materials under Commission’s proposal of 11 June 2013 Degree of protection

Types of documents

Under Commission’s proposal, degree of protection and earliest moment of possible disclosure26

Black list

Documents whose disclosure could ‘Absolute’ protection: jeopardize public enforcement efforts, Never disclosable by court specifically: order - Corporate leniency statements; - Settlement submissions

Grey list

Documents prepared for the purpose of the investigation, such as: - Replies to requests for information; - Statements of Objections; - Preliminary assessments (under Article 9 of Regulation 1/2003)

White list

Pre-existing materials not prepared in No protection:28 connection with the investigation, such as: Disclosable by court order - Written agreements; at any time - Texts of e-mails; - Minutes of meetings

Temporary protection: Disclosable by court order27 only after the authority in question takes a decision in the case or closes its file

26 Presumably the tenets that apply here would also be observed in other contexts, including for example the case where, pursuant to Article 15 of Regulation 1/2003, a national court requests that the Commission provide it with materials in its case file. 27 A court order would not be automatic; the national court would only issue such an order after satisfying itself that the request for evidence is proportionate. Proportionality in this context often boils down to whether a party has other reasonable means of its own to obtain the requested materials, and whether the information sought is of genuine relevance to the matter before the court. Illustrative in this regard is National Grid Electricity Transmissions Plc v ABB Limited and Others [2012] EWCH 869 (Ch), 4 April 2012. See especially para 39. 28 It appears that the proposal, if adopted, would necessarily relax point 40 of the Commission’s 2006 Leniency Notice, which states: ‘The Commission considers that normally public disclosure of documents and written or recorded statements received in the context of this notice would undermine certain public or private interests, for example the protection of the purpose of inspections and investigations, within the meaning of Article 4 of Regulation (EC) No 1049/2001, even after the decision has been taken.’ (footnote omitted) Documents received by the Commission ‘in the context of’ a leniency application include pre-existing documents as a matter of course. Another note of dissonance resulting from the liberal rule on pre-existing documents relates to point 26 of the 2004 Notice on cooperation between the Commission and the courts of the EU Member States, where the Commission commits to withhold from national courts information submitted by a leniency applicant unless the applicant consents. Since this confidentiality commitment is broad and covers pre-existing documents, the Commission intends at a later stage to align the latter Notice with the Directive. See page 15 and footnote 45 of the Explanatory Memorandum attached to the draft Directive (Annex I to this volume).

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxv  

Campbell and Feunteun also recount the experiences, particularly in Germany, in cases involving the consolidation of multiple claims. They discuss, among other cases, the experience of CDC in pursuing substantial bundled claims (amounting to hundreds of millions of euros) against companies that participated in the cement and hydrogen peroxide cartels. CDC’s special purpose vehicle model sidesteps collective action in the sense that the company assesses the value of multiple claims for damages, purchases them (with nominal up-front consideration coupled with deferred potential compensation whereby a percentage of any damages recovered is paid to the original assignors) and then seeks recovery in its own name while bearing the full risk of failure. With the support of the German Federal Court of Justice and lower German courts, the door is now open in Germany to the pursuit of consolidated claims in a way that partially overcomes the problem of ‘scattered harm’ and its consequence of under-supply in compensation claims.29 Finally, the authors discuss the spontaneous efforts of private litigants in Europe to settle claims for damages by means of settlement funds. Significant cases in this context, and notably the air passenger and marine hose cartels, provide examples of imaginative solutions aimed at achieving ‘global peace’, which include, among other things, the possibility of ‘turncoat’ cartelists furnishing ammunition in private proceedings against erstwhile co-conspirators. Donald Baker: ‘Trying to Use Criminal Law and Incarceration to Punish Participants and Deter Cartels Raises Some Broad Political and Social Questions in Europe’. Baker’s paper appears in this collection as a reminder that a full appreciation of how private enforcement in the EU fits within the broader picture requires a consideration of the criminal law environment as well as the administrative law environment. Are cartels a supreme evil in the rest of the world just as they are in the U.S.?30 As a starting point, Don Baker fully embraces the wisdom of criminal law penalties and incarceration for cartelists under the Sherman Act, and he gives good reasons to support his views.31 These include, 29 In the cement cartel case, for example, following appeals on threshold issues of procedure and admissibility (decided ultimately by the Bundesgerichtsof in CDC’s favour), Judge Ollerdiβen of the Landgericht Düsseldorf heard evidence on 10 October 2013 and a verdict on aggregated damage claims approaching 200 million euros is awaited. Four other lawsuits are pending in Germany, the Netherlands and Finland. However, while the CDC model can be promising where medium and large businesses wish to outsource the enforcement of claims reaching a certain value (due to high-volume purchasing of the cartelised product), relative transaction costs rise in the case of final consumers and small businesses, whose claims tend to be of modest or micro scale. CDC is not the solution for them. A similar point is made in Tim Reher, ‘Specific Issues in Cross-Border EU Competition Law Actions Brought by Multiple Claimants in a German Context’, in Paul Beaumont, Florian Becker and Mihail Danov, eds, Cross-Border EU Competition Law Actions, Hart Publishing, 2013, chapter 12, at page 161 (‘the assignment model is not feasible in cases of dispersed and small individual damages’). 30 U.S. law does however condone the supreme evil of antitrust in the case of purely outbound export cartels. 31 See, eg, OECD policy roundtable on Cartel Sanctions against Individuals 2003, DAF/ COMP(2004)39 (10 January 2005), including in particular the Background Note by Andreas Reindl at pages 15–28. The document is available at http://www.oecd.org/competition/cartels/34306028.pdf. See also Christopher Harding and Julian Joshua, Regulating Cartels in Europe, Oxford University

xxvi  Mel Marquis above all, the frequent disconnect between the pathology – an individual rational or not-so-rational actor choosing whether to collude – and the remedy, ie, the mere imposition of high corporate fines.32 Undoubtedly, some company officers and directors are deterred from unlawful collusion by the prospect of negative consequences flowing from the tarnished corporate reputation and possibly diminished share prices or other losses that may cause shareholders to hold them to account.33 But, as suggested earlier, given the low rate of detection of cartels, and given the extraordinary time lag that can elapse before the company must pay any penalty, many company officers and directors must feel secure when they roll the dice. The U.S. experience suggests that serious enforcement of criminal law under the Sherman Act goes a long way toward closing the deterrence gap.34 But irrespective of the wisdom of punishing natural persons for cartelizing markets, Baker’s assessment of the situation in Europe suggests clearly that the same attitudes do not prevail here. At the level of the EU, as Baker points out, individual criminal penalties for cartels are alien to the Treaty-based legal framework.35 That will not change soon, even if some of us would like to see the EU move in that direction when there is sufficient and hard-won intellectual and popular momentum. But even at the level of the Member States – and here Baker focuses on the experiences of the United Kingdom and Ireland – the record on individual punishment so far is unimpressive. The well-publicized story in the Press, 2010, chapter 11; Douglas Ginsburg and Joshua Wright, ‘Antitrust Sanctions’, 6(2) Competition Policy International 3–39 (Autumn 2010). 32 See also James Venit and Andrew Foster, ‘Competition Compliance: Fines and Complementary Incentives’, this volume, page 71 (‘The lack of individual penalties may lead to diverging incentives between firms and their employees as regards competition compliance. Where a decision maker faces individual rewards for generating profits (even those gained anticompetitively) but no penalties for violating the antitrust laws, his or her normative commitment to ‘following the rules’ may not be aligned with the goal of maximum competition compliance.’). 33 See ibid (discussing securities fraud litigation and derivative shareholder suits brought in U.S. courts under securities laws). 34 Arguably, this also raises the conceivable possibility that some employees managing globally active firms may intensify their quest for cartel profits in other countries (including relatively poor countries) in order to compensate for lost ‘opportunities’ in the United States. Paradoxically, insofar as this theoretical diversion phenomenon occurs, the strong consumer-regarding impact of the Sherman Act indirectly harm consumers elsewhere. Of course, the fault lies with the cartelists, not with cartel crusaders in the U.S. 35 There may however be other non-criminal possibilities under the current framework. Although the antitrust prohibitions in the Treaty are addressed to undertakings, there is an argument to the effect that the Treaty would not preclude the imposition of administrative fines or other non-criminal sanctions such as director disqualification on (non-undertaking) natural persons. The argument would be that non-criminal sanctions are necessary and proportionate to the goal of ensuring the effectiveness of the Treaty rules. This possibility of the Commission imposing administrative fines or other noncriminal sanctions without any revision of the Treaty is highlighted by Jim Venit and Andrew Foster in their contribution, ‘Competition Compliance’. Could the Commission impose such sanctions even without amendments to Regulation 1/2003? If it did so, there would likely be a legal fight raising interesting questions before the EU Courts as to the assumptions underlying the Regulation and as to the boundaries of the Commission’s powers. Notwithstanding the desirability of individually targeted sanctions, my own view is that, given the secondary law framework as it stands, such acts by the Commission would be ultra vires and possibly contrary to general principles of EU law.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxvii  

UK has been so unsatisfactory, in fact, that there is new legislation (Section 47 of the Enterprise and Regulatory Reform Act 201336) purporting to correct the flawed Section 188 of the Enterprise Act 2002.37 More generally, it is a stunning fact that, after all these years – at least 20 years of ‘hard core cartel’ rhetoric in Europe – price fixers do not serve jail time here. Those prison sentences that have been handed down have almost all been suspended (in this respect, Ireland is the leading EU jurisdiction,38 although recent legislative changes seem to invite judges to show no quarter39). Will legislatures and courts be ready, in the future, to turn the screws and get serious about incarcerating individuals? My answer is that this will indeed happen, but on a gradual basis – punctuated, perhaps, by occasional bombshell cases catching responsible executives quite off guard. As Baker reminds us, although criminal law penalties appeared in the Sherman Act from the very beginning – riding a populist wave, with no debate or reflection – it was not until watershed events transpired in the 1960s (the electrical equipment Section 47 ERRA appears at http://www.legislation.gov.uk/ukpga/2013/24/section/47/enacted. The UK’s bitter experiences under the Enterprise Act 2002 were largely due to a ‘dishonesty’ requirement in criminal cartel cases. This requirement has now been abolished (while excluding from prosecution unconcealed cartel behavior, oxymoronic as that may seem). However, as a concession apparently granted to overcome business interests fiercely opposed to the reform, the new provisions allow an accused colluder to exonerate herself if, prior to entering into or implementing a cartel arrangement, ‘she took reasonable steps to ensure that the nature of the arrangements would be disclosed to professional legal advisers for the purposes of obtaining advice about them before their making or implementation’. If UK courts apply this defense literally – the statute does not stipulate that the cartelist must have followed the lawyer’s (bad) advice for the defense to succeed – it will probably make a mockery of the cartel offense, hardly an intended consequence of the reform; on the other hand, particularly where collusion is pursued by rogue employees (unless they cunningly seek legal advice without tipping off their employers; this may become a more common risk as the defence becomes better known by businessmen), the defense may be immaterial. And of course, for what they are worth, administrative sanctions and remedies will still be available to the Competition and Markets Authority, and the wrongdoing company may still face exposure to private claims. For an early criticism of the ‘legal advice’ defense, see Peter Whelan, ‘Does the UK’s New Cartel Offence Contain a Devastating Flaw?’, UEA/CCP blog post, 21 May 2013, http://competitionpolicy.wordpress.com/2013/05/21/does-theuks-new-cartel-offence-contain-a-devastating-flaw/. Less pessimistically (ie, hoping that the UK courts will put a gloss on the defense), see also Andreas Stephan, ‘The UK’s New Cartel Offence: It Could Be Alright on the Day’, UEA/CCP blog post, 9 July 2013, http://competitionpolicy.wordpress. com/2013/07/09/the-uks-new-cartel-offence-it-could-be-alright-on-the-day/#more-836. 38 As of June 2013, prison sentences totalling 84 months for 10 individuals had been successfully secured in the Irish courts, but all sentences were suspended. See further Patrick Kenney, ‘Making Cartel Crime Not Pay’, speech of 27 September 2012, http://www.tca.ie/images/uploaded/ documents/2012-09-27%20Making%20Cartel%20Crime%20Not%20Pay.pdf. 39 See Kenney, ‘Crime Not Pay’, cited previous footnote, pages 6–10; Peter Whelan, ‘Strengthening Competition Law Enforcement in Ireland: The Competition (Amendment) Act 2012’, 4(2) Journal of European Competition Law & Practice 175–181 (2013) (interpreting the increase of maximum jail time from 5 to 10 years as a message to judges). This institutional ‘dialogue’ is perhaps in some respects mirrored in Canada. Up to and including 2012, cartelists in Canada convicted in criminal cases were generally able to avoid or reduce prison sentences, in part via plea agreements, and instead were given ‘conditional sentences’: community service tasks and/or house arrest (and individual fines have amounted to paltry sums). However, as a result of amendments to the Criminal Code introduced in March of 2012, Canada has made incarceration mandatory for persons convicted of cartel behavior. See, eg, Mark Katz, ‘Punishing Cartels in Canada: Is a ‘Sea Change’ on the Horizon?’, CPI Cartel Column, 31 January 2013, https://www.competitionpolicyinternational.com/assets/Uploads/ Cartel1-31-2013-1-1-1.pdf. 36 37

xxviii  Mel Marquis cartel cases) and 1970s (especially the 1974 amendment upgrading Sherman Act violations to felonies) that the criminal side of US antitrust law came of age. I submit that social change occurs at a rather faster pace today than it once did, and I don’t expect the ‘criminalization’ of cartel conduct in Europe will take 70 or 80 years to hit its stride; on the other hand, ce n’est pas demain la veille. While we wait patiently for change, inaction is not an acceptable option. Two observations should be made in this respect. First, the fact that individuals do not go to prison in Europe does not mean that individuals cannot be subject to targeted penalties. An obvious and sensible example is that Member States can provide for director disqualification orders, as jurisdictions such as the UK, Australia and now Ireland have done.40 Australia has also prohibited companies from indemnifying company officers for cartel behaviour, another sensible measure. Baker provides several other examples of administrative measures that can implement the ‘polluter pays’ principle better than the brute force of high corporate fines can do.41 Furthermore, as Baker points out, administrative sanctions can be adopted under a lower standard of proof, and they can potentially be adopted more quickly compared to criminal cases.42 The second observation is that the fact that cartelists are not going to jail in Europe does not mean that criminal laws are not being applied. Apart from the suspended sentences in Ireland, the authorities in some Member States have become quite active in imposing criminal fines on natural persons for breach of cartel laws. For example, in 2012, criminal fines were imposed on 31 individuals in Germany. In the Netherlands, six more unlucky individuals got the same treatment.43 This is perhaps the shape of things to come. James Venit and Andrew Foster: ‘Competition Compliance: Fines and Complementary Incentives’. Like Don Baker’s chapter, Venit and Foster stress the importance of rethinking the essentially one-dimensional approach to sanctions under EU competition law. As these authors state: ‘Despite the consistent imposition of increasingly higher fines, it is not clear that the Commission has successfully instilled a culture of compliance in Europe or elsewhere. In the absence of material consequences for individuals and additional affirmative obligations on undertakings, the incentives to develop a pro-active, firm-wide 40 On disqualification orders, see, eg, Andreas Stephan, ‘Disqualification Orders for Directors Involved in Cartels’, 2(6) Journal of European Competition Law & Practice 529–536 (2011); OFT, Director disqualification orders in competition cases (2010), http://www.oft.gov.uk/shared_oft/ business_leaflets/enterprise_act/oft510.pdf. 41 See page 59. See also Venit and Foster, ‘Competition Compliance’, this volume (discussed below in the main text). 42 For this reason he suggests that, where administrative and criminal sanctions co-exist, criminal cases should be reserved for particularly momentous cases, while run of the mill cases should be subject to sanctions easier to administer; calibrating enforcement in this way can potentially achieve both the wake-up call of possible imprisonment and an enhanced sense, on the part of actors deciding whether to collude, that the enforcer can process more numerous cases. 43 See Marco Slotbloom, ‘Individual Liability for Cartel Infringements in the EU: An Increasingly Dangerous Minefield’, Kluwer Competition Law blog post of 25 April 2013, http:// kluwercompetitionlawblog.com/2013/04/25/individual-liability-for-cartel-infringements-in-the-euan-increasingly-dangerous-minefield/.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxix  

sense of accountability may still not be sufficient.’44 Whereas Baker questions the deterrent effect of corporate fines and affirms the need for individual penalties on the basis of personal experience, Venit and Foster make similar points but link them more specifically to EU practice, referring for example to enforcement statistics (see section 2 of their paper). Like Baker, they recommend alternative means of enhancing deterrence and compliance, but they dwell in further detail on the possibilities of ‘complementary incentives’, in particular those arising from other areas of law (eg, securities law and rules against public disclosures that ‘cook’ the books) and from benefits that could conceivably be granted to first-in leniency applicants to reduce their exposure to private damages claims. As they point out, private enforcement may also encompass not just antitrust damages actions but also securities fraud litigation (where shareholders recover compensation from the corporation and/or the managers or directors engaged in fraudulent acts, including by way of company disclosures failing to acknowledge that profits have been artificially elevated by cartel overcharges) and shareholder derivative suits (where shareholders sue fiduciaries such as managers or directors for their misdeeds but any sums recovered in restitution are paid to the corporation). In section 3 of their paper, Venit and Foster refer to ‘undesirable side-effects’ that may follow from the reliance of the Commission on very high fines. On the one hand, it is claimed that high fines can stunt innovation and market entry. Without stronger empirical evidence one should perhaps be cautious about such conclusions. First of all, one would think that in most highly innovative sectors cartel behavior may be both illogical (since the drive for innovation is in principle to outpace rivals) and in any case difficult to sustain, although it is also undeniable that collusion has occasionally occurred in sectors where innovation is important. Second, the claim that high fines reduce or eliminate the ability of a cartelist to innovate presents an ex post view of the sanction, whereas from an ex ante perspective a prudent firm might want to preserve its capacity to innovate by eschewing illegal collusion and avoiding the related penalties. This latter argument, however, makes exactly the same assumptions about ‘rational’ firm behaviour that often seem to be mistaken: high corporate fines may mean that antitrust compliance is the best strategy for the firm, but that fact may not be internalized by individual corporate actors. Third, the cost of cartel fines is often financed after the fact by consumers, meaning that if innovation is a necessary parameter of competition in a given industry, the firms may be able to at least partially fund R&D through higher prices on today’s generation of products. Empirical evidence would also be necessary with regard to the authors’ claim that high fines inhibit market entry in industries of high fixed costs because they cut deeply into returns on investment. As a matter of policy, it would seem imprudent to begin checking whether fines will undermine the performance of cartelists and then to make adjustments to soften the blow, as this may engender moral hazard. But again the problem with the current ‘architecture’ of EU sanctions is a more fundamental one, if indeed it fails to secure sufficient compliance with the law. 44

Page 64.

xxx  Mel Marquis Drawing inspiration from U.S. law, and noting an international trend, Venit and Foster sensibly recommend that non-criminal individual penalties should be made available under EU law. The authors’ discussion in this respect is brief, and researchers should direct their creative energies toward the broader debate about ‘smart sanctions’ for cartel conduct in the EU.45 Absent Treaty amendments, how far can the EU legislator go with the rationale that non-criminal penalties for natural persons are required for the effective enforcement of the prohibition of cartels in Article 101?46 What exactly do the principles of effet utile and teleological Treaty (and now Charter47) construction demand? Venit and Foster also address certain issues that bear on the integration of public and private enforcement. One issue that can be highlighted here is the way the risk of exposure to civil liability can not only diminish the incentive to seek immunity, in the Pfleiderer sense; it can also enhance the incentive to do so. The clearest example is in the United States, where the ACPERA statute of 2004 de-trebles, and thus de-punitizes, damage recoveries under the Clayton Act as a reward for successful corporate amnesty applicants who also assist plaintiffs that pursue claims against co-cartelists.48 Given that in much of continental Europe49 the recovery of punitive damages is often an unthinkable taboo,50 it may be unrealistic to include in an EU Directive measures to de-punitize damages as a leniency sweetener.51 However, 45 Commentators have given thought to this, but there still seems to be ample room for discussion. For one notable paper discussing some of the issues, see Douglas Ginsburg and Joshua Wright, ‘Antitrust Sanctions’, cited above note 31. 46 As noted above, I doubt that the Commission could punish (non-undertaking) natural persons by way of administrative fines or other non-criminal penalties under the current legislative framework. On this view, the Commission could only have the option of imposing individually targeted penalties if Regulation 1/2003 were revised, as it one day inevitably will be. 47 Most notably, Article 47 of the EU Charter on Fundamental Rights guarantees the right to an effective remedy from an impartial tribunal. Given that the principle of effectiveness (effet utile) is a horizontal principle of EU law, it applies equally to the Charter (subject to the Charter’s various qualifications and provisos). 48 More precisely, the limitation of recovery is to ‘actual damages sustained’ (no trebling and no joint and several liability). In order to qualify, the company must actively cooperate in the lawsuit against the other cartel participants, for example by submitting relevant documents and making witnesses available for depositions and testimony. See Section 213(b) ACPERA. 49 With regard to the UK, see 2 Travel v Cardiff Business [2012] CAT 19, paras 448 et seq. (allowing exemplary damages in the amount of GBP 60,000; Albion Water v Dŵr Cymru Cyfyngedig [2013] CAT 6, paras 229 et seq. (possibility to claim exemplary damages affirmed as in Cardiff Business but the test is ‘very stringent’ (para 287) and plaintiff was denied such damages on the facts). Contrast Devenish Nutrition Ltd v Sanofi-Aventis SA (France) [2007] EWCH 2394 (Ch), [2008] EWCA Civ 1086 (denying such damages on ne bis in idem grounds where a fine had nominally been imposed by the Commission, notwithstanding a leniency reduction to zero)). The legislation on private damages actions proposed by the UK Government would specifically disallow the recovery of exemplary damages in collective redress cases, which may at least in part reflect the influence of Continental attitudes. 50 The aversion to punitive damages may be seen, for example, in Commission Recommendation 2013/396/EU on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law, cited above note 18, para 31 (recommending to the Member States that they prohibit punitive damages in the context of bundled claims arising from ‘mass harm situations’). 51 One might conceive of an analogous rule precluding recovery of pre-judgment interest, which would represent a substantial reduction of damages in cases of protracted infringements. Such a rule is simply not on the table.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxxi  

the institution of treble damages is not strictly necessary for the shaping of civil damages rules in a way that aims to preserve and even reinforce the attractiveness of leniency programs. Building on earlier ideas that appeared, for example, in connection with the 2008 White Paper,52 the Commission in its draft Directive of June 2013 has proposed to require Member States to introduce limited exceptions to the general rule of joint and several liability for cartelists. The proposal is clearly motivated by the need to press forward with stronger national systems of antitrust damages litigation without giving the immunity applicant cold feet, especially since immunity applicants may well be the first targets in plaintiffs’ line of fire.53 Article 11 of the draft Directive provides in pertinent part as follows: 2. Member States shall ensure that an undertaking which has been granted immunity from fines by a competition authority under a leniency programme shall be liable to injured parties other than its direct or indirect purchasers or providers only when such injured parties show that they are unable to obtain full compensation from the other undertakings that were involved in the same infringement of competition law. 3. Member States shall ensure that an infringing undertaking may recover a contribution from any other infringing undertaking, the amount of which shall be determined in the light of their relative responsibility for the harm caused by the infringement. The amount of contribution of an undertaking which has been granted immunity from fines by a competition authority under a leniency programme shall not exceed the amount of the harm it caused to its own direct or indirect purchasers or providers. [The Council seeks to suppress the second sentence.] 4. Member States shall ensure that, to the extent the infringement caused harm to injured parties other than the direct or indirect purchasers or providers of the infringing undertakings, the amount of contribution of the immunity recipient shall be determined in the light of its relative responsibility for that harm.

Article 11 of the Commission’s draft thus provides double insurance for successful first-in leniency applicants: a limited suspension of liability vis-à-vis cartel victims outside the vertical supply chain; and a limitation on the amount that a successful immunity applicant may have to pay as a contribution to joint tortfeasors (if the latter have paid more than their fair share) to those damages it caused within that 52 Cf. Commission, White Paper on Damages actions for breach of the EC antitrust rules, COM(2008) 165 final of 2 April 2008, page 10, final paragraph; Commission, Staff Working Paper accompanying the White Paper on Damages actions for breach of the EC antitrust rules, SEC(2008) 404 of 2 April 2008, paras 304–306. It has been suggested, though, that limiting leniency recipients’ follow-on liability for damages ‘is not necessary for the effective functioning of leniency. If there were a need to enhance the attractiveness of leniency, this could always be done by increasing the level of fines or other public penalties from which leniency applicants are granted immunity or reductions.’ Wils, ‘Relationship between Public Antitrust Enforcement and Private Actions for Damages’, cited above note 4, at 25. 53 See Jochen Burrichter and Enno Ahlenstiel, ‘Legal and Jurisdictional Issues – The German Perspective’, this volume, pp 99–100 (providing a vivid example of how a successful immunity applicant can find itself all alone in civil litigation with potential responsibility for damages caused by all its co-cartelists’ conduct as well as its own). See also recital 28 of the draft Directive of June 2013 (a decision (or the relevant part of a multi-party decision) of a competition authority finding that the successful immunity applicant has infringed the law may become final before decisions (or parts of the same decision) against the other cartelists become final, an asymmetry which again could ceteris paribus dampen the incentive to confess).

xxxii  Mel Marquis supply chain.54 These benefits for companies that waste no time coming clean are among the subtler features of the Directive, but they may have a significant practical impact in the agencies’ fight against cartels. Reasonably enough they also have the effect of shifting part of the ‘cost’ of compensating victims to infringers that fail to qualify for leniency. From an alternative and less positive point of view, it is said that Article 11(2) of the draft may dampen the incentive to self-report because of the proviso restoring the possibility of joint and several liability visà-vis a successful immunity applicant if an injured party shows that he is unable to recover full compensation from co-cartelists.55 The nightmare scenario for an immunity recipient would be to be sued for the entire harm caused by the cartel (ie, sued by direct and indirect purchasers but also by claimants outside the vertical chain invoking the exception in draft Article 11(2) above) only to find that seeking contribution from the other companies concerned is legally impossible because, in the meantime, the infringement decision as applied to all of those other companies (but not as applied to the leniency recipient, who did not lodge an appeal) has been annulled – eg, for failure to respect essential procedural requirements – by a final judicial decision and therefore ‘legally did not exist’.56 Geradin and Grelier suggest that the EU legislator could address this anomaly by eliminating the exception according to which joint and several liability for an immunity recipient is restored if victims are unable to obtain full compensation from its co-infringers.57 The problem just outlined does amount to a glitch in the proposal of June 2013, but one whose 54 Member States that so desire could go even further than what is ultimately required under the Directive. Hungary, for example, exempts successful immunity applicants from paying damages to any claimant in a follow-on action (ie, including direct or indirect purchasers or providers) unless the claimant is unable to obtain full compensation from the other co-infringers. See Section 88 D of the Hungarian Competition Act. 55 See Sebastian Peyer, ‘Is the New EU Private Enforcement Draft Directive Too Little Too Late?’, UEA/CCP blog post of 15 June 2013 (‘In the worst case scenario the leniency applicant has to await the end of all (!) private damages claims before he knows whether or not the claimants have obtained full compensation from other cartel members, or are likely to pursue him. This may reduce the incentives for firms to voluntarily disclose information to the agencies.’ A similar concern is expressed by Luís Morais in section 6.2 of his contribution to this volume (but on balance he regards as ‘very positive’ the way in which the draft Directive treats the specific issue of the civil damage exposure of successful immunity applicants; another positive assessment is provided by Veljko Milutinović in the postscript to his chapter). Peyer’s overall assessment is that the draft Directive neglects the core issues of costs, funding of claims and ‘class actions’. The omission of the latter two issues is deliberate, as the Commission has chosen to deal with them (satisfactorily or not) in Recommendation 396/2013, cited above note 18. Peyer’s other overarching point is that the slow construction of the draft Directive has been outpaced and made largely redundant by rapid developments at the national level. It is true, of course, that several Member States have either amended their procedural regimes to better address issues of private antitrust litigation or are at present contemplating such amendments. Nevertheless, when one considers the EU as a whole it is not clear that such matters are a priority for many Member States. Uncoordinated legislative activity across Europe, despite the potential it presents for creative solutions, would seem likely to exacerbate the existing fragmentation of regimes while leaving behind aggrieved but poorly resourced parties in the less proactive Member States. 56 See Pinar Akman, ‘Period of limitations in follow-on competition cases: the elephant in the room?’, CCP Working Paper 13-8 (2013), at 18–20 (quote at 18), discussing Deutsche Bahn AG v Morgan Crucible [2011] CAT 16; Geradin and Grelier, ‘Tip of the Iceberg’, cited above note 9, at 11–12. 57 See Geradin and Grelier, ibid, at 19–20.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxxiii  

relevance is probably limited to quite rare (not to say inconceivable) concrete cases. Furthermore, although Article 16(1) of Regulation 1/2003 appears to brook no decision by a national court that contradicts a Commission infringement decision, and since, in the scenario above, that decision remains unannulled as regards the non-appealing immunity recipient, a national court under the circumstances would surely – or rather it should surely request a preliminary ruling from the Court of Justice; and one may doubt whether the ECJ would insist on a reading of Article 16(1) so inflexible as to require the national court to ignore the judicially determined infirmity of the Commission’s decision.

Part II

Integrating Public and Private Enforcement in Europe. Legal and Jurisdictional Issues

Fred Louis: ‘Promoting Private Antitrust Enforcement: Remember Article 102’. This paper has two principal features. First, it seeks to redirect attention from the routine debate on the EU’s private enforcement initiatives, which tend to revolve around private claims brought in cartel cases, and to point out that the measures adopted by the EU will also have important consequences for litigating abuse of dominance cases. As a definitional matter, it is argued that the term ‘private enforcement’ should not even be used with regard to follow-on claims brought by private litigants, since it is in this case the dog wagging the tail – significant issues may be litigated, such as causality and damages, but it is the public authority that enforces the competition rules. Private litigation that follows an authority’s final decision merely draws the corresponding civil consequences. Second, an updated part of the paper reacts to the Commission’s proposed Directive on antitrust damages actions, published in June of 2013.58 If follow-on actions are removed from the concept of ‘private enforcement’, it may be that Article 102 cases (and/or cases alleging breach of equivalent national rules) constitute the main area of private enforcement. In a pure exploitative abuse scenario, private damages claims may not always be brought before national courts, although there is evidence of significant litigation in this context.59 Where those exploited by exorbitant prices are individual consumers or small or micro businesses, the familiar problem of dispersed harm may imply, in a manner similar 58 Some of the other chapters herein, including notably those of Luis Silva Morais and Veljko Milutinović, also provide early assessments of the Commission’s draft Directive. A small but growing number of analyses of the draft by practitioners are also emerging. See, eg, Anneli Howard, ‘Too little, too late? The European Commission’s Legislative Proposals on Anti-Trust Damages Actions’, 4 Journal of European Competition Law and Practice 455–464 (2013). Another edited work has also just become available: David Ashton and David Henry (eds), Competition Damages Actions in the EU (Edward Elgar, 2013). 59 See the Commission, Staff Working Paper accompanying the Report on the Functioning of Regulation 1/2003, SEC (2009)574 final of 29 April 2009, point 270. Philip Lowe also refers to the ‘many claims of exploitative abuse at national level’ in his contribution. See ‘Conclusions’, point 7.

xxxiv  Mel Marquis to many cartel cases, that unless an efficient method of pooling similar claims is available, the cost of litigation may be inordinately high for potential plaintiffs compared to the small stakes. Under those conditions, a customer suspecting that a dominant firm’s prices are excessive within the meaning of Article 102 might in some cases be better off seeking to persuade a competition authority to take the case. If the exceptional conditions of actionable exploitative conduct appear to be satisfied,60 the enforcer can possibly put enough pressure on the putative infringer to secure some kind of prospective price relief, although absent litigation or ADR the customer would have no means in this case of obtaining compensation for any illegal rents she may have paid to the dominant firm. As for exclusionary abuses, private enforcement is of potentially immense significance. First of all, it is rather exceptional for public enforcers to pursue full-blown exclusionary abuse cases all the way to an infringement decision. Many such cases are filtered out, either at an early stage or – in light of the complexity and controversial nature of such cases, not to mention the fact that prolonged investigations may become superfluous due to rapid market evolution – at an intermediate stage by means of an Article 9 decision or its national equivalent. The rarity of final infringement decisions in this context implies that private enforcement may be a better option for aggrieved parties, who may well be companies with large claims and sufficient means to vindicate them. (Of course, the helpfulness of such litigation depends to a large extent on whether the substantive law of abuse of dominance is sufficiently rational and clear that legitimate, aggressive competition can be distinguished from anticompetitive conduct within a tolerable margin of error; the ECJ’s guidance will be utterly crucial in this regard in the coming decades.) Furthermore, the nature of exclusionary conduct is such that expeditious intervention may make the difference between life and death in the market. But as Louis points out, it is difficult to get competition authorities to impose interim measures, above all at the level of the EU, where Article 8 of Regulation 1/2003 provides that such measures may be adopted solely at the discretion of the Commission. For these reasons, national courts are likely to be the port of call when time is of the essence. Of course, depending on the jurisdiction it may be difficult to secure rapid injunctive relief in court as well, but perhaps the principles of effectiveness and effective judicial protection imply that it should not be too difficult. With regard to the Commission’s draft Directive of June 2013 (as well as the Commission’s Communication on the quantification of harm in private damages actions and the corresponding Staff Working Document, each of which formed part of the June 2013 package), Louis offers a positive early evaluation (stating that the proposed Directive is ‘overall a very commendable attempt to facilitate private damages actions while seeking to maintain a balance between the rights of victims to obtain redress, the rights of defence and the need to avoid deterring 60 For extensive descriptive and normative discussions relating to the prohibition of exploitative conduct by dominant firms, see various contributions (eg, those of: Lars-Hendrik Röller; Emil Paulis; Marc van der Woude; and Amelia Fletcher) in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2007: A Reformed Approach to Article 82 EC, Hart Publishing, 2009.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxxv  

leniency applications […]’61). He underlines, in particular, the importance of the proposed rules on disclosure of evidence (Articles 5–8 of the draft text): ‘The system of court-ordered disclosure […] constitutes a true revolution for many if not most Continental legal regimes. The possibility to force the allegedly dominant defendant (as well as third parties) to divulge information is crucial to establish some balance and equality of arms in damages actions for abuse of dominance’.62 In part, the boldness of the proposal relates to the not-yet-precisely-resolved matter of how much protection from disclosure will be accorded to a defendant’s business secrets. As Louis points out, Article 5(4) of the draft Directive indicates that courts would have to walk the thin line between too much protection and not enough: ‘Member States shall ensure that national courts have at their disposal effective measures to protect confidential information from improper use [an undefined term] to the greatest extent possible whilst also ensuring that relevant evidence containing such information is available in the action for damages.’ Preliminary references to Luxembourg might be expected with respect to delicate issues such as this one. But while Louis generally praises the Commission’s proposal on damages actions, he reiterates his point concerning the need for swift injunctive relief in exclusionary abuse cases because the draft Directive does not address this.63 Considering that there remains significant uncertainty in relation to the private enforcement of Article 102, Louis compiles a list of issues that could be addressed in a guidance document to assist courts as well as litigants when relevant cases arise. Jochen Burrichter and Enno Ahlenstiel: ‘Integrating Public and Private Enforcement in Europe: Legal and Jurisdictional Issues – The German Perspective’. Together with the UK and the Netherlands, Germany has come to be known as one of the few EU Member States in which private antitrust litigation really happens. This positive development owes much to the amendments made in 2005 to the German Act against Restraints of Competition (ie, the GWB). Burrichter and Ahlenstiel begin their paper with a review of the most relevant provisions of the Act in this context, Sections 33(3) and 33(4). Together, these provisions provide a crucial basis for bringing damages actions before the German courts, and especially so in the case of follow-on actions. With regard to the latter, the German legislator took the daring, or even revolutionary step of declaring German courts that hear and adjudicate antitrust damages claims to be bound by any prior finding of an infringement either (i) by a competition authority of any Member State (or court acting as such), or (ii) by an appellate court that upholds such a finding in a final judgment. This rule now serves as the model that would apply to all Member States if the corresponding Page 90. Ibid. Cf. also Morais, ‘Legal Issues’, this volume, page 130, footnote 62 (similarly observing the innovative nature of the provisions vis-à-vis ‘Romanistic-Germanic systems’). 63 Injunctive relief is specifically addressed in Commission Recommendation 396/2013 (cited above note 18), but the Recommendation is geared toward ‘mass harm situations’; few exclusionary abuse cases would qualify. 61 62

xxxvi  Mel Marquis provision in the draft Directive of June 2013 is adopted without amendment by the Council and the Parliament (which seems unlikely).64 In the German courts and in Member State courts following the adoption of the Directive, plaintiffs would still have to satisfy the conditions of tort law including, in particular, fault on the part of the defendant, damages sustained, and causality.65 Nevertheless, this envisaged pan-European rule of mutual recognition would establish a remarkable system of cross-institutional and ‘diagonal’ trust. The particulars with regard to the German version of the rule are discussed in section II of the paper. In section III, the authors consider the position of leniency applicants in followon damages actions, and recall that successful leniency applicants under the EU leniency program (or under similar national leniency programs) may often be easy targets for follow-on claims because the decisions against them become final earlier than do decisions against co-cartelists who keep fighting their case. (The Bundeskartellamt does not address infringement decisions to companies that qualify for full immunity under the German leniency program,66 but the point remains relevant for leniency applicants later in the queue.) Combined with the normal rule of joint and several liability, the accelerated exposure of leniency applicants can put them in a very awkward position. As an illustration, Burrichter and Ahlenstiel discuss a follow-on case in the German courts that arose from the prominent EU investigation in the carglass cartel case. However, as noted above the draft Directive of June 2013 would attenuate this conflict between the private and public enforcement systems.67 Another conflict is discussed in section IV, namely that between access to evidence and the need to protect sensitive documents within the possession of competition authorities. In this context, the Pfleiderer case and a similar scenario in the coffee roasters case are mentioned, with the German courts in each case attaching somewhat more weight to the need for a safety zone for leniency applications than the Court of Justice did in Pfleiderer and Donau Chemie. 64 Article 9 of the draft Directive provides that ‘Member States shall ensure that, where national courts rule, in actions for damages under Article 101 or 102 of the Treaty or under national competition law, on agreements, decisions or practices which are already the subject of a final infringement decision by a national competition authority or by a review court, those courts cannot take decisions running counter to such finding of an infringement. This obligation is without prejudice to the rights and obligations under Article 267 of the Treaty.’ Recital 25 of the draft Directive states that the binding effect of national decisions established in Article 9 ‘should apply to the operative part of the decision and its supporting recitals’. (emphasis added) The reference to the decision’s supporting recitals seems to suggest that even the reasoning of the authority that adopted the decision, so long as it supports the operative part, should be treated as ‘binding’ (or more aptly, preclusive). To that extent, recital 25 seems to go beyond the position expressed by the Commission in 2000 with regard to the conflict rule that later became Article 16 of Regulation 1/2003. The Commission at that time stated that ‘the potential for conflict depends on the operative part of the Commission decision and the facts on which it is based’. Commission, Explanatory Memorandum attached to the 2000 Proposal for a Council Regulation on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, 2000 OJ C365E/284. 65 See, eg, Enron Coal Services Ltd. v English Welsh & Scottish Railway Ltd., [2011] EWCA Civ 2 (sectoral regulator had found an infringement but later damages action failed for failure to prove causation). 66 See OECD Roundtable on Private Remedies 2007 (2006), cited above note 11, at 307. 67 See above notes 64–66.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxxvii  

Although cross-border litigation issues concerning, for example, rules of personal jurisdiction, recognition of judgments, conflicts of laws and forum shopping were unfortunately not examined in depth at the Workshop, Burrichter and Ahlenstiel touch on some of these areas of interest in section VI of their chapter.68 With respect to personal jurisdiction, for example, they recall Articles 5(3) and 6(1) of Council Regulation 44/2001 (the ‘Brussels I’ Regulation on jurisdiction and enforcement), now reincarnated as Regulation 1215/2012.69 Article 6(1) is of particular importance in cross-border cartel cases (and not just those brought in Germany), as it enables a plaintiff to sue a group of cartelists in any Member State in which one of them is domiciled, assuming the claims against them are closely connected. The more geographically heterogeneous the group of cartelists, the wider the range of possible venues for litigation, and hence the greater the scope for ‘forum shopping’ – the great equalizer or the great scourge, depending on one’s point of view.70 The authors cite an example of a Belgian company (CDC) taking advantage of Article 6(1) to sue multiple defendants in Germany, with only one anchor defendant domiciled in the forum State.71 Since the sole defendant based in Germany settled out of court, the Landgericht Dortmund in June 2013 asked the European Court of Justice, inter alia, whether the anchor of Article 6(1) is still lodged deeply enough to justify consolidated administration of the case in that forum.72 As Burrichter and Ahlenstiel explain, the answer under German law would be that the Landergericht by virtue of perpetuatio fori retains jurisdiction notwithstanding the awkward procedural posture of the case. Assimakis Komninos: ‘The Relationship between Public and Private Enforcement: quod Dei Deo, quod Caesaris Caesari’. Komninos begins his chapter with a description of the goals of competition law enforcement, which in his view correspond to an injunctive-restorative-punitive triad of objectives that can be and are pursued in various ways. He rejects any neat categorization, or any ‘antinomy’, 68 For rich assortments of papers addressing such issues, see Beaumont, Becker and Danov, eds, Cross-Border EU Competition Law Actions, cited above note 29; Jürgen Basedow, Stéphanie Francq and Laurence Idot, eds., International Antitrust Litigation: Conflict of Laws and Coordination, Hart Publishing, 2012. For extended discussion and further references, see also Mihail Danov, Jurisdiction and Judgments in Relation to EU Competition Law Claims, Hart Publishing, 2011. 69 See 2012 OJ L351/1. Regulation 1215/2012 is scheduled to enter into force on 10 January 2015. In order to take account of developments linked to the future Unified Patent Court and to the Benelux Court of Justice, the Commission has proposed more amendments to the Regulation. See COM(2013) 554 final of 26 July 2013. 70 Seen in a positive light, forum shopping may be an indicator that claimants are enjoying the benefits of the internal market; the likely reality, however, is that the plaintiffs that have real choices about where to sue are likely to be only well-resourced companies. In this sense, the undesirability of (too much scope for) forum shopping is not because of a vague feeling that plaintiffs are somehow cherry picking and eroding national preferences but because effective access to judicial process can in practice only be realized by a privileged class of (potential or actual) litigants. 71 Analogous cases have been brought elsewhere as well, in particular in the UK. See Ian Forrester and Mark Powell, ‘Market Forces and Private Enforcement: A Start But Some Way Still To Go’, this volume. 72 See Case C-352/13, Cartel Damage Claims Hydrogen Peroxide SA (CDC) v Evonik Degussa GmbH and others, not yet decided.

xxxviii  Mel Marquis according to which private antitrust litigation necessarily advances the goal of corrective/restorative justice, while public enforcement is necessarily limited to injunctive and punitive functions. There are overlaps, and Komninos is at pains to point out the quasi-Smithian dual nature of private claims: their immediate relevance concerns the particular dispute they bring before courts; but they also form part of a system of justice that has a wider public purpose. At least two implications might follow from these considerations. First, if private antitrust litigation can contribute significantly to the general interest, then its non-existence – and even now in some Member States it is practically nonexistent – should not be accepted. Private enforcement as an institution should be cultivated and harnessed by means of appropriate mechanisms that capitalize on its virtues while mitigating or avoiding its less desirable characteristics. In the best conditions, the development of a culture of authentic corrective justice feeds into the development of a competition culture, which is one of those projects never fully attained, in any jurisdiction. Second, if the pursuit of private interest through litigation presents a complementary means of advancing the public interest, then arguments to the effect that public enforcement is categorically to be privileged over private enforcement may be based on faulty assumptions and should at least be tested. But these are my words; Komninos presents an even stronger version of this idea, arguing from the premise that although in the EU there is a hierarchy of supranational law over national law (established in the classical jurisprudence of the 1960s), there is no foundation for a hierarchy of public enforcement of the EU competition rules over private actions to recover damages for the breach of those rules.73 These modes of enforcement are ‘institutionally independent’. And though it is not stated explicitly, one can accept that national courts enforcing directly effective Treaty rules, and doing so under the duty of sincere cooperation imposed by Article 4(3) TEU (and bound moreover to respect the general principles of EU law), are in functional terms agents of the Union.74 According to Komninos, ‘[i]ntroducing 73 For this proposition Komninos also cites paragraph 40 of the Advocate General’s Opinion in Case C-360/09, Pfleiderer v Bundeskartellamt [2011] ECR I-5161. The Advocate General adds, and few could doubt, that as a strictly factual matter it is public enforcement in Europe, and not private litigation, that makes the most impact in terms of securing compliance with Articles 101 and 102. And, as noted earlier, private claims may be contingent on prior public actions in a way that does not really apply vice versa: no public investigation will live or die according to what occurs in civil or commercial litigation before national courts. 74 See, eg, Case C-344/98, Masterfoods Ltd v HB Ice Cream Limited [2000] ECR I-11412, para 56 (‘application of the Community competition rules is based on an obligation of sincere cooperation between the national courts, on the one hand, and the Commission and the Community Courts, on the other, in the context of which each acts on the basis of the role assigned to it by the Treaty’). Similarly, see Case C-199/11, Europese Gemeenschap v Otis NV and others, judgment of the ECJ of 6 November 2012, not yet reported, para 52. The Commission’s lawsuit for damages before the Rechtbank van koophandel (Tribunal de Commerce) of Brussels has been interpreted as the Commission’s response to the stalled efforts to bring forward EU-level proposals on private enforcement. See case note, Case C-199/11, Commission v Otis, 50 Common Market Law Review 1105, 1114 (2013) (‘The relevance of the Otis judgment should be assessed in view of the difficulties encountered by the Commission to achieve a minimum harmonization of the national procedural rules relevant for the private enforcement of competition law. In view of such difficulties, the Commission might have decided to explore a ‘new road’ to promote private enforcement of competition law in Europe; rather than ‘improving the

Well Integrated Public and Private Antitrust Enforcement in the EU 

xxxix  

a rule of primacy [of public enforcement] would be problematic because it would undermine the role of courts as enforcers of equal standing’.75 If it is objected that Article 16(1) of Regulation 1/2003 establishes the hierarchical superiority of the Commission by precluding national courts from adopting decisions running counter to the Commission’s actual or contemplated decisions, Komninos replies that Article 16(1) and the jurisprudence on which it is based do not relate to any institutional hierarchy but rather give expression to the supremacy of EU law over national law. Also relevant here is the principle that national courts generally have no competence to rule on the validity of any instrument of EU law.76 With regard to the principle of primacy, Article 288 TFEU states that ‘[a] decision which specifies those to whom it is addressed shall be binding only on them’. One might therefore claim that a literal reading of Article 288 leaves national courts free to adopt decisions incompatible with a Commission decision, since the latter is addressed to (in the antitrust context) undertakings, and not courts. EU law cannot ‘take precedence’ over national law where they do not come into conflict. This interpretation is certainly a cheeky one, and its validity is highly doubtful in particular where a Commission decision imposes a fine, since Article 299 TFEU provides that an act of the Commission imposing a pecuniary penalty on persons other than states ‘shall be enforceable’ in national court in accordance with the relevant rules of civil procedure (which must be applied in conformity with Article 4(3) TFEU); although this obligation flows directly from Article 299 of the Treaty and not from the decision, it nonetheless endows the decision with binding authority vis-à-vis the national court in which enforcement is sought. However, quite apart from the defects of the ‘no conflict, hence no operation of primacy’ interpretation described above, the rule established in Masterfoods and enshrined in Article 16(1) of the Regulation environment’ in which private enforcement of EU competition law takes place in national courts through the harmonization of national procedural rules, in Otis the Commission ‘tested the possibility’ to directly start a damage compensation action in a national court.’ It is not explained why the Commission would have found it unnecessary to seek damages on behalf of the EU for losses sustained from collusive tendering if efforts to harmonize national rules on damages actions had proceeded more smoothly.). For discussion of some of the open procedural issues following the Otis case, see Ruchit Patel and Paul Stuart, ‘Now the Commission Wants Compensation Too … The Commission as Private Damages Claimant and its Implications’, CPI Antitrust Chronicle, July 2013 (1). Page 146. The limits of judicial competence in this respect are clear from Case 314/85, Foto-Frost v Hauptzollamt Lübeck-Ost [1987] ECR 4199, para 15 (national courts ‘do not have the power to declare acts of the Community institutions invalid. […] [T]he main purpose of the [preliminary reference procedure, now provided for in Article 267 TFEU] is to ensure that Community law is applied uniformly by national courts. That requirement of uniformity is particularly imperative when the validity of a Community act is in question. Divergences between courts in the Member States as to the validity of Community acts would be liable to place in jeopardy the very unity of the Community legal order and detract from the fundamental requirement of legal certainty.’). See also ibid, paras 17–19 (pointing out that in a preliminary reference proceeding the institution that authored the act in question has the opportunity to defend it before the Court; but leaving open the possibility for an exception to the Foto-Frost rule where a national court has to decide whether to adopt interim measures; on the strict conditions that must apply when a national court seeks to rely on the exception, see Case C-465/93, Atlanta Fruchthandelsgesellschaft mbH v Bundesamt für Ernährung und Forstwirtschaft [1995] ECR I-3761). 75 76

xl  Mel Marquis (‘no decisions running counter’) does not derive from the primacy of EU law77 but is based, originally, on the principle of sincere cooperation.78 As the Court states in Masterfoods, ‘[w]hen the outcome of a dispute before the national court depends on the validity of [an existing Commission decision subject to appeal before the EU Courts], it follows from the obligation of sincere cooperation that the national court should, in order to avoid reaching a decision that runs counter to that of the Commission, stay its proceedings pending final judgment in the action for annulment by the Community Courts, unless it considers that, in the circumstances of the case, a reference to the Court of Justice for a preliminary ruling on the validity of the Commission decision is warranted.’79 Recently, the Court described the Masterfoods rule, partly in light of the exclusive jurisdiction of the EU Courts (in most cases80) to review the legality of the acts of the institutions, as a ‘specific expression of the division of powers, within the EU, between, on the one hand, national courts and, on the other, the Commission and the EU Courts’.81 For its part, the Council has attached to Article 16 of Regulation 1/2003 the label of ‘Uniform application of Community competition law’, which is also of course a central objective on which the judge-made doctrine of primacy is based. But from the point of view of the Court, although the Masterfoods rule is a conflict rule, it is one that appears to flow 77 This misperception persists even in recent scholarship. See case note (cited above note 74), 50 CMLR at 1116 (‘In Masterfoods, the requirement for the national courts to comply with the Commission Decision was justified by the supremacy of the EU legal order and by the need to guarantee a consistent enforcement of EU competition rules throughout the EU Member States.’ (emphasis added)). 78 Although the principle of sincere cooperation is formally a reciprocal obligation, national courts cannot help but notice that under EU law there is no ‘reverse’ or symmetrical Masterfoods rule requiring the Commission to abstain from adopting a decision running counter to an actual or contemplated decision by a national judge. See Masterfoods v HB Ice Cream, cited above note 74, para 48. This asymmetry does not constitute a formal hierarchy but may in practice resemble one quite closely. 79 Masterfoods, ibid, para 57 (and para 59). See also Steven Preece, case note, Masterfoods Ltd (t/a Mars Ireland) v HB Ice Cream Ltd (C-344/98) [2000] E.C.R. I-11369 (ECJ), 22 European Competition Law Review 281–288, at pp 284–285 (‘[T]he ECJ in Masterfoods has not established that a Commission decision will bind the national courts, at least not in the strict sense of the term. The ECJ’s finding seems to be based upon the general principles of E.C. law and arguably makes a great deal of logical sense. National courts must apply the general principles of Community law when they are implementing Community law. Furthermore, Article 10 [EC] applies to all institutions of the Member State government, including the national courts. A judgment of a national court on the direct effect of Articles 81(1) or 82 [EC] is a national measure implementing Community law and the Member State court therefore has an obligation to comply with the principle of legal certainty. A judgment which fails to comply with this principle will infringe Community law, by virtue of Article 10.’ (emphasis in original; footnotes omitted)). The Delimitis rule, according to which a national court must not adopt a decision that would conflict with a decision contemplated by the Commission (also enshrined in Article 16 of the Regulation) was based on the requirement of legal certainty to which Preece refers. See Masterfoods, ibid, para 51. See also Veljko Milutinović, ‘The ‘Right to Damages’ in a ‘System of Parallel Competences’: A Fresh Look at BRT v SABAM and its Subsequent Interpretation’, this volume, footnote 66. In addition to legal certainty, the principle of sincere cooperation is also relevant: that principle implies that a national court ‘should, in order to avoid reaching a decision that runs counter to that of the Commission, stay its proceedings [or seek a preliminary ruling from the ECJ]’. Masterfoods, para 57. 80 See above note 76 in fine. 81 Commission v Otis, cited above note 74, para 54.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xli  

from institutional roles, assignments and interplay rather than from primacy.82 Of course, once the Commission adopts a decision the latter is an act under the Treaties and benefits from primacy vis-à-vis any rule of national law,83 as is true of (to use anachronistic terms) ‘first-pillar’ and ‘third-pillar’ EU law in general. But – for purposes of establishing the scope of the ‘binding’ effect (ie, the preclusion/estoppel effect) of a Commission decision, and conversely the margin for manoeuvre on the part of a national court hearing claims for damages, when would a decision of the latter ‘run counter’ to the decision of the Commission? Komninos explains that a national court is only obliged to avoid ruling in such a way as to prevent an addressee of a Commission decision from complying with the operative part of the decision; only in this scenario is there a ‘real conflict’,84 because only the operative part of the decision can produce legal effects.85 With regard to the very live issue of the legal effects of the conclusions contained in a decision (to the extent that they are not repeated in the operative part), on which I should think the ECJ sooner or later will have to speak, Komninos believes the parties to national proceedings should remain free to contradict those conclusions and to persuade the national court to do likewise. ‘An unqualified binding effect would essentially subjugate private to public enforcement; it would also withdraw from the ambit of national courts a substantial part of competition law disputes, in particular those referring to the infringement of the antitrust norm […].’86 With regard to the Commission’s idea of ‘universalizing’ Masterfoods so that (as in the German example) the conflict rule would also apply with respect to the final infringement decisions of national competition authorities and national review courts, it would be better, according to Komninos, if a finding of infringement made by a competition authority gave rise to a presumption of antitrust liability that could then be rebutted before the national court. 82 What can rightly be claimed is that the duty of sincere cooperation and the ‘division of powers’ to which the Court refers in Otis share, with the principle of primacy, a fundamental preoccupation for the (ideal of a) uniform application of EU law. 83 See, eg, Case C-119/05, Lucchini [2007] ECR I-6199 (Commission’s decision finding state aid to be incompatible with the Treaty was an act of Community law and thus prevailed over a national rule of res judicata; a national court’s decision barring the Italian State’s attempt to recover the illegally granted aid on the basis of res judicata was therefore itself unlawful, as the national court had failed to give full effect to the Community act by disapplying the contrary national rule or at least by seeking a preliminary ruling from the ECJ as to the validity of the Commission’s decision). 84 Cf. Lars Kjølbye, case note, Case C-344/98, Masterfoods v HB Ice Cream, 39 Common Market Law Review 175–184 (2001), at page 182 (‘[…] it could be argued that it is of significance that the Court of Justice does not employ the term ‘conflict’ but rather the term ‘running counter’ or ‘allant à l’encontre’ in the original French version. This term might be broader than the term ‘conflict’, which in the strict legal sense of the term would require that the operative parts of the instruments are irreconcilable. It is submitted, however, that such an interpretation would go too far. It is likely that the Court only had in mind the true legal conflicts that would put into question the uniform application of Community competition law and that, consequently, the obligation established in Masterfoods is limited to conflicts at the level of the operative parts of the instruments.’). 85 On the latter proposition that only the operative part of a decision can create legal effects, see Case T-138/89, NBV and NVB v Commission [1992] ECR II-2181. 86 Page 148.

xlii  Mel Marquis He may therefore be disappointed to see that Article 9 of the draft Directive of June 2013 would impose a seemingly unqualified obligation (without prejudice, of course, to Article 267 TFEU) on Member States to give preclusive effect, in the sense of Masterfoods,87 to final infringement decisions of competition authorities and review courts (but not to decisions merely contemplated, à la Delimitis, as compliance by courts would likely be quite burdensome).88 To make matters ‘worse’, as it were (at least from the perspective of the independence of private litigation and the courts), recital 25 of the draft Directive suggests that, not only must the operative part of the national decision be respected, so too must the recitals of the decision that support it. If recital 25 of the draft text were applied literally, and if the recitals supporting the operative part of a national decision were sufficiently comprehensive, national courts could perhaps become, as Komninos fears, ‘mere assessors of damages’. However, like other institutions, courts have survival instincts. If the Commission’s version of Article 9 were preserved in the final text, it could be expected that any perceived assault on their decisionmaking power would provoke defensive strategies, subtle or less subtle, so as not to give up too much ground.89 On the other hand, the European Parliament and the Council are both seeking to dilute Article 9; but if they do this the final text might be more conservative than what Komninos advocates. The Council, in particular, prefers simply to scrap the Masterfoods concept as far as cross-border cases are concerned. While a finding of an infringement by NCA X could be used as irrefutable evidence before a court of the X’s own Member State, courts of other Member States would merely be required to treat X’s decision as admissible evidence; not even a rebuttable presumption would be established. If this approach is adopted the Directive will be irrelevant for a dozen Member States, which already fit that pattern. Of course, Member States would in any event be able to go beyond this modest approach, for example by adopting a German-style rule or by adopting a presumption along the lines of what Komninos suggests. (For its part, the Parliament has discussed a presumption that could be rebutted where an infringement decision of an authority from another Member State is tainted by error or failure to respect procedure.) 87 The draft Directive uses the deceptively soft term ‘probative’ effect. As pointed out by Bruno Lasserre in his contribution to this volume, ‘the civil judge would still be responsible for adjudicating a number of pivotal issues, in particular by deciding [in the context of a collective action] whether or not to certify a group of claimants, by checking whether certain firms which were not addressees of the NCA decision can also be held liable and whether group structures have evolved, by assessing the causal link between the infringement and the injury, and by setting the level of damages’ (pp 321–322). 88 Other aspects related to the subject of the cross-border ‘binding effect’ of final infringement decisions in the courts of other Member States are discussed below in connection with the chapter written by Veljko Milutinović. See notes 205–207 and accompanying text. 89 On the possibility that the decisions of national judges may be guided in part by the perceived need to correct or deviate from (due process) shortcomings in the Commission’s infringement decisions, see Ian Forrester and Mark Powell, ‘Market Forces and Private Enforcement: A Start But Some Way Still to Go’, this volume.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xliii  

Komninos also considers the relationship between EU action in the field of private enforcement and leniency programs. First, he accepts the desirability of limiting the civil liability of successful immunity applicants, in particular because the joint and several liability of the remaining cartelists should provide adequate insurance that injured parties will obtain full compensation (thus ensuring compliance with the ‘constitutional’ principles spelled out in Courage and Manfredi). He would even exclude recovery of damages by direct and indirect purchasers (or suppliers, in the case of a buyers’ cartel), again on the joint and several liability rationale, a rule that could be disapplied in case the other cartelists turned out to be ‘judgmentproof’ or living on faraway tropical shores. It may be presumed that he embraces the somewhat toned down – but still bold – proposal of June 2013, which as noted above would require national laws to shield a successful immunity applicant from claims made by injured parties other than its direct or indirect purchasers or providers unless such injured parties show that they cannot obtain full compensation from the other infringers.90 As a second point concerning the relationship between private enforcement and leniency programs, Komninos also weighs in on the ECJ’s judgment in Pfleiderer. I will merely note here that his reaction to Pfleiderer is more positive than mine. He correctly identifies the rigidities that might have posed problems if the Advocate General’s categorical approach had been adopted; but he does not consider alternative options besides the solution proposed by the AG and the unstructured balancing of interests required of national judges by the Court of Justice. Komninos is not alone in defending Pfleiderer, however: Luís Morais considers that the chorus of criticism for the judgment is unfair because, in the absence of a stronger legal base, the Court could not go much further than it did.91 In my view, the ECJ only made a poor solution worse in its judgment in Donau Chemie.92 Arguably, whatever may become of the other features of the draft Directive, one can expect that a better defined framework for the courtordered disclosure of evidence in the possession of competition authorities will emerge in the final legislation, and that at least in some Member States this will improve the chances that a plaintiff’s claim will be assessed on its merits and not by operation of procedural rules. Some commentators, and some within the Parliament, read Donau Chemie expansively and conclude that, since it prohibited a far-reaching rule under Austrian law, it equally precludes the EU legislator from granting absolute protection against disclosure for corporate leniency statements and settlement submissions, which is precisely what the Commission has proposed See above notes 51–52. Morais, ‘Legal Issues’, this volume, page 125. Case C-536/11, Bundeswettbewerbsbehörde v Donau Chemie AG and others, judgment of the ECJ of 6 June 2013, especially para 48, which states: ‘It is only if there is a risk that a given document may actually undermine the public interest relating to the effectiveness of the national leniency programme that non-disclosure of that document may be justified.’ On a strict interpretation of this paragraph, the implication is that a presumption applies in favour of disclosure; if this reading is correct it means that, until such time as the EU adopts legislation to regulate access to evidence, a national judge, when in doubt, must resolve the uncertainty against protection for the given document. On a different reading (ie, not mine), the loose wording (what is a ‘risk’?) may be interpreted as the Court giving national courts substantial discretion, which would soften or neutralize the presumption just described. 90 91 92

xliv  Mel Marquis in Article 6(1) of the draft Directive. Although this syllogistic interpretation is not downright implausible, I doubt that Pfleiderer and Donau Chemie necessarily must be taken as controlling precedents that pre-determine the way the ECJ would rule if it were called on to consider the validity of Article 6(1) in its present form.93 But there is also reason to doubt that the ECJ would ever have to decide that precise issue: preliminary indications are that the European Parliament itself subscribes to the broad reading of the above case law and is seeking to relax the relevant provision so that corporate statements and settlement submissions would only enjoy relative and not absolute protection.94 93 My caution on this point is largely linked to the premise on which Pfleiderer and Donau Chemie were decided, ie, that national judges are expected to balance interests to determine disclosure in the absence of governing EU rules. This conditional reading of the case law is said bluntly to be incorrect by, among others, Christian Kersting, ‘Removing the Tension Between Public and Private Enforcement: Disclosure and Privileges for Successful Leniency Applicants’, 5(1) Journal of European Competition Law and Practice 2–5 (2014), at 3. (‘[It cannot] be argued that the ECJ explicitly ruled against the background of ‘the absence of EU rules governing the matter’ […] The ECJ has based its decisions on the primary law principle of effectiveness. This also binds the EU legislator. The Member States cannot be forced by secondary EU law to introduce the very measures which primary EU law forbids them to introduce.’ (emphasis in original; footnotes omitted) Apart from the technical and subsidiary point that Article 6(1) is not ‘the very measure’ that was challenged under Austrian law in Donau Chemie, Kersting may be looking at the issue through a narrow lens. The future Directive is expected to soften a quite restrictive tradition of evidence disclosure – ie, a rigidity Member States have the prerogative to maintain, absent EU rules, under the principle of procedural autonomy – and thereby introduce a significantly more liberal regime. In this respect, the Commission’s proposal rules on evidence disclosure enhance access to justice to some extent, even in their present graduated form. I am therefore more sympathetic to a discussion put forward by Hirst, which seems to illustrate the ambiguity of the present situation: ‘[T]he statements of the [ECJ in Donau Chemie] do lead to the conclusion that, were it essential for evidencing a damages claim, victims ought to have access to leniency applicants’ self-incriminating corporate statements. This in turn raises the question of whether the [Commission’s] own proposal, absolutely prohibiting any access to corporate statements, would fall foul of the case law. In its defence, the Commission may point to the other initiatives taken in the proposal which facilitate claims, including the opening up of disclosure rules.’ Nicholas Hirst, ‘Donau Chemie: National Rules Impeding Access to Antitrust Files Liable to Breach EU Law’, 4 Journal of European Competition Law and Practice 484–486 (2013), at 486 (emphasis added). A more principled argument could also be raised by drawing on the point made by Milutinović in the final chapter: if the right to damages derives from the imperative of effectiveness of the competition rules, as it certainly seems to do in the case law, then it may be fallacious to conclude that the risk of undermining the effectiveness of the derivative right (and the principle of effectiveness does apply to the right to damages, and not only to the competition rules that constitute the foundation for that right – see Manfredi; see also Article 3 of the proposed Directive) should be accorded as much or more weight than the risk of undermining the effectiveness of Article 101 more generally by tampering with the device that secures, better than other tools by far, the exposure of cartels. Having said all of that, as noted immediately below in the main text, the self-assured interpretation adopted by Kersting may never be tested judicially if the European Parliament succeeds in diluting Article 6(1) of the draft text. 94 A possible middle ground, for those who are concerned with preserving adequate incentives for leniency applicants but who also feel that absolute and permanent protection from disclosure for corporate leniency statements and settlement submissions denies injured parties their right to damages, is to relax the degree of protection for such materials (by replacing the principle of permanent protection with a ‘grey list’-style, temporary protection) while simultaneously reinforcing the degree of (qualified) insulation of successful leniency applicants from liability, in particular vis-à-vis their direct and indirect purchasers. For the modalities of such a solution, see Caroline Cauffman, ‘The European Commission Proposal for a Directive on Antitrust Damages: A First Assessment’, MEPLI Working Paper Series 2013-13 (October 2013), at § 4.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xlv  

Luís Silva Morais: ‘Integrating Public and Private Enforcement in Europe: Legal Issues’.95 Like the Komninos chapter, this one too is rich and thoughtful. It is a bit longer than most of the other chapters, and the reader might more easily discern the structure from the following crude ‘table of contents’: (i) introduction; (ii) foundations relating to the interplay between public and private enforcement; (iii) access to information in the context of public enforcement (ie, access to the enforcer’s case file and to sensitive contents of infringement decisions) and its interaction with private enforcement; (iv) leniency programs and their reciprocal impact on private enforcement; (v) incentives that could encourage private enforcement in cases unlikely to be addressed through public enforcement; and (vi) a possible incrementalist approach to foster private antitrust litigation without undermining public enforcement. Not every section of the chapter can be recounted and discussed here; only a few points will be made. Morais’ nuanced argument might be depicted as one that: praises the Commission for abandoning the occasionally wild-eyed ambitions of the 2005 Green Paper and turning to a less ambitious agenda (eg, an agenda which leaves deterrence to the public authorities and which relies to some extent on soft law); while, at the same time, criticizing the Commission for not being ambitious enough, or more precisely for neglecting a range of subtle measures that may gradually strengthen private enforcement not at the expense of, but with the assistance of public enforcement mechanisms. As a normative theme running through the chapter, Morais argues that public enforcement should remain the dominant driver of European competition law, and that private enforcement should play a ‘strictly complementary and subsidiary’ role.96 This conception of the relationship between the two modes of enforcement implies that it is inappropriate to recast or reformulate the foundations of competition enforcement as we know them in Europe, but Morais acknowledges that proper coordination mechanisms may be needed to ensure that public and private enforcement are working with and not against each other. The introduction to the chapter (section 1) provides a narrative of the development of EU initiatives in the field of private antitrust litigation since Courage was decided in 2001. It also discusses fundamental questions relating to EU action in this area, including not least the issue of which legal base or bases are appropriate for such action.97 The draft Directive of June 2013 invokes 95 The chapter written by Morais actually precedes that of Assimakis Komninos. I shuffle the order slightly here to facilitate the flow of the discussion. 96 Page 110. 97 For discussion of the possible range of legal bases, and with reference to the Commission’s (withdrawn) draft Directive of 2009, see also Mel Marquis, ‘Cartel Settlements and Commitment Decisions’, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2008: Antitrust Settlements under EC Competition Law, Hart Publishing, 2010, at page lxix, footnote 148. More recently, discussion of the Commission’s choice of a dual legal base for the proposal of June 2013 is provided by Cauffman, ‘A First Assessment’, cited above note 94, at § 3.2 (discussing the ECJ’s case law on the legitimacy of dual/multiple legal bases (and on the criteria that apply where Article 114 is invoked) and concluding, given the relatively flexible approach of the ECJ where it is clear that no attempt is being made to circumvent the Treaty’s essential procedures and where the

xlvi  Mel Marquis Articles 103 and, ‘transcendently’, 114 TFEU98 as its twin legal base, as the Commission considers that the Directive ‘pursues two equally important goals which are inextricably linked, namely (a) to give effect to the principles set out in Articles 101 and 102 of the Treaty and (b) to ensure a more level playing field for undertakings operating in the internal market, and to make it easier for citizens and businesses to make use of the rights they derive from the internal market’.99 Morais also refers to the possible relevance of Article 81 TFEU.100 The introduction is also concerned, as other contributions to this volume are, with how private enforcement is to be understood, since it is not a monolithic concept. In section 1.4.1 of the chapter, Morais provides his taxonomy of the different kinds of claims that might be pursued with the aim of seeking damages, or injunctive relief, or, on the side of defendants, the claims that may have to be raised on the basis of antitrust law when they are accused of breaching contractual obligations or intellectual property rights. Similar to Fred Louis and others, Morais considers provisions in question are mutually compatible, that the draft Directive does not appear unlawful from that point of view). 98 Article 103 provides for the adoption of EU directives and regulations that ‘give effect’ to the principles contained in Articles 101 and 102. Article 114 provides for the adoption of EU measures, where appropriate for the establishment and functioning of the internal market, which approximate national provisions of law in a given field. The subtext to the Commission’s use of a twin legal base is that Article 103 contemplates the adoption of EU measures exclusively by the Council of Ministers, which is only obliged in that case to consult the Parliament. Needless to say, the Parliament insists that it must be involved as a co-equal legislator so far as harmonizing measures in the field of private antitrust litigation are concerned, and politically the Commission had little choice but to cast the latest initiative as an internal market instrument. (For further background, see Marquis, cited previous footnote.) To achieve this, the Commission relies on and discusses at length the (growing) fragmentation of procedural rules at the national level, which tilts playing fields on both supply and demand sides of markets, and even portrays the diverse regimes of the Member States as interfering with the EU right of cross-border establishment. See pp 9–10 of the Explanatory Memorandum attached to the draft Directive (Annex I to this volume). The Commission also explains that the scope of the draft Directive exceeds the limits of Article 103 because it requires minimum rules at the national level that pertain not just to infringements of the EU antitrust rules but also the antitrust rules of the Member States, to the extent that they are applied (in parallel with the EU rules) to agreements and practices that affect trade between Member States. See ibid, page 10. 99 See Explanatory Memorandum, ibid, page 8. 100 Article 81, which could conceivably be relevant for future EU initiatives in the field of private enforcement, provides that the Union will develop judicial cooperation in civil matters that have ‘crossborder implications, based on the principle of mutual recognition of judgments and of decisions in extrajudicial cases’. This may entail the adoption, via the ordinary legislative procedure, of measures that approximate national laws. Such action is envisaged in particular when necessary for the proper functioning of the internal market, and the range of possible measures concern: (a) the mutual recognition and enforcement between Member States of judgments and of decisions in extrajudicial cases; (b) the cross-border service of judicial and extrajudicial documents; (c) the compatibility of the rules applicable in the Member States concerning conflict of laws and of jurisdiction; (d) cooperation in the taking of evidence; (e) effective access to justice; (f) the elimination of obstacles to the proper functioning of civil proceedings, if necessary by promoting the compatibility of the rules on civil procedure applicable in the Member States; (g) the development of alternative methods of dispute settlement; and (h) support for the training of the judiciary and judicial staff.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xlvii  

that too much attention is reserved for follow-on damages actions when the real complementarity between public and private enforcement lies in what Morais calls private enforcement stricto sensu: stand-alone actions (eg, relating to rules on the abuse of dominance) and the sorts of defensive claims described above. With regard to private enforcement stricto sensu, Morais stresses the importance of commercial arbitration, under the shroud of which significant antitrust disputation occurs. It is well known that in the context of arbitration, the principles driving the EU antitrust rules are a matter of public policy, and that an arbitral award incompatible with those rules should be unenforceable notwithstanding the usual rule precluding judicial review of the award’s merits.101 It is poorly known what transpires within an arbitrator’s dark sanctum, other than anecdotally or on the rare occasions when parties partially or totally waive confidentiality. This may result in a submerged legal order, bound by principles of EU law but where the application of those principles tends to elude verification,102 and where the compilation of empirical data and the drawing of lessons become difficult or impossible. Of course, it is the very nature of arbitration and some of its inherent advantageous characteristics – opacity and inoculation in ordinary circumstances from judicial review – that can be problematic. Easy answers are unlikely to be found, but Morais suggests that more could be done in this regard, consistent with both the fundamental EU jurisprudence and national prerogatives in this area, and perhaps without upsetting the institutions of arbitration too much.103 Without 101 See Case C-126/97, Eco Swiss v Benetton [1999] ECR I-3055 (adoption by an arbitrator of an award incompatible with Article 101 TFEU (and ex hypothesi with Articles 102 and 106) obliges national courts, on grounds of public policy, to set aside or refuse to enforce the corresponding arbitral award if domestic law requires them to observe national rules of public policy); Joined Cases C-430 and C-431/93, Jeroen van Schijndel and Johannes Nicolaas Cornelis van Veen v Stichting Pensioenfonds voor Fysiotherapeuten [1995] ECR I-4705 (national courts must in principle apply Articles 101, 102 and 106 at their own initiative where those rules are relevant to a civil case but the parties fail to invoke them; but not if introducing the issue of a possible breach would oblige them to go beyond the scope of the dispute defined by the parties and to rely on facts and circumstances other than those pleaded by the party with an interest in having the relevant competition rules applied). A variety of related literature is cited in Morais’ chapter. For further discussion and references to literature relating to arbitration of EU (and US) antitrust claims, see, eg, Gordon Blanke, ‘EU Competition Arbitration’, in Luis Ortiz Blanco, ed., EU Competition Procedure, 3rd edition, OUP, 2013, chapter 29; Colette Downie, ‘Will Australia Trust Arbitrators with Antitrust?’, 30(3) Journal of International Arbitration 221–266 (2013); Assimakis Komninos, ‘Arbitration and EU Competition Law’, in Basedow et al., eds., International Antitrust Litigation, cited above note 68, chapter 9; Gordon Blanke and Phillip Landolt, eds., EU and US Antitrust Arbitration, Wolters Kluwer, 2011; Mihail Danov, Jurisdiction and Judgments in Relation to EU Competition Law Claims, Hart Publishing, 2011, chapter 7. 102 It has been pointed out, however, that arbitration is not used, at least nowadays, as a subterfuge to facilitate cartel activity. See Laurence Idot, ‘Arbitration and Competition’, OECD background note, DAF/COMP(2010)40, pp 51–86, at page 53 (with reference to Jacques Werner, ‘Application of Competition Laws by Arbitrators: The Step Too Far’, 12(1) Journal of International Arbitration 21–26, at 23 (1995)). 103 ‘I believe that some incremental measures or steps may reinforce the important status […] of arbitration procedures in order to foster stricto sensu private enforcement of competition rules. Such measures include, eg, the encouragement, through soft harmonization impulses, of legislative changes of national laws on arbitration that give more weight to public policy considerations […] as an element of possible annulment or non-enforceability of arbitral awards […]. Another possible measure could be the extension or adaptation of Articles 15(1) and 16(2) of Regulation 1/2003 in

xlviii  Mel Marquis knowing if Morais would agree, in my view it remains desirable, as several Workshop participants believed in 2001, for the Court of Justice to retreat from its Nordsee judgment104 and to allow arbitrators to submit preliminary references, perhaps with unique admissibility and confidentiality criteria (and perhaps to the General Court), if they consider it useful for the resolution of a dispute.105 The wisdom of opening up such an option is arguably underscored by the substantial economic significance of modern commercial arbitration.106 Although arbitrators are not bound by the duty of sincere cooperation107 and would not be formally obliged under EU law to respect the ruling issued by the ECJ (or the GC), an award that failed to comply with such a ruling would be unenforceable by the courts of the Member States, to whom the duty of sincere cooperation clearly does apply.108 (As an aside, the arbitration of antitrust claims under U.S. law has been the subject of significant seismic activity, especially as regards class arbitration order to cover information and opinions to be addressed to arbitral courts and a systematic effort to identify and collect arbitral awards in which competition law issues have been discussed and considered as a basis for the final decision.’ Morais, ‘Legal Issues’, page 117 (emphasis in original; footnote omitted). 104 Case C-393/92, Nordsee Deutsche Hochseefischrei GmbH v Reederei Mond Hochseefischerei Nordstern AG & Co. KG [1982] ECR 1095, paras 10–12. See also Case C-125/04, Denuit, Cordenier v Transorient – Mosaïque Voyages et Culture [2005] ECR I-923, paras 13–16. 105 See Claus-Dieter Ehlermann and Isabela Atanasiu, eds., Effective Private Enforcement of EC Antitrust Law, cited above note 1, at pp 295, 297–298 and 300, interventions of Carl Baudenbacher, Walter van Gerven and Jürgen Basedow, respectively. See also Baudenbacher, ‘Enforcement of EC and EEA Competition Rules by Arbitration Tribunals Inside and Outside the EU’, in ibid, pp 341 et seq., at pp 358–360. Of course, the Court of Justice is not fond of directly reversing its own jurisprudence. Alternative options, including ‘indirect’ preliminary references, are discussed by Assimakis Komninos, ‘Assistance to Arbitral Tribunals in the Application of EC Competition Law’, in ibid, pp 363 et seq., at pp 367–379. 106 For the view that the ECJ’s position in Nordsee should be maintained, see, eg, Idot, ‘Aribtration and Competition’, cited above note 102, page 67. As Idot remarks: ‘Contrary to a portion of the [legal scholarship], we are not in favour of a reversal. Besides it being difficult on a practical level to impose an additional burden on the Court of Justice [ie, to exacerbate its workload], it does not seem justified to us for two reasons. In the first place, in case of difficulty, at some time or another one will go before the national judge and the latter may [submit a preliminary reference to the Court], as was illustrated in the Commune d’Almelo and Eco Swiss decisions. This may [occur only after some delay], but in any case, it makes it possible to preserve the uniformity of European Union law, which remains the objective of [the preliminary reference procedure]. In the second place, the immense majority of arbitration in Community law concern competition disputes and, in this case, it is cooperation with the competition authorities which must be developed if the need to do so is felt.’ These are valid points (and the last point about expanding the involvement of competition agencies is not necessarily incompatible with overturning Nordsee), and some tradeoffs seem undeniable. However, with respect to the concern that the ECJ (or GC) would receive too many requests for rulings, there may be practical mechanisms available that could avoid docket overload. For example, as suggested in the main text, the admissibility of a request for a preliminary ruling could be made subject to stricter standards than the relatively liberal standard that applies under Article 267 TFEU and the corresponding jurisprudence. Furthermore, the preliminary rulings issued to arbitrators could in general be decided by a single judge, unless the responsible judge saw fit, in light of the importance of the dispute, to have the case transferred to a chamber. 107 See, eg, Komninos, ibid, pp 369–370. 108 See ibid, page 301, where Mario Siragusa remarks that, ‘[i]f the arbitration tribunal were to ignore the answer received from the ECJ to their request for a preliminary ruling, then the award is not enforceable by the courts of the EC Member States’.

Well Integrated Public and Private Antitrust Enforcement in the EU 

xlix  

and as regards the ‘bindingness’ of arbitration clauses. This subject is taken up in following chapter by Hawk and Seaton – see below.) In sections 3 and 4 of his chapter, Morais discusses the tension between the need for public enforcers to protect confidential materials (eg, sensitive information in their files, and in particular, leniency materials) and the need for private litigants to have sufficient evidence to vindicate meritorious claims, whether access is sought through court order, Pfleiderer or National Grid-style, or by way of the transparency provisions and exceptions thereto contained in Regulation 1049/2001 (read in light of Articles 15(1) and 15(3) TFEU and Article 42 of the EU Charter of Fundamental Rights), where the jurisprudence of the Court of Justice in the field of merger control (Editions Odile Jacob and Agrofert, establishing a rebuttable presumption against disclosure of ‘external’ documents, ie, those not generated internally by the Commission’s services) and its contrast with the way the General Court has handled analogous issues in antitrust cases may imply future adjustments.109 The basic argument put forward, consistent with Morais’ overall position, is that in the presence of a conflict the risks posed to the effectiveness of public enforcement by easy access to evidence should weigh heavily in the balance. From that general orientation, Morais discusses the Pfleiderer and Donau Chemie cases and concludes (at section 4.3.2 in fine), as many if not most do, that legislators are better placed than judges to reconcile the opposing long-term needs of public (and private) enforcement and the short-term access to justice considerations that may justify the disclosure of certain materials to particular claimants. With regard to the legislative proposal now on the table, it has already been mentioned that Morais takes a generally critical view,110 in particular because he considers that important details are lacking and because, as he seems to intimate, a relatively liberal regime for post-investigation access to sensitive documents (the so-called ‘grey list’ category) could be risky. However, Morais also recognizes that the draft Directive incorporates some ‘important solutions’ aimed at enhancing the possibility that victims of antitrust infringements may obtain relief in national courts. Sections 6.2 and 6.3 provide a brief point-by-point reaction to the Commission’s proposed text and a series of suggestions as to other measures that could be employed, either by means of amendment to the draft or by means of other, essentially soft or informal techniques (relying, in particular, on 109 This would be so if, for example, the Court of Justice decides, in Case C-365/12 P, Commission v EnBW Energie Baden-Württemberg, to annul the judgment of the General Court. For discussion of these issues, see Küllike Jürimäe, ‘The Interaction between EU Transparency Policy and the Enforcement of EU Competition Law: Who Should Strike the Balance and How Should it be Struck?’, in Philip Lowe and Mel Marquis, eds., European Competition Law Annual 2012: Public Policies, Regulation and Economic Distress, Hart Publishing, forthcoming; Gaëtane Goddin, ‘Access to Documents in Competition Files: Where Do We Stand, Two Years after TGI?’, 4(2) Journal of European Competition Law and Practice 112–120 (2013). See also Gaëtane Goddin, ‘Recent Judgments Regarding Transparency and Access to Documents in the Field of Competition Law: Where does the Court of Justice of the EU Strike the Balance?’, 2(1) Journal of European Competition Law and Practice 10–23 (2011). 110 See above note 25 and accompanying text.

l  Mel Marquis soft harmonization across the Member States) that build, gradually, on the EU’s nascent acquis in this area. Barry Hawk and Yolaine Seaton: ‘U.S. Antitrust Arbitration’. With a title such as this the natural point of departure is the 1985 Mitsubishi judgment of the U.S. Supreme Court,111 which capsized the then-conventional judicial wisdom that antitrust law was not a proper subject of commercial arbitration. In Mitsubishi and subsequent cases the Supreme Court, interpreting the U.S. Federal Arbitration Act (FAA), has ascribed to Congress a strong policy in favor of arbitration where contracting parties have agreed, even in boilerplate language, that disputes arising from their contractual relations are to be decided finally and exclusively by an arbitrator. Such compulsory arbitration clauses are thus, according to this judicial philosophy, to be broadly construed. The liberal approach to arbitrability, with its strong ‘freedom of contract’ bent, sets the stage for a particular trajectory of jurisprudence highlighted by Hawk and Seaton in section IV, the centerpiece of the chapter. This section discusses the arbitrability of class action claims and the enforceability of class action waivers where arbitration is the stipulated method of dispute resolution.112 If I may jump right to the end of the section, the authors there conclude: ‘In sum, AT&T Mobility LLC v Concepcion strongly suggests that although the FAA can be regarded as a statute favoring arbitration, the FAA disfavors class arbitrations.’113 In AT&T Mobility,114 a husband and wife had concluded an agreement with a mobile phone company for the sale and servicing of cellular phones. The agreement obliged each customer to pursue any claims solely through arbitration, and solely in an ‘individual capacity’. The amount of money at stake for the married couple was about 30 dollars (a sales tax on what were supposed to be ‘free phones’), but they brought a class action against AT&T in a California federal district court. When the defendant company sought to compel arbitration exclusively on the basis of individual claims, first the District Court and then the Ninth Circuit found the class action waiver to be unconscionable and thus unenforceable under California state law. In a five-to-four opinion written by Justice Scalia, the Supreme Court held that under the circumstances the FAA pre-empted California law; not even 111 Mitsubishi Motors Corp. v Soler Chrysler-Plymouth, Inc., 473 U.S. 614 (1985), effectively overturning the judgment of the Second Circuit in American Safety Equipment Corp. v J.P. Maguire & Co., 391 F.2d 821 (1968). The finding that international arbitral tribunals could properly decide claims under the Sherman Act was echoed, with regard to domestic arbitration, in Gilmer v Interstate/Johnson Lane Corp., 500 U.S. 20 (1991). 112 Although the main focus of the chapter is on the use of arbitration in the context of private disputes, Hawk and Seaton also discuss, in section VII of the chapter, the use of arbitration by the federal enforcers. Overall, arbitration has been used sparingly in this context, for reasons explained by the authors. This is especially true as regards the U.S. Federal Trade Commission, whose powers are circumscribed by particular principles of constitutional law. 113 Page 171. 114 AT&T Mobility v Concepcion, 563 U.S. ___ (2011). For further discussion of the case, see Mark Mandich, ‘AT&T v Concepcion: The End of the Modern Class Action’, 14 Loyola Journal of Public Interest Law 205–236 (2012).

Well Integrated Public and Private Antitrust Enforcement in the EU 

li  

unconscionability – the residual paternalism reserved under state contract law used historically as a corrective for dramatic asymmetry of bargaining power – could withstand the overriding will of Congress.115 Notwithstanding the conclusion drawn by Hawk and Seaton on the basis of the views of a majority of the Supreme Court – that the FAA disfavors class arbitrations – the Second Circuit decided to interpret AT&T Mobility much more narrowly when deciding the American Express case in 2011. According to the Second Circuit, AT&T Mobility turned on the conflict between state and federal law, and state law had to submit. Distinguishing AT&T Mobility in this way enabled the Second Circuit to conclude, in American Express, that a class action waiver was void as a matter of federal law because, given the low value of the plaintiffs’ claims relative to the high cost of obtaining the economic expert evidence that would be required to substantiate the alleged adverse competitive effects of tying under Section 1 of the Sherman Act, it would be economically irrational to pursue them on an individual basis. This reasoning drew on dicta contained in Mitsubishi according to which an arbitration clause should be deemed void if its application would preclude the vindication of federal statutory rights. We now know, in light of the Supreme Court’s judgment in American Express v Italian Colors,116 decided on 20 June 2013, that Hawk and Seaton were better than the Second Circuit at reading tea leaves. In this remarkable judgment, Justice Scalia and a 5-3 majority rebuke the Second Circuit; ignore a well-reasoned amicus brief underlining that arbitration agreements precluding ‘effective vindication’ of federal antitrust claims should be deemed contrary to public policy;117 and invite businesses in countless industries, to the extent they have not yet done so, to immunize themselves from class litigation by incorporating waivers systematically in the contracts they sign with small businesses and individual consumers incapable of bargaining over terms and conditions. In a part of the opinion that tests the line between serious and disingenuous, the Supreme Court opines that while the ‘effective vindication’ passage in Mitsubishi sought to avoid prospective waivers of a claimant’s right to pursue statutory remedies, ‘the fact that it is not worth the expense involved in proving a statutory remedy does not constitute the elimination of the right to pursue that remedy’.118 Justice Scalia attributes great weight to the fact that there was no 115 Unconscionability remains a possible argument, however (as do other contract defences such as fraud, duress, etc.), with respect to the question of whether there is a valid agreement to arbitrate in the first place (ie, not the more specific question of whether compulsory bilateral arbitration is unconscionable or contrary to the public policy of a US state). This follows the FAA itself, which states at 9 U.S.C. § 2 that a written provision in any commercial contract whereby the parties agree ‘to settle by arbitration a controversy thereafter arising out of such contract […] shall be valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract’. (emphasis added) 116 American Express Co. et al. v Italian Colors Restaurant et al., 570 U.S. ___ (2013). 117 The brief was prepared by a number of eminent scholars and it united traditional defenders of robust private enforcement, such as Eleanor Fox, with others that have been critical of private antitrust litigation, including notably Herb Hovenkamp. 118 American Express v Italian Colors, cited above note 116, page 7 of the slip opinion (emphasis in original). See also ibid, page 4 (‘[T]he antitrust laws do not guarantee an affordable procedural path to the vindication of every claim.’)

lii  Mel Marquis hand-wringing about the absence of class action litigation in the four decades that elapsed between the adoption of the Sherman Act and the introduction of class action procedures (FRCP 23) in 1938. This anachronistic reasoning is buttressed by what is perhaps a more respectable argument. As Scalia writes: The regime established by the [Second Circuit] would require – before a plaintiff can be held to contractually agreed bilateral arbitration – that a federal court determine (and the parties litigate) the legal requirements for success on the merits claim-by-claim and theory-by-theory, the evidence necessary to meet those requirements, the cost of developing that evidence, and the damages that would be recovered in the event of success. Such a preliminary litigating hurdle would undoubtedly destroy the prospect of speedy resolution that arbitration in general and bilateral arbitration in particular was meant to secure. The FAA does not sanction such a judicially created superstructure.119

The latter argument is independent of the formalistic proposition that the right to pursue a remedy and the feasibility of doing it may be neatly distinguished. One can even read the quoted passage above as a frank admission that what the Supreme Court has really done in American Express is to decide the tradeoff between the benefit of a quick res judicata at the lowest private and social/ judicial cost, on the one hand, and the individual and aggregate costs of the non-compensation of antitrust-related harm and, more generally, lower-intensity antitrust enforcement.120 One question that springs from the judgment is whether the way individual claims have in practice been traded off is a faithful expression of congressional will. One cannot say with certainty, but what is known is that some legislators were disturbed by the Supreme Court’s judgment in AT&T Mobility, and must be equally concerned by American Express. Pending bills in Congress are designed to substantially redraw the lines laid down by the Court in its recent case law interpreting the FAA,121 but the bills are by no means bipartisan efforts. Fierce and possibly fatal opposition from Republican lawmakers can be expected.

Ibid, page 9 of the slip opinion. Of course, American Express would not reduce to zero the intensity of antitrust enforcement, in those sectors where waivers of class action proceedings are used in compulsory arbitration agreements and where individual claims do not justify the cost of pursuing them vigorously, since federal and/ or state antitrust authorities – not being bound by arbitral decisions – might intervene in some cases. 121 For the proposed ‘Arbitration Fairness Act’, see S.878 in the U.S. Senate and H.R. 1844 in the House of Representatives. Section 2 of S.878, for example, states that ‘[a] series of decisions by the Supreme Court of the United States have interpreted the Act so that it now extends to consumer disputes and employment disputes, contrary to the intent of Congress’. Further, ‘[a]rbitration can be an acceptable alternative when consent to the arbitration is truly voluntary, and occurs after the dispute arises’. A new Section 402 of the FAA, if it were to survive and be enacted in its current form, would provide that ‘no predispute arbitration agreement shall be valid or enforceable if it requires arbitration of an […] antitrust dispute […]’. (emphasis added) 119

120

Well Integrated Public and Private Antitrust Enforcement in the EU 

Part III

liii  

Options for Collective Redress in the European Union

J Thomas Rosch: ‘Designing a Private Remedies System for Antitrust Cases – Lessons Learned from the U.S. Experience’. As the title suggests, in this chapter Tom Rosch draws on the experience he has gained since he started practising law in 1965 in order to identify the flaws of private antitrust enforcement in the U.S. (which have led to the well-known ‘equilibrating tendencies’ of the federal courts to raise substantive and procedural bars ever higher, for private plaintiffs and agencies alike122) and to collect his recommendations for the EU as it strives to promote more effective damages actions and collective redress. His observations and advice may be summarized as follows. Main flaws of private antitrust enforcement in the United States First, Rosch considers, like many on both sides of the Atlantic, that treble-damage antitrust class actions in the United States are out of control.123 He accepts that plaintiffs and especially plaintiffs’ attorneys require adequate incentives to bring legal actions if private claims are to be effective. But he recalls that, in general, the incentives to bring individual claims under the Sherman Act seemed to be sufficient in the days before class actions became the norm. Under Section 4 of the Clayton Act, the plaintiff was able to recover three times the loss she sustained plus costs and attorney’s fees. Section 4 is indeed potent even when it is not combined with the additional leverage gained when a class action is brought. On the other hand, Europe has no Section 4, which suggests that comparisons should be considered with caution. Furthermore, as Rosch explains very well, the in terrorem effect of a treble-damages class action under federal law, which famously induces defendants to settle cases that sometimes may not be meritorious in order to avoid an even larger payout following trial by juries that may sometimes be bewildered by antitrust law, is potentially amplified by the possibility of state-law based class actions brought by indirect purchasers against the same defendants. Since ‘passing on’ theories are disallowed in federal court, the combination of claims by direct purchasers who may have passed on overcharges and claims by indirect purchasers under state law may lead to, or threaten to lead to, exposure to damages well beyond the trebling foreseen in Section 4 of the Clayton Act and hence in some cases, ‘extortionate settlements’.124 In short, the lack of coordination 122 In section III of the chapter, Rosch discusses some of the famous judgments in which the Supreme Court has refashioned antitrust law and litigation in the United States, including, eg, Twombly, Trinko and Credit Suisse. As has been highlighted by various commentators, the hollowing out of American antitrust law by the Supreme Court has spilled over (or has even ‘slopped over’, in American parlance) and has to some extent raised the cost of enforcement for the federal antitrust agencies, even though the concerns driving the ‘hollowing’ effect may not apply in all respects in the context of public enforcement. 123 A similar view is advanced vigorously by James Keyte, ‘Collective Redress: Perspectives from the U.S. Experience’, this volume. 124 Page 185.

liv  Mel Marquis between the judicial interpretation of federal antitrust law by the Supreme Court and the more populist antitrust legislation in a number of U.S. states has resulted in a dysfunctional private enforcement system. The second flaw of the U.S. system is the rule according to which antitrust defendants that do not settle a case and are ultimately found to have infringed the Sherman Act are jointly and severally liable and have no right to any contribution from defendants that did settle.125 The pre-trebled amount of a settlement will be subtracted from the damages the plaintiff may recover from the non-settling defendants, but the latter will bear the rest of the damages burden even if it exceeds the trebled amount of the harm they caused. Ceteris paribus, this would appear to be another element that distorts the choice of whether or not to settle a case. The third flaw relates to the asymmetric costs of automatic discovery rules, the brunt of which is again borne by defendants. Those who have experienced document reviews in the United States may appreciate how wasteful much of the discovery process can be in terms of manpower, time and money. The need to review electronic files adds a further layer of cost and effort. Rosch’s concern about the asymmetry of costs, which tends to favor plaintiffs, should be considered in light of the equally sensible principle that defendants are often more apt to possess evidence and information that is relevant to the merits of the case. Nevertheless, the U.S. discovery rules seem poorly designed in light of their purpose. The final flaw highlighted in this chapter is said to be the ‘opt out’ approach used in class litigation in the United States where damages are sought. Rosch clearly describes the main disadvantage of ‘opt out’ class actions: if some potential members of the plaintiff class are well-resourced businesses, they may indeed opt out and proceed with separate lawsuits against the same defendants, thus multiplying procedures and costs for (courts and) defendants, including in cases where a defendant in fact exonerates itself in the class action but must then continue defending itself.126 As Rosch explains, ‘[b]ecause the opt-out mechanism does not distinguish between the needs of individual consumer plaintiffs and corporate plaintiffs, the class action vehicle fails to achieve its purpose of minimizing an antitrust defendant’s exposure to multiple lawsuits and multiple liabilities’.127 Recommendations for the European Union Taking account of the above criticisms, Rosch’s recommendations for a more balanced system of private antitrust enforcement are these: 1. Avoid ‘opt out’ class actions. Rosch will be pleased that the Commission has advised Member States, in Recommendation 2013/396, to base collective action instruments on ‘opt in’ procedures, unless they can justify a deviation See Texas Industries, Inc. v Radcliff Materials, Inc., 451 U.S. 630 (1981). The Supreme Court has not provided guidance as to whether, as a matter of constitutional law, plaintiffs who opt out of class litigation must have the opportunity to re-litigate a case if the class loses in court or if the case settles. See Rosch’s discussion of Ticor Title Insurance v Brown. 127 Page 187. 125 126

Well Integrated Public and Private Antitrust Enforcement in the EU 

lv  

on grounds of sound administration of justice. But, undoubtedly conscious of the incentivization problem that arises where plaintiffs must opt in rather than opt out, Rosch suggests that if a plaintiff chooses not to opt in to the collective action she will surrender any claim she may have under EU law. In a sense, this may restore an incentive that could be lacking, either where a plaintiff is sophisticated and has sufficient resources to litigate separately, or where a plaintiff’s stakes in the case are small or trivial. The advantages to such a ‘one shot’ system would be enhanced certainty and the elimination of the risk of duplicative litigation. However, it is not clear that EU Member States, or, in the future the EU itself if it should ultimately proceed with harmonizing legislation, would be able to introduce the system envisaged by Rosch without breaching the European Convention on Human Rights and the EU Charter of Fundamental Rights and Freedoms. 2. Do not recognize standing for indirect purchasers in circumstances where their claims may result in multiple liability. This recommendation flows from the problem described above, ie, the combination of (i) a federal rule in the U.S. precluding defendants from claiming that the plaintiff has not sustained the claimed harm because the latter passed on some or all of the illegal overcharge to its own purchasers (which thus tends to reinforce the incentives of direct purchasers to bring claims under the antitrust laws, while tolerating a deficit of compensation for some injured parties), and (ii) a rule under the laws of many states which, incoherently with (i), allows indirect purchasers to sue, including by way of class actions. This risk of a defendant paying out more than once, coupled with the risk of unjust enrichment on the part of direct purchasers, does not have an exact parallel in Europe. Under EU law, the Court of Justice has insisted that ‘any person’ harmed as a consequence of the breach of Article 101 or Article 102 (which in a given case might even extend to non-purchasers outside the vertical chain) must be entitled to claim damages they have sustained, including any lost profits.128 The most likely claimants – sophisticated indirect purchasers (eg, purchasers that purchase a product that incorporates a cartelized input) and direct purchasers – are thus given de jure equal standing for purposes of admissibility, subject to the relatively light substantive burden borne by an indirect purchaser to show that he paid an overcharge passed on to him.129 Potentially, this ‘democratic’ admissibility rule could, if left in isolation, 128 The landmark judgments, Courage and Manfredi, have dealt with claims arising from restrictive agreements. The judgments are also broad enough to cover abuse of dominance scenarios. (The principle that in Europe indirect purchasers must have standing to claim damages (insofar as they are legally able, taking account of other national (substantive) rules, to bring such claims), which derives from the ‘any person’ standard laid down in those judgments, would be codified by Article 13 of the draft Directive published on 11 June 2013.) The ‘private enforcement’ of Article 106 (typically read together with Article 102) is also possible. See Case C-242/95, GT-Link A/S v De Danske Statsbaner (DSB), [1997] ECR I-4449; and recital 3 of the draft Directive. 129 See Article 13(2) of the draft Directive (onus of showing that the overcharge was passed on is deemed to be discharged where the indirect purchaser has demonstrated that the defendant infringed

lvi  Mel Marquis present significant risks of duplicate litigation and hyper-restitution.130 It is true that the Court has recognized the validity of Member State rules designed to avoid unjust enrichment,131 which would include rules allowing defendants to advance ‘passing on’-type arguments and thereby defeat or weaken claims brought by direct purchasers. However, assuming that all Member States implement on a more harmonized basis the passingon defense132 as foreseen in the draft Directive of 11 June 2013,133 the the law; that this resulted in an overcharge for the direct purchaser; and that he (ie, the indirect purchaser) purchased the subject products to which the infringement relates, or derivatives thereof). See also Article 15(1) of the draft Directive, which seeks to coordinate damages actions by claimants at different levels of the supply chain (including those brought by direct purchasers) by providing that national courts hearing an action for damages brought by an indirect purchaser must take into account, when applying Article 13(2), any damages actions (and/or resulting judgments) related to the same infringement but brought by claimants at other levels of the supply chain. Article 15(2) would preserve the rights and obligations of courts under Article 30 of the ‘Brussels I’ Regulation on jurisdiction and enforcement (Council Regulation 44/2001, now re-numbered, when it takes effect on 10 January 2015, as Regulation 1215/2012). According to the latter provision, parallel actions brought by claimants at different levels of the supply chain are considered to be related and courts may either adjourn proceedings to avoid irreconcilable judgments or consolidate the claims to resolve the coordination problem. 130 See Geradin and Grelier, ‘Tip of the Iceberg’, cited above note 9, at 12–13. On the other hand, if safeguards can keep this risk within tolerable bounds, the recognition of standing for indirect purchasers can be useful, for example in cases where the incentives of direct purchasers to enforce their claims are dampened – either because they pass on most or all of any overcharge they have paid, or because they do not wish to disrupt commercial relations with their suppliers. See Reindl, Secretariat Note, cited above note 11, at 14 (recounting the arguments raised in an OECD roundtable by Andy Gavil). 131 See Courage, para 30 (with references to case law); Manfredi, paras 98–99. In a non-antitrust context, see also Case 199/82, Amministrazione delle Finanze dello Stato v SpA San Giorgio [1983] ECR 3595, para 13. 132 It has been contended that passing-on is not properly conceptualized as a defense but rather should be regarded as a mitigation of damages to be taken into account when assessing – subsequent to a finding of liability – the proper quantum of damages. See Frank Maier-Rigaud, ‘Towards a European Directive on Damages Actions’, forthcoming, Journal of Competition Law and Economics. A version of the paper dated 22 July 2013 was consulted at http://papers.ssrn.com/sol3/papers.cfm?abstract_ id=2296843. 133 See Article 12 of the draft Directive (assigning to the defendant, unsurprisingly, the burden of proving that the claimant passed on the illegal overcharge). An exception to the general recognition of passing-on arguments is provided for in Article 12(2), which precludes such arguments if it is ‘legally impossible’ for the person at the next level of the supply chain to claim compensation for the harm suffered from the passing on of the overcharge, the rationale for this exception being, of course, to avoid creating a legal zone of impunity, even if the effect may be to overcompensate direct purchasers. (On this point, see, eg, Reindl, cited above note 11, at 5–6.) The reference to ‘legal impossibility’ may be deliberately elastic (and vague) in light of the diversity of procedural rules across the Member States. However, the primary example of legal impossibility is where an indirect purchaser would be excluded from recovering damages – assuming the principle of effectiveness is respected – on tort law grounds of foreseeability or remoteness. As recital 30 of the draft Directive states: ‘The court seized of the action should […] assess, when the passing-on defence is invoked in a specific case, whether the persons to whom the overcharge was allegedly passed on are legally able to claim compensation. While indirect purchasers are entitled to claim compensation, national rules of causality (including rules on foreseeability and remoteness), applied in accordance with the principles of Union law, may entail that certain persons (for instance at a level of the supply chain which is remote from the infringement) are legally unable to claim compensation in a given case. Only when the court finds that the person to whom the overcharge was allegedly passed on is legally able to claim compensation will

Well Integrated Public and Private Antitrust Enforcement in the EU 

lvii  

following scenario might arise. In one Member State a court may decide that a defendant has not met its burden of showing pass-on and thus grants damages to a direct purchaser, while the court of another Member State rules that a customer of that direct purchaser is entitled to damages because the overcharge was passed on.134 The result – multiple liability – is essentially the same as problem noted by Rosch in the U.S. context, albeit in a different configuration. It remains to be seen whether the Directive, if adopted as is, provides adequate means to avoid this. Much will depend on Article 13 of the draft, which would require Member States to ensure that their courts can estimate how much of an overcharge has been passed on to an injured party, which could provide a basis for damage deductions where orders to pay compensation have already been secured against the same defendant by other claimants at a different level of the supply chain.135 Article 13 is to be read together with Article 15(1), according to which national courts should be required under national law to ‘take due account’ of relevant judgments or actions that have been brought which would raise the risk of overpayment. With regard to the passing-on defense in the Directive, it should be noted incidentally that the defense can be only partial, if only part of the overcharge was passed on; and significantly, the fact that an overcharge was applied by a direct purchaser to its own customers may have resulted in diminished sales and possibly lost profits for the direct purchaser, for which the defendant may be liable.136 it assess the merits of the passing-on defence.’ The passing-on defense would therefore be available only to those defendants that withstand this judicial screen. The approach just outlined has been criticized by Josef Drexl in the following terms: ‘This […] approach […] collides with fundamental principles of civil procedure including the one on inter partes effect of private court proceedings. The Proposal does not require that these subsequent purchasers have become parties [to] the proceedings. Nor does the Proposal take account of the fact that these requirements raise major issues of evidence and, therefore, would also require rules on the burden of proof.’ Drexl, ‘The Interaction between Private and Public Enforcement in European Competition Law’, draft presented at the EUI in Florence on 27 June 2013, page 15. Drexl is undoubtedly right to warn that inquiries into the hypothetical legal position of indirect purchasers could raise major issues of evidence. It is not clear, though, from an EU law perspective, why the fact that the proposed EU rule would collide with the traditional rule of inter partes effects of litigation (also a basis for Drexl’s criticism of Article 15(1) of the draft) constitutes a compelling reason why the judicial check would be inappropriate. The risk of upsetting national procedural traditions of this kind is surely outweighed by (and as a strict legal matter could not even be weighed against) the risk that the passing-on defense could prove to be too powerful and thus negate, in the ‘legal zone of impunity’ scenario to which I referred above, the effectiveness of Article 101 or Article 102 (at least where public enforcement does not pick up the slack). See Geradin and Grelier, ‘Tip of the Iceberg’, cited above note 9, at 12–13. See Daniele Calisti and Luke Haasbeek, ‘The Proposal for a Directive on Antitrust Damages Actions: The European Commission Sets the Stage for Private Enforcement in the European Union’, CPI Antitrust Chronicle, August 2013 (1), at 5. 136 See Article 14(1) of the draft Directive (rules set forth in Chapter IV on the passing-on defense ‘shall be without prejudice to the right of an injured party to claim compensation for loss of profits’). See also section 4.4 of the Explanatory Memorandum, at page 17 (‘[W]here a loss is passed on, the price increase by the direct purchaser is likely to lead to a reduction in the volume sold. That loss of 134 135

lviii  Mel Marquis 3. Avoid a rule that makes defendants jointly and severally liable but precludes them from seeking, from any settling defendants, contribution for their share of liability. As Rosch explains, such a rule, which again can result in overpayment by non-settling defendants (assuming the settling defendant pays to the settling claimant an amount less than that defendant’s full liability, leaving non-settling defendants on the hook for the difference), may be unnecessary, at least in terms of maintaining proper incentives for plaintiffs and their lawyers to invest in the litigation. From a quite different perspective, however, it might be argued that a defendant considering whether to settle will have less incentive to do so, ceteris paribus, if he knows the settlement will not relinquish him from subsequent residual liability depending on the outcome of pending or prospective court claims. If significant value is attached to consensual settlement, as it is in Europe, a tradeoff favoring incentives to settle may be made at the potential expense of non-settling defendants. This tradeoff seems to be embodied in Article 18 of the draft Directive.137 However, the impact of the tradeoff – that is to say, the risk that co-infringers would be obliged to pay more than their share of the harm they caused – is mitigated in the following sense. With regard to the court action against the non-settling defendants, the injured party’s claim would be reduced, pursuant to Article 18(1), by the settling defendant’s share of the harm caused to that injured party. 4. Permit discovery only upon a showing of good cause, and deem good cause to be shown where the party seeking discovery cannot obtain the relevant information from law enforcement authorities. Such a rule presupposes that the authority has a case file and that it is therefore more efficient to obtain the documents or other materials needed from the authority than from the other party. One would think that such a rule would be inapplicable in circumstances where public authorities have not taken any investigative steps, such as may be the case in private contract disputes, or in some private claims of abuse of dominance, etc. But such scenarios would apparently be excluded by Rosch’s recommendation number 5, to wit: profit, as well as the actual loss that was not passed on (in the case of partial passing-on) remains antitrust harm for which the injured party can claim compensation.’). The possibility to recover lost profits caused by an antitrust infringement is in fact guaranteed by the seminal case law cited earlier. In addition to the ‘without prejudice’ proviso in Article 14(1), the right to claim lost profits would be codified by Article 2(2) of the draft Directive as part of the concept of full compensation. 137 Article 18(1) provides that ‘Member States shall ensure that, following a consensual settlement, the claim of the settling injured party is reduced by the settling co-infringer’s share of the harm that the infringement inflicted upon the injured party. Non-settling co-infringers cannot recover contribution from the settling co-infringer for the remaining claim. Only when the non-settling coinfringers are not able to pay the damages that correspond to the remaining claim can the settling co-infringer be held to pay damages to the settling injured party.’ (emphasis added) Article 18(2) provides that, ‘[w]hen determining the contribution of each co-infringer, national courts shall take due account of any prior consensual settlement involving the relevant co-infringer’. See also recitals 40 and 41 of the draft Directive.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lix  

5. Limit private actions to those that follow public enforcement actions in the EU. As Rosch explains, ‘[h]aving someone else do the work (including gathering the relevant evidence […]) should be a powerful incentive for plaintiffs’ antitrust attorneys to invest in private collective action litigation’.138 Rosch’s concern for establishing an adequate risk/reward ratio that makes suits for compensation worth the effort is understandable. However, this does not provide a rationale that would explain the surprisingly categorical exclusion of stand-alone actions, such as those mentioned in the preceding paragraph. 6. Award to prevailing plaintiffs not only damages for their losses but also prejudgment interest dating back to the time the infringing conduct or transaction began. According to Rosch, ‘[b]y making an award of prejudgment interest mandatory, the EU (and its Member States) could help ensure that most damage awards equal or exceed treble-damages awards in the United States, which would promote convergence on the private enforcement front’.139 The common-sense notion that damages should include pre-judgment interest is consistent with the EU case law (in particular, Manfredi) and is reflected in Article 2(2) of the draft Directive, which states that ‘[f]ull compensation shall place anyone who has suffered harm in the position in which that person would have been had the infringement not been committed. It shall therefore include compensation for actual loss and for loss of profit, and payment of interest from the time the harm occurred until the compensation in respect of that harm has actually been paid.’140 It is empirically an open and complex question, however, whether Rosch is right to suppose that the foregoing definition of compensable damages will ensure that most damage awards match or exceed treble-damages awards in the U.S.141 7. Avoid ‘loser pays’ rules when designing the allocation of parties’ litigation costs. In order to maintain proper litigation incentives, Rosch says, it is better if a plaintiff bears his own costs but no more than this, so that he is not deterred from bringing a meritorious suit because of the prospective risk that if he lost the case he would lose far more than his investment. However, this final recommendation goes against the general European grain. Most Member States are comfortable with ‘loser pays’ rules (and with the inevitable ex ante uncertainty associated with them), since a painful Page 194. Page 195. 140 Similarly, see recital 11 to the draft Directive. It therefore seems that Article 2(2) would require payment of both pre-judgment and post-judgment interest as essential components of full compensation. 141 Cf. Commission, Impact Assessment Report (Annex II to this volume), para 67, footnote 58 (‘[O]n average, single damages with pre-judgment interest can be said to equate roughly to double damages without pre-judgment interest.’). A realistic way of considering the issue is that the recovery of pre-judgment interest in a successful damages suit restores the victim to the status quo ante, ie, it compensates her for the true cost of her injury. One may therefore question whether terms such as ‘double damages’ or ‘treble damages’ in such circumstances are really an unhappy epithet that distorts discussion. 138 139

lx  Mel Marquis loss is commonly perceived as filtering out frivolous lawsuits; and the draft Directive does not call on the Member States to innovate in this area, even through the introduction of, for example, cost protection orders at a court’s discretion. Indeed, apart from a passing reference in Article 8, and in contrast to the Commission’s 2005 Green Paper and 2008 White Paper, the text is quite silent on the subject of costs.142 One must acknowledge that in several Member States there is significant flexibility that enables judges – usually on grounds of equity or because the other side has acted unreasonably – to deviate from the general rule and to exempt an unsuccessful plaintiff from payment of a prevailing defendant’s costs.143 It is equally true, however, that courts tend to decide matters of costs when the litigation is concluded, which could be many years after the initial ‘investment’ decision was made by the claimant.144 It may be appropriate, when contemplating future iterations of the Directive on antitrust damages actions, to revisit the issue of the adequacy and diversity of national rules in relation to cost allocation and, relatedly, the issue of court fees.145 Following his appearance at the June 2011 Workshop, Rosch proceeded, in a speech in September of that year, to put forward some additional views.146 The tenor of that speech is seemingly even more skeptical of private antitrust claims than that of the paper that appears in this volume. However, an important qualification which appears in footnotes 2 and 82 of the text of that speech, 142 The text has in fact been criticized for ducking cost issues. See Peyer, cited above note 55. The 2005 Green Paper and the accompanying Staff Working Paper had identified the costs of court actions as being among the main obstacles to the enforcement of legitimate private antitrust claims. Taking account of stakeholder views collected following the Green Paper, the 2008 White Paper and the ‘white’ Staff Working Paper were somewhat reluctantly conservative with regard to ‘loser pays’ rules. See, eg, paras 243, 245 and 252–263 of the Staff Working Paper; and see Horst Butz, ‘Integrating Public and Private Enforcement in Europe: Issues for Courts’, this volume (Commission’s suggestion of cost orders derogating exceptionally from the usual loser pays rule ‘is quite opposite to [Germany’s] whole traditional system of court fees’). The issue of how ‘loser pays’ rules are properly structured from the perspective of the effectiveness of EU law may ultimately be considered by the Court of Justice, but the Court may be hesitant to require radical changes unless a significant clutch of Member States spontaneously reshape their own rules first. The prospects for this are not terribly bright. 143 See the 2008 Staff Working Paper, ibid, paras 255–259 (referring to flexible cost rules in Finland, Italy, the UK, France, Germany; and citing also Article 69 of the Rules of Procedure of the European Court of Justice, which allows the Court to deviate from ‘loser pays’ in exceptional circumstances). 144 See ibid, paras 260–261. In light of the general tendency of courts to decide costs at the end of procedure, the 2008 White Paper (section 2.8 in fine) and Staff Working Paper stated, quoting the latter (para 261 in fine), that ‘Member States are encouraged to provide national courts with the possibility to issue cost orders derogating from the normal cost rule, preferably upfront in the proceeding. Such cost orders would guarantee that a claimant, even if unsuccessful, will not have to bear all costs incurred by the other party.’ 145 See the 2008 Staff Working Paper, paras 262–263, pointing to the possibility that court fees in some Member States are disproportionately high and stating that ‘Member States are encouraged to set their court fees in an appropriate manner so that they do not constitute a disincentive for antitrust damages claims’. On the possibility of reducing court fees in Germany where a plaintiff demonstrates financial hardship, see Butz, ‘Issues for Courts’, this volume. 146 See J Thomas Rosch, ‘Does the EU Need a System of Private Competition Remedies to Supplement Public Law Enforcement?’, speech, Christ Church, Oxford, 23 September 2011.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxi  

is that Rosch was ‘addressing only the stated goal of using private damages actions and collective redress mechanisms as a means of supplementing public enforcement of EU competition law’.147 It emerges (subtly) from his remarks that the ‘supplementing’ of which he speaks (significantly, he does not use the term ‘complementing’) concerns the reliance on private enforcement to achieve a level of deterrence beyond that resulting from the efforts of competition authorities. His multifaceted critique of private enforcement, which he considers to have run amok in the United States, should be understood in that light. By contrast, with regard to private enforcement as a means to ensure the compensation of victims of anticompetitive behavior, Rosch says ‘I don’t take issue with this goal (except to note that private, treble damages actions can lead to overcompensation for reasons I have previously stated […])’.148 James Keyte: ‘Collective Redress: Perspectives from the U.S. Experience’. In this chapter Keyte sings harmony with Tom Rosch with regard to most criticisms of class action litigation, in the U.S., which he presents as a ‘natural experiment’ with lessons for the EU. The main lesson is the old adage: be careful what you wish for. Keyte’s overall argument appears to depend on the following dilemma. First, consistent with the observations of Rosch, he stresses the importance of having adequate incentives that will appeal to the animal spirits of plaintiffs and their lawyers; if the incentives are insufficient, the goal of ensuring fair compensation for victims will remain illusory because the available tools will languish unused. And yet, in Keyte’s view, when the incentives are strong enough – which they may well be if an opt-out class action with high stakes is possible – then the train will go off the rails and the spoils of lawsuits and out-of-court settlements will foster the litigation culture that most would prefer to avoid.149 The message is clear: the risks of pervasive class actions and their pathologies are too substantial; the EU would be better advised not to tempt fate. In presenting his argument, Keyte reviews the prerequisites that must be satisfied in order to secure court permission to take a class action forward. These famous prerequisites (‘numerosity’, commonality, typicality and adequacy of representation) are set forth, as he recalls, in Rule 23(a) of the Federal Rules of Civil Procedure. Rule 23(c) requires the party representing the class to serve notice on absent class members to make them aware of the litigation, to advise them of their rights, and give them the opportunity to opt out and, in doing so, avoid being bound by legal decisions made in the context of the subject litigation. Keyte argues that, despite the filters embedded in Rule 23, frivolous suits still ‘‘sneak’ by – and defendants are still forced to settle these frivolous claims’.150 The pressure to settle dodgy claims that by some type I error have been certified Ibid, page 2, footnote 2. Ibid. 149 See, eg, page 197 (‘[I]t is important to view the United States class action model through the lens of the litigation culture it has created.’). 150 Page 198. 147 148

lxii  Mel Marquis by the judge results from the dysfunctional elements of U.S. procedural rules, as highlighted by Rosch (see above) and by Keyte.151 But whereas Rosch emphasizes the difficulty of making it past the Rule 23 hurdle,152 Keyte points out that ‘many cases are now routinely settled before a class has even been certified, and parties agree to settlement without even considering the procedural thresholds […] imposed by Rule 23’.153 The relatively low chance of getting a certification denied does seem to have conduced to pre-certification settlement in the past,154 and even today this may occur occasionally, for example if a defendant is keen to avoid the significant costs of pre-certification discovery; but it may also be that, nowadays, defendants will see the certification stage as a key opportunity to challenge plaintiffs’ evidence.155 More generally, and looking forward, potential defendants will often have means at their disposal to avoid unmeritorious settlements of the kind Keyte describes if they take advantage of the Supreme Court’s rulings in cases such as AT&T Mobility v Concepcion (judgment of 27 April 2011)156 and American Express v Italian Colors (judgment of 20 June 2013),157 which uphold the enforceability of class action waivers in arbitration agreements, as discussed above. In other cases, where no arbitration clause applies, plaintiffs may still face formidable obstacles. The main obstacles in many antitrust cases will be the scrutiny of the merits of the case at certification stage (ie, taking account of the requirements of the Clayton Act158), and the application of Rule 23(b)(3), which requires would-be class plaintiffs to show that common issues predominate and that class litigation is superior to other possible methods of resolving the dispute. Beyond these, there is also the increasingly rigorous application of Rule 23(a), as illustrated in Wal-Mart v Dukes (judgment of 20 June 2011),159 where the 151 See Keyte’s list of factors at page 199. Whereas Rosch suggests that ‘loser pays’ rules present too much of an obstacle to the filing of potentially meritorious claims, Keyte would be delighted to import such rules into the U.S. litigation context. 152 Amendments to FRCP 23 entered into force in 2003 and accentuated the responsibility of courts to carry out a thorough examination of the class certification criteria. The importance of a ‘rigorous analysis’ of the aptitude of a case for class certification is illustrated by In re Hydrogen Peroxide Antitrust Litigation, 552 F.3d 305 (3d Cir. 2008), and a similar trend has unfolded in most federal Circuits. Although it has been said that ‘[o]ne of the critiques of private class actions is that judges too readily ‘rubber stamp’ settlements and fee requests that they are required as a matter of law to review carefully’ (Deborah Hensler, ‘Goldilocks and the Class Action’, 126 Harvard Law Review Forum 56, 59–60 (2012)), in the specific context of antitrust, at least, rubber-stamped class certification is no longer accepted by appellate courts. See, eg, Steven Bizar and Allison Khaskelis, ‘Wal-Mart v Dukes: A Non-Event for Antitrust Defendants’, 26 Antitrust 25, 27 (Fall 2011). 153 Page 199 (emphasis in original). 154 Cf. Bizar and Khaskelis, cited above note 152, at page 28 (‘With many courts seemingly unwilling to afford the defendants’ opposition to the class motion the careful scrutiny defendants desired, the strategy debate on joint defense group conference calls often centered on whether it even made sense to oppose class certification at all.’). 155 As observed above in note 152, the federal courts today are more likely than they were in the past to engage in a rigorous analysis of the evidence offered by the prospective class representative and whether it satisfies the prerequisites of Rules 23(a). 156 Cited above note 114. 157 Cited above note 116. 158 To prevail under the Clayton Act a plaintiff must prove that a violation of the antitrust laws caused him injury, and must prove the amount of damages sustained. 15 U.S.C. § 15. 159 Cited above note 18.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxiii  

Supreme Court in effect instructed federal courts to be tougher when applying the ‘commonality’ test. This is not to suggest that the problems identified by Keyte are about to vanish. But the tide seems to have turned, and the future trajectory in the U.S. is likely to be one of decline: the rampant class action litigation that motivates Rosch and Keyte’s chapters will likely be looked back upon as a high water mark.160 In any event, in cautioning against the excesses of the American example, Keyte in his conclusion leaves a sliver of hope for the EU. ‘Perhaps,’ he says, ‘it may be as simple as reducing damages to single or double; or in some manner capping contingency fees; or precluding certain financing arrangements that support a plaintiffs’ bar.’161 These are suggestions that will go down easy for many in Europe. Brian Facey and Brian Rosner: ‘Collective Redress for Cartel Damages in Canada’. Facey and Rosner here provide a chapter-sized mini-treatise on antitrust class actions brought before the Canadian provincial courts, including the jurisdictions where the battles are mostly fought – Ontario, Quebec and British Columbia. The fundamental legal base for such actions is section 36 of the Competition Act, which in its first paragraph establishes a cause of action for any person who has suffered loss or damage caused by a violation of the criminal provisions of the Act, which include classic cartel behavior (‘conspiracy’), bidrigging and misleading advertising. As background to this chapter, a trend had seemingly developed in Canada marked by several denials of class certification following the 2003 judgment of the Ontario Court of Appeal in Chadha v Bayer, which may have seemed to be the beginning of the end of indirect purchaser suits in Canada.162 However, especially since 2010 the Canadian courts appear to be moving in a direction quite different from that of the federal courts in the United States, where the class action apparatus is increasingly under strain from recent U.S. Supreme Court judgments (see discussion above). Indeed, following a significant judgment by the Supreme Court of Canada in 2013, despite a rule precluding defendants from claiming a passing-on defense against direct purchasers, the offensive use of passing-on to support suits by indirect purchasers is fair game in Canada.163 The Supreme Court seems to accept this apparent contradiction (and by the same token shuns the 160 See also Robert Klonoff, ‘The Decline of Class Actions’, 90 Washington University Law Review 729–838 (2013). 161 Page 204. 162 Chadha v Bayer Inc., [2003] O.R. 3d 22 (C.A.), leave to appeal to the Supreme Court of Canada refused. As Facey and Rosner explain, since the Supreme Court soon thereafter decided that a defendant had no possibility under Canadian law of claiming a pass-on defense against direct purchasers in tort or restitution, this logically (ie, under the logic of Hanover Shoe and Illinois Brick) foreclosed the offensive use of pass-on theories to support indirect purchaser claims. In Canada there is no parallel to the ‘Illinois Brick repealer’ problem, which, as lamented by Tom Rosch and others, leads to actual or potential overexposure of antitrust defendants in the U.S. The corollary to Canada’s approach is that – again, in accord with Illinois Brick – direct purchasers may seek to recover the full extent of the overcharge, irrespective of any possible pass-on. 163 See Pro-Sys Consultants Ltd. v Microsoft Corporation, 2013 SCC 57.

lxiv  Mel Marquis logic of Illinois Brick) because of its greater concern for ensuring that wrongdoers cannot maintain any ill-gotten gains. Private enforcement in this sense thus contributes prominently to the goal of deterring the infringement of the Canadian competition rules. As Facey and Rosner point out, judicial ‘hospitality’ to class actions also reflects the lighter statutory conditions for class certification that apply in Canada under the class action legislation adopted by the various provinces, such as Ontario’s Class Proceedings Act of 1992.164 One significant statutory difference, for example, is that unlike FRCP 23(b)(3), which in general limits class certification in U.S. federal antitrust cases to those where the common issues raised by the claims of the class members predominate over individual issues (a criterion distinct from the prerequisite of ‘commonality’ under Rule 23(a) – see above), in Canada the Class Proceedings Act requires common issues but has no ‘predominance’ test. From the perspective of class plaintiffs, it might perhaps be argued that the lower certification threshold is something of a counterpart to Canada’s rules on damages in antitrust cases, which are limited to single damages only. In the case of infringements of substantial duration the incentive to bring suit would probably be adequate in any event, as pre-judgment interest may be recovered; other cases may be more ambiguous. The most wide-ranging part of the paper is section III (‘Procedural considerations’), which discusses numerous practical issues that come up in Canadian class litigation. The authors cover jurisdiction over claims and the main principles governing class actions, including: the criteria applied for class certification; discovery, including cross-border discovery; the opt-out, opt-in and hybrid models used in different provinces; recovery of litigation costs; third-party funding; standing; the occasional certification of ‘national’ and ‘international’ classes and related constitutional constraints; the relatively more relaxed certification of cases for purposes of settlement; the recognition and enforcement in Canada of foreign settlements and judgments; the (non)enforceability of arbitration agreements purporting to forbid class litigation; and coordination between plaintiff’s firms in Canada and plaintiff’s firms in the United States. Running through the wide gamut of these issues, Facey and Rosner highlight not only significant differences between Canada and the U.S., for example as regards the use of class action waivers and compulsory arbitration clauses in contracts of adhesion (where the Supreme Court of Canada interprets provincial consumer protection legislation as overriding and rendering void such prospective waivers); but also differences from one province to the next, a reflection of Canada’s federal political and legal structure. This divergence often concerns procedural rules somewhat technical in nature but with potentially important consequences. On the other hand, as far as class actions are concerned, some semblance of coherent application of the Class Proceedings Act in different cases by different courts is arguably facilitated by three frequently invoked statutory objectives: judicial 164 In addition to Ontario, each of the other Canadian provinces has its own legislation (or, in the case of Prince Edward Island, equivalent rules) that govern(s) the administration of class proceedings.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxv  

economy, access to justice and, most intriguingly or controversially from a foreign perspective, ‘behavior modification’.165 Mario Siragusa: ‘Options for Collective Redress in the EU’. Siragusa’s contribution dates from around the time the Commission was soliciting comments in 2011 on how to proceed with an EU initiative, binding or otherwise, in the field of collective redress.166 It also provides a useful look back at the Commission’s controversial ‘non-proposal’ of 2009, an initiative never officially launched and quietly laid to rest, very likely by the Commission’s top leadership.167 Together with the following chapter by Silvia Pietrini (see below), Siragusa’s paper serves as an introduction to the Recommendation that the Commission ultimately issued in June 2013, which I have already characterized as a strategically timid step in the direction of a future European landscape in which each Member State will have legislation or other rules enabling aggrieved consumers to bring some form of collective legal action against undertakings causing them injury, including but not limited to antitrust-related harm.168 Siragusa accomplishes essentially two things in this chapter. At one level he reviews the central issues raised by the prospects of instituting forms of collective redress in Europe169 and synthesizes the clashing views of big business (eager 165 The role of behavior modification was discussed in a 1990 report by an advisory committee to the Attorney General, which served as a basis for the 1992 Act. According to the report, the objective of behavior modification related to a need to promote a ‘sharper sense of obligation to the public by those whose actions affect large numbers of people’. Quoted in Michael Eizenga, Dany Assaf and Emrys Davis, ‘Antitrust Class Actions: A Tale of Two Countries’, 25 Antitrust 83, 89 endnote 11 (Spring 2011). 166 In response to the Commission’s consultation, the ‘bindingness’ issue was presented by many stakeholders as an issue of subsidiarity. As Siragusa explains, ‘many respondents […] pointed out that several Member States have already enacted national class actions laws, and […] superimposing an EU model of class action would fail to respect the principle of subsidiarity’. Page 240. Although subsidiarity is frequently depicted as a one-way argument opposed to centralized solutions, the principle in fact cuts two ways, and it does not necessarily follow from the existence of national solutions in a large number of Member States (ie, around two-thirds of them) that subsidiarity dictates a non-binding initiative. Excessive and chronic fragmentation can equally support subsidiarity arguments in favor of binding common rules. 167 See sections 2 and 5.3.1 of the paper. The withdrawn draft Directive of 2009 is included as Annex III to this volume. See also Marquis, ‘Cartel Settlements and Commitment Decisions’, cited above note 97. 168 For the proposition that the Commission’s ‘retreat’ to opt-in mechanisms as the strongly recommended default approach is strategic in nature, see above note 18 in fine. 169 The scope of the chapter extends beyond the subject of collective redress, and thus beyond Siragusa’s title. For example, it covers cross-cutting issues such as the quantification of damages (on which, see also the chapter by Komninos; and for discussion from an economic perspective of various Italian appellate judgments in follow-on actions, see Pierluigi Sabbatini, ‘Interesse private e interesse pubblico al risarcimento del danno antitrust’, 12(2) Mercato concorrenza regole 335, 349–357 (2010)) and access to evidence as essential parts of private damages claims, irrespective of whether they are brought on a bundled or individual basis. Another cross-cutting issue, the passing-on defense, is also discussed, with particular reference to Italian case law and to the application of Article 1227 of the Civil Code (claimant’s contribution to own injury and duty to mitigate harm). Siragusa also briefly refers to ‘negative declaratory actions’, sometimes called ‘Italian torpedoes’ (even if equivalents certainly exist elsewhere), which have not yet been used extensively in antitrust cases but which provide firms facing the strong prospect that they will be sued for damages the possibility to act as ‘plaintiffs’ and to

lxvi  Mel Marquis to dilute the effectiveness of collective action mechanisms) and of those entities and individuals who, often rationally ignorant/apathetic and/or lacking bargaining power, are most likely to need the possibility of collective redress if effective access to justice is to be served.170 Behind these diverging interests that need to be reconciled there lurks, if I may, the status quo- or inertia-bias that takes the code name ‘national constitutional and legal traditions’. On the other hand, collective redress systems can present risks, not just for business insofar as a disorderly system can create unjust and disproportionate costs if unmeritorious claims are encouraged, but also for consumers with valid claims, insofar as ill-conceived rules can incite class representatives to exploit those they are supposedly protecting and to raid any sums acquired on behalf of the latter through judgments or out-of-court settlements. Any reasonable proponent of collective redress would hope for and welcome solutions finely honed to minimize both kinds of risks. Most Europeans are in full consensus that the U.S. model minimizes neither. But I must repeat myself and add that ‘opt out’ should not be reductively equated to the complex constellation of rules in the U.S. (considered also in light of the judicial rebellions they have provoked) that makes us all apprehensive. At a second level, Siragusa presents the early experience of Italy with its own form of collective redress, the ‘consumer class action’ established by Article 140-bis of the Italian Consumer Code. This provision enables consumers to band together (including in the form of associations or committees) and to sue collectively, on an opt-in basis, for breaches of contract or for torts occurring after 15 August 2009. In such a case the court at an early stage screens the case to make sure the claim is not manifestly unfounded; that the class members are not divided by conflicting interests; that the rights alleged to have been infringed appear to be identical; and that the first claimant is able to protect the interests of the class in an adequate manner. If one or more of these conditions fails the action is declared inadmissible. If however the case is admitted, the court proceeds to set the ground rules for its continuation, including for example rules requiring the first claimant to provide notice to other class members and the deadline by which the latter may elect to opt in. Those who do not opt in may bring separate actions but once the opt-in period expires any such actions must be pursued individually. As Siragusa explains, there have in fact been cases brought as consumer class actions under Article 140bis of the Code. Siragusa refers to five such cases, although none of them was an antitrust dispute (see section 4.2 of the chapter). One may conclude that this early sampling of cases is inconclusive, as all but one of them were dismissed at the admissibility stage, for various reasons; in these cases it was therefore impossible to know for certain whether, if the admissibility conditions had been fulfilled, how ask a court to declare that they bear no liability. The Italian torpedo is thus a significant strategic tool; it enables such ‘plaintiffs’ to turn the tables on the other party and to influence, in conjunction with the Brussels I Regulation, the determination of which court will exercise jurisdiction over the case. 170 See in particular page 241 (summarizing the contrasting visions of collective redress advocated in the public consultation of 2011 by trade associations on the one hand and consumer organizations on the other).

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxvii  

the collective action dynamics would have unfolded. Would the full range of injured plaintiffs have signed on, hoping that their rights would be vindicated through the class action mechanism? The experience with an ‘opt in’ consumer representative action in the UK was disappointing,171 and that single-track approach has now been duly discarded by the UK Government. If the Italian approach proved to be more promising it would be a significant case study now that ‘opt in’ is, notwithstanding some wayward Member States, Europe’s officially sponsored solution. However, expectations are that Article 140-bis will produce few tangible results. Silvia Pietrini: ‘The Future of Collective Damages Actions in Europe’. In the first part of this chapter, Pietrini too reviews the milestones passed in the walk-athon toward harmonized collective redress rules in Europe. We now know that this process will be limited, for the foreseeable future, to soft harmonization, due not least to the fact that some Member States still resist the idea of binding EU rules in this area.172 However, the position advanced by Pietrini in this chapter, which was completed prior to the Commission’s adoption of Recommendation 2013/396, is that soft measures will only result in the perpetuation of divergent approaches across the Member States, and that this will ‘reinforce a disturbing trend toward multi-tiered justice in Europe’.173 For this reason she emphasized the need for a serious, binding initiative, and finds that the ensemble of mechanisms proposed by the British Government for the development of private damages in the UK is persuasive and should be adopted as a broader model for the EU. As noted earlier,174 the envisaged UK model will include exclusive jurisdiction by a court of competition specialists (ie, the CAT); judicial discretion about whether to proceed on an opt-out or opt-in basis; an active role for the court in certifying a class, including a preliminary review of the merits of the claims; a limitation of the opt-out rule, where it is used, to claimants domiciled within the UK (while those outside the UK may opt in); and various other safeguards, such as ‘loser pays’ cost rules absent exceptional circumstances, a ban on exemplary damages and a ban on contingency fees (without prejudice to the possibility of using conditional fee arrangements and after-the-event insurance). In addition, the UK system will provide for the possibility of (judicially approved) ‘opt out’ collective settlements 171 See above note 22, and for discussion see, eg, Barry Rodger, ‘A License to Print (Monopoly) Money? Replica Football Kit and Toys and Games, Resale Price Maintenance and the Competition Act 1998’, in Barry Rodger, ed., Landmark Cases in Competition Law: Around the World in Fourteen Stories, Wolters Kluwer, 2013, chapter 13, pp 339–343; Erik Werlauff, ‘Class Action and Class Settlement’, cited above note 14, at page 175 (less than 0.0008 percent of affected consumers, ie, a thousand out of about two million, opted in to the consumer representative action); and citing its own experience in the same case, see Which?, Consultation Response to European Commission Consultation on Collective Redress (prepared by Deborah Prince), dated 26 April 2011, for example at pp 7–8 (referring to the UK’s opt-in rule under section 47B of the Competition Act 1998 as a ‘fundamental flaw’). 172 See Commission, ‘Towards a European Horizontal Framework’, cited above note 18, at page 6 (reaction of Member States to the 2011 ranging from ‘support’ for binding EU rules to ‘strong scepticism’). 173 Page 267. 174 See above notes 16 and 22.

lxviii  Mel Marquis analogous to the 2005 Dutch Act on the collective settlement of mass damages claims,175 subject to a territoriality limitation on the ‘opt out’ element, in line with the one described above. Additional features of an eclectic European approach might be inspired by other European and non-European jurisdictions. For example, conscious of the general aversion in Europe to third party funding of collective actions, Pietrini suggests public funding devices, and points to the example of the Quebecois Class Action Assistance Fund, a program that has worked successfully and without incentivizing abusive litigation (the Commission, however, has not gone down this road176). In sum, Pietrini would push for a more assertive approach than the one the Commission has in fact adopted in its Recommendation. Indeed, Pietrini advocates not only binding harmonization rules based on minimum criteria and residual diversity; she would prefer to go all the way, crystallizing a collective action regime in a directly applicable European regulation (see section II of the chapter). Some in Europe would undoubtedly bristle at this, and perhaps point to Article 67(1) TFEU, which states that ‘[t]he Union shall constitute an area of freedom, security and justice with respect for fundamental rights and the different legal systems and traditions of the Member States’ (emphasis added),177 which would appear to call, at least in principle, for some allowance for variation at the national level. Most of those who would not fall back on the ‘do nothing’ or the ‘soft harmonization’ options would probably assume that a directive would be the natural choice of instrument for establishing common rules on collective redress, since, under Article 288 TFEU, a directive only binds Member States as to the results to be achieved while leaving the forms and methods in the hands of the national authorities, so long as these forms and methods are compatible with Union law. Pietrini rejects this conventional wisdom on the ground that, instead of promoting harmonization, directives have often institutionalized a subtle fragmentation of the internal market.178 It may be true that regulations are superior instruments in terms of clarity, predictability and effectiveness, as Pietrini says, and they are no doubt more apt for pursuing the ideal of a uniform application of Union law. However, the long-running battle the Commission has fought over the prospect of overriding cherished traditions of civil procedure at the national level See above notes 14 and 15, and accompanying text. See Commission, ‘Towards a European Horizontal Framework’, cited above note 18, at page 15 (‘the Commission does not find it necessary to recommend direct support from public funds, since if the court finds that damage has been sustained, the party suffering that damage will obtain compensation from the losing party, including their legal costs’). 177 Similarly, see the European Parliament’s Resolution of 2 February 2012, ‘Towards a Coherent European Approach to Collective Redress’, 2013 OJ C239E/32, para 16 (underlining the ‘need to take due account of the legal traditions and legal orders of the individual Member States’). 178 See page 268 (impact of harmonization directives ‘has not been the creation of a single, consistent and coherent body […] of law common to all the EU Member States; instead, there are now 27 national rules on doorstep selling, distance selling and so on’, quoting Christian TwiggFlesner, ‘Good-Bye Harmonisation by Directives, Hello Cross-Border Only Regulation? – A way forward for EU Consumer Contract Law’, http://ec.europa.eu/justice/news/consulting_public/0052/ contributions/309_en.pdf, page 5). 175 176

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxix  

suggests that, when the question of adopting EU legislation to govern collective redress reasserts itself a few years hence, the choice of legal instrument will be determined as much by political tactics and compromise as by logic or efficiency, or by what the best means to achieve effective access to justice would be.

Part IV

Drawing Lessons and Conclusions

John Ratliff: ‘Integrating Public and Private Enforcement of Competition Law: Implications for Courts and Agencies’. In this chapter Ratliff covers a variety of points where public and private enforcement intersect, including in particular: the stark differences between EU-level administrative investigations and appeals and national-level litigation; the relevance of and difficulties posed by the long time periods that elapse (ie, periods stretching to 20 years or more), in the context of follow-on antirust damages claims, from the time material events occur to the time the facts are tried before a court; access to the files of competition authorities (including, inter alia, access to the Commission’s file via Regulation 1049/2001 and the rapidly evolving related case law); confidentiality issues, particularly as concerns which information must properly be redacted from the Commission’s published decisions; the ‘binding effect’ of decisions by national authorities; the insulation of successful immunity applicants from the full brunt of followon damages claims; and the idea of the reduction of public fines to account for and incentivize the payment of compensation by investigated firms to parties that have suffered harm, for example in the form of a trustee-administered fund. I will comment incidentally on only one of the many points just recited. The comment pertains to the third part of the chapter, which should be highlighted for its discussion of the Pfleiderer case including not just the ECJ’s judgment but also the post-reference follow-up decision of the Amstgericht Bonn of 18 January 2012 and the judgment of the UK High Court in National Grid v ABB Limited, decided 4 April 2012. Among the notable points made is the way the ECJ’s judgment in Pfleiderer should be interpreted: ‘It may be argued that the Court had sympathy for the protection of the leniency materials and qualified that only to the extent that it considered that there might be a case for access, in the event that it was ‘excessively difficult or manifestly impossible’ for the plaintiff to bring his action otherwise.’179 Following this line, reference is made to ‘the high standard set by the Court with the ‘excessively difficult or manifestly impossible’ test’.180 From my own reading of the judgment I don’t think it can be excluded that at least some of the Court’s judges had sympathy for the need to limit access to sensitive leniency materials. However, I wonder whether one may draw from the judgment a test of this kind. The duty of the national judge under Pfleiderer, 179 180

Page 285. Ibid.

lxx  Mel Marquis it seems, is to balance opposing interests and only where the balance tips toward non-disclosure of the materials sought must he or she test that conclusion by verifying whether the party seeking disclosure would in consequence be deprived, or effectively deprived, of the right to recover damages resulting from the breach of EU law. By contrast, if the balance favours the claimant in the first place, then under Pfleiderer the principle of effectiveness of EU law – if it is reduced, as the ECJ seemingly would have it, to the question of whether a particular claimant’s right to damages would be abridged, as opposed to effectiveness of the law in broader and more dynamic terms – is in a sense redundant. But I have also raised the possibility that in the Donau Chemie judgment (decided several months after Ratliff prepared his chapter) the ECJ may arguably have created a presumption in favor of disclosure operating to the benefit of claimants where the balance of interests would otherwise be inconclusive.181 Having said all that, nuances as to the precise interpretation of Pfleiderer may potentially become of merely historical significance if the draft Directive of June 2013, or some modified version of it, introduces a new framework to assist national judges – and if that framework survives any challenge that might be brought against it on grounds of the effectiveness of the Treaty’s competition rules.182 Ian Forrester and Mark Powell: ‘Market Forces and Private Enforcement: A Start But Some Way Still To Go’. The overarching point of the chapter by Forrester and Powell seems to be that, just as in the days of Regulation 17/62 and its nowquaint idiosyncrasies, there remains, as we step forward into a world of enhanced private enforcement, the possibility and likelihood that decision-making by national courts will reflect perceptions about the legitimacy of EU-level procedures and the degree to which the latter deliver correct and just outcomes. Of course, national courts cannot tailor the substance of Treaty rules as the U.S. Supreme Court might do with U.S. antitrust law (for example, as concerns its scope of application) if that court considers that treble-damage class actions and asymmetric discovery lead to settlements completely unconnected to the merits of the underlying claims. But there are nevertheless more subtle tactics that national courts in Europe might employ to achieve desired results in a manner that eludes both appeals and, in the most dramatic case, infringement proceedings against the Member State concerned. As Forrester and Powell state in their introduction, ‘[n]ational courts will seek to escape the constitutional consequences of European competition law if the consequences of strictly applying it seem unpalatable’.183 The tactics that could be used are undoubtedly many, and the authors discuss, as one means available, the way a national court might interpret Commission decisions. National courts are of course precluded from taking decisions running counter to the operative parts of Commission decisions, which must also mean that corresponding elements of the decision have to be respected if failure to do so would call into question those 181 182 183

See above note 92. See the discussion above at note 93 and accompanying text. Page 297.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxxi  

operative parts.184 But the weight given by the national court to the other parts of a Commission decision – those that do not have irrefutable ‘probative’ effect, to use the language of the draft Directive – may depend on the perceived quality and fairness of the ‘federal’ procedures used in Brussels.185 This constitutes the authors’ last point in the paper as well. Between fore and aft, the authors cover selected topics that include the passingon defence and the line between too much collective redress and not enough. But another subject highlighted by Forrester and Powell, which may be characterized as their second principal point, is that – in the absence of any hard EU legislation, and while we swim in a small sea of reports, resolutions, consultations and now travaux préparatoires – private antitrust enforcement in Europe has in fact already come of age.186 The genie is out of the bottle in England and Wales, in the Netherlands and in Germany, and it may soon be roaming free elsewhere in Europe too.187 Forrester and Powell refer to a ‘shift in attitudes’, and to the rise of a competitive European ‘marketplace’ for the litigation of private antitrust claims. In this regard they present a roundup of various means by which such litigation is financed, including private investment and litigation insurance. One wonders what sort of backlash will occur if and when the phenomenon is caricaturized by critics as an abusive means of circumventing the ‘loser pays’ principle, cherished as a great sentinel keeping out frivolous litigation. Already the standard assumption is that third-party funding must be properly regulated when it makes collective legal action possible.188 The point of the authors is that, as a cause and consequence of the competitive marketplace, private antitrust enforcement – especially businessto-business litigation – is indeed expanding. That expansion feeds into the argument outlined above because, as private antitrust claims and the engagement of national judicial institutions in antitrust disputes become ever-more prevalent, the problem of flawed procedures at the EU level may have unexpected impacts Cf. above notes 84–88 and accompanying text. The possibility of a perceived legitimacy problem at the level of the Commission could also conceivably lead to more indirect ‘spillover’ adjustments, even unconscious ones, at national level. For example, such problems could affect the way national courts treat non-binding principles and methods contained in Commission guidelines, which could in turn affect litigation outcomes. 186 John Ratliff likewise points, in his contribution, to significant levels of litigation in a number of Member States. See also Geradin and Grelier, ‘Have We Only Seen the Tip of the Iceberg?’, cited above note 9. The submerged part of the iceberg represents the litigation of private damages claims without any assistance from EU rules, which is said to be ‘mushrooming’. To put all of these portraits in perspective: in the period 2006–2012, the number of Member States from the EU-28 reporting zero follow-on actions brought on the basis of a Commission infringement decision was 20. See Daniele Calisti and Luke Haasbeek, ‘The European Commission Sets the Stage’, cited above note 135, at 3; Commission, Impact Assessment Report (Annex II to this volume), para 52. Furthermore, even in Member States where private antitrust litigation flourishes, it may be necessary to consider how that litigation is composed. In Germany, for example, only 20 percent of the cases brought before the national courts concern claims for damages or unjust enrichment; cases most often aim at securing the nullity of contract clauses or obtaining injunctions in a business-to-business context. See Keske, ‘Collective Redress’, cited above note 18. 187 Taking the UK as the leading forum in the EU for follow-on damages actions (typically following a finding of infringement by the European Commission), the authors provide a helpful chart listing 25 such cases before the UK High Court and the UK Competition Appeal Tribunal. 188 See Commission Recommendation 2013/396, paras 14–16 and recital 19. 184 185

lxxii  Mel Marquis in an ever-growing number of cases and may build into a quiet cacophony of competition law as applied. Bruno Lasserre: ‘Integrating Public and Private Enforcement of Competition Law: Implications for Courts and Agencies’. It is well known that Europe’s efforts to establish a competitive market order made it necessary to have competition rules embedded within the framework of a common market, and that, partly because of the Commission’s broad powers and because of the central role given to the competition rules by the Court of Justice, competition enforcement became Europe’s first genuine ‘supranational’ policy.189 However, it is equally clear that supranationalism was not a destination or end-point for Europe in this context but a dimension of what has become, now more than ever, a multilevel regulatory landscape. The enforcement of competition law has become a significant part of national regulatory activity, with certain Member States leading the way. ‘Multi-levelism’ bears risks, including in particular, fragmentation and incoherence. But as Lasserre points out in this chapter, the competition field in Europe is characterized, in the main, by a process of convergence. This process may be thought of as both an input for and an output of a socialization process, or of a construction of culture. The idea of convergence in Europe should not be oversold, as differences between Member States will probably always persist, for as long as the EU persists. Furthermore, it is arguable that, since the emergence of the question ‘What has competition ever done for us?’, roughly associated with debates over the Constitutional Treaty that never was – and in light of the global economic malaise, competition policy in Europe is more politicized than ever before, another guarantee of heterogeneous preferences as opposed to a wide consensus on what competition law and policy should mean and how they should be executed.190 Interestingly, Lasserre is confident that the convergence process, which has generally been a public sphere phenomenon, has managed to pave the way for bolder steps in the private sphere. As he says, ‘EU antitrust law has become a privileged and fitting area for the development of European standards for collective redress’.191 With regard to collective redress, Lasserre points to a substantial change of scenery between 2004, when only three Member States (Portugal, the UK and Sweden) had collective redress mechanisms, and 2011, when the number had nearly quadrupled (and as of today it has nearly septupled). But in many Member States it is still not clear whether the legal possibility to bring collective damages actions for antitrust infringements is translating into actual recovery of 189 Lee McGowan and Stephen Wilks, ‘The first supranational policy in the European Union: Competition policy’, 28 European Journal of Political Research 141–169 (1995). 190 Attitudes toward collusion may be a way of illustrating this point. One could speak of a consensus in the United States as to the opprobrium attached to classic cartels. In Europe, while competition authorities are on the same page it is still not clear that a broad consensus has emerged that collusion should be condemned with severity or treated, loosely speaking, as a felony as opposed to a misdeameanor. 191 Page 317.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxxiii  

compensation for victims of those infringements. The combination of instrument design, institutions and shared value codes or habits within Member States varies substantially, and as many realize, the effectiveness of collective redress can vary widely in different parts of Europe. The difference between Member States where collective actions can really secure adequate outcomes and Member States that simply do not have the legal means to aggregate claims is starker still. Lasserre underlines the point that the status quo is unacceptable, above all because of the problem highlighted earlier: while some cartel victims – namely, those businesses with sufficient money and resources – can shop until they find a European jurisdiction in which to file a collective action, common citizens with low-stake claims are economically speaking deprived of any real possibility of recovering their losses if they do not reside in a Member State with an adequate system of collective redress. The market mechanism of forum shopping therefore appears to provide only suboptimal solutions that fall short from a variety of perspectives, whether it is the right to an effective remedy or the establishment of a common Union-wide area of justice or, in some cases (ie, where public authorities do not pick up the slack) the effectiveness of the competition rules. Lasserre thus has no doubt that the EU is well advised to act resolutely in the field of collective redress and to establish common rules that can ensure satisfactory outcomes in this area in all Member States. Wisely, Lasserre sees past the deceptive dichotomy of opt-in/opt-out and explains that ‘intermediate options might draw their inspiration from the proposals of the British Civil Justice Council or from the Danish example, which leave some degree of discretion to the judge on a case-by-case basis depending on the size of the class, its representativeness, the level of financial stake, and the status of the claimant’.192 As we have seen, the UK Government is moving in this general direction; the European Commission has decided not to do so and rather has established an improvident hierarchy that privileges opt-in class actions as a blanket principle. On a somewhat different issue, Lasserre advocates private litigation that puts predominant emphasis on cases brought on the heels of infringement decisions by competition authorities. He does not seem to go so far as to suggest that stand-alone actions are best forgotten, but he firmly subordinates them to follow-on claims, perhaps not surprisingly for the head of an agency.193 In this context one should guard against categorical views, since public authorities are neither infallible194 nor blessed with Page 326 (footnotes omitted). Pointing to a negative experience of the Conseil (now the Autorité) de la concurrence, Lasserre states that ‘stand-alone actions may undermine the efficiency of antitrust enforcement [and legal certainty] if a concurrent administrative investigation by an NCA is already underway and if it has led to on-site inspections’. Page 320. 194 For example, type II errors are possible. If one considers a Commission decision, national courts cannot validly adopt decisions running counter to it. However, the Commission does not adopt ‘constitutive’ exemption decisions (although Article 10 non-infringement declarations are possible). Therefore, when the Commission makes a type II error, there is (putting aside Article 10) no decision to which the decision of a national court would run counter. The nature of the type II error might be a failure to detect an infringement, or an erroneous finding of no infringement. However, the incidence of each set of errors is likely to be quite low. In the first scenario, if the Commission initially fails 192 193

lxxiv  Mel Marquis limitless resources.195 The cases that will sooner or later slip through the public net are probably those which require significant time and energy on the part of enforcers but which defy predictable outcomes due to their complexity, their rigorous evidentiary burdens or the difficulty they pose in fashioning relief. Abuse of dominance cases may sometimes fit this description of high-hanging fruit, and indeed at the national level there is significant stand-alone litigation in this area.196 The point is perhaps more muted if one is only considering the necessity of standalone private litigation in the collective redress context, since most (but not all) claims arising from abuse of dominance allegations will take the form of businessto-business litigation.197 Lasserre also advocates the extension of the Masterfoods rule (discussed above198) to decisions of national competition authorities, which again accords with his role as an enforcer. Since a rule of this kind would facilitate followon claims, he must be pleased to see that Article 9 of the Commission’s draft Directive would transpose Masterfoods in the way he suggests. Lasserre adds, though, that the ‘no decisions running counter’ rule does in itself guarantee success in court unless other evidentiary challenges are met, which may go beyond the standard tort prerequisites of causation and damages, depending on the circumstances. He also weighs in on other related issues such as whether it should be the role of competition enforcers to quantify anticompetitive harm, particularly with respect to the damages of particular victims, as opposed to aggregate harm. The answer, in his view, is that there should be no such duty but that furnishing such information in infringement decisions is often very desirable and can lead to ‘downstream’ efficiencies if generating it is not too costly. Another issue touched on briefly is access to evidence held by competition authorities in the EU, when such information is sought not only by litigants in European courts but also by plaintiffs suing in third countries; as with most other authors, Lasserre considers EU legislative action to be a logical necessity following the ECJ’s judgment in Pfleiderer. Finally, is it unduly severe and hence unjust to punish an undertaking for infringing competition law when it has already, or surely will, pay substantial damages to plaintiffs in private litigation? As we saw to detect a cartel, for example, but a private case is launched, the Commission will undoubtedly become aware of the alleged infringement, possibly via receipt of leniency applications or from media reports. The national court would then be precluded from taking a decision that would run counter to any decision the Commission may be contemplating, and may stay the proceedings – effectively turning the case into a follow-on action. As for the second scenario, it is difficult to know exactly how often anticompetitive agreements and practices are misdiagnosed as benign; but to be generous to competition agencies, it may be assumed that this is a rare event. The practical conclusion is therefore that type II errors in this context will tend to arise, if at all, from resource constraints and not from the mistakes described above. 195 On the limited resources of public agencies, see Philip Lowe, ‘Conclusions’, this volume, point 7. 196 See Fred Louis, ‘Promoting Private Antitrust Enforcement: Remember Article 102’, this volume. 197 On the possibility of and difficulties with consumer-to-business litigation in the context of excessive pricing cases, see above note 60 and accompanying text. 198 See above notes 77–88 and accompanying text.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxxv  

above, Ottervanger speaks of ‘disproportionate double deterrence’,199 and other commentators have raised similar concerns. According to Lasserre, however, ‘the right to compensation for a tort is a fundamental principle of law. It would therefore be illegitimate – and probably contrary to ECHR principles – to grant full or partial ‘civil immunity’ to a firm just because it has already been fined in antitrust administrative proceedings.’200 He explains furthermore that crediting a company in administrative proceedings for damages paid or to be paid in private proceedings improperly confuses the aims of public and private enforcement. If the aims are different and independent, as they should be held to be, then the two types of proceedings must not be substitutes and a zero-sum outcome would be illogical. Horst Butz: ‘Integrating Public and Private Enforcement in Europe: Issues for Courts’. This concise but revealing chapter highlights the legal and cultural distance between the sphere of national traditions and the sphere of EU law with its imperialistic DNA, programmed to interact with, hybridize and at the limit even displace those traditions unless the conflict can be mediated by doctrineshaping and political finesse. Ticking off a 10-point list of issues that emerge from Europe’s longed-for turn toward private enforcement, Judge Butz repeatedly asserts that however much certain reforms may or may not be wise or expedient, their path is obstructed by deep-rooted procedural and constitutional principles, at least in Germany. The issues he addresses are the following: the implications of significant court fees for plaintiffs and the discretion of judges to reduce them; locus standi in class action cases; the usefulness of agency decisions in national litigation given asymmetric information about evidence; the access courts and litigants have to pertinent documents, and limits thereto; burden and standard of proof; the use of expert (economic) evidence; the difficulty of, but also the variety of means used for, quantifying damages; the famous tension between plaintiffs’ access to leniency materials and enforcers’ tight grip on them; and finally the rarely discussed problem of how complex, multi-player antitrust litigation presents such enormous challenges for a trial judge that Dworkin’s Hercules himself, let alone an inexperienced judge, might find it a daunting task.201 With this inventory of 199 Tom Ottervanger, ‘Designing a balanced system: Damages, Deterrence, Leniency and Litigants’ Rights’, this volume, at page 20. 200 Page 324. On the other hand, Lasserre seems open to the introduction of some finely honed mechanism that limits the liability of a successful immunity applicant, provided that it does not leave claimants without adequate compensation. This issue too has now been addressed by the draft Directive, although it is unknown whether the approach espoused by the Commission (see above notes 52–54 and accompanying text) will be retained in the final version of the legislation. 201 In its VEBIC judgment, which concerns a challenge brought against an infringement decision by the (now historical) Belgian Competition Council, the ECJ seems to acknowledge that it is only the adversarial nature of such proceedings that serves to counter the risk that a national court could be overwhelmed by the complexity of a competition law case if the proceedings were ex parte and that its capacity for independent judgement could thus be compromised. As the Court states, ‘if the national competition authority is not afforded rights as a party to proceedings and is thus prevented from defending a decision it has adopted in the general interest, there is a risk that the court before which the proceedings have been brought might be wholly ‘captive’ to the pleas in law and arguments

lxxvi  Mel Marquis issues Butz puts a useful perspective on the conflicts that will either erupt or be submerged (only to surface later, perhaps in the context of preliminary references) as the 2013–2014 legislative procedure unfolds. Short as it is, the chapter functions as its own adept summary. Philip Lowe: ‘Conclusions’. Another brief contribution is provided by my conference co-organizer, who once again demonstrates his skill at capturing the salient points made during our day and a half of discussions in Villa La Fonte. I leave Philip’s conclusions for the reader to consider directly. Let us turn without delay to the next and final contribution, a provocative epilogue for this book.

Part V

Private Damages Claims and the Elusive Future

Veljko Milutinović: ‘The ‘Right to Damages’ in a ‘System of Parallel Competences’: A Fresh Look at BRT v SABAM and its Subsequent Interpretation’.202 Milutinović’s chapter serves two functions. First, it furnishes a fitting conclusion to the present collection. Second, it contains, as a last-minute postscript, an initial reaction to the Commission’s proposal of June 2013 on private damages actions. The essay begins with the observation that the system of parallel competences associated with the ECJ’s Delimitis and Masterfoods judgments (which permit national courts, subject to their options or obligations under Article 267 TFEU, to hear cases raising competition issues in parallel with or in anticipation of public enforcement by the Commission) and the Courage-fertilized right to damages for breaches of EU competition law have led to complexity and confusion. For example, a system of parallel competences seems to contradict a Hart-oriented concept of a (supranational) legal ‘order’, and thus is something of a hangover from the less orderly system of public international law. But Milutinović points to other problems as well. His discussion is organized around the following themes: the ‘binding’ legal effect of the so-far-unused ‘non-infringement decisions’ that the Commission (and none but she) can adopt under Article 10 of Regulation 1/2003, and other issues arising from the binding effect of public enforcement decisions; the relationship between leniency programs and follow-on damages claims in national courts; and limitations on such claims (or, conversely, limitations on limitations, such as curbs on permissible rules on standing). Some parts of the put forward by the undertaking(s) bringing the proceedings’. Case C-439/08, Vlaamse federatie van verenigingen van Brood- en Banketbakkers, Isjbereiders en Chocoladebewerkers (VEBIC) VZW [2010] ECR I-12471, para 58. However, it appears from the remarks of Judge Butz that at least in some cases heard by trial courts, adversarial argumentation is not in itself a sufficient guarantee that a case will be manageable. 202 For extensive discussion of issues arising from the Courage-based private litigation of the EU antitrust rules, see the author’s monograph, The Right to Damages under EU Competition Law: from Courage v Crehan to the White Paper and Beyond, Kluwer, 2010.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxxvii  

story, including, for example, the author’s discussions of access to evidence under Plfeiderer and CDC Hydrogene Peroxide, will be skipped here. The history of the system of parallel competences and of the right to damages is traced back to the case mentioned in the title of the chapter, BRT v SABAM,203 which is the standard citation for the (bold) proposition that Articles 101 and 102 are of sufficient precision, clarity and ‘unconditionality’ to qualify for direct effect under EU law (‘direct rights […] which the national courts must safeguard’).204 It is in the BRT judgment that the ECJ sets down its ‘foundational myth’ of a double sphere of enforcement, public and private, in which each is in principle of equal rank. Cracks in the myth appeared in Delimitis, however, since the possibility of dual enforcement brought with it a need to resolve private-public conflicts, and as noted earlier in this chapter it is national judges, unsurprisingly, who must yield to (contemplated or actual) Commission decisions, and not the reverse. Notwithstanding this set of conflict rules, the purportedly equal stature of public and private enforcement was later, as Milutinović says, ‘completed through the ‘Courage turn’ and its rights-based discourse’, in part by transposing principles from the Francovich and Brasserie du Pêcheur judgments to the domain of horizontal liability (ie, liability for breach of duty as between persons, as distinct from state liability) for infringement of EU competition law. The introduction of a rights discourse in Courage, which was not strictly necessary in order for the ECJ to decide the case, appears to be a critical juncture in the sense that private remedies are discovered in the penumbras and emanations of the Treaty itself – which means, among other things, that the EU legislator can promote them and shape their pursuit in a variety of ways, but generally it cannot abridge them (one can even imagine breaches engaging vertical liability here, though such cases are likely to be hypothetical). Nor may they be compromised by national (constitutional or ordinary) laws, or by national courts. However, while these consequences may logically follow from the notion of equal stature, the latter approach in the author’s view fails to reflect the fact that the effectiveness of private enforcement, and more precisely the ‘right to damages’, is derived from, instrumental to and conditional on the effectiveness of Articles 101 and 102. After explaining why, if the Commission ever adopted an Article 10 decision it would indeed have preclusive legal effects in a national court despite its ‘declaratory’ nature, Milutinović discusses what he calls merely ‘apparent’ versus ‘real’ problems relating to the ‘binding effect’ concept. The merely apparent problems concern fears about the independence of the judiciary, in a division of powers sense, and about the possibility that creating a system where an infringement decision by any NCA must be respected by courts in other Member Case 127/73, BRT v SABAM [1974] ECR 51. Ibid, para 16. As Milutinović points out, though, it was not in BRT but in Delimitis (para 50) that the ECJ made it clear that national courts are competent to apply the full substantive content of Article 101(1). Milutinović also explains that the direct effect of Article 101(1) could not have been inferred from Article 101(2) because nullity has the character of a sanction imposed in the public interest, and not of a (waivable) remedy for breach and personal injury. 203 204

lxxviii  Mel Marquis States could turn the courts into mere assessors of damages, to use the coined phrase.205 A more serious problem with the binding effect of national decisions is its uneasy fit with due process requirements.206 An infringement of Article 6 ECHR (Article 47 EU CFR) might potentially arise because a civil defendant before a court that must respect an infringement decision of another Member State’s competition authority is seemingly deprived of its right to a fair trial, at least as to the allegation of infringement, by an independent and impartial tribunal (which would not seem to be the case, however, if the decision has been upheld on appeal by a court with full jurisdiction to review factual and legal findings; and the most serious infringement decisions are appealed most of the time207). Furthermore, putting the ‘independent tribunal’ issue aside, and considering that the national competition authorities exhibit astonishing diversity, it is only by way of legal fiction that it can be said all the relevant enforcement systems follow a common fundamental rights standard; doubts remain, even post-Menarini, precisely because of that diversity. This explains why the European Parliament is presently considering a relaxation of the European Commission’s proposed rule, which would make all such NCA decisions binding in private litigation, with no emergency brake.208 (And as noted earlier in this chapter, the Council’s position is the most timid of all, as its favored rule would go no further than requiring Member States to ensure that cross-border final infringement decisions are admissible as evidence.) Milutinović would take a different approach by launching a thorough review of respect for fundamental rights standards in all the public enforcement systems across the Member States; to mess about with the ‘binding effect’ rule, without a more comprehensive review, is to leave the source of the problem unaddressed. 205 The cross-border ‘binding effect’ of final infringement decisions are also discussed above at notes 87–89 and accompanying text. 206 See also, eg, Stefano Grassani, ‘The Binding Nature of NCA Decisions in Antitrust Followon Litigation: Is EU Antitrust Calling for Affirmative Action?’, CPI Antitrust Chronicle, August 2013(1). For the contrary view, ie, the view that a cross-border ‘binding effect’ would not undermine fundamental rights protection, see Wils, ‘Relationship between Public Antitrust Enforcement and Private Actions for Damages’, cited above note 4, at 17. 207 Grassani argues, however, that in reality there may be cases in which review courts fail to engage in sufficiently rigorous review of an NCA’s infringement decision; he points to the Italian administrative courts in this regard (where they control a decision’s ‘legality’, as opposed to reviewing fines, where unlimited jurisdiction applies) and states that ‘the attribution of a binding nature to an NCA’s decisions – as Article 9 of the Proposal suggests – should be measured against the extent of such judicial review; and to how, in practice, the latter review is run by courts in their daily activity. If appellate judges are empowered to merely monitor the legality of [an] NCA’s decisions, lacking (or being reluctant to exercise) the ability to fully second-guess or retry ex novo the case, then the right to a fair antitrust trial at the ‘upstream’ public enforcement stage is not wholly safeguarded.’ Ibid, at 5 (footnote omitted). If the application of a weak version of contrôle de légalité results in a defective finding of infringement, and if Article 9 of the draft Directive endows that decision with preclusive effect, then the lower standard of judicial review may effectively be exported to Member States, such as Germany or the UK, where judicial control can be much more rigorous. This appears to create some scope, as Grassini argues, for forum shopping. See ibid, page 4. 208 As far as an emergency brake is concerned, see the Staff Working Paper accompanying the Commission’s White Paper of 2008, at para 162 (and cf. Article 45 of Regulation 1215/2015). In the June 2013 proposal, the idea of an exception to the rule of preclusive effect is gone.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxxix  

The author’s conclusion states that ‘EU antitrust law should be allowed to develop in a manner that best suits the effectiveness of its provisions – which must have primacy over rights derived from that effectiveness (eg, the ‘right to damages’).’209 He says that the binding effect of NCA decisions is ‘socially useful’ and ‘should not be precluded by intellectual constructs that seek to make public and private enforcement ‘equal’ at all costs.’ The primary intellectual constructs to which he refers are the system of parallel competences and the right to damages. These must be seen as derivative, not self-justifying; when one peers through the myth of equality and the teleological ‘retrodictions’ of jurisprudential accretion, a number of legal principles in the field of private enforcement which are now almost taken for granted may appear to rest on shaky foundations. As for the elusive future, Milutinović’s postscript on the proposed Directive published by the Commission in June 2013 contains further perceptive commentary that I leave to the reader’s meditations and pleasure. Conclusions regarding the proposed private enforcement package of June 2013 Regardless of how the Council and the European Parliament proceed to modify the draft Directive published by the Commission in June of 2013 (and on the assumption that they will indeed adopt some compromise version of the text), two trends can be expected to continue. First, private claims for damages will continue to be brought before courts with greater frequency in cases where the alleged victims of anticompetitive behavior are large or medium-sized businesses that have the means to pursue their own cases and do not really need a collective redress mechanism,210 or where (in some Member States where the law so allows) the claims of such businesses have been transferred to a claims aggregator in exchange for consideration and a percentage of any recovery. That trend can also be expected to extend ‘horizontally’, spreading from Germany, the Netherlands and the UK to a few other Member States, perhaps not many.211 This horizontal effect will be in part a spontaneous spillover that would likely occur in the absence of the Directive, particularly if Article 19(1) TEU were given serious consideration;212 but the Directive’s most useful provisions will likely lead to a 209 Emphasis in original. The sentiment also reflects arguments made in the author’s monograph, cited above note 202. 210 See Keske, cited above note 18, page 7 of the draft paper (discussing the claims brought by Deutsche Bahn and a number of municipalities in Germany in the rail track cartel case). 211 Where damages actions arise in different EU Member States involving the same alleged infringements by the same defendants, the separate claims may well be joined under the ‘Brussels I’ Regulation, and such consolidated actions may well proceed in those Member States (currently three of them) that already have a ‘head start’ in private antitrust litigation. To some extent, this tendency may be one factor among others that curbs the horizontal spread of private damages litigation. See Howard, ‘Too little, too late?’, cited above note 58, at 464 (forecasting ‘a concentration of jurisdiction in certain preferred Member States in any event, thwarting the Commission’s aspirations of an internal market for competition litigation across the EU’). 212 The second sub-paragraph of Article 19(1) TEU requires Member States to provide ‘remedies sufficient to ensure effective legal protection in the fields covered by Union law’.

lxxx  Mel Marquis wider and deeper spillover than otherwise might occur. In turn, the increase in litigation before the courts of the Member States will yield an increasing number of requests for preliminary rulings by the ECJ. Ideally, the resulting judgments will be coherent and provide clarity with regard to subjects both within and beyond the scope of the Directive-to-be.213 The feedback loop institutionalized by Article 267 TFEU will thus be able to develop further; and, particularly as those preliminary references increasingly prompt the ECJ to speak on substantive issues, a significant impact on public enforcement (and on enforcers’ ubiquitous soft law) can be expected as well. The development of coherent and systemically sensitive substantive principles is particularly important – as if this needed to be said – since wrongheaded liability rules can be even more problematic when they contaminate a system encompassing active private as well as public enforcement.214 As a tentative prediction, with secondary law in place, the ECJ will be emboldened to enunciate rules and principles to guide private antitrust litigation even as regards interstitial matters left untouched or barely regulated by the Directive, as opposed to merely ‘re-affirm[ing] the Rewe effectiveness case law’ and waiting for further action by the EU legislator.215 More generally, across Europe and compared to past experience, control over how to frame and resolve competition law issues will migrate, in some measure, from competition authorities in part to courts and in part to the parties that drive and shape litigation, a shift that may well influence the ways in which authorities and courts operate.216 To mitigate the possibility of disruptive fragmentation, particularly where judicial expertise remains underdeveloped, institutional cooperation – particularly in the form of participation, in appropriate cases, by the national competition authorities and by the European Commission in national litigation – may become ever more important.217 Yet this does not diminish the need for effective public enforcement as such, which remains of the highest interest for the public. The reasons for this are many but they relate, inter alia, to the links between public and private enforcement as highlighted in this and other chapters herein, and to the fact that even a perfectly functioning private enforcement system fails to address issues 213 Coherence has become, increasingly, a significant challenge for the Court of Justice, with its various chambers and configurations. But at the same time, coherence at the ECJ has perhaps never before been so important, now that a greater number of interlocutors (exhibiting, inter alia, extreme linguistic diversity) are making their entrance on stage. The risk of fragmentation is described by Gerber, ‘Private enforcement of competition law’, cited above note 4, at 450 (‘The goals and concepts of competition law are likely to become both less clear and less coherent. […] Rather than having one administrative office and, perhaps, one or two courts using concepts and articulating and interpreting goals and norms, private enforcement is likely to mean that many voices will use concepts and participate in the process of defining the goals of the system. This is likely to lead to less consistency in conceptual usage and less clarity in the articulation of goals.’). 214 See Reindl, Secretariat Note for OECD, cited above note 11, at 10–11; and in more detail, Richard Epstein, ‘The Coordination of Public and Private Antitrust Actions’, in ibid, 36–45. 215 Less optimistically, see Howard, ‘Too little, too late?’, cited above note 58, at 464. 216 Cf. Gerber, ‘Private enforcement of competition law’, cited above note 4, at 448 (trend toward greater incidence of private antitrust enforcement ‘will mean that administrators no longer control the operation and development of competition law. […] [C]ourts that make decisions in these cases will influence legal development and effectiveness as much or more than administrators.’). 217 See Reindl, Secretariat Note, cited above note 11, at 12.

Well Integrated Public and Private Antitrust Enforcement in the EU 

lxxxi  

such as the deadweight loss caused by anticompetitive conduct, or the complex task of developing coherent policies. Second, the trend of a growing gap between the foregoing claimants and other alleged victims of anticompetitive harm – chiefly small businesses and individual consumers – can also be expected to persist. For this latter category of unfortunates, the Directive and its overall positive features will be of limited impact. Abstracting from reality, so to speak, we might even imagine a definitive version of the Directive requiring the Member States to introduce the most plaintiff-friendly procedural rules in the world. Without sufficient economic incentives for small antitrust damages claimants to go to court, it is difficult to see how the objectives of full compensation and meaningful access to justice for this group have any remote chance of being achieved.218



218 Since it seems unlikely that the EU, or the Member States acting spontaneously, will address this incentive problem in the foreseeable future, it seems worthwhile to consider and discuss further possible second best mechanisms (or third-best, to be technical, since litigation and ADR should be pathological and compliance the norm) that may serve as imperfect proxies for restoration of loss for dispersed, small-damage victims. One such mechanism – ‘public compensation’ – is proposed in Ariel Ezrachi and Maria Ioannidou, ‘Public Compensation as a Complementary Mechanism to Damages Actions: From Policy Justifications to Formal Implementation’, 3 Journal of European Competition Law and Practice 536–544 (2012) (discussing the benefits of authorizing competition authorities to order infringing undertakings to compensate victims – not as a means of obtaining a fine discount but in addition to the final as calculated under the standard criteria of the authority concerned). See also Claus-Dieter Ehlermann, ‘Introduction’, in Ehlermann and Atanasiu, eds., Enforcement of Prohibition of Cartels, cited above note 1, at xxxix.

Andrew I Gavil*

Designing Private Rights of Action for Competition Policy Systems: The Role of Interdependence and the Advantages of a Sequential Approach

Introduction The spread of competition policy enforcement systems over the last two decades has spawned a great deal of commentary on institutional design. Given the initial focus on public enforcement, it is not surprising that much of that literature has focused on the choices faced in establishing enforcement agencies, defining their powers, and integrating them into existing administrative, regulatory, and judicial systems.1 Newly formed agencies looked to the more established jurisdictions, such as the European Commission, the United States, the United Kingdom, Canada, and Australia, both for their examples and, more actively, for guidance in drafting statutes and guidelines and for training. Donor agencies also interjected themselves, funding technical assistance projects around the globe and, on occasion, requiring the adoption of a competition law as a precondition to aid. As a consequence of this relative burst of activity, today the International Competition Network – which itself is just over a decade old – can boast more than 100 members. The competition policy systems that have emerged combine substantive and institutional features from the more established jurisdictions, adapted to local cultures and conditions. There is at once convergence, but also divergence, in this new global competition policy community. While new competition policy regimes were sprouting in some parts of the world, other more established ones were maturing. That maturation process included consideration of greater reliance on private rights of action to supplement the private enforcement system, but also had other goals, such as promoting public awareness of the values of competition and facilitating * Professor of Law, Howard University School of Law. 1 See, eg, Eleanor M Fox, ‘Antitrust and Institutions: Design and Change’, 41 Loyola University of Chicago Law Journal 473 (2010); William E Kovacic, ‘Rating the Competition Agencies: What Constitutes Good Performance?’, 16 George Mason Law Review 903 (2009); William E Kovacic, ‘Institutional Foundations for Economic Legal Reform in Transition Economies: The Case of Competition Policy and Antitrust Enforcement’, 77 Chicago-Kent Law Review 265 (2001); D Daniel Sokol, ‘Antitrust, Institutions, and Merger Control’, 17 George Mason Law Review 1055 (2010); Philip J Weiser, ‘Towards an International Dialogue on the Institutional Side of Antitrust’, 66 New York Annual Survey of American Law 445 (2011).

4  Andrew I Gavil the development of a broader ‘culture of competition’. As is true for public enforcement, the success of a private right of action to enforce competition laws is deeply dependent upon the existence of conditions that generally favor the rule of law, and more particularly, an established, well-functioning judicial system. Given modern competition policy’s focus on economic analysis, private rights of action for competition law infringements also require an infrastructure that can support and manage complex litigation, including highly differentiated legal and economic expertise. Moreover, as the topic of our conference correctly suggests, private rights must be ‘integrated’ into the larger public competition policy system, which may require re-calibration to take into account the effects of private rights on the public enforcement system. Private rights also must be considered in light of the features of the general litigation system, such as the rules of procedure and evidence. Beginning with the Green Paper issued in 2005,2 the European Commission opened a dialogue on the wisdom of, and challenges of creating, private rights of action for damages attributed to competition law infringements. The Green Paper was followed, after comment, by the White Paper in 2008.3 More recently, in February 2011, the Commission sought public consultation on collective redress, initiating the latest stage of the dialogue on facilitating private rights of action.4 In this paper, I will discuss recent efforts to promote private rights of action in Europe and how those efforts can take account of the larger goal of integrating private rights into existing public enforcement systems. First, I will emphasize the ‘interdependence’ of competition policy systems generally, and, more particularly, the interdependence of the various components of private enforcement regimes. To illustrate these points, I will discuss the relationship among (1) the private right of action for indirect purchasers (often consumers); (2) the availability of a mechanism for collective redress; and (3) access to economic evidence, including especially expert economic witnesses. As some including the Commission have observed, indirect purchaser rights have little independent value without a procedural device for enforcing them, such as ‘collective redress’. But can a means of pursuing collective redress alone be successful given the economic content of the substantive 2 European Commission, Green Paper on Damages Actions for Breach of the EC Antitrust Rules, COM (2005) 672 (19 December 2005), http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CO M:2005:0672:FIN:EN:PDF. See also Commission Staff Working Paper, Annex to the Green Paper on Damages Actions for Breach of the EC Antitrust Rules, SEC(2005) 1732 (19 December 2005), http:// ec.europa.eu/competition/antitrust/actionsdamages/sp_en.pdf. 3 European Commission, White Paper on Damages Actions for Breach of the EC Antitrust Rules, COM (2008) 165 final (2 April 2008), http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM: 2008:0165:FIN:EN:PDF. See also Commission Staff Working Paper, SEC(2008) 404 (2 April 2008), http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=SEC:2008:0404:FIN:EN:PDF. 4 See Commission Staff Working Document Public Consultation: Towards a Coherent European Approach to Collective Redress, SEC (2011) 173 (4 February 2011), http://ec.europa.eu/justice/ news/consulting_public/0054/ConsultationpaperCollectiveredress4February2011.pdf.  See also European Parliament, Resolution of 2 February 2012 on ‘Towards a Coherent European Approach to Collective Redress (2011/2089(INI)), http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-// EP//TEXT+TA+P7-TA-2012-0021+0+DOC+XML+V0//EN.

Designing Private Rights of Action for Competition Policy Systems 

5  

prohibitions of European competition law, unless reforms also provide for effective means for collecting and presenting economic evidence? It seems unlikely. Second, I will examine how, given the interdependence of various elements of the private right ‘system’, jurisdictions can best approach the task of introducing and integrating reforms designed to remove or reduce identified impediments to private rights of action. Two paths to reform will be considered: ‘holistic’ and ‘sequential’. Contrary to what might be perceived as the case, I will examine the US experience and argue that private rights of action developed sequentially over decades, not all at once, and that their development continues today in the US in the courts and legislative bodies. From this experience, I conclude that there is much to commend the sequential approach now being followed by the European Commission. Despite the interdependence of various components of successful system of damages actions, the ‘shock therapy’ of more holistic reforms – trying to address all of the pieces of the system at once – is not required and is likely to lead to unintended and unanticipated consequences. Finally, I will turn to the role of public enforcement agencies in assisting – and monitoring – the evolution of private damages actions. I will conclude that the public system has an important role to play in birthing the private system, and that like any good parent it needs to nurture, monitor, and even occasionally intervene to insure the well-being of, this new child until it reaches maturity. Otherwise, the salutary goals of the private action – providing an added means of deterrence, an effective means for securing compensation, and promoting a deeper culture of competition in Europe – may not be achieved. In addition, if the US experience is repeated, the potential price for not doing so may be judicial adjustments to doctrine implemented in response to the perceived excesses of private actions that will in turn constrain public enforcers. I will conclude, however, that the relevance of the US experience in this regard – and hence the risk of a similar kind of backlash to private damages actions in the courts of the EU – is limited given significant differences between the US and European litigation systems.

I. System interdependence Competition laws do not exist in a vacuum and are not self-executing. The commands and prohibitions contained in competition law statutes and regulations are constituent parts of larger ‘competition policy systems’. Whether they are effective depends on their goals, the specific features of that system, and how effectiveness is measured. Typical systems are comprised of prohibitions, some kind of public enforcement agency or agencies with investigatory powers, a remedial scheme that specifies the consequences of infringements, and a judicial system that, under specified and standardized rules of procedure and evidence, reviews the actions of the agency. These various elements, however, are highly

6  Andrew I Gavil interdependent. If prohibitions are not well-specified, even expert agencies will be challenged to enforce them in a consistent manner. Conversely, a wellspecified law is unlikely to be effective without a well-functioning enforcement agency. The same will be true of a prohibition that lacks effective sanctions, be they criminal or civil. Additional institutional features like leniency programs seek to boost effectiveness by increasing detection rates. Agencies and private plaintiffs alike also need means of obtaining access to the information necessary to prove infringements. And courts must have the institutional expertise necessary to review and intelligently evaluate the economic evidence presented to them. The three most typically identified goals of competition systems are deterrence, compensation, and remediation. Deterrence is often the primary goal of public enforcement systems. But public, and to a larger degree private, enforcement systems also may seek to provide compensation for victims, something that governments do not always do, even if they have a system of civil fines or disgorgement. A third goal is often remediation – providing means for halting offensive conduct and perhaps correcting for its adverse competitive effects, as with equitable and injunctive relief.5 As is discussed in the next section, one of the complicating factors in calibrating the private right of action in Europe is that it reflects the distinct supplemental goal of promoting a ‘culture of competition’. Calibrating the various components of a system to realize these goals is a complex process that is informed by each system’s unique goals. ‘Decision theory’ provides an additional, economic way of thinking about calibration. Decision theory has come to exert significant influence on how we think about system design and evaluate system performance. It prompts us to inquire about both error and direct or administrative costs. Assuming that no system is flawless, all systems will have some incidence of error and there will be varying consequences associated with those errors. Decision theory demands that we examine those errors and their likely consequences. Given any specific design, what will be the incidence and consequences of false convictions and false acquittals? Sweeping per se prohibitions can eliminate all false acquittals, but, depending on their scope, may tend to increase the incidence of false convictions. Conversely, we can reduce the incidence of false convictions by adopting very permissive standards of conduct, but that will tend to increase the incidence of false acquittals. Although it receives less attention, decision theory also demands that we consider the direct costs of enforcing any specific prohibition. What will it take for the parties and institutions to resolve a dispute given the prohibitions and other features of the system? If, for example, a lenient burden of proof provides certainty of condemnation, but raises concerns about false positives, one possible cure might be to raise the burden of proof by, perhaps, demanding greater economic proof of competitive harm. In that event, however, it is also fair to ask whether the benefit from the reduction of false positives will be outweighed by 5 For a more comprehensive discussion of the range of goals that may be served by a private right of actions, see Edward D Cavanagh, ‘The Private Antitrust Remedy: Lessons from the American Experience’, 41 Loyola University of Chicago Law Journal 629 (2010).

Designing Private Rights of Action for Competition Policy Systems 

7  

the cost of lost certainty and the increased processing costs that may follow from using a more demanding standard of proof. Similarly, if a kind and quantity of evidence is required as a matter of law to establish an offense or defense, it is also fair to ask whether the litigation system provides adequate means for gathering and presenting that evidence to decision-makers. The Interdependent Elements of the Competition Policy System

Procedural Rules Detection Substantive Legal Rules

Institutions & Rule of Law

Decision Theory Economic Presumptions Costs of Error Costs of Administration

Antitrust Injury and Standing Penalties Remedies Costs of Enforcement

Deterrence Compensation Remediation

These are not mere theoretical issues. Over the last thirty years, the antitrust decisions of the US Supreme Court have reflected greater awareness of the importance of false positives and the over-deterrence they can cause. The Court has also openly expressed its concern that the combination of treble damages, class actions, broad access to discovery, and reliance on jury trials might create incentives for plaintiffs to bring weaker cases that coerce defendants into settling to avoid the costs and risks associated with complete adjudication. To diminish the occurrences of false positives, it has largely relied on two techniques: limiting the standing of private parties6 and imposing more demanding burdens of pleading, production, and proof.7 There can be little doubt that raising the standards of proof has increased uncertainty and significantly increased the costs of deciding antitrust cases. But the Court has expressed almost no concern for the loss of certainty that had been associated with now abandoned per se rules and lenient burdens of proof and the increased party and institutional costs associated with reliance on more demanding 6 Brunswick Corp. v Pueblo Bowl-O-Mat, Inc, 429 US 477, 489 (1977) (private plaintiff must demonstrate ‘antirust injury’); Illinois Brick Co v Illinois, 431 US 720 (1977) (barring indirect purchasers); Associated Gen’l Contractors of Cal, Inc. v Cal State Council of Carpenters, 459 US 519 (1983) (limiting standing for remote injuries not proximately caused by defendant’s conduct). But see Reiter v Sonotone Corp, 442 US 330, 343 (1979) (overcharges attributable to an antitrust violation constitute injury to a consumer’s ‘property’); and Blue Cross Blue Shield of Virginia v McCready, 457 US 465, 472 (1982) (acknowledging consumer standing to challenge reduced competition flowing from exclusionary conduct). 7 Andrew I Gavil, ‘Burden of Proof in US Antitrust Law’, in Wayne Dale Collins (ed), Issues in Competition Law and Policy, Vol I, pp 125 et seq, ABA Antitrust Section, 2008.

8  Andrew I Gavil standards of proof. Neither has it been concerned with the possibility that the incidence of false negatives could increase due to cost and lack of access to the necessary proof. In effect, the Court has presumed that the benefits of a lower incidence of false positives outweigh the possible costs of these other effects. In part, that belief was animated by an unspoken assumption that markets will be selfcorrecting, and hence false negatives will not be as costly to competition as false positives. Today, courts and commentators are also expressing concern about the costs of administering competition law systems. For example, in the US, increased expectations for precise economic evidence have significantly boosted the cost of litigating antitrust cases. While that has benefited the large law firms and economic consultancies that often defend such cases, it has likely made it far more costly on average to resolve antitrust claims and made their outcome less certain. In the case of private damage actions, system interdependence prompts three areas of inquiry, which might be referred to as ‘foundations’, ‘mismatch’, and ‘backlash’.8 Reflecting an appreciation for the first concern of interdependency, the Green Paper focused largely on ‘foundations’, and asked whether the building blocks – the prerequisites – for a successful private action for damages existed within the context of the legal systems of EU member states.9 Hence, it identified various impediments to private damage actions, such as access to evidence, fault and causation requirements, standing, costs of litigation, and jurisdictional rules, and posed various reform options for each. After further study and consultation, the White Paper took the next step, suggesting specific solutions for some of the impediments that had been identified and examined in the Green Paper. But the Green and White Papers also demonstrated sensitivity to the problem of ‘mismatch’ – a primary concern in calibrating the private enforcement component of a competition system. For example, the White Paper observed that under the European Court of Justice’s Courage and Crehan and Manfredi decisions, the right of victims to seek compensation is guaranteed, and that includes indirect purchasers.10 But as a practical matter, it also acknowledged that the procedural components necessary to enforce those rights were lacking. One can argue convincingly that indirect purchaser rights have little value without some kind of procedural mechanism for collective redress. Many indirect purchasers will be The problem of backlash is addressed in Part III below. Another specific area of interdependency discussed by the White Paper was the interaction of leniency programs and private damages actions against cartels. See White Paper, cited above note 3, § 2.9. 10 See White Paper, cited above note 3, at 2, 4, citing Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297, and Joined Cases C-295 to C-298/04, Manfredi v Lloyd Adriatico Assicurazioni [2006] ECR I-6619. The overwhelming majority of US states now agree, and grant indirect purchasers the right to sue for antitrust violations under state antitrust law, but as is widely understood today, under the Supreme Court’s decision in Illinois Brick Co v Illinois, 431 US 720 (1977), those same indirect purchasers are barred from suing under the federal antitrust laws. See generally Andrew I Gavil, ‘Thinking Outside the Illinois Brick Box: A Proposal for Reform’, 76 Antitrust Law Journal 167 (2009); Andrew I Gavil, ‘Antitrust Remedy Wars Episode I: Illinois Brick from Inside the Supreme Court’, 79 St John’s Law Review 553 (2005). 8 9

Designing Private Rights of Action for Competition Policy Systems 

9  

consumers, and their individual injuries from a cartel, for example, or from abuse of dominance, might be small. Yet the collective financial harm to consumers can be very substantial. This, of course, is one of the principal justifications for creating collective redress.11 Consider, however, that a further premise of the Green and White Papers was that there now exists an established body of European Union-wide competition law that is ready to be enforced by private parties. The content of that law, however, has evolved towards greater economic sophistication. In fact, reversals in the courts attributable to a lack of economic support in part led the Commission to create the Office of the Chief Economist. If those same substantive standards are applied to private actions, will plaintiffs, including indirect purchasers allied together through a collective redress mechanism, be in a position to prevail without access to business records and strategic planning documents from defending firms (some degree of what is called ‘discovery’ in the US system)? Again, the White Paper addressed this mismatch problem, calling for a greater degree of access to the evidence necessary to evaluate competition law infringements.12 As I have written elsewhere, however, it stopped short of fully endorsing the use of expert economic evidence, which may prove to be a significant continuing impediment for private damages actions, both in establishing liability and proving damages. Who will play the role in private cases of the ‘Chief Economist’?13 As discussed here, ‘mismatches’ of expectations and means can undermine the effectiveness of reforms, such as private rights of action. If cases prove difficult to win, parties and lawyers alike will abandon the effort. More broadly, however, mismatches can corrode public confidence in the rule of law and, in the case of competition law, public understanding of, and commitment to, the underlying principles of a market-based economy. Quality calibration, therefore, is an essential goal of any reform effort intended to integrate private rights of action. There is likely no single, optimal way to calibrate every system. Many different combinations of prohibitions, penalties and other remedies, and institutional design are possible, and may achieve a different mix of deterrence, compensation, and remediation.14 While some have proven to be objectively ineffective (for example, agencies that lack autonomy and prohibitions in countries with weak rule of law traditions tend to be less effective), ‘optimal design’ is likely a mostly a relative, not an objective aspiration. See White Paper, cited above note 3, § 2.1. See White Paper, cited above note 3, § 2.2. See Andrew I Gavil, ‘The Challenges of Economic Proof in a Decentralized and Privatized European Competition Policy System: Lessons from the American Experience’, 4 Journal of Competition Law and Economics 177 (2008). 14 Illinois Brick, for example, can be understood as prioritizing deterrence over compensation. Faced with the challenges of calculating pass-on, the Court concluded that concentrating all rights of action in the hands of direct purchasers would maximize the incentive to challenge illegal conduct and achieve maximum deterrence, even though it might not be the optimal rule from the point of view of compensation, because it barred indirect purchasers from suing even though pass-on surely occurred in some cases. 11

12 13

10  Andrew I Gavil

II. Holistic and sequential approaches to integrating private rights of action into public competition law enforcement systems With the Green and White Papers, as well as the consideration of mechanisms of collective redress, the European Commission has embarked upon what might be described as a ‘sequential’ approach to reform. Having identified a set of impediments to private actions, it has elaborated on specific proposals for addressing them that, as noted in Part I, evidence appreciation for the interdependence of the various elements of its competition policy system. But it has not done so with the expectation that a comprehensive, European-wide solution will be instantly adopted, or that even a significant number of member states will embrace broad-ranging reforms. It has chosen more of an incremental approach, relying primarily on its role as an advocate for competition policy reform. One might be inclined to criticize such an approach in light of what has already been said here of interdependence and the collective list of impediments identified in the Green Paper. Why not seek a comprehensive, EU-wide reform based on the White Paper’s recommendations? First, and wholly aside from the obvious response that it would not be politically feasible, there are significant dangers associated with such multiple, simultaneously implemented reforms. Precisely because of interdependence, it is difficult to predict in advance how multiple reforms, carried out simultaneously, will interact with each other, as well as the other elements of the system. For example, as the White Paper acknowledged, it is important to ask what effect amplification of the private right of action will likely have on leniency applications. If the availability of leniency currently improves detection, but fear of private suits inhibits firms from coming forward in the future, would an increase in private damages actions have the unintended consequence of diminishing the incentives to submit leniency applications, undermining cartel detection and reducing cartel prosecution? Similarly, if cartels were pursued through collective actions brought by indirect purchasers, would the combination of private damages actions and collective actions undermine the incentive of cartel participants to particulate in a leniency program? On the other had, will the threat of effective private actions provide additional deterrence, reducing the incidence of cartel formation and hence the importance of the leniency program? With too many new parts in motion, it might be difficult to isolate cause and effect to answer these kinds of questions. Moreover, even if ‘the perfect’ could be known in advance, as the saying goes, it should not be the ‘enemy of the good’. In other words, even if a sequential approach is a practical concession to political reality and practicality, it may be a defensible one, and it may move the system in the right direction. A decided advantage of sequential change over system-wide calibration is that it permits observation and assessment, as in a controlled experiment, of the impact of each

Designing Private Rights of Action for Competition Policy Systems  11   step. Over- and under-correction can more easily be observed, diagnosed, and addressed. So does the US experience suggest otherwise? Perhaps surprisingly, it does not, largely owing to a possible gap between the perception and reality of the private enforcement system. First, it is historically incorrect to suggest that the antitrust private right of action in its current form has existed since the modern US system’s inception in 1890. While it is true that the current statutory authorization of a treble damages private right of action for persons injured by an antitrust violation originated in the Sherman Act of 1890,15 it would be inaccurate to say that the private right as it exists today originated in 1890. Such a view disregards the ‘systems’ nature of the private right of action. Like any field of law, to flourish competition law requires a complete infrastructure, and such infrastructures take time to evolve. In truth, private treble damage antitrust actions were not common in the United States until the 1960s – seventy years after the private right of action was created. What was missing before then? First, antitrust as a field of study and legal practice was slow to develop. The broad-based educational, legal, and economic infrastructure necessary to nurture and then monitor the field simply did not exist before the 1950s. Although there were some early scholarly articles about antitrust that appeared in law reviews at the turn of the twentieth century, and William Howard Taft authored an important early study of the field in 1914, the first antitrust casebook for use in law schools was not published until 1937. In addition, the field of industrial organization economics, including the Chicago School’s brand of law and economic analysis, advanced significantly following World War II at mid-century. Not surprisingly, therefore, the Antitrust Section of the American Bar Association was not established until 1952, when it held its first ‘annual meeting’.16 Second, the field of complex litigation did not develop until mid-century. Until 1938, there were no uniform federal rules of civil procedure, and the modern class action rule was not promulgated until 1966, followed by the basic framework for the modern discovery rules in 1970. Similarly, there were no uniform federal rules of evidence until 1975. Indeed, despite the presumed enticement of treble damages and attorneys’ fees, there were few if any private antitrust cases filed in the federal courts until the Electrical Equipment Antitrust cases of the 1960s. Only then did all of preconditions to a fully functioning antitrust system coalesce so that the incidence of private antitrust civil actions became significant.17 One direct and important further procedural consequence of those cases was the creation in 1968 of the Judicial Panel on Multidistrict Litigation, which facilitated the coordination In 1914 it was relocated to Section 4 of the newly adopted Clayton Act, 15 USC § 15. For a timeline showing the history of the founding and evolution of the Antitrust Section, see Special Insert, ABA Section of Antitrust Law 50th Year Milestones 1952–2002, Antitrust, Summer 2002 (following page 46). 17 See Douglas H Ginsburg and Leah Brannon, ‘Determinants of Private Antitrust Enforcement in the United States’, 1 Competition Policy International 29, 32 (Autumn 2005) (where Figure 1 shows level of private and public antitrust cases filed in federal courts from 1945–2000). 15 16

12  Andrew I Gavil of multiple, related antitrust cases filed in different federal districts around the country.18 Before those cases, the phenomenon of large volume (more than 2000 private cases were triggered by the alleged price-fixing conspiracy), multiple, related follow-on cases was simply unknown. The private right of action in the US, therefore, did not sprout in its fullest form the day the Sherman Act became law. There was no ‘system’ to support it. That system, including an expert bar and an interested academy, took decades to form and many of its pieces, such as class actions and modern discovery rules, developed without any specific focus on antitrust, even though it was collaterally and profoundly affected by them. Collectively, therefore, the construction of the modern US private right of action is correctly understood as having been sequentially developed – over decades. Second, although the US private antirust system certainly has its excesses, those excesses can be easily exaggerated, and often are. It is very difficult and very expensive today for plaintiffs to successfully bring private antitrust cases. Like running the hurdles, plaintiffs must vault over demanding pleading standards,19 screens for necessary expert testimony,20 increasingly demanding standards for class certification,21 and heightened burdens of proof.22 These are not cases lightly instituted for ‘quick settlements’ as is frequently suggested by irate defendants, their lawyers, and their economists. As a result, the number of private civil cases filed in the federal courts has steadily declined over the last forty years on average and is now roughly half of what it was in the 1970s despite a burgeoning federal civil docket. In recent years, on average there have been 600–800 private civil antitrust cases filed in the federal court system out of a total civil docket of roughly 270,000 cases. The numbers have been declining, however, and in the two most recently reported years, 2010 and 2011, there were only 544 and 475 new federal private antitrust cases filed.23 In truth, of course, these numbers can understate the import of the cases, which can be very large, long-lasting, and challenging. In addition, focusing solely on federal statistics understates the number of private antitrust cases actually filed each year. Because a majority of states have rejected Illinois Brick as a matter of state antitrust law, indirect purchaser cases can be filed in significant numbers in state courts. Even with those caveats, however, would the European Commission be troubled if the volume of private cases filed throughout the Member States in say 2020 was 600? It seems unlikely. 18 See 28 USC § 1407. For a contemporary discussion of the cases, see Phil C Neal and Perry Goldberg, ‘The Electrical Equipment Antitrust Cases: Novel Judicial Administration’, 50 ABA Journal 621 (1964). 19 See, eg, Bell Atlantic Corp v Twombly, 550 US 544 (2007). 20 See, eg, Daubert v Merrell Dow Pharmaceuticals, Inc., 509 US 579 (1993). 21 See, eg, In re Hydrogen Peroxide Antitrust Litigation, 552 F.3d 305 (3d Cir. 2008). 22 See, eg, Leegin Creative Leather Products, Inc v PSKS, Inc, 551 US 877 (2007) (resale price maintenance); Verizon Communications Inc v Law Offices of Curtis V Trinko, LLP, 540 US 398 (2004) (refusals to deal); Brooke Group Ltd v Brown & Williamson Tobacco Corp, 509 US 209 (1993) (predatory pricing). 23 See Administrative Office of the United States Courts, Annual Report of the Director, Judicial Business of the Federal Courts, Table C-2A (2011), http://www.uscourts.gov/uscourts/Statistics/ JudicialBusiness/2011/appendices/C02ASep11.pdf.

Designing Private Rights of Action for Competition Policy Systems 

13  

III. Nurturing private rights: the continuing role of public agencies As noted earlier, public competition enforcement agencies may have to live with the consequences of poorly conceived private actions. If a ‘market’ develops for private actions – if parties and lawyers alike perceive that they are economical to initiate – it will be hard to predict which kinds of cases will be initiated. If courts do not view the cases brought favorably, they may respond as have courts in the US by elevating standards of pleading and proof, imposing screens that require very specific kinds of proof that may not always be reasonably available, and generally developing skepticism about the wisdom of competition law as a mechanism for regulating business conduct. In short, unleashing a private market for private rights of action could trigger ‘backlash’ that would very much defeat the purposes of promoting them in the first place. As former US Federal Trade Commission Chairman William E. Kovacic once cautioned, ‘an expansion of private rights could lead judicial tribunals to adjust doctrine in ways that shrink the zone of liability’, which may ‘encumber public prosecutors’ enforcing the same prohibitions.24 If for no reason other than self-interest, therefore, the agencies will want to monitor the development and use of private actions. But, as noted above, the risk of backlash can be overstated if based on the US experience. It is evident that in limiting the scope of US antitrust law, the US Supreme Court has been influenced not merely by the treble damage remedy, but by the specter of protracted and costly discovery and the threat of class action damages coercing unwarranted settlements. The oft-criticized US litigation ‘toxic cocktail’ consists of contingency fee arrangements, the right of prevailing plaintiffs to recover their attorneys’ fees, treble damages, broad discovery, class actions, and jury trials. These features of the US system simply do not exist in Europe and are not likely to result – certainly not to the same degree – from the reforms being considered. Beyond self-interest, however, lies the greater importance of developing competitive economies. In their role as advocates of that goal, public agencies will want courts and private parties to get the economics right. The incentives of the private rights market and the litigation system, however, may not always support that result – indeed, it may not even be the chief aim of the litigants. Plaintiffs and defendants alike may be inclined to seek any advantage they can through legal process, even at the price of distorting the ultimate result for competition. Agency intervention, therefore, may be critical, especially in the early days of development. As Bill Kovacic has observed, ‘[a]n expansion of private rights in the EU and the Member States … is likely to alter the way in which public competition agencies spend their resources’.25 Agencies will not be well advised to be bystanders as private rights develop. 24 See William E Kovacic, ‘Private Participation in the Enforcement of Public Competition Laws’ (15 May 2003), http://www.ftc.gov/speeches/other/030514biicl.shtm. 25 Kovacic, ibid.

14  Andrew I Gavil Agency intervention might take many forms. Here are four ways that agencies might encourage thoughtful and productive development of the private right of action beyond advocating reforms that facilitate the filing of damages actions. Monitoring Private Action Activity. There will be no effective way to evaluate the quantity or quality of private actions if they are not tracked in some form. Because it will generally be more difficult to identify and study competition law cases retroactively, a competition case tracking system could be implemented that would build a data base of information designed to assist statistical as well as qualitative analyses of private cases. The first order of business will be to accurately identify them. Because such actions will be filed in the courts of Member States, a successful monitoring program would likely require partnering arrangements with National Competition Authorities (‘NCAs’). It would be useful to know, for example, not just how many cases are filed, but the general nature of the alleged infringements (eg, cartels, abuse of dominance, dealer restraints), the specific conduct being challenged (eg, price fixing, refusals to deal, bundled rebates, resale price maintenance), and the remedy being sought (eg, damages or injunctive relief). Together, the Commission and the NCAs could share the task of tracking and evaluating the private cases. And from the broader set of cases, subsets could be identified for further study. This kind of tracking could also aid in determining whether various types of interventions would be advisable. Case Intervention. One of the more obvious devices available to the Commission and the NCAs would be an amicus curiae brief program. This has long been a staple of agency advocacy in the US, and has often been used to rein in poorly conceived antitrust cases, and, less frequently in recent years, to support plaintiffs being confronted with unreasonable interpretations of law or demands for elevated standards of proof. Intervention might also be especially valuable with respect to the design of remedies.26 Finally, the agencies might want to offer their views if defendants urge courts to adopt other calibrating devices, such as limits on evidence or standing requirements. Direct Assistance. Monitoring private activity could suggest a need for more aggressive intervention. Just as was true in the early days of agency development in the wake of the collapse of the Soviet Union and the expansion of competition agencies in Central and Eastern Europe and elsewhere in the world, the European Commission and the NCAs might find it advisable to offer a form of ‘technical assistance’ to private lawyers, economists, and advocacy groups interested in pursuing private litigation. This would not likely take the form of specific case support. Rather, the agencies might play the role of teacher, offering instruction on competition law requirements and perhaps even assistance in prospective case evaluation, helping parties to identify ‘good’ and ‘bad’ cases for challenge and advising on possible remedial solutions. 26 See, eg, Kovacic, cited above note 24 (‘Competition agencies may need to spend more time preparing amicus curiae submissions in private cases to guide the courts in their treatment of substantive doctrinal issues and, in some instances, the proper design of remedies’.).

Designing Private Rights of Action for Competition Policy Systems 

15  

Trust But Verify. If the level and nature of competition actions give rise to concerns – if a significant proportion of the cases brought simply are not worthy and threaten to inhibit competition – a more aggressive type of intervention could be considered. Although not common in competition law, in other areas of regulation private parties can be required to seek and receive approval from the appropriate government agency to initiate a civil action. In the US for example, private cases alleging employment discrimination are screened and cannot be initiated unless the government issues a ‘right to sue letter’. This provides the government with a very direct way of screening and ensuring at least the plausible value of specific, proposed civil suits. Between the White Paper and the public consultation on collective redress, the European Commission has assumed a leadership role in exploring the need for more private damages actions, identifying impediments to their use, and proposing solutions for facilitating their future growth. Having assumed that role, the Commission might also deem it valuable to undertake concrete steps to monitor the progress of any reforms.

Conclusion Intra- and inter-system calibration will be a challenge for years to come as the global competition policy system evolves and matures. In the case of private rights of action, jurisdictions will need to carefully consider the need for private enforcement as a complement to public enforcement, the goals sought to be achieved, and the receptivity of the judicial system to the needs of competition cases. As I have discussed, although the success of private rights ultimately will turn on many different components, it is not necessary, and might even be counterproductive, to undertake broad-based reform efforts all at once. While a sequential approach may not fully succeed in the short run, it will permit careful integration and perhaps avoid judicial and political backlash. Markets will play an important role in the process, but public agencies also must keep a watchful eye and play an important role. New drivers may not yet possess the requisite skill to drive Ferraris, so the keys should not be turned over to them without proper training and some adult supervision.

Tom Ottervanger1

Designing a Balanced System: Damages, Deterrence, Leniency and Litigants’ Rights

I. Setting the scene 1. What do we mean by ‘private enforcement’? In order to properly understand the impact of EU competition law, its contextual basis in EU law must first be briefly examined. In the seminal case of Van Gend & Loos the Court of Justice ruled that ‘the Community constitutes a new legal order of international law...the subjects of which comprise not only Member States but also their nationals... Community law therefore not only imposes obligations on individuals but is also intended to confer upon them rights...These rights arise not only where they are expressly granted by the Treaty but also by reason of obligations which the Treaty imposes in a clearly defined way upon individuals as well as upon the Member States’. This case, one of the most celebrated cases ever decided by the Court, introduced the principle of ‘direct effect’ of Union law. When certain conditions are fulfilled, this doctrine allows provisions of Union law to be relied on by individuals before national courts in cases against other individuals or against governments. Provisions of Union law which impose clear, precise and unconditional obligations are, in other words, directly applicable. The Court of Justice has held that the competition rules are among the rules that are, by their very nature, directly applicable. Case law has demonstrated that the direct application of the competition rules in the national legal order can have a great impact on relations between individuals, creating rights which the courts must safeguard. The Court, in Courage v Crehan, affirmed that this is the case, even where an undertaking that is a party to a (vertical) contract invokes the nullity of that same contract because it breaches Article 101 TFEU. In this same case, it was held that undertakings that had suffered a loss caused by a restrictive practice in violation of the competition rules could claim damages and that national law should provide for a remedy for such individuals – irrespective of their position, either as a customer or as a party to an anticompetitive practice. It has now been clearly established by the Court and by the fundamental reform of 2004 (Regulation 1

Europa Institute, University of Leiden; Allen & Overy LLP, Amsterdam.

18  Tom Ottervanger 1/2003) that national courts, in administrative, civil or, where appropriate, even criminal proceedings have the full power to apply the Union competition rules. So what does private enforcement in this context actually mean, bearing in mind that, notwithstanding the private law notion of nullity in paragraph 2 of Article 101, competition law is after all conceived as public economic law? In the context of competition law, the concept of private enforcement refers to a wide variety of actions under private law where the rules are relied upon before a national civil court, or before arbitrators, as a shield or as a sword, be it as the core issue or as just as one of various arguments raised by the plaintiff or the defendant. A party may try to escape a contractual obligation by invoking, as a defence, the nullity of a contract or a clause. Also, a party may base a claim for restitution on alleged nullity. Quite often a party in need of interim relief seeks an injunction, either under Article 101 or under Article 102, to bring anti-competitive conduct to an end. Last but not least, a party, typically a direct or indirect customer of a ‘member’ of a cartel, may institute an action for damages, although until a few years ago in practice such claims have always been relatively rare. Damage claims usually concern follow-on actions in cartel cases. This latter category, follow-on actions for damages allegedly suffered by victims of anti-competitive conduct for which the defendants have already been fined, is the main subject of this contribution. The Commission’s early activism in this matter is notable; as far back as 1966 the Commission had already published a report on the possibility to sue for damages caused by violations of EU competition law under the (tort) laws of the then six Member States. More generally, the Commission has for quite some time advocated a more frequent application of the competition rules by the national courts. In fact, the recitals of the 2003 Modernisation Regulation confirm that national courts have an essential part to play in the application of the competition rules. In some landmark cases such as Courage, Manfredi and Pfleiderer the Court held that the right to claim damages discourages anticompetitive practices and ‘can make a significant contribution to the maintenance of effective competition…’. The debate regarding the direct effect of Union law, referred to above, is primarily about better safeguarding the rights of private parties, while at the same time serving the public interest of safeguarding the full effectiveness of the competition rules and thereby the principle of undistorted competition in the internal market.

2. What’s in a name? Nevertheless, I would submit that the term private enforcement as used today notably for follow-on cartel damage claims is somewhat confusing. The word ‘enforcement’ (as in ‘enforcement agency’) suggests not only some higher public interest goal but also an element of potential deterrence and punishment. More appropriate terminology for the direct application of the competition rules in all sorts of private litigation between undertakings might be private or civil remedies,

Designing a Balanced System 

19  

civil litigation or civil action, or when specific reference is made to follow-on actions for cartel damages: private damage claims or damage actions. The expression ‘civil enforcement’ may be misleading for American lawyers – in the US, if I understand correctly, civil enforcement, or civil proceedings, are what in continental Europe would be referred to as administrative procedures, as opposed to criminal prosecution. The current debate largely focuses on obtaining redress for losses suffered in those cartel cases where, under the rules for public enforcement, the competition authority has established in a final decision that there has been an infringement and where fines have already been imposed on the defendants. Such private actions are markedly different from public enforcement, notably in that private plaintiffs are not at all concerned with the general interest or with competition policy issues (eg, relating to the interpretation of the exceptions of Article 101, paragraph 3) or about the precedential value of judgments. Instead, a private plaintiff claiming to have suffered damages is concerned only with his own private commercial interest as homo economicus. Of course, as mentioned earlier, there is a certain element of public interest attached to it. But that is not unique to competition laws. The possibility for victims of, for example, environmental damage to sue for compensation shall certainly contribute to the public interest of safeguarding the effectiveness of the laws concerned. However, deterrence may be a side effect but is not the goal of such private actions.

3. Two sides of the same coin? This brings us to a preliminary policy point worth briefly addressing. Is it desirable that the Government or a competition agency pro-actively promotes private action as a way to support its own public enforcement? In this regard, are private and public enforcement really two sides of the same coin? Is it desirable from a policy point of view to accumulate public and private liability, ie, fines and damages, in order to deter illegal behaviour as much as possible? One of the principal notions inspiring the introduction of Regulation 1/2003 was that, next to the national agencies, national courts should bear part of the burden of enforcing the rules in private cases. The Commission could then concentrate its limited resources on the most serious infringements. From this perspective, private action at the national level can be seen as strengthening enforcement through decentralization and by substituting one method for the other, not necessarily by accumulating liability. On the other hand, the origin of the Green Paper (2005) and the White Paper (2008) lays not only in assisting victims of illegal conduct in effectively enforcing a right to compensation; but, more importantly, in increasing deterrence by adding to the punitive element of fines the additional risk of having to compensate ‘private’ harm. Facilitating damage claims is considered helpful to encourage compliance and developing a culture of competition.

20  Tom Ottervanger It goes without saying that harmonization of private (substantive and procedural) law in certain areas is useful, if not indispensable, to create and maintain the internal market. A coherent European approach to ‘collective redress’ of cartel damage may in this respect indeed be required. Perhaps, as in other areas such as environmental law or product liability, further measures may be necessary to facilitate individual legal proceedings. However, it is highly contentious whether deterrence – as a main incentive for compliance – should be the primary purpose of such efforts; and whether it is right for the agency in its press releases to actively stimulate private parties to sue for damages in follow-on actions. Such encouragement by the agency suggests, albeit implicitly, that harm has indeed been done even though actual harm is not a prerequisite for the prosecution of an infringement. There may very well be violations of Article 101, for example relating to the exchange of sensitive business information, that cause little or no harm. After all, in its decision imposing a fine the Commission does not necessarily deal in any great detail with the issue of effect, impact or harm. Moreover, accumulation of public and private enforcement, ie, of fines and damage awards, may lead to disproportionate double deterrence. The objectives of public enforcement in the general interest and of followon damage actions in a very private inter partes setting are different; and, consequently, so are the roles of the competition authorities on the one hand and the national civil courts on the other. They should not be confused. Generally speaking I would argue that public enforcement is to be preferred where, as often is the case in competition law cases, the transaction costs for a victim of an alleged infringement are high – because of lack of information, because the norms are rather vague, and because of the complex (economic) analysis that may be required. Enforcement under such circumstances should not be ‘privatized’. It would be wrong, I believe, to compare the situation in Europe with that in the US. It appears to be the case that in the US roughly 90% of the cases are private and often have a pretty strong deterrent effect in that treble damages are claimed – most cases, so it seems, are settled, the nuisance value being high. Already the threat of claims may deter not only unlawful but also lawful and even desirable cutting edge competition. This means that across the Atlantic enforcement is, and has always been, to a large extent a ‘private’ phenomenon with a punitive and therefore public element attached to it. Is the DOJ actively promoting private damage claims, or are the incentives such that encouragement is not necessary?

4. Main issues of interaction between public and private action. ‘Designing’ and ‘integrating’ public and private action concerning follow-on damage claims suggests positive action by the legislator or, where possible, on an administrative level through Guidelines issued by competition authorities. As briefly mentioned, the two forms of action should not be confused. Moreover, the authorities should in my view exercise self-restraint in promoting private

Designing a Balanced System 

21  

(collective or individual) damage litigation as a way to encourage compliance. It does not appear necessary therefore to design an overall system at the Union level that facilitates and enhances private follow-on damage claims while preserving public enforcement. Follow-on damage claims are, it is submitted, not some form of policy instrument for attaining general interest objectives through further deterrence. They are simply tort actions. What is necessary though, in any event, is to identify those issues arising where the two courses of action may interfere or interact. The most pressing, I would submit, are the somewhat overlapping issues of leniency and access to documents. These will be briefly discussed below, followed by some comments concerning the binding effect of Commission decisions in civil follow-on cases and the Dutch example of collective redress.

II. Balancing interests 1. Leniency The leniency system is seen as the centrepiece of today’s competition enforcement. Whistleblowers receive lenient treatment in exchange for providing incriminating evidence. The Commission’s leniency programme has proven to be very successful but it does raise a number of questions in relation to civil litigation by alleged victims of a cartel. These questions relate to the position of the whistleblower as a defendant in a civil case as well as to the disclosure of leniency documents to the plaintiffs. First of all, there is the prisoner’s dilemma of confessing wrongdoings in exchange for immunity (or for a reduction of fines) while at the same time, perhaps, triggering actions for damages as a result of the mere fact of disclosing and admitting an infringement. The same applies, to a certain degree, for settlements and ‘undertakings’. This can be a very difficult (strategic rather than ethical) choice indeed, irrespective of the question, discussed below, of whether or not certain documents may be disclosed to private litigants seeking damages. The latter question receives a lot a attention in law journal articles, court cases and statements by competition authorities while the first issue is often ignored. But the mere fact that a party is revealed in a Commission decision to be the party that blew the whistle or otherwise admitted involvement in the cartel, and that it is unlikely to appeal the decision, makes that party vulnerable to damage claims (and, where joint and several liability fully applies, vulnerable even to claims for the aggregate damage, including harm caused only by the party’s co-cartelists). This no doubt already today reduces the attractiveness of the leniency programme. The 2008 White Paper toyed with the idea of providing rewards or incentives to leniency applicants as regards both their public and civil liability, such as a

22  Tom Ottervanger reduction not only in the fine but also in the damages claim and/or the removal of joint and several liability (the latter approach being not so much a reward stricto sensu but rather a limiting of liability to ‘his’ share of the damage) or some other form of limiting liability to only direct victims of the leniency applicant. These ideas would require significant changes to the civil law systems of the member states, either through creation of an EU regulation or a directive specific to competition. In the second place, there is an obvious conflict between, on the one hand, the effectiveness and attractiveness of leniency programmes and, on the other hand, the effectiveness of private enforcement, which results from disclosure to private claimants of documents created by leniency applicants. Seen from this perspective, encouragement of private enforcement may undermine public enforcement. If the agencies wish to maintain the undeniable ‘success’ of their whistleblower programmes, a solution needs to be found for dealing with leniency applications and documents. What therefore remains, whether or not the legislator would limit the liability for damages, is the exclusion of discoverability of so-called corporate statements. The Commission’s policy, as outlined in the 2008 White Paper and as defended also in US litigation, is never to disclose corporate statements. Thus far it has succeeded in avoiding court orders forcing it to disclose documents relating to the origin of the case. This is not only a matter of fairness but also one which, according to the competition authorities, serves the interests of public enforcement. If these statements that are crucial for detecting infringements and initiating investigations did not benefit from protection, so the argument goes, leniency would become less attractive. There is certainly some truth in this but at the same time one may question whether claimants really need these statements to bring their case. In follow-on actions, the existence of the infringement as well as the role of the leniency applicant already follows from the decision which is binding on the civil courts as will be discussed below. The first case before the Court specifically dealing with the conflict between public and private enforcement is the preliminary reference by a German court in the 2011 Pfleiderer case (Case C-360/09). Pfleiderer is a purchaser of products supplied by members of a cartel that had been fined by the German competition authority. It sued for damages and had applied for access to the file, including the leniency applications. Access to certain documents had been refused and in appeal the national court asked the Court of Justice essentially whether disclosure would undermine the effectiveness of EU competition law. The Court confirmed that disclosure of leniency documents could compromise the effectiveness of leniency programmes; although a claimant is not precluded from access, it is ‘for the… (national) courts…on the basis of their national law to determine the conditions under which such access must be permitted or refused by weighing the interests protected by EU law’. This judgment does not give concrete guidance and calls for decisions on a case-by-case basis with an outcome that may differ from country to country. No

Designing a Balanced System 

23  

doubt concerned by this judgment, the network of EU competition authorities (ECN) in an unusual resolution of 23 May 2012 explained their position that while civil damages can make a very significant contribution to the enforcement of the competition rules and while such claims and leniency programmes are complementary tools, the protection of leniency material is necessary to ensure the effectiveness of leniency programmes. So, following the open-ended Pfleiderer judgment, the competition authorities do not want to leave any doubt as to the fundamental importance they attach to the protection of leniency material. It is worth noting that only one day earlier, on May 22, the General Court in EnBW Energie Baden-Wurttemberg vs Commission (Case T-344/08), decided on an appeal against a rejection by the Commission of a request for access to documents. The request by a third party was based on Regulation 1049/2001 and concerned most documents, including leniency material, relating to cartel proceedings (the ‘Gas insulated switchgear’ – GIS – decision). The Court held, first, that the Commission was not entitled to deal with a request on the basis of categories, such as a category of documents provided in connection with a leniency application, unless such documents are manifestly covered in their entirety by one of the exceptions of Regulation 1049. As a rule, therefore, each and every document must be examined individually, unless in exceptional cases such individual examination proves particularly burdensome. Secondly, the Court decided that most commercially sensitive information in the documents, including the leniency material, dated from well over five years earlier and did not deserve any protection. Interestingly, three weeks later, on June 14, the Court of Justice decided the case Commission v Editions Odile Jacob (Case C-404/10 P), also dealing with a request by a third party for access to documents on the basis of Regulation 1049/2001. This time it did not concern a cartel (and therefore did not relate to leniency material) but the Commission’s file in a merger control case. The Court held that, in line with the jurisprudence of the General Court, the Commission must provide an explanation as to how access to a document could specifically and actually undermine the interest protected; however, it added that the Commission may base its decisions in that regard on ‘general presumptions which apply to certain categories of documents as similar general considerations are likely to apply for disclosure relating to documents of the same nature’. Such general presumptions are, according to the Court, based on applicable State aid as well as merger control procedures because in both areas there is legislation containing provisions on access to information. This general presumption, which applies irrespective of whether the request for access concerns a procedure that is already closed or still pending, may be rebutted by demonstrating that a specific document is not covered by it or that a higher public interest justifies disclosure. Arguably, even though there is no formal legislation dealing with leniency materials, the same principles could apply here.

24  Tom Ottervanger

6. Binding effect Another topic worth discussing when dealing with private enforcement, and in particular follow-on damage actions, is the binding effect of final Commission decisions. Following the Court’s jurisprudence in Masterfoods (Case C-344/98), Regulation 1/2003 provides in Article 16, which concerns the uniform application of the competition rules, that national courts cannot take decisions that run counter to decisions adopted by the Commission concerning the same infringement. This is interesting because, in a way, it establishes the supremacy of public over private enforcement; of the Commission over national civil courts operating in a very different environment. Claimants in a follow-on action can rely on the (final) Commission decision in actions against the addressees of that decision and ‘only’ have to show damage and causality. National courts are bound by the decision. The addressees of the decision cannot challenge the decision in litigation before the national court on public policy, fair trial or other grounds. Any shortcomings in the decision and in the due process in reaching the decision can therefore only be remedied in appeals to the General Court and the Court of Justice. It is questionable whether this situation is perfect given that the administrative procedure before the Commission and before the appeal courts, whatever the standard of proof, differs greatly from civil litigation. For example with respect to the hearing of witnesses and the possibility to (cross-) examine leniency applicants who testified against the defendant. Moreover, the Union courts when reviewing a Commission decision do not necessarily fully examine all the facts and in practice, for example, do not interrogate leniency applicants to ascertain whether their eagerness to obtain immunity (or a fine reduction) may have led to some exaggeration in their testimony.

7. Collective redress: the Dutch example? Recognizing the different functions of public and private enforcement, and recognizing that provisions for collective redress may be useful in cases where the loss per claimant is small so bundling aids efficiency (even if damages will ultimately be paid by consumers and/or shareholders, as is true for the costs of litigation if the defendant loses), one of the main issues that has arisen from the comments on the 2011 Commission’s Public Consultation on collective redress concerns the choice between ‘opt-out’ and ‘opt-in’. Many commentators reject opt-out class actions whereby the group of claimants consists of all those who have not expressly opted out of the action – as opposed to opt-in actions where the group is limited to those victims that have expressly stated their wish to be included. Opt-out systems are rejected largely because such systems would lead to higher costs, would risk ‘overcompensation’ (if not all potential class members eventually claim damage), would pose problems of

Designing a Balanced System 

25  

representation and would not compensate a successful defendant for the costs of winning a (sometimes frivolous) case brought against it (because most individual claimants are simply not known). There is a danger that EU law, if an opt-out system were introduced, would suffer from the excesses prevalent in the system of US class actions, particularly with relation to its contingency fees, discovery procedures, and multiple damages. It is interesting to note that the opt-out system developed in the Netherlands has been very successful and could serve as a model or source of inspiration for the EU. It was introduced in 2005 by the ‘Act on the collective settlement of Mass Damages’, the provisions of which form part of the Dutch Civil Code and the Code of Civil Procedure. It allows for court-approved mass settlements that bind an entire class of injured parties unless they opt-out within a certain period. Its strength and at the same time its weakness is that it is confined to settlement cases. Most of the objections to collective redress are avoided through a largely consensual process. An association representing an injured class (but which under Dutch law cannot sue for damages) first negotiates and reaches a settlement agreement with the defendant. This agreement is submitted to the Amsterdam Court of Appeal which in turn determines whether the representative body is indeed representative and whether the agreement is reasonable, taking all circumstances into account (such as for example the magnitude of the damage, speed of payment, the costs of litigation in the absence of a settlement etc.). The settlement becomes binding on those individual injured parties that do not opt-out. Five major mass settlements have been approved thus far, including cases where most parties concerned were established in other countries. An extensive study entitled ‘Collective Redress in Antitrust’ and commissioned by the European parliament, published in June 2012, analyses different systems of collective redress and assesses the pros and cons. It strongly recommends the adoption of an antitrust specific measure in the form of a regulation. The report acknowledges that there are no clear reasons to prefer one model to another but prefers an opt-in solution, mainly because opt-out does not seem politically achievable except perhaps in certain limited circumstances (it seems that the study did not take into account the Dutch model, for what it is worth). At the time of writing it is too early to tell how this saga, which started more than a decade ago, will play out.

Scott Campbell and Tristan Feunteun*

Designing a Balanced System: Damages, Deterrence, Leniency and Litigants’ Rights – A Claimant’s Perspective

1. Introductory comments In recent years there has been a drive, both at national and supra-national levels within the EU, to develop a competition enforcement system in which private actions for damages work alongside and in harmony with public enforcement of EU competition rules to the best effect for consumers and for the economy and society as a whole.1 The debate was started by the European Commission publishing its Green and White Papers on damages actions for breaches of EC competition rules, in 2005 and 2008 respectively, and these initiatives have stirred vigorous discussion across the EU on two related issues: first, is it desirable to increase the extent of private damages and other actions for breach of the competition rules and, secondly, how should this be done without leading to ‘US-style excess’? The general response to the first question appears to be cautiously positive: there is recognition that the EU must move forward from the ‘state of total underdevelopment’ of private claims for competition damages identified by the Commission’s report in 2004.2 Regarding the second question, there is also general agreement on the need to avoid ‘US excess’. But there is much less agreement on how best to achieve both these goals. The Commission is now preparing to move to the next stage of this debate, in the form of a legislative response to be proposed during the course of 2013 – as informed by the Collective Redress Consultation that was launched in February 2011.3 At this juncture, it is worth considering, for example, why greater private * Scott Campbell is a Partner, and Tristan Feunteun is a Solicitor, at Stewarts Law LLP in London. All views and any errors or omissions in this chapter are their own. 1 The Commission has taken a number of steps since 2004 to stimulate the debate on that topic. See Green Paper – Damages actions for breach of the EC antitrust rules, SEC(2005) 1732 of 19 December 2005 (the ‘Green Paper’); White Paper on damages actions for breach of the EC antitrust rules, SEC(2008) 404/ SEC(2008) 405/ SEC(2008) 406 of 2 April 2008 (the ‘White Paper’); and Oxera, Quantifying antitrust damages – Towards non-binding guidance for courts, December 2009 (the ‘Oxera Report’). 2 Ashurst, Study on the conditions of claims for damages in case of infringement of EC competition rules, 31 August 2004 (the ‘Ashurst Report’), p 1. 3 Towards a Coherent European Approach to Collective Redress, SEC (2011) 173 (the ‘Collective Redress Consultation Paper’).

28  Scott Campbell and Tristan Feunteun enforcement of EU competition law is thought desirable to complement the enforcement efforts of the public authorities (in particular to outline clearly for all stakeholders the benefits it can bring). What exactly is meant by the much impugned ‘US excess’? Does it exist and, if so, what may be its causes? How might an increased level of private law claims within an improved pan-European system best be brought about?

1.1 Advantages of private damages actions The Commission and other public authorities (the UK Office of Fair Trading, for example4) see increased exercise of the rights of redress of private victims of anticompetitive behaviour5 as a necessary complement to their own public enforcement efforts. There appear essentially to be three reasons for this: 1. the primary aim of private enforcement is to achieve redress for those affected by cartels and other breaches of competition law. In contrast, the aims of public enforcement are to promote the public good and so to condemn conduct which infringes public law by imposing a public law penalty to deter further infringement not only by the acknowledged cartel members, but also by other businesses. Both private and public enforcement seek to promote economic efficiency by improving the functioning of the market. There is a public policy recognition that both aims need to be pursued in a balanced way – and the realisation that this is not happening entirely effectively at present; 2. successful damages claims (especially those on a compensatory basis) increase the deterrent effect of the competition law regime overall. Seeking to ensure that the compensation and fine to be paid negate any benefit the cartelist might have gained from committing the economic tort or breaching the EU competition rules may be equally beneficial for the public good; and so 3. the resources available to private parties can be used as a complement to the (necessarily finite) resources available to public authorities to pursue infringers and maintain market competitiveness. Greater private enforcement should, then, give public competition enforcers greater freedom to prioritise their activity where they think it will do most good. As a general rule, private redress should be independent of and complementary to enforcement by public bodies. Private redress should not impede or otherwise 4 See OFT, Response to the European Commission’s White Paper, Damages actions for breach of the EC antitrust rules (July 2008), OFT 1006. 5 EU law requires the courts of Member States to provide a remedy in damages to victims of infringements Articles 101 and 102 TFEU. See Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297, paragraph 26. This right to damages includes compensation for lost profits. See Case C-295-298/04 Manfredi v Lloyd Adriatico Assicurazioni SpA [2006] ECR I-6619, para 95.

Designing a Balanced System – A Claimant’s Perspective 

29  

jeopardise enforcement by public bodies, and the pursuit of private redress must necessarily (by virtue of the secretive nature of cartels) be informed by – and be subordinate and subsequent to – public enforcement.

1.2 Collective redress Fully effective access to redress for cartel victims is only likely to be possible through some form of collective action. Although some victims will be able to pursue claims on an individual basis, the large majority of them will only be able to do so if the inevitable risks are spread over a significant number of claimants. The introduction of mechanisms of collective redress is of ‘priceless’ value to the enforcement of EU law. The Collective Redress Consultation Paper correctly defines collective redress as the ability to take action when there is harm caused by an infringement of any EU legislation creating substantive rights.6 The key to collective redress is the creation of effective mechanisms for injured parties to access judicial systems in order to assert rightful claims for violation of EU law. This condition exists throughout a number of the legal disciplines: competition, mass tort, financial services, consumer law and environmental law. The unifying factor in all of these matters is the presence of a wrong committed by an individual, company or a group of individuals or companies causing similar injuries to similarly situated individuals arising out of a common course of misconduct.

1.3 Is there ‘US excess’? Notwithstanding the recognition within the EU that breaches of competition law often require a collective form of redress, the lingering European perception of the US class action system seems to be that large, feral ‘classes’ of claimants roam the United States seeking out unfortunate companies to hold to ransom. It is therefore perhaps worth recalling some of the features of the US system: 1. damages awards are set by jury in antitrust cases: this means that they can be unpredictable and on occasion have exceeded even the plaintiffs’ damages claims where the jury feels that the defendants’ behaviour has been particularly egregious; 2. at the federal level and in most state courts, the award handed down in antitrust cases by the jury after trial is automatically trebled – the courts have no discretion in this regard, a situation which strongly contrasts with that of the European courts; 3. although the US uses an ‘opt-out’ class system for civil damages claims in the antitrust field, the scope of the class is certified by the court (now, antitrust 6

Collective Redress Consultation Paper, cited above note 4, at 4.

30  Scott Campbell and Tristan Feunteun cases are almost always adjudicated at the federal level – thereby promoting consistency across the US) because it is in the interests of justice for the case to be brought on behalf of that class; and 4. the vast majority of cases in the US lead to settlements short of trial: for early settlement, the amounts paid are usually based on some estimate of the total actual damage caused by the cartelists to the class – rather than the treble damages that would be awarded if the case came to trial. Such settlements have to be approved by the court following scrutiny. The US system is, then, more subject to judicial control than is generally recognised. However, it may be the combination of features in the US system – in particular the unpredictability of jury awards and the automatic trebling of damages awarded – that elicits criticism from Europeans and that leads to the concerns expressed about elements of the US system being adopted in Europe. The ‘opt-out’ class device of itself does not lead to the issues which commentators now recognise are being addressed in the US system. Other countries – notably Canada and Australia – have introduced opt-out classes without attracting the opprobrium reserved for the US system, and it can be suggested that this is precisely because they do not have the additional features of automatic treble (punitive) damages, and juries are generally not involved with awarding them. And yet in both jurisdictions the opt-out class has proven to be a highly effective way of ensuring that victims of cartels (in particular) are able to spread the inevitable risks of litigation so as to obtain redress on a more level playing field against large and well-resourced defendants.

2. An emerging European alternative: Government sector-led efforts 2.1 Collective redress 2.1.1 EU level On 5 October 2010, the European Commissioners for Justice, Competition and Consumer Policy took stock of the Commission’s various approaches to collective redress mechanisms to date and foreshadowed the Collective Redress Consultation with a number of reflections on the best way forward for EU-wide collective redress by way of a Joint Information Note.7 The consultation document itself relating to the Collective Redress Consultation further reflected the Commission’s views on the matter. 7 Towards a Coherent European Approach to Collective Redress: Next Steps, SEC(2010) 1192 (the ‘Joint Information Note’).

Designing a Balanced System – A Claimant’s Perspective 

31  

The Joint Information Note reaffirmed the Commission’s view that collective redress is an important instrument to strengthen the enforcement of EU law within both the context of today’s enlarged (and enlarging) EU and the greater extraterritorial ambit of EU law. However, the Joint Information Note correctly observes that existing forms of collective redress in the EU designed to compensate groups of victims harmed by illegal business practices vary widely throughout the EU. Stakeholders that had responded to the various consultations on the Commission’s Green and White Papers as well as DG SANCO’s Green Paper8 warned against an inconsistency between the different Commission initiatives on collective redress and pleaded for the development of more coherent approach. Therefore, the three Commissioners’ objective was stated to be to ensure from the outset that any future proposal in this field, while serving the purpose of ensuring a more effective enforcement of EU law, fits well into the EU legal tradition and into the set of procedural remedies already available for the enforcement of EU law9.

The Commissioners went on to set out a series of core principles to guide EU initiatives on collective redress: 1. any initiative on compensatory collective redress should first and foremost ensure that any right of injured parties to compensation can be effectively and efficiently vindicated; 2. parties should have the possibility to resort to a collective consensual resolution of their dispute; 3. the rules on European civil and procedural law should work efficiently for collective actions, and judgments should be enforceable throughout the EU;10 and 4. adequate means of financing should be available to allow citizens and businesses to have access to justice. The Commission’s Joint Information Note also emphasises the importance of avoiding the risk of abusive litigation, citing the US class action model as an example of a system giving rise to abuse. Therefore, effective safeguards should be put in place to curtail any economic incentives giving rise to the bringing of abusive claims. The Joint Information Note highlights the ‘loser pays’ principle11 as a potential safeguard against abusive claims. These principles were put out to consultation under the Collective Redress Consultation. The feedback received was collated in the Report entitled Towards a Coherent European Approach to Collective Redress, which was published on 12 January 2012. This Report advises that the unsuccessful party should bear the costs in order to avoid the proliferation DG SANCO, Green Paper On Consumer Collective Redress COM(2008) 794. Collective Redress Consultation Paper, cited above note 4, at 5. Efforts, notably the abolition of exequatur under the 2010 Commission Proposal on Reform of the Brussels I Regulation, should yet help regarding this latter point. 11 The ‘loser pays’ principle is a common feature of the English legal system under which the ultimately unsuccessful party in litigation stands liable to bear both its own costs and the successful opponent’s costs (including court fees, lawyers’ fees and disbursements). 8 9

10

32  Scott Campbell and Tristan Feunteun of unmeritorious claims. However, the Report does note that rules on costs are a matter for Member States themselves. While claimants – at the time of writing – await the Commission’s eventual conclusions as to the appropriate form(s) of collective redress that could fit into the EU legal system and into the legal orders of the 27 EU Member States, there has been increasing activity – both in the courts of Member States and by way of private settlements among cartelists and their victims – involving parties to disputes adapting existing mechanisms and devising new approaches to resolving their disputes, often on a collective, cost-effective and EU-wide basis. Outlined below are a series of examples of private solutions to competition law disputes, which already adhere to each of the Commission’s core principles relating to collective redress.

2.1.2 UK level In the UK, the Department for Business, Information and Skills (BIS) launched in April 2012 a consultation on reforming private actions relating to competition law infringements.12 This contained several key proposals, including both an opt-out regime for collective actions, but also a rebuttable presumption regarding cartel overcharges. The latter will be considered in section 2.2.2 below. BIS’ idea for an opt-out collective redress mechanism is most likely to be suited for individual consumers and specially-designated representative bodies, such as consumer associations. It may also be applied to small and medium-sized enterprises (SMEs). There is concern over the current dearth of UK mechanisms (that are proportionate, balanced and cost-effective) that are available for individual consumers and SMEs to seek redress for cartel infringements. It is likely that large corporations will not be subject to such an opt-out regime, and will be able to continue to pursue their own cartel damages actions according to their own business strategies – as evidenced by the increasing number of such cases before the High Court and Competition Appeal Tribunal.

2.2 Drafting of Commission decisions; passing on defences and overcharge presumptions 2.2.1 EU level While claimants in private damages actions can clearly rely on the Commission’s infringement decisions in the context of ‘follow-on’ damages claims,13 12 Department for Business, Innovation and Skills, ‘Private Actions in Competition Law: a Consultation on Options for Reform’ (April 2012), https://www.gov.uk/government/uploads/system/ uploads/attachment_data/file/31528/12-742-private-actions-in-competition-law-consultation.pdf. 13 The following statement appears at the foot of the Commission’s press releases relating to infringement decisions: ‘[a]ny person or firm affected by anticompetitive behaviour as described in

Designing a Balanced System – A Claimant’s Perspective 

33  

infringement decisions do not, apart from a few rare and tangential examples, render a finding as to the effects of the anticompetitive behaviour on downstream market participants. Therefore, while the infringement decision can be used as a sword by a prospective damages claimant, it can equally be used as a shield by the cartelist defendant inasmuch as the defendant may state, ‘the Commission’s decision made no findings as to the cartel having had any effect’ – the often deployed ‘zero overcharge’ argument. Within this context, it is encouraging that the Oxera Report has assembled a tool kit of possible economic analyses that can be applied in assessments of the effects of anticompetitive behaviour.14 This tool kit was largely transposed into the Commission’s 2011 consultation paper on quantifying cartel overcharges and other anticompetitive effects.15 It is understood that the outcome of this process will be a non-binding guidance paper issued to the courts of Members States. This is a very useful and worthwhile exercise. However, on its own, economic guidance for courts is not enough to combat the ‘zero overcharge’ argument which faces claimants (and generates the need for costly and time consuming economic analyses for both claimants and defendants). A statement in the Commission’s infringement decision laying down a presumption that the infringement generated anticompetitive effects on the downstream market would be a most valuable contribution to private enforcement of competition law.

2.2.2 UK level In the UK, BIS’s idea of a rebuttable presumption of a 20% overcharge was intended to help prospective claimants, both by enabling them to estimate the likely benefits of seeking redress, and perhaps more importantly by shifting the burden of proof to defendants. One example of a jurisdiction with such a rebuttable presumption is Hungary, where a presumption of an overcharge of 10% for cartel claims was introduced in 2008.16 However, doubts regarding the merits of a one-size-fits-all approach were voiced in response to the consultation,17 and it is likely that the proposal for such a rebuttable presumption will be dropped. this case may bring the matter before the courts of the Member States and seek damages. The case law of the Court and Council Regulation 1/2003 both confirm that in cases before national courts, a Commission decision is binding proof that the behaviour took place and was illegal. Even though the Commission has fined the companies concerned, damages may be awarded without these being reduced on account of the Commission fine. The Commission considers that meritorious claims for damages should be aimed at compensating, in a fair way, the victims of an infringement for the harm done.’ Cited above note 1. Draft Guidance Paper on quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty on the Functioning of the European Union (17 June 2011). 16 Section 88/C of the Hungarian Competition Act, adopted on 3 June 2008. 17 Oxera, ‘Presuming Too Much? The UK Consultation on Private Actions in Competition Law’, Agenda (August 2012), http://www.oxera.com/Oxera/media/Oxera/downloads/Agenda/Competitionlaw-private-actions.pdf?ext=.pdf. 14 15

34  Scott Campbell and Tristan Feunteun

3. An emerging European alternative: Regulator- and court-led efforts 3.1 Leniency programmes and disclosure: A balancing act between the Commission and courts 3.1.1 Public policy background Apart from the still evolving European approach to collective redress which will, it is anticipated by the Commission, require legislative activity later in 2013 to become enacted, there are examples of areas where the Commission might clarify or modify its current practice to the benefit of claimants in private damages actions. Indeed, in a resolution on 2 February 2012, the European Parliament called for the Commission to examine thoroughly the appropriate legal basis for any measure aiming to assist collective redress taken in the wake of the consultation. Some suggested approaches are proffered below. The consensus that the leniency programmes, not only of the Commission18 but also of national competition authorities, are essential to public cartel enforcement and should be protected is surely right. It must therefore be right to exclude the compulsory disclosure of documents created solely for the purpose of a leniency application, whether those documents are held by the authority or by private parties. Similar considerations would apply to documents created solely for the purpose of the Commission’s settlement procedure.19 However, it is at all times important to bear in mind that, despite the co-operation which the leniency applicant has given to the public enforcement process, the cartel of which it may have been a key member was still responsible for serious and unlawful economic damage to the victims. Therefore, it would be helpful if the limits of this protection could be made more explicit. In particular a clear statement of the Commission that: the protection does not cover pre-existing documents held by the defendant cartelists at all but only those created for the purpose of the leniency application or settlement proceedings; and nothing prevents voluntary disclosure by the creator of the leniency document or settlement statement after the authority’s investigation has ended (and the Statement of Objections or similar proposed decision issued) if they wish to do so in the context of settlement discussions with private claimants. It is understood that, in the wake of the recent case law outlined at section 3.1.2 18 Commission Notice on Immunity from fines and reduction of fines in cartel cases, 2006 OJ C298/11. 19 Commission Regulation (EC) No 622/2008 of 30 June 2008 amending Regulation (EC) No 773/2004, as regards the conduct of settlement procedures in cartel cases, 2008 OJ L171/3; Commission Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel cases, 2008 OJ C167/1.

Designing a Balanced System – A Claimant’s Perspective 

35  

below – in particular the Pfleiderer case, that the Commission will propose legislation over the course of 2013 to clarify its position. Disclosure between the parties should be the principal method of obtaining information to support a competition law claim. Disclosure by the Commission and any other national competition authority should be an exception rather than the rule. Nevertheless, there may be circumstances where a leniency applicant or other cartelist defendants may wish to disclose documents – for instance, the confidential version of a Commission infringement decision – voluntarily to claimants or their advisers in the context of discussions surrounding the settlement of claims for compensation. It may be unhelpful if, in this context, the Commission (or indeed any other public enforcement body) were to object to this, for example, on the grounds that disclosure for this purpose might be against the public interest. There are likely to be strong efficiency-based arguments to enable private actions for redress to proceed in parallel with any public enforcement action where possible. In claimants’ experience, voluntary evidence disclosure by leniency applicants and other defendants is often the key to ensuring that settlement negotiations can progress properly and be appropriately supervised by the courts.

3.1.2 Recent case law In light of all this, there have been several recent cases regarding access to leniency that ought to provide some encouragement to cartel damages claimants. Pfleiderer AG, a victim of the German decor paper cartel, sought access to the leniency application made by a cartelist to the Bundeskartellamt in order to aid its pursuance of a cartel damages claim. The Administrative Court in Bonn referred the question to the European Court of Justice, which rejected a general prohibition on disclosure of leniency applications. Instead, the ECJ held that, given that Member States enact and apply such cartel leniency procedures, it is for the national courts to balance the public interest of uncovering and effectively prosecuting cartels with the need to facilitate a third party’s interest in disclosure in order to exercise its right to compensation. The ECJ emphasised that: (i) the national court must have regard to equivalence/effectiveness; and (ii) before ordering disclosure (inspection) by the third party, the national court must ask if there are other sources of information that are equally effective. Following the ECJ’s decision in June 2011, the Bonn court in January 2012 denied Pfleiderer access to the file, citing that disclosure would have compromised the purpose of the investigation. The Pfleiderer judgment was taken into account by the English courts in National Grid in April 2012, where the High Court permitted limited disclosure of both the Commission’s and NCA’s files. But Pfleiderer has been widely criticised, and the Commission has sought to clarify the interaction of public and private enforcement at both EU and national levels – most notably making written

36  Scott Campbell and Tristan Feunteun submissions to the High Court (published on the Commission’s website), on Mr Justice Roth’s invitation, during the National Grid proceedings. A more encouraging judgment was given by the European General Court in the CDC Hydrogen Peroxide judgment in December 2011, which annulled the Commission’s refusal to grant a damages claimant access to the contents list of the Commission’s file regarding the hydrogen peroxide cartel.20 Indeed, in that case the General Court noted that leniency programmes are not the sole way of ensuring compliance with EU competition law, and that damages actions before the courts of Member States can make a significant contribution to that objective.

4. An emerging European alternative: Private sector-led Efforts 4.1 Bundling of multiple claims: Examples from Germany and the Netherlands As it stands, in some other EU jurisdictions it is possible to assign multiple claims to a special purpose vehicle that is entitled to bring damages proceedings on behalf of its assignors. In circumstances where this is possible, it potentially provides a means of bundling together multiple claims under the control of one claimant. This may help circumvent the traditional ‘opt-in’ character of litigation in most European jurisdictions. Where the cause of action is common to the bundled claims, redress may be sought collectively. One example comes from Germany. The applicant, Cartel Damages Claims SA (‘CDC SA’), a Belgian corporation, purchased damages claims from twenty-eight customers of participants in a cement cartel (a further eight claims have since been added), which was fined by the German Federal Cartel Office in 2003. In 2005, CDC brought an action for damages incurred between 1993 and 2002 (now amounting to € 176 million) against the six participants in the cartel before the Regional Court of Düsseldorf. The defendants tried to obtain a suspension of the damage proceedings until the end of their appeals against the decision of the Bundeskartellamt. However, under a court order of 9 March 2006 the Regional Court of Düsseldorf rejected the requests for suspension. The defendants’ appeals against this order have were also rejected by the Higher Regional Court of Düsseldorf by a court order of 3 May 2006.21 The defendants then sought to contest the admissibility of the action by arguing that, first, the class was insufficiently precise regarding both membership 20 21

Case T-437/08, CDC Hydrogene Peroxide, [2011] ECR II-8251. OLG Düsseldorf, Ref. No. VI-W (Kart) 6/06.

Designing a Balanced System – A Claimant’s Perspective 

37  

and exact amount of damages claimed, and second, that the assignment of the underlying claims to CDC SA was a sham arrangement which violated both German civil procedure and German rules regarding the provision of legal services. However, by interlocutory judgment of 21 February 2007, the Regional Court of Düsseldorf confirmed the admissibility of the action. On 14 May 2008, the First Cartel Senate of the Higher Regional Court of Düsseldorf upheld this decision.22 The latter decision was then contested by Dyckerhoff AG (one of the defendants) before the Federal Court of Justice. On 7 April 2009, the German Federal Court of Justice dismissed this appeal and confirmed that CDC SA’s damages action is admissible.23 The Regional Court of Düsseldorf proceeded with deciding on the merits of the case. However, the case has since become mired in interlocutory appeals on matters other than CDC SA’s assignment-based business model – for example, as regards jurisdiction and, at the latest hearing of the Regional Court of Düsseldorf on 17 January 2013, as regards the period of limitations. Nevertheless, this series of judgments is particularly noteworthy as it may pave the way for future actions by CDC SA or other similar special purpose vehicles in Germany, or at least Düsseldorf.24 Indeed, only a few days after the 2009 German Federal Court of Justice decision, CDC SA brought a second claim, for € 600 million excluding interest, on behalf of 32 claimants affected by the hydrogen peroxide cartel. Similar claim mechanisms are also being used in the Netherlands in damages claims relating to the air cargo and elevator and escalators cartels. Under such mechanisms, a number of victims of these cartels have assigned their claims for damages to a stichting, ie, a Dutch special purpose vehicle, to pursue their claims in the Dutch courts.

4.2 Market-wide private settlement funds: Examples from the air passenger and marine hose cartels A number of novel mechanisms for the distribution of damages have arisen from private settlements of claims based on breaches of competition law. For example, in 2007 a £ 73.5 million settlement was reached on behalf of UK air passengers in the price-fixing class action lawsuit brought against British Airways and Virgin Atlantic in the US. Under the terms of the settlement, passengers who purchased OLG Düsseldorf, Case No. VI U (Kart) 14/07. BGH, Case No. KZR 42/08. 24 One other development that will cement Germany’s position as one of the leading European jurisdictions in which to bring cartel damages actions – after (in the authors’ opinion) England and then the Netherlands – is the ability of indirect purchasers to bring actions. This was made possible by the German Federal Supreme Court on 29 June 2011, when it ruled that cartel members could raise the ‘pass on’ defence; the rights of indirect purchasers to damages are expected to be put on a statutory footing early in 2013. 22 23

38  Scott Campbell and Tristan Feunteun tickets from either airline for specified long-haul flights between 11 August 2004 and 23 March 2006 are able to claim a partial refund. Passengers who purchased tickets from the airlines in the UK are entitled to a refund of 33.3 percent of the fuel surcharge they paid on each flight segment purchased. Passengers who purchased tickets in the US will receive their partial refund (also 33.3 percent of the fuel surcharge paid on each flight segment) in dollars rather than pounds out of a separate $73 million US settlement fund. This unprecedented settlement represents the first instance in which such a case has been resolved simultaneously under English law and US law and has provided compensation to classes of purchasers from both the US and the UK. More recently, Parker ITR S.r.l. (‘Parker’) has made a global offer to settle claims arising from the marine hose cartel to all purchasers of marine hose on the basis of settlement terms. The offer to settle applies not only to all purchasers of marine hose from Parker but also to all other purchasers of marine hose from Parker’s co-cartelists or companies that operated the Parker business prior to 31 January 2002. However, the settlement terms differ depending on whether purchases were from Parker or not. In return for giving up rights to litigate against Parker, purchasers can claim against the settlement fund. An independent expert assisted by an independent economist will determine how much of the fund goes to each claimant but there is a presumption that direct Parker purchasers will be entitled to 16 per cent of purchases during the settlement period unless they passed on the loss. Purchasers settling with Parker will also benefit from Parker’s cooperation in the form of production of witnesses, interviews, depositions and documents in relation to proceedings against other cartelists and a guarantee against payment of adverse costs in civil damages proceedings against those other cartelists.

5. Concluding remarks There is a clear need for action at the European level to assist and support greater take-up of private civil damages claims against cartels. It is hoped that the Commission will continue to ensure that, within the broad framework of encouraging more businesses and consumers to claim redress collectively where they are the victims of cartels, complementary Member State initiatives in this area are also encouraged and undertaken. Europe now has an excellent opportunity to create an effective redress system which both safeguards the rights of defendants and encourages greater private enforcement of competition law. Nevertheless, while the settled position of the Commission, courts and the various stakeholders in the collective redress process is gradually arrived at, claimants around the EU are developing and utilising innovative means of seeking collective redress based on the existing legal frameworks available to them. These

Designing a Balanced System – A Claimant’s Perspective 

39  

examples of innovation are likely to continue at least until such time as any draft Directive gets transposed by Member States. Indeed, such examples may not only form good route-maps for those drafting the Directive as to how best to approach the collective redress agenda, but may well continue as a complement to any such Directive – as one form of the private enforcement that complements the EU’s vigorous regulatory enforcement regimes.

Donald I Baker1

Trying to Use Criminal Law and Incarceration to Punish Participants and Deter Cartels Raises Some Broad Political and Social Questions in Europe

Even though we have gradually achieved a broad global consensus that naked cartels are bad, we have no corresponding consensus on how bad they are or exactly what are the best ways to punish and deter them. Of course, the general consensus on cartels as evil has been largely generated within the international community of antitrust enforcers and their political allies, while questions about legal remedies and criminal punishments confront broader social and legal assumptions and therefore any major changes tend to require broader political support in most democratic societies. Recognizing this reality, I am going to try to look at the punishment and deterrence situation in Europe through the eyes of a friendly American who generally believes that the modern Unites States (US) efforts to use criminal law as a device to punish conspirators and deter would-be cartel participants has been quite successful. One sees some apparent confirmation of this hopeful premise in the International Air Carriers Fuel Surcharge cases—where the big US carriers have essentially escaped prosecution in the US, the European Union (EU), and the rest of the world, while their foreign competitors have been prosecuted and heavily fined by many antitrust authorities. At the very least, the vigorous criminal enforcement of the Sherman Act by the US Department of Justice (DOJ) has clearly generated a heightened awareness among US executives that they face the serious personal risk being detected and then imprisoned if found to be engaged in even run-of-the-mill cartel activities. The apparent success of the vigorous US efforts naturally causes thoughtful observers to ask: would criminalization be an effective way to deter and punish cartel conspirators in other countries? However, when we ask the question this way, we may be asking what is at best a subsidiary question so far as many European Member States are concerned. Thus we may miss that the more fundamental underlying question about effective deterrence of cartel activities, rather than a particular method for trying to enhance it. In European terms, the question really is: does the imposition of very heavy fines only on corporate 1 Senior Partner, Baker & Miller PLLC, Washington, DC, and Adjunct Professor of Law at The George Washington University Law School. Mr. Baker acknowledges with special thanks the extensive efforts of Edward A Jesson, at that time a J.D. candidate of the University of Miami School of Law, in helping to produce this paper.

42  Donald I Baker shareholders by the European Commission and many National Competition Authorities adequately curb the incentives and temptations of some corporate officers and employees to engage in profitable cartel activities? If the answer is ‘no’, then what potential alternatives might be available to raise the stakes for would-be wrongdoers? This is a serious question that deserves a lot of further thought. I admit to being skeptical about the adequacy of corporation-only remedies, however stringent, as the sole deterrent of activities that individuals may find personally profitable. Rather, I believe that much cartel activity is engaged in by individuals who hope for bigger bonuses, promotions or other benefits from making a business unit or enterprise look more profitable than it would otherwise be. Often the shareholders (and sometimes the senior management) are not privy to such covert efforts – and the wrongdoers generally do not worry about the shareholders’ interests when they decide to undertake such activities. To be clear, my focus in this paper concerns the use of criminal sanctions against individuals who participate in cartel activities. In some jurisdictions, the legislature has imposed criminal sanctions against both individuals and enterprises (eg, the US and Ireland), while in others the criminal prohibitions only apply to individuals (eg, the United Kingdom (UK)). I do not regard that choice as especially significant so far as enterprises are concerned because under either system the offending enterprise is subject to the same type of monetary sanctions, albeit with a higher burden of proof being required if it is a criminal offense. I am not arguing against imposing serious corporate fines for cartel violations. These are necessary to give the enterprise a serious stake in educating, supervising, and monitoring employee performance. I just do not believe such fines alone can provide a complete answer to the ongoing cartel problem. Thus, for me, the fundamental question for Europe becomes: how can we make conspiracy seem a more risky course for the kinds of opportunistic individuals who go on engaging in these activities? For them, how can we get their attention and raise their perceived risks of being discovered and personally penalized for continuing to do what they may have been doing or years? The American experience is of course relevant for Europe, but the American approach need not (and probably should not) be treated as the only available avenue open to antitrust enforcers and national parliaments seeking to raise the stakes for cartel culprits. To be effective, a legal mandate will need to be reasonably related to the culture of a society and its legal institutions. Moreover, any remedy must be regularly and visibly enforced in order for it to be perceived as a personally painful risk for individual employees and executives in an industry.

Criminalising Cartels in Europe  43  

A. Politics and public psychologies are important but are often different Competition is frequently not a popular political cause among many elements in a democratic society. The introduction of increased competition can disrupt the existing interests of enterprises and their employees, while the consumers and innovators who benefit from competition are often not overly aware that open markets are the source of said benefits. United States. In the US, the political dynamics are different from elsewhere because most important antitrust law has resulted from our periodic bouts of populist outrage at the perceived economic villains of the day, rather than being the product of careful study by some learned commission. Thus American antitrust has always seemed to have a strong moral dimension less apparent elsewhere. Populist frustrations generated the Sherman and Clayton Acts as prominent national landmarks, which represent political responses to perceived abuses of Standard Oil, the other ‘trusts’, and other high-visibility villains. That type of populist-driven antitrust reappeared in the 1974 upgrade of Sherman Act violations to million dollar corporate felonies (also punishable by three year jail terms for individuals) and in recurring Congressional efforts to pressure the DOJ to bring a case against OPEC. Populism reappeared in 2004, albeit in a more muted form, when the maximum corporate and individual fines for Sherman Act violations were increased to $100 million and $1 million respectively, while the maximum jail sentence was increased to 10 years. Thus, for a majority of Americans, imbued with the ‘robber barons’ lore, pricefixing is accepted as being a form of covert theft—and thus, sending antitrust conspirators to jail, like other white collar thieves, is seen as the appropriate remedy. Moreover, American jurors would be more likely than many Europeans, or Asians for that matter, to assume that a conspiracy was inherently improper and hence the only relevant question in a trial is whether any particular defendant actually conspired with others. Even with this generally favorable political and public psychology, US society took 70–80 years to broadly accept regular imprisonments as the normal remedy for convicted Sherman Act felons. Europe. Competition has been a high level public goal for over half a century since the Treaty of Rome was adopted, because creating an integrated common market has been seen as a rational way of reducing the risk of future war and promoting economic progress by breaking down national frontiers and other historic barriers to integration and efficiency. But competition has generally been a top-down goal – with well-educated politicians and senior government officials as the most visible advocates of antitrust and other pro-competitive policies (eg, prohibitions on state aids to favored industries). At the same time, cartel activities (long assumed to be a way of life) were not an early enforcement priority of the then newly-created European Commission in the 1960s and 1970s, as the agency tended to focus on vertical territorial restraints and abuses of market dominance.

44  Donald I Baker Even though antitrust enforcement against cartels has gained priority and prominence during the last two decades, cartels nonetheless have been lumped in with other antitrust violations by the public and the politicians who view such offenses as a ‘regulatory’ problem subject to regulatory remedies (ie, fines and prohibition orders against the participating enterprises). Nor does there seem to be much of a sense that smaller businessmen who fix prices in local markets deserve to be treated as serious criminal targets, as they now are in the US.2 Thus, there does not yet seem to be the kind of broad basis of political support (or acceptance) that is probably necessary to generate support for criminalization on any broad scale in Europe.

B. The european experience with antitrust criminalization The obvious starting point for a discussion is that the European Commission’s antitrust jurisdiction under the Treaty extends only to ‘undertakings’ and there is no EU competence in the area of criminal law. Hence any effort to deal with the problem of individual antitrust conspirators must be done at the Member State level, regardless of whether criminalization or other legal remedies are employed. This jurisdictional reality necessarily dictates that any criminal enforcement against individual participants in a major international cartel would require close cooperation and coordination among the Commission, a national competition authority (‘NCA’), and the criminal prosecutors in the relevant Member State. At the very least, this jurisdictional split complicates the amnesty process that is so central to modern anti-cartel enforcement. To date, the European efforts to create criminal laws that deal with cartel participants have been largely confined to two Member States with a common law tradition – the UK and Ireland – even though there has been considerable discussion of the issue in a number of other Member States. Moreover, the enforcement history remains quite limited, especially in the UK. The United Kingdom. In 2002, the UK enacted the Enterprise Act that included a criminal prohibition for individuals participating in a cartel, while retaining an administrative system for prosecuting enterprises.3 This statute included substantial criminal sentences for individual violators – including an unlimited See, eg US v Foley, 598 F.2d 1323 (4th Cir. 1979) (price fixing by local real estate brokers). That the UK adopted its first criminal antitrust statue so recently may help to explain why so much public debate and controversy has been generated in the UK by the efforts (of the Home Office and the US authorities) to extradite a British citizen and resident in a DOJ cartel investigation. This particular case has gone all the way to the House of Lords (the UK’s highest court), which denied extradition based on the lack of a parallel antitrust criminal statute in the UK at the relevant time (although the government argued that conspiracy to defraud had long been a common law crime). Norris v Government of the United States of America [2008] 2 WLR 673. (HL). Ultimately, Ian Norris was extradited to the US for obstruction of justice, based on document destruction, rather than for antitrust misdeeds, where the US Government was ultimately unsuccessful before the British courts. 2 3

Criminalising Cartels in Europe  45   maximum fine and maximum jail sentences of five years. The basic criminal prohibition provided ‘[a]n individual is guilty of an offence if he dishonestly agrees with one or more other persons to make or implement, or to cause to be made or implemented, arrangements … relating to at least two undertakings …’, involving various cartel-like purposes (price-fixing, bid-rigging, supplylimitation, customer allocation, or product-market division).4 Thus, the prosecutor must prove beyond a reasonable doubt that a defendant had behaved ‘dishonestly’ in engaging in the prohibited activity. The ‘dishonesty’ standard provided in Section 188(1) of the Act was seen from the outset as being a serious hurdle to prosecutors and has thus far proven to be the case. Even before the Enterprise Act was passed the Department of Trade and Industry (today the Department for Business Innovation and Skill) emphasized the importance of effective deterrence in its modestly-titled White Paper ‘Productivity & Enterprise: A World Class Competition Regime.’5 The White Paper stated that the UK criminal offence must be ‘ … actively applied so that its deterrent effect is genuinely felt …’. The White Paper also discussed reasons why fines do not provide an effective deterrent against cartels and why, therefore, individual criminal sanctions needed to be authorized against individual cartel members.6 Meanwhile, a report prepared for the Office of Fair Trading (‘OFT’) in anticipation of the Enterprise Act (the so-called, ‘Penrose Report’)7 went as far as to predict that there would be in the range of six to ten prosecutions a year as a result of the new Act. Thanks to the dishonesty standard and related difficulties in prosecuting cartel violations, this optimistic prediction has come nowhere close to fruition (thus with the inevitable loss of the hoped-for deterrence effect of the Act). The controlling dishonesty standard was announced in 1982 by The Court of Appeal (Criminal Division) in Regina v Ghosh.8 It provided that, for an individual to be convicted under the dishonesty standard, a jury must determine that a twopart objective/subjective test was satisfied. That test is: Enterprise Act 2002 – Section 188(1) & (2). Department of Trade and Industry, Productivity & Enterprise: A World Class Competition Regime (CM 5233, 2001). 6 The paper provides several reasons as to why the fines at their 2001 level were not high enough to provide an effective deterrent against cartels. ‘If fines are to deter firms and their executives effectively, they need to be set at a level which is greater than the expected gains from participating in a cartel. US evidence shows that cartels often raise prices by around 10%. Increasing prices will have some dampening effect on demand so a cartelist might increase its profits by a smaller proportion. Conservatively, they might do so by around 5%. If the cartel operates for six years (as the average US cartel is thought to do), then the total benefit might be 30% of annual turnover. The Competition Act 1998 allows fines to be imposed at this level, up to 10% of annual turnover in the relevant market for a maximum of three years. But not all cartels will be caught: in the US, estimates suggest that only a sixth of cartels are detected. The expected fine for the would-be cartelist is the probability of being caught multiplied by the likely fine (ie a sixth of 30%), around 5% of annual turnover. Faced with very high likely benefits arising from engaging in a cartel the expected fine is unlikely to act as a meaningful deterrent.’ 7 Anthony Hammond and Roy Penrose, Proposed criminalisation of cartels in the UK, OFT365 (November 2001), http://www.oft.gov.uk/shared_oft/reports/comp_policy/oft365.pdf. 8 Regina v Ghosh [1982] 2 ALL ER 689. 4 5

46  Donald I Baker 1 … . according to the ordinary standards of reasonable and honest people what was done was dishonest. [and] 2. … the defendant himself must have realised that what he was doing was by those standards dishonest.9

The picture was further complicated under the Enterprise Act by the House of Lords decision in the famous 2008 extradition case, Norris v Government of the United States of America10 where Their Lordships held that proof of secret price fixing alone was not in and of itself proof of a ‘dishonest’ agreement unless it was accompanied by aggravating features other than simply concealing the agreement. ‘[T]here are problems with the notion that mere secrecy can of itself render the price-fixing agreement criminal. It is not as if secrecy is always necessary for a price-fixing agreement to be effective, or that it is the secrecy which causes the purchaser loss … it must be the alleged dishonesty that causes the loss.’11 The first criminal guilty pleas under the Enterprise Act came in the somewhat unusual Marine Hoses case in 2008.12 This alleged cartel involved bid rigging for the supply of specially manufactured rubber hoses that were to be used in off shore drilling projects around the world. The three named defendants (Peter Whittle, Bryan Allison, and David Brammar) were arrested in the United States and, when faced with lengthy custodial sentences in the US, agreed to unique plea agreements with the US DOJ.13 The terms of the plea agreements were that each defendant’s agreed term of US imprisonment would be reduced by one day for each day of imprisonment imposed upon him following his conviction in the UK under the Enterprise Act. Put otherwise, so long as each defendant pled guilty in the UK and was sentenced for as long as he would have been sentenced in the US, the defendants would not have to spend another day in a US prison. On June 10, 2008, the defendants appeared before Southwark Crown Court, where two defendants were sentenced to 3 years each, and the third to 30 months, even though the DOJ plea agreements had provided for slightly shorter sentences for all three defendants. The Court of Appeal (Criminal Division), apparently influenced by the DOJ plea agreements, reduced these enhanced sentences and reverted to the original, shorter, sentences in the DOJ plea agreement. Had the DOJ agreement not been in place, would the Court of Appeal have reduced the sentences even further, or even implemented a non-custodial sentence? In any event, the Marine Hoses case, with its unique origins, represents the only custodial sentences so far imposed under the Enterprise Act. In April 2010, the UK began its first trial under the new criminal prohibitions against four British Airways Executives accused of participating in the Ibid at 696. Norris v Government of the United States of America and Others [2008] UKHL 16. 11 Ibid at 60. 12 Regina v Whittle [2008] EWCA Crim 2560. 13 Donald I Baker, ‘The UK marine hoses prosecutions: beachhead or aberration?’ Global Competition Review, August/September 2008. 9

10

Criminalising Cartels in Europe  47   international Air Freight Fuel Surcharges cartel. A month later the embarrassing result was a high-visibility acquittal because the prosecutors had been unable to timely produce some late-discovered documents. The Chairman of the Office of Fair Trading almost immediately responded that ‘Lessons have already been learnt’ and emphasized that, [T]he OFT does not regret bringing these proceedings. They have sent an important signal to business of the risks of engaging in cartel activity and hopefully have helped practitioners to convince their clients that the risks of criminal charges in hardcore cartel cases are real, not imaginary. Research indicates the very significant deterrent effect of the prospect of criminal sanctions.14

Taking a less positive view, a former Commission cartel enforcement official (who is English) argued that, ‘The [British Airways] trial exposed a deep fault line running through the whole [UK] criminalization project, linking the requirement of dishonesty and built-in reliance on immunity as the driver of investigations and prosecution.’15 His point was that ‘An individual is [only] guilty of an offense if he dishonestly agrees to commit or implement [a defined set of cartel violations’;16 and the question of ‘dishonesty’ is a question for the jury under whatever contemporary moral standard it accepts as controlling. The author then adds, ‘Such shared standards may still exist when it comes to stealing or fraud, but as surveys show, there is no community consensus that price fixing equals turpitude.’17 I agree that this need to prove ‘dishonesty’ beyond a reasonable doubt in a criminal court is likely to seriously dilute the potential usefulness of the Enterprise Act as an effective anti-cartel weapon, because it will reduce the potential prosecutions to a very few cases. Indeed this ‘dishonest’ question has been squarely raised by the UK Department for Business Innovation and Skills (‘BIS’) in its recently published paper ‘A Competition Regime for Growth: A Consultation on Options for Reform’. Emphasizing that the Government’s goal is to ‘improve the robustness of decisions and strengthen the regime’ and to ‘support the competition authorities in taking forward high impact cases,’18 in Section 6 of the paper BIS examines four policy options for dealing with the statutory ‘dishonesty’ standard in UK criminal cartel prosecutions. BIS’ preferred choice is ‘option 4’ which would be (i) to remove the dishonesty element, and (ii) to redefine the offence so that it does not include 14 Philip Collins, ‘New decade, new Government—Reflections on possible evolution of the UK’s competition and consumer regimes’, speech, London, 20 May 2010 (emphasis added). 15 Julian Joshua, ‘Shooting the Messenger: Does the UK Criminal Offense Have a Future?’, The Antitrust Source 1, 2 (August 2010). Mr Joshua recognizes that, ‘A strong case can be made for criminalizing cartels’ (at 19). 16 Section 188(1) of the Enterprise Act 2002 (emphasis added). 17 Joshua, cited previous footnote, at 5 (citing Andreas Stephan, ‘Survey of Public Attitudes to Price-Fixing and Cartel Enforcement in Britain’, 5 Competition Law Review 123 (2008)). 18 Department for Business Innovation & Skills, A Competition Regime for Growth: A Consultation on Options for Reform (March 2011), page 2, https://www.gov.uk/government/uploads/system/ uploads/attachment_data/file/31411/11-657-competition-regime-for-growth-consultation.pdf.

48  Donald I Baker agreements which are made openly. The rationale behind BIS choosing this option is that it strikes a good balance between ‘excluding from the scope of the offence the kinds of agreement that might have countervailing benefits under the civil antitrust prohibitions; and differentiating the offence from those prohibitions to reduce the risk that the offence would be categorized as ‘national competition law’ (in which case it would not be possible to prosecute whenever there was a parallel European commission investigation).’19 The Republic of Ireland. The Republic of Ireland was one of the first countries outside of the US to implement any form of comprehensive criminal antitrust sanctions. Ireland has had in place various competition acts since before it joined the ‘Common Market’ in 1973. In 1991, Ireland enacted the ‘Competition Act’ replicating Articles 85 and 86 EEC, but ‘[o]ne of the most consistent criticisms of this legislation was the inadequacy of the enforcement procedures.’20 This then led to the amended Competition Act in 1996. The amended act followed the Sherman Act model of criminalizing the conduct of both enterprises and individuals found to be in breach of the Competition Act; both were subject to fines and individual executives were subject to imprisonment for up to five years. In 2002, Ireland’s competition laws were once again amended and enhanced by the Competition Act of 2002 that repealed both the 1991 and 1996 Competition Acts and criminalized infringements of what had since become Articles 81 and 82 EC. Moreover, the maximum sentence for so-called ‘hard core’ cartel activities was increased. The 2002 Act defined ‘hard core cartel activity’ in Section 6(2) as price fixing, limiting output or sales, or allocating market shares or customers. ‘Hard core cartel activity’ was punishable under Section 8(1) of the Act, which covers both enterprises and individuals. The penalties imposed on an individual defendant include: (a) on summary conviction─ (i) a fine not exceeding € 3,000, 21 and/or (ii) imprisonment for a term not exceeding 6 months. (b) on conviction or indictment─ (i) a fine not exceeding whichever of the following amounts is the greater, namely, € 4,000,000 or 10 percent of the turnover of the individual undertaking in the financial year ending in the past 12 months,22 and/or (ii) imprisonment for a term not exceeding 5 years

As you can see from the above statute, the severity of the penalty imposed depends on whether or not the conviction is the result of a ‘summary’ offence or an ‘indictable’ offence. When the Oireachtas (Ireland’s National Parliament) passes a criminal law it also decides on whether a violation of that law will be a summary offence, an indictable offence, or both (as was done in the Competition Act). Ibid, page 61. Peter Charleton and Marguerite Bolger, ‘The Competition (Amendment) Act, 1996: Extending the Criminal Law’, 3 The Bar Review 214 (1998). 21 For Corporate fines, see Part 2, Section 8(1)(a)(i). 22 For Corporate Fines, see Part 2, Section 8(1)(b)(i). 19 20

Criminalising Cartels in Europe  49   Thus in Ireland there are two ways criminal antitrust offences can be tried: (1) a ‘summary’ offense is tried in the lower court (District Court) before a judge only, and (2) an ‘indictable’ offense is tried in the higher courts (Circuit Criminal Court, Central Criminal Court) before a judge and jury. The initial decision whether to charge a summary or indictable offence is made initially by the Competition Authority (usually in conjunction with the Director of Public Prosecutions), but this determination can be overturned by the Judge in the court where the case is brought. At the present time the Irish Competition Authority and the Director of Public Prosecutions have secured 33 convictions stemming from the various antitrust infringements. Four of those were summary offences, subject to the less severe penalties, and the remaining 29 were procured by the way of indictment, therefore opening the defendants up to the maximum liability under the 2002 Act. Stemming from those 33 convictions fines have been issued exceeding € 600,000. Of those 33, 18 have been handed down against undertakings (all of which received fines), and the remaining 15 convictions were handed down to individuals. All 15 individuals received fines and 10 received custodial sentences. It is important to note, however, that those 10 custodial sentences were all suspended sentences ranging from between 3 to 12 months.23 Despite the lack of convictions thus far, during Director of Public Prosecutions v Denis Manning, the sitting Justice of the Central Criminal Court emphasized that: ‘I see no room for a lengthy lead in period before jailing convicted persons becomes commonplace under this legislation.’24

C. The US experience: punishing individuals seems almost an accident of history Criminalization of antitrust offences originally occurred in the US without any serious debate. In the late 1880s, there was huge political momentum to do something about ‘The Trusts’, but few had a very clear idea of exactly what ‘a trust’ was or how many organizational variations there were, let alone exactly how to deal with any such entity.25 And all this intense concern arose before the concept of the modern administrative state had arisen. The framers of the Sherman Act wanted to do something more than just give the government the power to prohibit future conduct by the defendant trust; they also wanted the government to be able 23 The same thing has apparently been true in Canada, which has had criminal antitrust liability since 1889, but Canadian judges have generally entered suspended or other non-custodial sentences. 24 See Director of Public Prosecutions v Dennis Manning, discussed more extensively below at footnote 45 and accompanying text. 25 Anti-monopoly fervor in the US exploded during the latter half of the 19th Century as a result of the Industrial Revolution. See John J Flynn, ‘Criminal Sanctions Under State and Federal Antitrust Laws’, 45 Texas Law Review 1301, 1302 (1967).

50  Donald I Baker to punish these imagined villains for their past misdeeds. In 1889–90, the way to do that was to make the conduct criminal, and so that is what the Congress did.26 By then many US states had enacted their own antitrust laws, showing the way as many of those laws imposed criminal penalties.27 Thus, when the Sherman Act was passed in 1890, it was essentially an assumed conclusion among legislators that antitrust violations should be subjected to criminal penalties. Ironically, Senator John Sherman himself was the only identifiable voice against criminalization.28 Ultimately, the Sherman Act, with criminal penalties as a principal means of punishment, passed almost unanimously by a vote of 51–1 in the Senate and 242–0 in the House of Representatives. The Sherman Act prohibition was principally directed at business entities, but the statutory phrase chosen, ‘any person’, was a concept broad enough to cover individuals as well as almost any form of business organization. Still, the criminal prohibition was fairly modest—a mere misdemeanor with a maximum fine of only US $5000 and a maximum jail sentence of 1 year.29 Thus, the American system of punishing individuals was more or less thrown in, but without much debate, as an adjunct to punishing corporations. However, regularly subjecting individual wrongdoers to criminal liability would not become a pillar of modern antitrust enforcement for at least seventy years. Today, everything is totally different. Sherman Act violations are felonies for which individual conspirators are regularly jailed under an amended statute that permits jail sentences of up to ten years as well as a personal fine of up to US $1,000,000.30 How did this complete change come about in the second half of the 20th Century? A vital breakthrough on criminal liability for individuals came with the famous Electrical Equipment indictments and pleas in the late 1950s.31 These cases exposed an extraordinarily broad set of conspiracies that resulted in electrical utilities paying hundreds of millions of dollars too much for equipment and consumers ultimately paying even more for electricity purchased from rate-base regulated utilities. The size of the conspiracy and the prominence of the companies involved 26 Interestingly, the first Canadian antitrust law, passed in 1889, made all violations criminal. Combined Investigations Act ch. 41 1889 SC16. At that time, there was apparently no other national antitrust statute in existence. 27 Ibid. 28 See 21 Cong. Rec. 2604 (1890). 29 Sherman Act, ch. 647, 26 Stat 209 (1890) (current version at 15 USC §§ 1–3 (1994)). 30 15 USC §§ 1–2 (2004). In 1974, the maximum individual fine was increased to US $100,000, and the maximum corporate fine was increased to US $1million. See Antitrust Procedures and Penalties Act, Pub L No 93-528, § 3, 88 Stat. 1706, 1708 (1974) (current version at 15 USC §§ 1–3 (1994)). In 1990, the maximum individual fine was increased to US $350,000, and the maximum corporate fine was increased to US $10million. See Antitrust Amendment Act of 1990, Pub L No 101–588, § 4, 104 Stat. 2879, 2880 (1990) (codified at 15 USC §§ 1–3 (1994)). The increase in jail sentences to 10 years and fines to US $1million for individuals were contained in Antitrust Criminal Penalty Enhancement and Reform Act of 2004, Pub L 108-237 (2004), 118 Stat 665 (2004). The maximum corporate fine was also increased to US$100 million. 31 See Donald I Baker, ‘The Use of Criminal Law Remedies to Deter and Punish Cartels and BidRigging’, 69 George Washington Law Review 693, 705 (2001). Although the first jail sentence for a Sherman Act violation was slightly earlier, the Electrical Equipment cases were the watershed because they were so highly visible to the business public and their lawyers.

Criminalising Cartels in Europe  51   caught the public attention. A number of individual participants—executives of some of America’s most prominent manufacturing companies—were publicly dispatched to federal penitentiaries based on guilty pleas. The business press widely covered the whole process, and, for the first time, potential price fixers perceived themselves as being at risk of actually going to jail.32 This perceived risk of incarceration is critical. In the mid-1970s, with the passage of the 1974 felony statute,33 which had raised the maximum jail sentence to three years, the DOJ began to place a heavy emphasis on indicting individuals and seeking jail sentences, for the very purpose of emphasizing to wrongdoers and potential wrongdoers that they were seriously at risk. As the head of the DOJ Antitrust Division at the time, I was at the centre of this effort and very much believed in it.34 The stakes went up even further with the passage of the Sentencing Guidelines in 1987.35 These Guidelines treated antitrust felonies as very serious offenses. Based on the ‘point system’ used to punish corporate criminal behavior, the Guidelines make imprisonment an immediate remedy for the sentencing court in an antitrust case. As with corporate defendants, different factors may affect the fine or the number of months (or years) that the individual defendant is sentenced.36 In 1990, the maximum fine for an individual was increased to $350,000, while the maximum corporate fine was raised to $10 million.37 Finally, in 2004, Congress seriously raised the stakes again for individuals engaged in Sherman Act violations with a new maximum jail sentence of 10 years and the maximum individual fine of $1 million.38 It is interesting to note that all of the key criminalization landmarks (in 1890, 1974, and 2004) were signed by Republican Presidents, and the ten year jail term for antitrust felons was enacted in 2004 by a Republican controlled Congress. This is interesting because the Republicans are the party generally more inclined to be anti-regulation and pro-business. Strong criminal penalties for antitrust conspirators are clearly not a subject on which there is any real partisan division in America. 32 For a detailed report on the GE/Westinghouse/Allen Bradley, et al., price-fixing trial and outcome, see Richard Austin Smith, ‘The Incredible Conspiracy,’ Fortune, April 1961: 132–137; May 1961: 161–164. See also Charles A Bane, The Electrical Equipment Conspiracy: The Treble Damage Actions, New York Legal Publications, 1973. 33 Antitrust Procedures and Penalties Act, Pub L No 93-528, § 3, 88 Stat. 1706, 1708 (1974). As noted above, the maximum individual fine was increased six fold to US $100,000, while the maximum corporate fine was increased twentyfold to US $1 million. 34 See Justice Department speech by Donald I Baker, ‘To Make the Penalty Fit the Crime: How to Sentence Antitrust Felons,’ 20 Nov 1976, reprinted in Antitrust & Trade Rep No 790, at D-1 (Nov 23, 1976). 35 US Sentencing Commission, Sentencing Guidelines and Policy Statements (1987). 36 Ibid. § 2R1.1(b)–(d). 37 See Antitrust Amendment Act of 1990, Pub L No 101-588, § 4, 104 Stat. 2879, 2880 (1990). 38 See Antitrust Criminal Penalty Enhancement and Reform Act of 2004, Pub L 108-237 (2004), 118 Stat. 665 (2004). As noted above, the maximum corporate fine was also increased to US $100 million.

52  Donald I Baker

D. Criminalization of cartels in the US allows use of enhanced investigational tools and processes The drafters of the Sherman Act clearly intended to create a common law system of antitrust enforcement (rather than a code-centered administrative system) by virtue of the sparse language of the statute. Prosecutors and generalist federal judges applying vague statutes were left with huge discretion in defining legal wrongs in pragmatic terms, while fact-finding was to be done by juries of ordinary citizens. This is fundamentally different from the code-centered administrative systems that are used for antitrust enforcement in most of the rest of the world. In the process, Congress opened up now-familiar institutional arrangements and enforcement tools that Americans take for granted but are generally not available to antitrust enforcers elsewhere. This is why, even when a criminal antitrust statute is enacted (as in the UK and Ireland), the NCA is generally faced with an even more difficult task than their DOJ counterparts in conducting a criminal investigation. Combined Functions. It also turns out to be significant that the antitrust civil and criminal enforcement functions were combined in the Attorney General’s hands when the Sherman Act was passed in 1890, and this arrangement has remained unchanged. Thus the DOJ Antitrust Division (which was finally created by the then new Franklin D. Roosevelt Administration in 1933) makes its own prosecution decisions and does not need the concurrence of the local US Attorney to empanel a grand jury or bring a case in his or her District. Elsewhere in the world civil and criminal law enforcement responsibilities have usually been divided when criminal cartel and bid rigging penalties were enacted. Civil enforcement and investigational responsibilities are entrusted to an administrative agency, while ultimate responsibility for bringing and trying a criminal case are with a public prosecutor’s office. This logical division can have some important institutional implications – because it will almost inevitably dilute a competition agency’s influence vis-à-vis the business community and the bar. Traditional prosecutors (with whom the competition agency may or may not have a positive working relationship) are likely to be more cautious in bringing criminal prosecutions against antitrust conspiracies – particularly given unfamiliarity with competition law issues, other enforcement priorities, or understandable concerns about the higher burdens of proof required in a criminal case.39 Finally, combining both civil and criminal enforcement in a single agency avoids a problem of running an amnesty program whenever two different agencies have responsibility for prosecuting enterprises and individuals. Neither acting alone may be able to give sufficient certainty and comfort to would-be applicants in many cases. This issue may become more important in Europe if the Commission is dealing 39 For example, I understand that, in Ireland, the Competition Authority has recommended quite a few criminal cases that the Director of Public Prosecutions’ office has not been willing to process.

Criminalising Cartels in Europe  53   with enterprises, while the NCAs are investigating cartel violations by individual executives. Investigational Tools. DOJ prosecutors have investigational tools that are unlikely to be available to prosecutors elsewhere. Most notably, the secretive US grand jury system enables DOJ to piece together a credible criminal cartel case even when documentary evidence is limited or benign.40 Prosecutors use the grand jury process by playing individual targets off against each other. Only the government lawyers know exactly what information they have and from whom it came, and what information they still need to make a case – or to make a stronger or broader case. The result is definitely a ‘dog eat dog’ environment highly favorable to the prosecutors. Other investigational tools are also available. For example, in 2006, Congress amended the wiretapping statute to include antitrust violations to the list of crimes that could be investigated by court-authorized wiretaps.41 Encouraging Whistleblowing. Since it was adopted in 1993, the DOJ Antitrust Division’s corporate amnesty program has been much copied around the world, but the effectiveness of this program for DOJ has been seriously enhanced with an effective criminal enforcement program against individual executives.42 The ‘stay out of jail’ card, available only to the first-in leniency applicant, adds an intensely personal dimension for some senior corporate officials who are likely to be involved in making the decision whether the corporation should turn in its allies and seek amnesty. Thus, the US process appears quite different from that in countries that only punish enterprises for cartel violations, or where two different authorities are involved in prosecuting the enterprises and the individuals.43 There the involved individuals have little incentive to work hard to recall awkward facts about meetings and understandings; they, like their employers, hope that the whole thing blows over without serious fines being imposed. This basic reality helps explain why the DOJ investigators are generally in a better position than many foreign agency case handlers to piece together the details of a conspiracy in situations involving very little documentary evidence of what transpired and when.

40 Use of the grand jury in the US is guaranteed by the Bill of Rights (‘No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentation or indictment of a Grand Jury’). The grand jury requirement was apparently put into the Bill of Rights, as a safeguard for liberty, because colonial era grand juries would sometimes refuse to indict individuals for saying or doing politically popular things when the Crown prosecutors sought to proceed against them. Today, the grand jury system is generally seen as a prosecutorial tool for obtaining evidence without the presence of counsel for suspects or witnesses. 41 USA PATRIOT Improvement and Reauthorization Act of 2005 § 113(g)(3), Pub L No 109-177, 120 Stat 192, 210 (2006). See 18 USC § 2516(1)(r) (Supp. IV 2004). 42 US Department of Justice, Corporate Leniency Policy (1993), available in its current form at http://www.justice.gov/atr/public/guidelines/0091.htm. 43 This will be a serious issue in Europe, whenever the Commission is prosecuting the enterprise(s) and an NCA (or, worse still, several NCAs) are prosecuting individual executives criminally, or even administratively.

54  Donald I Baker

E. The wisdom of punishing individuals The European Commission and most EU Member States now severely punish enterprises for cartel violations, but what distinguishes the United States is (i) its serious commitment to using incarceration to punish individual conspirators, and (ii) its successful antitrust amnesty program (first adopted in 1993) to greatly increase cartel participants’ perceived risk of actually being caught. As I have indicated, the case for punishing individual wrongdoers seems relatively strong to those of us who have been involved in the process. First, illegal conspiracies do not exist in the abstract; active participation by particular individuals is essential to the success of any conspiracy. Second, the participating individuals (and sometimes those who supervise them) presumably anticipate higher compensation (such as salary, bonuses, or commission) based on inflated cartel-generated economic performance, or, at the least, expect corporate advancement if the cartel efforts are successful in enhancing the revenues and profits to the relevant businesses. The deterrence-based rationale for severely punishing individuals for cartel violations remains a largely US phenomenon. It is interesting that outside the US, the state has been more willing to punish individuals who have defrauded the state (eg, by bidrigging) or an employer (eg, by embezzlement) than those who steal from consumers via a cartel. Perhaps, because the enterprise can benefit financially a from cartel violation, it seems appropriate for the state to limit penal efforts to heavy administrative fines against the enterprise whose employees have misbehaved. Additionally, in a case involving major embezzlement or fraud, the amount of public loss may be a lot more obvious than when a conspiracy manipulates markets and fleeces numerous consumers out of individually smaller sums. Outside the US, there may be a stronger (and perhaps sometimes more appropriate) assumption that a company’s leaders have participated, or at least have encouraged subordinates to participate, in some clearly illegal cartel activities in order to enhance corporate profits. Stated another way, perceived antitrust risk is a more central part of US corporate culture—even for senior executives—and the price of being caught is well understood to be more painful than elsewhere. In my experience, some US corporate officials being transferred to Europe have been advised on arrival by foreign colleagues (but not lawyers) that ‘rules are different over here’ on dealing with competitors. Where there is top-down encouragement or participation, the leaders involved should of course be punished directly, while the corporation should also be heavily punished to make clear such activities are contrary to the corporation’s interest. However, most of my experience concerns conspiratorial activities that have been hidden from the company’s senior management and in-house lawyers by their wrongdoing employees – who are generally ambitious individuals seeking promotions, reputation enhancement, bonuses or other indirect gains from appearing to have been more successful commercially than they really were.

Criminalising Cartels in Europe  55   Thus, if I am correct, these antitrust conspirators’ psychology may be analogous to that so famously illustrated by the rogue traders at Barings and Société Générale, who apparently would not have reaped the immediate financial benefits if their fraudulent gambles had succeeded rather than failed. The cartel conspirators to whom I have been exposed never seem to have worried about the corporate shareholders being socked with large fines and civil damages as a result of their activities. Rather, to the extent that they have worried, their concern was about being exposed and punished themselves. Still, this approach to deterrence is not uniquely American. Rather, there seems to be a slowly growing realization elsewhere that incarceration is a very useful deterrent that ought to be used. Thus, in April 2010, when the UK OFT was just beginning its first criminal trial (against the four British Airways executives), a partner in a major London solicitors firm predicted that, ‘[m]aking individuals responsible for competition law infringement is likely to become increasingly common … . There is a growing feeling that imposing ever larger fines on businesses is neither sustainable nor effective in ensuring compliance with the competition law rules.’44 An outstanding example of such thinking occurred recently in Ireland, where no convicted cartel defendant had yet been incarcerated, when Justice McKechnie (then) of the Central Criminal Court stated: In my view there are good reasons as to why [the] court should consider the imposition of custodial sentences in such cases. Firstly, such a sentence can operate as an effective deterrent in particular where if fines were to have the same effect they would have to be pitched at an impossibly high figure. Secondly, fines on companies may not always guarantee an adequate incentive for individuals within those firms to act responsibly. This particular point may not, in some circumstances, have the same force where individuals are concerned. Thirdly, a knowledge that courts will regularly make use of a custodial sentence may act as an incentive to people to offer greater co-operation in cartel investigations against, and quite frequently against their employers. Fourthly, prison, in particular for those with unblemished pasts, for those who are respected within the community, and for those who are unlikely to re-offend can be a very powerful deterrent and finally, the imposition of a sentence for the type or category of persons above described can carry a uniquely strong moral message. Accordingly they are, in my view, some very powerful reasons to custodise an individual who has been found guilty under the 2002 Act. In this context I would like to state clearly and categorically that I see no room for a lengthy lead in period before jailing convicted persons becomes commonplace under this legislation.45

In a particularly thoughtful article, Wouter P.J. Wils, then of the Commission’s Legal Service, asked the question: Is Criminalization of EU Competition Law the Answer?46 His answer is basically ‘yes’ – provided that five conditions were met (i) 44 ‘BA Executives Begin Cartel Trial’, Global Competition Review, 12 April 2010 (electronic edition), www.globalcompetitionreview.com/news/article/28258/. 45 See Direct of Public Prosecutions v Dennis Manning, quoted in Annual Report of the Competition Authority 2007, pp 8–9 (emphasis added), http://www.tca.ie/images/uploaded/documennts/2007_ Annual_Report.pdf (last accessed 6/1/11). 46 See 28 World Competition 117 (2005).

56  Donald I Baker a dedicated investigator and prosecutor, (ii) adequate powers of investigation, (iii) judges and juries willing to convict, (iv) adequate political and public support, and (v) continued punishment of enterprises for hard core cartel violations. I entirely agree with this analysis—but question how many Member States (let alone the EU as a whole) can meet Dr. Wils’ conditions. This is of course a situation that could change over time (as it did in the United States during the 20th Century).

F. Making the penalty fit the crime The accepted wisdom in the US that we impose serious criminal sanctions against those who commit antitrust violations is to deter similar conduct by others. As a consequence, in the United States, where we now have had a serious antitrust criminal enforcement system in effect for several decades, we do not tend to see serial violators (those individuals who repeat the same offense after having been caught once). Rather, we see some different individuals and enterprises violating the law because they think they can get away with it, and they can see substantial economic advantage in trying to do so. While this is today’s wisdom, it took over 60 years for the US to start imposing any criminal sentences on price fixers, and about 85 years before such sentences were imposed quite regularly. Today’s fundamental US conception of jail for white collar criminals, as deterrence vis-à-vis future violators as reinforced by the Federal Sentencing Guidelines, does not necessarily fit easily within the social and jurisprudence assumptions accepted outside the US. Even where EU Member States impose criminal penalties for white collar crimes (eg, for embezzlement and frauds), jail sentences are apparently less routinely imposed or are shorter compared with the US practice. And so far, no custodial sentences for antitrust violations have been imposed anywhere in the EU, except for the unusual Marine Hoses sentences, where incarceration in the UK was a condition of the defendants not being imprisoned in the US. The Americans (including myself) tend to be firm advocates of criminalizing cartel violations, because our experience teaches us that sending individuals off to jail gets a lot of attention and is an effective deterrent. Yet we must be mindful of the basic truth that law in general, and criminal law in particular, must reflect the generally accepted values of a society. The United States famously relearned this basic lesson during the fourteen years of Prohibition (1919–1933), when it sought to criminalize the sale or possession of alcohol.47 Instead of generating compliance, the US created widespread disrespect for law and many profitable entrepreneurial opportunities for bootleggers and speakeasy operators.48 I strongly believe in the 47 This unwise nationwide prohibition was mandated by the 18th Amendment to the US Constitution, ratified in 1919. 48 During the 1920s, the American writer Ring Lardner offered a number of wonderfully cogent observations about this great American experiment: ‘After all, Prohibition is better than no liquor at all.’ (1924); and ‘7 or 8 years ago…most cities had a law that you must close your saloon at 11 o’clock

Criminalising Cartels in Europe  57   basic lesson from Prohibition: it is better to have no law than a law that cannot be enforced (or where the society is unwilling to fund enforcement of it). With this reality in mind, I have become ever more impressed by the reluctance of some of my very thoughtful civil lawyer friends about the idea of imposing criminal liability and potential incarceration on individual cartel conspirators. These are professionals who understand competition law at least as well as I do and recognize cartels as an economic hazard. Still they just do not tend to see imprisonment as an appropriate or effective remedy for cartel conspirators. Why this is so seems less than obvious. Most EU Member States would provide criminal liability for an embezzler or fraudulent securities trader who steals funds from his employer, and some important Member States have longstanding criminal liability for bid rigging. Is it that stealing a lot of small sums from consumers or customers somehow seems less reprehensible than stealing from an employer or the state? Or is it that, if the cartel conspiracy is successful and goes undetected, it is profitable to the corporation (which therefore may have encouraged it)? Which raises the intriguing question: would the famous rogue traders in the Barings and Société Générale cases have escaped liability if their fraudulent escapades had not caused economic disasters for their employers? There may be broader issues as well. The typical cartel conspirator is a nonviolent first time offender, and many societies do not tend to imprison such individuals. And, the idea of using jail to deter others rather than just punish the defendant seems to have a lot less traction in Europe than the US (which may help explain why the aggregate imprisonment rates for all crimes is so much higher in the US). Where this leaves me is with an enhanced sense that antitrust champions ought to be searching for alternative forms of deterrence vis-à-vis potential cartel participants in any Member State where incarceration for cartel conspirators would run counter to its culture and where criminal prohibitions probably would not be regularly enforced by criminal prosecutors.

G. Returning to the basic issue of deterrence Any system that is based on deterrence – and fairness – must have clear rules. Ordinary business actors must be able to understand the difference between right and wrong, and their lawyers must be able to give unequivocal legal advice. The cartel area seems to be one where this is possible. The anti-cartel rules can focus several types of (generally secret) agreements among competitors: • Agreements not to compete for any customers on price or some element of price (for example, credit terms). or 12 o’clock or 1 o’clock. Now days according to the law, they ain’t no saloons so they can and do stay open as long as they feel like.’ (1923)

58  Donald I Baker • Agreements not to poach each other’s existing customers. • Agreements on price floors. • Agreements not to compete for any customers, old or new, in each other’s agreed geographic markets. • Agreements on who will be the winning bidder in a bidding situation (with the other participants often putting in so-called ‘complementary bids’ to make auction look legitimate). • Agreements to limit output.49 Each of these examples is capable of being seen and understood by a business executive or salesperson as illegal before he takes any action to enter into or implement such an agreement. Thus, there is still a mens rea requirement, but the burden on the prosecutors should not be excessive (eg, highly subjective variables like ‘dishonesty’ are best avoided). Effective deterrence also requires that those who might be tempted to take illegal action believe that there is some reasonable probability of their being caught and that, if so, the consequences are likely to be personally painful. The use of a wellpublicized amnesty program that encourages any co-conspirator to turn informant is a useful tool for enhancing this sense of perceived risk. The presence of criminal remedies – particularly against individuals – seems to me to have seriously enhanced the successful DOJ amnesty program and hence is an important part of achieving the level of deterrence we now seem to have with major corporations in the United States. Risk of conviction and imprisonment provides a powerful inducement for one co-conspirator to inform on other co-conspirators. Consistency and comprehensiveness in enforcement are very important. The United States has provided a fairly high level of funding for antitrust criminal enforcement by the DOJ and a uniquely advantageous prosecutorial system of grand jury investigation. For criminal laws against individual cartel participants to really matter, enforcement must be frequent and highly visible. It has taken almost a century in the US for incarceration to become routine for cartel participants who were caught. By contrast, having a criminal law against a profitable activity is unlikely to be effective as a deterrent if the normal prosecutions are so infrequent as to appear more like random lightning strikes or prosecutorial vendettas.

H. Developing alternative enforcement avenues to promote deterrence It may well be that the ongoing European dialogue about deterring conspiratorial cartel activity by individuals has been too tightly focused on criminalization, 49 These categories are quite similar, but slightly broader than those specified in Section 188(2) of the UK Enterprise Act, discussed above at footnote 4 and accompanying text.

Criminalising Cartels in Europe  59   because criminalization is the route that the US has successfully pioneered since 1974 and publicly advocated in international fora (such as OECD and the ICN). Today, governments potentially have available a whole range of administrative processes and remedies which did not exist, even conceptually, when the US Congress decided without much debate to treat individual Sherman Act violators as criminals. Antitrust enforcers and advocates probably ought to be studying the administrative remedies that have been created and employed in other sectors (eg, regulation of banks and securities markets) where regulators seek to prevent dishonest conduct by individual, rent-seeking wrongdoers. Because deterrence remains an essential goal in anti-cartel enforcement, the more practical question for the Member States (and ultimately for the Commission) is this: what administrative remedies could be regularly applied to individual wrongdoers in ways that the business community would perceive as personally painful? And how could such administrative remedies be applied on a sufficiently regular basis to be perceived as a genuine risk by the target audience? So, the challenge is not only to generate sufficient actual and anticipated pain, but also to communicate the risk of being caught and punished to those who are willing to take legal risks to further their careers. Administrative remedies generally will demand a lower legal standard of proof and are much more within the control of the enforcement agency than are criminal prosecutions, especially in civil law systems. A reasonably staffed agency may be able to bring many more such cases and thereby increase the ‘probability of punishment’ factor in the deterrence equation. At this point, I cannot pretend to have developed a comprehensive list of potential administrative sanctions that might be used against an individual found by the agency to have engaged in illegal cartel conduct. Therefore, I offer what I hope will serve as an illustration and also as a catalyst for more thinking by those familiar with the administrative processes and practices in different Member States: –

A substantial administrative fine, especially when it could be coupled with an effective prohibition against the employer reimbursing the defendant (the Netherlands has started imposing these kinds of administrative fines in antitrust cases); – Prohibition against receiving deferred compensation or other employerprovided retirement benefits; – Prohibition against future employment for the same company or even in the same industry for a period of years (this kind of remedy has been regular feature of US SEC enforcement); – Prohibition against serving as an officer or director of a public company (the director-disqualification was a supplemental remedy created by s. 204 of the UK Enterprise Act of 2002, amending the Director Disqualification Act of 1986).

60  Donald I Baker I do not think that the ‘criminalization’ versus ‘administrative’ issue should be treated as a binary ‘either/or’ question. Even in a jurisdiction where a criminal antitrust law has been adopted (as in the UK, Ireland, Canada, or Australia), criminal enforcement need not necessarily be the only way to punish individual antitrust violators just because that is the way it has been in the US for more than 120 years. Rather, suitably-designed administrative sanction(s) may be a more effective weapon against some individual wrongdoers in some cartel cases. Where a Member State enacts both administrative and criminal remedies for antitrust cartel violations, the enforcement agency would have to make a prompt decision (eg, soon after receiving an amnesty application) whether to pursue a criminal case or an administrative remedy.50 It could rationally decide to use the criminal route when it had a potentially strong high-visibility case, while using administrative proceedings in a majority of the more ordinary cases against smaller or less picturesque cartels. In other words, with their generally lower burden of proof, administrative sanctions might enable the enforcement agency to bring many more cases—and thereby increase the risk-of-prosecution factor for would-be wrongdoers. At the same time, the most blatant cases could justify the enforcement resource expenditure necessary to send a high visibility message to the market.

I. Conclusion The European Commission is the most important antitrust enforcement agency in Europe, quite possibly the world. Yet the Commission seems to have remained conspicuously silent on what I believe to be one of the most important antitrust issues of the day – namely, whether punishing individuals as well as undertakings would be likely to substantially strengthen antitrust enforcement against cartels. We know that, even with the successful amnesty programs in place as an effective enforcement tool, cartels continue to operate in many industries because those involved make a positive risk/reward calculation. As a result, European enterprises continue to be well-represented (perhaps even overrepresented) in the lists of global conspirators. The fact that Commission lacks jurisdiction over individuals is not a reason for remaining silent on this question. With private damage litigation, the Commission has taken a vigorous leadership position since 2004 in urging Member States to establish private damage remedies. Commissioner Kroes was a strong and frequent advocate, while emphasizing increased deterrence as a major reason for encouraging private litigation. DG Competition established an energetic working 50 As explained earlier, in Ireland, the Authority has to make an initial decision of whether to charge a summary offense (with lesser penalties) or an indictable offense to be tried before a jury. However, as I understand the Irish practice, a judge can overturn this choice.

Criminalising Cartels in Europe  61   group in 2004 under the leadership of a senior official, Emil Paulis; they produced a Green Paper in 2005; and then, after receiving many comments, they created a White Paper and detailed Staff Commentary in 2008. Throughout the process, the Commissioner and the working group members urged the Member States to create private antitrust remedies and modify their legal systems in some respects in order to make private enforcement more effective. The process has generated a lot of dialogue and some controversy within the EU and at the Member State level. This is the kind of systematic effort that I would respectfully urge the Commission to employ again in encouraging Member States to develop more or better remedies to deter individuals from participating in potential cartel conspiracies. As I have suggested, the focus should not just be about whether to criminalize such conduct or not; it should include a thoughtful search for effective administrative alternatives that might curb the optimistic instincts of individual antitrust wrongdoers. As with the private damage project, DG Competition has the ability to develop constructive ideas and generate substantial dialogue about them. Finally, as the post-Sherman Act history illustrates in the US, longer time dimensions are needed, because the public has to gradually get used to the idea of strongly punishing individuals for crimes that have previously gone unpunished. Even if incarceration is the most effective remedy, administrative remedies may be easier to enforce on a regular recurring basis, thus enhancing deterrence; and ultimately the regular use of such remedies may help to educate lawyers, judges, and the public about the idea of applying criminal sanctions (including incarceration) against individual cartel conspirators. Doing so would clearly be based on choices made – often differently – by the parliaments of the Member States. Since the jurisdiction of the European Commission only extends to undertakings, it would apparently be necessary to amend the Treaty to provide the Commission any jurisdiction against individuals – which I think would be a very good idea. This is no doubt a question for another day. But if there were other substantive areas of law/regulation where giving the Commission enforcement jurisdiction over individual misconduct would be likely to generate positive enforcement results, then would it not be wise to seriously consider the admittedly difficult treaty-amending process in Brussels?

James S Venit and Andrew L Foster*

Competition Compliance: Fines and Complementary Incentives

1. Introduction This paper considers the effectiveness of the current European competition enforcement approach in deterring hardcore antitrust violations. It also explores the availability and attractiveness of complementary compliance incentives under the anti-fraud provisions of US securities laws and rules relating to the fiduciary obligations of corporate directors. These measures have the potential for introducing additional deterrent mechanisms that may prove as or more effective than punitively high antitrust fines given the individual civil sanctions they may entail and the strong due diligence culture embedded in securities law and corporate governance. The measures considered include financial penalties or personal civil liability for individuals under the securities laws, compliance-oriented remedies imposed in the context of shareholder derivative suits, and executive pay clawbacks under US financial services legislation. A deterrence system that puts all its enforcement eggs in one basket and relies exclusively on deterring and punishing antitrust infringements by fining delinquent companies may give rise to unwanted side-effects. This is because the exclusive reliance on fines for punishment and deterrence requires that the amounts levied must be sufficiently high that they cannot simply be written off as a cost of doing business.1 As EU competition fines have moved inexorably higher in recent years, they may also have begun to produce some unwanted effects, including the risk of driving firms out of business (and thus out of competition altogether), inhibiting the ability of firms to fund innovation, and imposing disproportionate penalties on firms whose revenues are needed to defray high fixed costs. The imposition of high fines also has the effect of intensifying scrutiny on the Commission’s * James Venit is a Partner at Skadden, Arps, Slate, Meagher & Flom LLP in Brussels. Andrew Foster is an Associate working in Skadden’s Brussels and Beijing offices. 1 See, eg, Joaquin Almunia, ‘Compliance and Competition Policy’, Speech/10/586 of 25 October 2010, at p 2 (‘Companies need to understand the cost of non-respect of competition rules, the cost of a cartel, and this cost must be larger than what they hope to gain from entering into the price-fixing or other similar illegal behaviour, in the first place.’); Wouter PJ Wils, ‘Optimal Antitrust Fines: Theory and Practice’, 29 World Competition 1, 11 (2006) (arguing that a fine must ‘weigh[] sufficiently in the balance of expected costs and benefits to deter calculating companies from committing antitrust violations’).

64 

James S Venit and Andrew L Foster

procedures, and potential evidentiary and due process concerns, that have become an increasing target for public criticism as fines have moved up the Richter scale. Despite the consistent imposition of increasingly higher fines, it is not clear that the Commission has successfully instilled a culture of compliance in Europe or elsewhere. In the absence of material consequences for individuals and additional affirmative obligations on undertakings, the incentives to develop a pro-active, firm-wide sense of accountability may still not be sufficient. True cultural transformation may be effected through additional affirmative obligations rather than the mere imposition of ever heavier fines. In particular, incentives to comply could be reinforced through the imposition of penalties or personal civil liability for individuals or by enhancing firms’ incentives to self-investigate and comply. As for individuals, there seems to be an emerging consensus among commentators and practitioners that the credible threat of individual punishment is an important deterrent. In addition, in the US, the Securities and Exchange Commission (the ‘SEC’) has begun to use US securities laws to expand the potential scope of individual exposure for antitrust wrongdoing, while the Dodd-Frank Wall Street Reform and Consumer Protection Act may add a new tool for competition compliance, in the form of executive pay clawbacks. Private enforcement, which has been underutilized in the EU, may also provide additional incentives for strengthening corporate antitrust compliance. Such measures may include the enhancement of leniency programs through the reduction of civil damages for whistle-blowers, a measure that remains within the antitrust cultural framework, or exposure to costly and reputation-damaging securities fraud and shareholder derivative suits which inject a new cultural dimension in domains whose watchwords are due diligence, fiduciary duty and full disclosure. While the US has sought to enhance the effectiveness of its leniency program (and thereby alleviate the need for very high fines through increased detection) by ‘de-trebling’ the private damages obtainable from successful leniency applicants, the absence of treble damage systems in Europe and the Commission’s treatment of private enforcement as purely compensatory may eliminate ex ante the need for, and benefits of, this approach in the EU. However, private enforcement in the form of securities fraud litigation and shareholder derivative suits may impose additional affirmative obligations on firms and, more importantly, on their officers and directors, to eliminate anticompetitive behavior proactively, given, respectively, possible exposure for failing to disclose anticompetitive profits that distort a company’s earnings or either negligence or malfeasance in failing to prevent serious antitrust misconduct in the first place. Such actions appear to be increasingly common in the US, and their transplantation to the EU is something that may being to attract attention at either the national or EU level.

Competition Compliance: Fines and Complementary Incentives 

65  

2.The deterrent effect of corporate fines The current EU deterrence system relies exclusively on the imposition of civil fines imposed on firms. Regulation No 1/2003 provides for fines on ‘undertakings and associations of undertakings’ of not more than 10% of total turnover in the preceding business year where the undertaking intentionally or negligently infringes Articles 101 or 102 TFEU.2 The Regulation states explicitly that decisions imposing fines ‘shall not be of a criminal law nature’,3 and no fines on individuals have ever been imposed, although they are theoretically possible.4 However, the EU makes no provision for individual penalties such as director disqualification or the use of incarceration as a penalty, although it permits individual Member States to do so.5 Corporate fines in the EU must therefore do the work of what might otherwise be accomplished through a varied arsenal of deterrence mechanisms. Both the Commission and the EU Courts have acknowledged that, in addition to punishment, one of the primary purposes of competition fines is to deter both individuals and undertakings from violating the antitrust laws.6 In this context, Commissioner Almunia has stated that ‘[t]he level of fines should therefore be determined by the objective to achieve a sufficient level of deterrence’.7 The deterrent effect of fines requires first, a credible threat of detection and prosecution, accompanied by a penalty which ‘weighs sufficiently in the balance of expected costs and benefits to deter calculating companies from committing antitrust violations’.8

2 Article 23(2), Council Regulation (EC) No 1/2003 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty (16 December 2002), 2003 OJ L1/1. 3 Ibid., Article 23(4). 4 See Wouter PJ Wils, ‘Is Criminalization of EU Competition Law the Answer?’, 28 World Competition 1, 14 (2005) (‘If the undertaking found to have committed a violation were found to consist of an unincorporated business (a single trader, with or without employees, who has not incorporated his or her business, or a professional exercising his or her profession alone and unincorporated, or several natural persons operating a single business without any employment relationship between them and without any form of legal person), the Commission would necessarily have to address its fining decision to the natural person or persons operating the business. However, in the more than one hundred decisions in which the Commission has imposed fines under Regulation No 17 and Regulation No 1/2003 up to now, this situation has never occurred: up to now, all fines have been imposed on companies or other legal persons.’). 5 Member States providing for director disqualification include the UK; Member States providing for custodial sanctions include the UK, Ireland, France, etc. 6 Joaquin Almunia Interview, 34 World Competition 1, 1 (2011) (‘The fine aims to achieve both individual and general deterrence. It serves to sanction the undertaking for its illegal behaviour but also to deter other undertakings from conduct which is in breach of EU antitrust rules.’). See also Case 41/69, ACF Chemieforma [1970] ECR 61, para. 173 (confirming that competition fines «ont pour but de réprimer des comportements illicites aussi bien que d’en prévenir le renouvellement») (emphasis added). The ECJ has also stated that it is «open to the Commission to consider that it is appropriate to raise the level of fines so as to reinforce their deterrent effect’. Joined Cases 100-103/80, Musique Diffusion Française and Others v Commission [1983] ECR I-11005, para. 81. 7 Almunia Interview, cited above note 6, at p 1. 8 Wils, Optimal Antitrust Fines, cited above note 1, at p 11.

66 

James S Venit and Andrew L Foster

According to one respected commentator, Wouter Wils, deterrence achieved solely through the use of fines ‘will work if, and only if, from the perspective of the company contemplating whether or not to commit a violation, the expected fine exceeds the expected gain from the violation’.9 Achieving a ‘sufficient’ level of deterrence using only corporate fines requires the Commission to increase the size of the fine levied to a level at which the fine cannot simply be written off as a cost of doing business.10 Additional elements work to raise the required size of the fines over the bare level necessary to overcome a hypothetical undertaking’s internal ‘pricing’ strategy. First, because it appears that ‘people consistently tend to overestimate the probability of good things happening to them, and underestimate the probability of bad things happening to them’, in the context of internal ‘pricing’ decisions ‘companies and individual decision-makers within them will tend to overestimate the gain from the antitrust violation and to underestimate the probability of being caught’.11 As a result, the level of fines has to be adjusted upward to counter this effect. Second, under the deterrence approach, the antitrust fine is ‘a sanction for doing what is forbidden’,12 and therefore the size of the fine often incorporates some punitive measure reflecting societal disapproval of the conduct in question. Given that the expected anticompetitive profits gained from cartel activity can be very high and the expectation of being detected will usually be underestimated, corporate fines must also achieve very high levels to ensure the necessary degree of deterrence. A recent study examining cartel activity between 1990 and 2008 found that fines have increased an average of 8% annually, with a 107% increase due to the introduction of the Commission’s 2006 Fining Guidelines.13 In a concrete example, Wouter Wils has noted a significant divergence between fines imposed on the Belgian and Dutch beer cartels in which, for comparable infringements in similar markets, Interbrew received a fine of € 46 million in 2001, while Heineken was fined € 219 million in 2007.14 9 Ibid, p 12. In this regard, the expected fine should equal ‘the nominal amount of the fine discounted by the probability that a fine is effectively imposed’. 10 See, eg, Almunia, cited above note 1, at p 2 (‘Companies need to understand the cost of nonrespect of competition rules, the cost of a cartel, and this cost must be larger than what they hope to gain from entering into the price-fixing or other similar illegal behaviour, in the first place.’); Wils, Optimal Antitrust Fines, cited above note 1, at p 11; Scott D Hammond, ‘The Evolution of Criminal Antitrust Enforcement Over the Last Two Decades’, Speech at the 24th Annual National Institute on White Collar Crime (25 February 2010) at p 4 (‘If the potential penalties that can be imposed upon cartel participants are not perceived as outweighing the potential rewards of participating in a cartel, then the fine imposed becomes merely part of the cost of doing business.’). 11 Wils, Optimal Antitrust Fines, cited above note 1, at pp 16–17. 12 Ibid, at p 14. 13 John Connor and Douglas Miller, ‘Determinants of EC Antitrust Fines for Members of Global Cartels’ (2009; latest draft dated 6 March 2013), http://ssrn.com/abstract=2229358. 14 Wouter PJ Wils, ‘The Increased Level of EU Antitrust Fines, Judicial Review, and the European Convention on Human Rights’, 33 World Competition 1, 10 (2010); see Commission Decision 2003/569/EC of 5 December 2001 in Case IV/37.614 – PO/Interbrew and Alken-Mars, 2003 OJ L200/1, and Commission Decision of 18 April 2007 in Case COMP/37.766 – Dutch beer market, 2008 OJ C122/1.

Competition Compliance: Fines and Complementary Incentives 

67  

What is clear is that competition fines now regularly run into the hundreds of millions of euros. The ten most recent cartel decisions have imposed total fines ranging from € 61.1 million (Calcium Carbide 2009) to € 799 million (Air Cargo 2010), with an average total of € 328 million for all participants collectively.15 The ten highest fines on individual undertakings for cartel violations have all been imposed in the ten years since 2001 (and all but one since 2007),16 with an average amount of € 466 million and a high of € 896 million.17 The Commission’s willingness to impose high fines has extended to abuse of dominance cases as well, as has been demonstrated by the fines levied on Intel in 2009 (€ 1.06 billion)18 and Microsoft in 2008 (€ 899 million)19 even though there is far less unanimity amongst economists and lawyers as to what conduct should be deemed to constitute an abuse of dominance than there is concerning the perniciousness of hard core cartels.20

15 Washing Powder (April 2011) € 315.2 million; LCD Panels (December 2010) € 648 million; Air Cargo (November 2010) € 799 million; Animal Feed Phosphates (July 2010) € 176 million; Bathroom Fittings (June 2010) € 622 million; DRAM (May 2010) € 331 million; Plastic Additives (November 2009) € 173 million; Power Transformers (October 2009) € 67.6 million; Concrete Rebar (September 2009) € 87.3 million; Calcium Carbide (July 2009) € 61.1 million. 16 Hoffman-La Roche (2001) Vitamins € 462 million; ThyssenKrupp (2007) Elevators/Escalators € 480 million; Siemens (2007) Switchgear € 397 million; St. Gobain (2008) Carglass € 896 million; Pilkington (2008) Car Glass € 370 million; Sasol (2008) Candle Waxes € 318 million; E.On (2009) Gas € 553 million; Suez (2009) Gas € 553 million; Ideal Standard (2010) Bathroom Fittings € 326 million; Air France/KLM (2010) Airfreight € 310 million. 17 This fine was imposed on car glass manufacturer St. Gobain. See Commission Decision of 12 November 2008 in Case COMP/39125 – Carglass, 2009 OJ C173/13. 18 Commission Decision of 13 May 2009 in Case COMP/37.990 – Intel, 2009 OJ C227/13. 19 Commission Decision of 27 February 2008 in Case COMP/37.792 – Microsoft, 2009 OJ C166/20. 20 From a deterrence standpoint, levying high fines on Article 102 behavior may not be appropriate. First, there is no broad consensus as to the conduct that merits characterization as an abuse. Second, if the deterrence calculation for an expected fine should equal the nominal amount of the fine increased by a multiplier to account for those cases that go undetected, then the amount necessary for deterrence may arguably be very different for Article 101 and Article 102 conduct. See Wils, Optimal Antitrust Fines, cited above note 1, at p 12. Cartel behavior by its very nature must be concealed and is not easily susceptible of detection. In contrast, conduct constituting abuse of dominance (including, eg, conditional rebates, exclusivity agreements, tying agreements and predatory pricing) is often communicated to customers and may even be publicized by the company itself. In addition, in the case of Article 102 conduct, ascertaining whether behavior is likely to result in a fine ex ante can be complicated by the legal uncertainty surrounding Article 102 and the use of difficult economic tests such as estimating customers’ ‘contestable share’ for an ‘as efficient competitor’ test or making evaluations of appropriate cost measures to evaluate the presence of predatory pricing. Given the lack of consistency between, for example, US and EU approaches to dominant firm conduct, the imposition of very high fines may also raise issues in connection with the principle of proportionality of penalties as articulated in Article 49(a) of the Charter of Fundamental Rights of the European Union, which provides that ‘the severity of penalties must not be disproportionate to the criminal offense’.

68 

James S Venit and Andrew L Foster

3. Potential problems with very large fines The magnitude of current fine levels requires consideration of the potential dangers that very large fines may entail. Here undesirable side-effects may include driving firms from the market, reducing their ability to innovate, and disproportionately punishing firms whose revenues must defray very high fixed costs. In addition, the use of very high fines in a system that has been criticized for its procedural unfairness may have the effect of aggravating due process and evidentiary complaints. These negative effects should be weighed against the benefits of imposing high fines and any such assessment should consider whether the use of very high fines alone has affected corporate mindsets and induced a culture of pro-active competition compliance.

(a) Bankruptcy It has been suggested that if fines high enough to ensure reliable deterrence were in fact imposed ‘many of the companies concerned would be forced into bankruptcy’.21 In addition to potential negative structural effects, bankruptcy may also have undesirable social effects, such as punishing not only the stakeholders and decision-makers of the undertaking, but also its employees, suppliers, customers, creditors and tax authorities.22 The European Commission’s 2006 guidelines on setting fines (‘Fining Guidelines’), which acknowledge these potential problems, provide that fines may be evaluated with regards to their ‘specific social and economic context’ and further state that reductions can be granted if the fine ‘would irretrievably jeopardise the economic viability’ of the undertaking.23 The General Court has noted that these concerns may be relevant only when payment of the fine would lead ‘to an increase in unemployment or deterioration in the economic sectors upstream and downstream’ of the market.24 The court went on to observe that ‘the fact that a measure taken by a Community authority leads to the insolvency or liquidation of a given undertaking is not prohibited as such by Community law’.25 Under this view, the imposition of a fine may be justified in driving a firm out of business despite the fact that doing so would result in a more tightly concentrated, transparent, and less competitive marketplace, regardless of the social context or upstream or downstream economic effects. Recognition of these downsides has led Commissioner Almunia to state publicly that the Wils, Optimal Antitrust Fines, cited above note 1, at p 19. Ibid. Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003, 2006 OJ C210/2 at ¶ 35. 24 Joined Cases T-236/01, T-239/01, T-244/01 to T-246/01, T-251/01 and T‑252/01, Tokai Carbon Co. Ltd et al. v Commission [2004] ECR II-1181, at para. 371. 25 Ibid at para. 372. 21 22 23

Competition Compliance: Fines and Complementary Incentives 

69  

Commission has ‘no interest in driving companies out of business.’ Indeed, of the 32 applications made in 2010 for a cartel fine reduction on grounds of inability to pay, the Commission granted nine (28%).26

(b) Innovation and fixed costs Very large fines may also have the effect of stifling innovation. A formerly cartelized sector that is currently paying € 1.38 billion in fines, such as the car glass industry, or € 648 million in fines, such as LCD panel manufacturers, is less able to spend that money on research and development or innovation. A related issue arises in respect of those industries that have very high fixed costs which require hundreds of million or billions of euros of investments prior to the realization of any profits. Construction of a microchip fabrication, for example, can cost USD 3 to 5 billion,27 while development of a new pharmaceutical may cost an average of USD 1.3 billion.28 Firms engaged in such industries must use revenues both to defray these high fixed costs as well as to pay for future investment. The imposition of very high competition fines based on turnover may disproportionately impact these industries, stifling innovation and chilling entry.

(c) Procedural perceptions While some have argued that ‘if additional monetary penalties were indeed required for optimal deterrence, these could be provided for in an administratively much cheaper and more reliable way by increasing the fines imposed in the public enforcement proceeding’,29 this approach overlooks the fact that centralizing the imposition of very high fines into a single body may also intensify concerns about the impartial and fair administration of the proceedings that result in large fines. Given the current debate surrounding due process30 and the use of evidence in Commission proceedings,31 having a variety of deterrence-enhancing tools at 26 See Joaquin Almunia, ‘Recent developments and future priorities in EU competition policy’, Speech/11/243 of 8 April 2011. As some commentators have observed, however, granting reductions for an inability to pay the fines may also have the effect of creating fines that are ‘lower than those required to deter’. Wils, Optimal Antitrust Fines, cited above note 1, at p 19. 27 Christopher Mims, The High Cost of Upholding Moore’s Law, Technology Review (Mass. Inst. Tech.) May/June 2010, http://www.technologyreview.com/computing/25141/. 28 Morgan Stanley, Pharmaceuticals, 20 January 2010, 2. 29 Ibid at 10–11. 30 See, eg, Ian S Forrester, ‘Due Process in EC competition cases: a distinguished institution with flawed procedures’, 34 European Law Review 817 (2009). 31 See, eg, James S Venit, ‘Human All Too Human: The Gathering and Assessment of Evidence and the Appropriate Standard of Proof and Judicial Review in Commission Enforcement Proceedings Applying Articles 81 and 82’, in Claus-Dieter Ehlermann and Mel Marquis (eds), European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review in Competition Cases, Hart Publishing, 2011, pp 191–253.

70 

James S Venit and Andrew L Foster

hand may be preferable to relying exclusively on administratively imposed fines. In addition, a wider arsenal of deterrents may create more effective deterrence than a system in which deterrence rests solely on the strength of a single decision, which can be appealed and annulled on review and which may be invalidated as a result of a material procedural defect.

4. Has fining undertakings instilled a pro-active compliance culture? Despite the increased reliance on, and prevalence of, very high fines, it is not clear that firms are adopting a culture of strong antitrust compliance. In the absence of penalties for individuals and additional affirmative obligations on firms and those in positions of responsibility within them, a forceful sense of accountability may still be lacking. For example, since the introduction of the Commission’s new leniency notice in 2006, the number of undertakings implicated in cartel activity has remained stable or increased, from 45 in 2007, to 37 in 2008, to 40 in 2009, to 69 in 2010.32 While the introduction of the new leniency program itself appears to have led to increased detection of cartels (the Commission prosecuted 157 undertakings between 2000 and 2004, compared with 207 between 2005 and 2009)33 the stable or increasing number of decisions suggests that firms may not be reacting to higher fines by abandoning cartels. Many legal scholars and antitrust practitioners have concluded that effective deterrence must include at least the credible threat of individual sanctions. Wouter Wils, one of the two EU Commission Hearing Officers and a former member of the Commission’s Legal Service, has persuasively argued that the ‘effective deterrence of cartels requires a combination of monetary sanctions on companies and individual penalties, in particular imprisonment’.34 Scott D. Hammond, the US Deputy Assistant Attorney General for Criminal Enforcement in the Antitrust Division of the DOJ, has similarly asserted that ‘the most effective way to deter and punish cartel activity is to hold culpable individuals accountable by seeking jail sentences’.35 The adoption of custodial criminal sanctions does not currently appear to be a realistic option at the EU level:36 EU law does not provide for individual penalties such as director disqualification or incarceration, and the EU has never 32 Commission, Cartel Statistics (9 June 2011), http://ec.europa.eu/competition/cartels/ statistics/ statistics.pdf. 33 Ibid. 34 Wouter PJ Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for Damages’, 32 World Competition 1, 10–11 (2009). 35 Hammond, cited above note 10, at pp 11–12. 36 Commissioner Almunia has indicated that ‘enforcement against undertakings remains the core principle at EU level’ and that individual penalties such as incarceration would not be feasible as there are no EU criminal courts. Almunia, cited above note 1, at p 2.

Competition Compliance: Fines and Complementary Incentives 

71  

imposed a civil fine on an individual for anticompetitive conduct. This lack of individual penalties may lead to diverging incentives between firms and their employees as regards competition compliance. Where a decision-maker faces individual rewards for generating profits (even those gained anticompetitively) but no penalties for violating the antitrust laws, his or her normative commitment to ‘following the rules’ may not be aligned with the goal of maximum competition compliance.

(a) Corporate compliance programs The US actively encourages the development of a rigorous compliance culture by incentivizing the adoption of well-designed antitrust compliance programs. Under the US Sentencing Guidelines, an undertaking with a well-designed compliance and ethics program may qualify for reduced penalties, as long as the employees committing the violation were not ‘high-level personnel’.37 The US Guidelines recognize that ‘[t]he failure to prevent or detect the instant offense does not necessarily mean that the program is not generally effective’, providing maximum incentive for firms to institute and enforce such a program. In contrast, the European Commission offers no incentive to adopt a compliance program, other than the avoidance of the large fines that accompany competition law infringements that are brought to light. In particular, the Commission does not consider the existence of a compliance program to constitute a mitigating circumstance relevant to the determination of the appropriate level of a fine.38 Some commentators approve this approach, observing that ‘[i]f fines are set at the level required for deterrence, namely above the expected gain from the violation multiplied by the inverse of the probability that a fine will be imposed, companies will already have all the necessary incentives to prevent the commitment of antitrust violations.’39 However, by taking this approach the Commission may also have passed up an opportunity to reward undertakings which have made a genuine attempt to transform their internal compliance cultures.

37 US Sentencing Commission, 2010 Guidelines Manual (November 2010), at § 8B2.1 (‘Effective Compliance and Ethics Program’). 38 See, eg, Almunia, cited above note 1 (‘I am often asked whether companies should be rewarded for operating compliance programmes when they are found to be involved in illegal commercial practices. The answer is no. There should be no reduction of fines or other preferential treatment for these companies. As already mentioned, we reward cooperation in discovering the cartel, we reward cooperation during the proceedings before the Commission, we reward companies that have had a limited participation in the cartel, but that, I think is enough.’). See also Commission Decision of 3 December 2003 in Case COMP/38.359 – Electrical and mechanical carbon and graphite products, 2004 OJ L125/45; Case T-31/99, ABB v Commission [2002] ECR II-1884, para. 221; Case T-224/00, Archer Daniels Midland v Commission [2003] ECR II-2597, paras. 280–281. 39 Wils, Optimal Antitrust Fines, cited above note 1, at p 24.

72 

James S Venit and Andrew L Foster

(b) Custodial sanctions and personal liability for individuals There would appear to be an emerging consensus that a system including personal liability not only creates the credible threat of punishment required for deterrence, but also has the effect of accelerating settlements and enhancing the effectiveness of leniency programs. Measures sanctioning individuals also create a strong sense of affirmative personal accountability for antitrust compliance, which fines on an undertaking alone cannot instill. In the US, for example, there exists a variety of individual penalties, all of which appear to enhance the deterrent effect of the antitrust laws. First, individuals violating the Sherman Act may be criminally liable for fines up to USD 1,000,000 or imprisonment not exceeding ten years, or both.40 Indeed, the individual fines can even exceed the statutory USD 1,000,000 limit under the Comprehensive Crime Control Act, which provides alternative maximum sanctions of twice the gross gain or twice the gross injury caused.41 While the availability of individual fines is unquestionably important – the US Sentencing Guidelines specify that ‘[s]ubstantial fines are an essential part of the sentence’ for individuals – the Antitrust Division of the US Department of Justice ‘has long emphasized that the most effective way to deter and punish cartel activity is to hold culpable individuals accountable by seeking jail sentences’.42 As a general policy, the Antitrust Division seeks to indict at least one individual from each indicted organization in a hardcore cartel case, and the average sentence for imprisonment has also been increasing, from eight months in 1999 to 24 months in 2009. In that same time period, 223 individuals have been sentenced to incarceration in cases prosecuted by the Antitrust Division.43 And it appears that these sentences have also increased the effectiveness of the US leniency program. For example, between 2000 and 2002, the Antitrust Division initiated an average of 26 Grand Jury investigations each year. However, between 2003 and 2009, the number of investigations initiated increased to 35.6 each year. The number of overall cartel investigations opened increased as well, from 88.0 per year between 2000 and 2002 to 97.6 per year between 2003 and 2009.44 Individual sanctions in the US are now having an increasing effect on foreign nationals. In May 1999 in the Vitamins case, the Antitrust Division first entered into a plea agreement with a Swiss executive that called for the imposition of jail time for a foreign national who had participated in an international cartel.45 Between 1999 and 2007, more than 40 foreign defendants served, or were serving, prison sentences in the US for participating in an international cartel or for obstructing an investigation of an international cartel.46 15 USC § 1. 18 USC § 3571(d). Hammond, cited above note 10, at pp 11–12. 43 DOJ workload statistics, http://www.justice.gov/atr/public/workload-statistics.html. 44 Ibid. 45 Hammond, cited above note 10, at p 7. 46 Ibid. 40 41 42

Competition Compliance: Fines and Complementary Incentives 

73  

While the US leads in imposing custodial penalties with regularity, there appears to be an emerging trend as other jurisdictions increasingly impose jail time for individual violations of competition laws. In 2008, three executives were sentenced to lengthy jail terms in the United Kingdom for participation in the Marine Hose cartel, marking the first jail sentences for a cartel offense under the 2002 Enterprise Act.47 On 12 March 2009, new legislation was passed in Canada making hard-core cartel behavior per se illegal and increasing the maximum term for imprisonment from five to 14 years.48 Similarly, the Australian Parliament introduced a criminal cartel offense effective 24 July 2009 which provides for imprisonment of up to ten years.49 Other jurisdictions such as Chile, the Czech Republic, Greece, Mexico, the Netherlands, New Zealand, Russia and South Africa have also recently adopted, or are considering, legislation that will criminalize cartel offenses.50

(c) Availability of individual sanctions in the EU There is no reason to believe that the EU could not introduce individual civil penalties to increase personal accountability for competition compliance. Although it should be recalled that Article 23(5) of Regulation 1/2003 explicitly provides that the European Commission’s antitrust fining decisions ‘shall not be of a criminal law nature’,51 thus precluding the EU from pursuing custodial sanctions absent reform of the Treaties, this would not appear to prevent the Commission from pursuing either civil fines against individuals or other incentivizing measures such as director disqualification. Under the European Charter of Human Rights, outside the ‘hard core of criminal law’, penalties can be imposed, ‘in the first instance, by an administrative or non-judicial body combining investigative and decision-making powers’.52 Given the greater deterrent effect of individual sanctions, the Commission’s enforcement efforts could be strengthened by such an approach.

47 See Enterprise Act 2002 § 188 (making it a criminal offence for individuals to engage dishonestly in certain horizontal agreements such as price fixing and market sharing with a term of imprisonment of up to five years and/or an unlimited fine). 48 Budget Implementation Act (Bill C-10), taking effect 12 March 2010. 49 Terry Calvani and Torello Calvani, ‘Custodial sanctions for cartel offences: An appropriate sanction in Australia?’, 17 Competition and Consumer Law Journal 119, 123 (2009). 50 Hammond, cited above note 10, at pp 11–12. 51 Regulation 1/2003, Article 23(5). 52 Ibid, at p 16.

74 

James S Venit and Andrew L Foster

(d) Director disqualification and securities actions In June 2010, the OFT published new guidance on the use of director disqualification as an individual penalty for cartel offenses in the UK.53 The OFT has not yet sought to disqualify directors, but ‘[s]purred on by studies which show that individual deterrence is the key to influencing corporate conduct’, the OFT appears to be expanding its toolbox.54 Under the Company Director Disqualification Act 1986 (as amended by the Enterprise Act 2002), a UK court must make a Competition Disqualification Order against an individual if (i) he or she serves as a director for an undertaking which breaches either Chapter I or II of the Competition Act 1998 or Articles 101 or 102 TFEU; and (ii) the director’s conduct renders ‘him or her unfit to be concerned in the management of a company’. The Competition Disqualification Order can have a maximum duration of 15 years. As noted above, the OFT has not yet sought to disqualify a director, but its recent publication of new guidelines on enforcement indicates a willingness to make use of this individual sanction as a mean of enhancing individual accountability for antitrust compliance. In the US, there also appears to be an emerging SEC administrative practice which could lead to increased individual exposure for antitrust violations. By way of background, in the US, the securities laws prohibit persons or entities from making false or misleading statements in connection with the purchase or sale of securities. The Securities Exchange Act of 1934 (the ‘’34 Act’) imposes liability on any person or entity that intentionally or recklessly makes material false or misleading statements in connection with the purchase or sale of any security that causes an investor who relies on that statement to suffer a financial loss. In the context of civil litigation, the ‘34 Act gives shareholders the right to bring a private action in federal court to recover damages the shareholder sustained as a result of securities fraud. Shareholder plaintiffs in a securities fraud case have often sought to recover on the theory that earnings statements have been misstated and share values artificially inflated by a company’s failure to disclose its (profit-enhancing) anticompetitive behavior. The Securities Act of 1933 (the ‘’33 Act’) imposes strict liability on issuers of securities that make false or misleading statements in connection with a registered public offering of securities. The ’33 Act also imposes liability on directors or officers of the issuer who negligently make false or misleading statements regarding material facts, such as approving a false and misleading registration statement, in connection with a public offering of securities. Actions under the US securities law are scarcely the exclusive province of private actors. The SEC itself has the primary responsibility for enforcing the 53 Office of Fair Trading, Director disqualification orders in competition cases: An OFT guidance document (2010), http://www.oft.gov.uk/shared_oft/business_leaflets/enterprise_act/oft510.pdf. 54 See Luis Gómez and Grant Murray, ‘Getting personal: director disqualification for competition law violations’, in Baker & McKenzie, eds, European Legal Developments Bulletin (Autumn 2010), vol 22 no 4, at p 48.

Competition Compliance: Fines and Complementary Incentives 

75  

laws it oversees, and has begun to develop an administrative practice linking antitrust violations and securities fraud. Thus, in July 2010, the SEC settled a complaint against Dell using an antitrust theory of harm establishing this linkage. In its complaint, the SEC had alleged, inter alia, that individual officers and directors of Dell, including its Chairman and its CEO, had violated Section 17(a) of the Securities Act of 1933 by omitting to state a material fact necessary for the correct understanding of Dell’s earning statements and engaging in a practice operating ‘as a fraud or deceit’ on share purchasers.55 Specifically, according to the SEC complaint, the individuals concerned had failed to disclose allegedly ‘secret’ payments by Intel (which the European Commission had found violative of Article 102 TFEU), when ‘[i]t was these payments rather than the company’s management and operations that allowed Dell to meet its earnings target’.56 While the undisclosed source of the profits in the Dell case were allegedly the result of anticompetitive behavior (conditional rebates or exclusivity payments paid by an allegedly dominant undertaking), it should be noted that the SEC did not allege that Dell itself had committed any antitrust violation by accepting the payments. Rather, the securities law infringement resulted only from the failure to report the payments.57 In the Dell Settlement, Chairman Michael Dell and CEO Kenneth Rollins – without admitting any wrongdoing – each paid USD 4,000,000, while other executives also settled for as much as USD 3,000,000 apiece.58 The Dodd-Frank Wall Street Reform and Consumer Protection Act in the U.S. includes a provision which may also enable the SEC to use the securities laws to stiffen antitrust compliance. The Dodd-Frank Act requires every public company to adopt a policy under which, in the event of a corporate restatement of earnings, the company will recover or ‘claw back’ from current and former executives ‘any incentive-based compensation, for the three years preceding the restatement, that would not have been awarded under the restated financial statements’.59 Any failure to do so would result in delisting from the U.S. exchanges. In addition, the Sarbanes-Oxley Act also contains a claw back provision of more limited scope, applicable only to the CEO and CFO, which is triggered if the restatement is the result of misconduct.60 It remains an open question whether either of these two provisions could be used where earnings are inflated by anticompetitive conduct, to claw back individual salaries, bonuses and other compensation of high-placed executives who have been complicit in, or have concealed, anticompetitive See 15 USC § 78j(2), (3). See SEC Press Release 210–131, SEC Charges Dell and Senior Executives with Disclosure and Accounting Fraud, (22 July 2010), http://www.sec.gov/news/press/2010/2010-131.htm; Complaint, Securities and Exchange Commission v Dell Inc. et al., Case: 1:10-cv-01245 (D.C.C. 22 July 2010). 57 Under the same ‘failure to disclose’ theory, any earnings statements or required SEC filings which contain revenues inflated by anticompetitive conduct of which a director was aware but failed to disclose could provide another means by which to impose liability on individuals for anticompetitive conduct. 58 Dell, Inc. also paid USD 100 million in the settlement, representing a significant, additional potential corporate cost that could arise for an undertakings’ non-compliance with the antitrust laws. 59 Public Law 111–203 § 954(b)(2). 60 15 USC § 7243. 55 56

76 

James S Venit and Andrew L Foster

conduct. Executives may have an argument that these provisions should not apply because typically, as a matter of accounting, the disclosure that earnings arose from anticompetitive conduct does not require a restatement of historical financial statements. While the actual numbers in those financial statements are ‘accurate,’ courts have nevertheless found the financial statements to be misleading because the company’s disclosed explanations for its sources of revenues in its public filings or public statements were false.

5. Using private enforcement to place additional affirmative obligations on undertakings In addition to individual accountability, corporate culture may be modified through the imposition of additional affirmative obligations on undertakings to investigate and self-enforce antitrust compliance. Private enforcement may provide one means of strengthening such affirmative obligations. Such incentives may include reduction of civil damages for whistle-blowers (which enhances the effectiveness of leniency programs but also incentivizes undertakings to detect and report cartel behavior), or exposure to expensive securities fraud and shareholder derivative suits for a failure to detect and report anticompetitive conduct that has a material impact on corporate earnings statements.

(a) Reduction of civil penalties for leniency applicants Deterrence could theoretically be enhanced through the reduction of potential civil damages for successful leniency applicants. This would have the effect of making entry into the program more attractive (which would in turn enhance the probability of detection, thereby increasing the deterrent effect and lowering the need for reliance on very high fines). In the US, under the Antitrust Criminal Penalty Enhancement and Reform Act of 2004, the ability of private claimants to obtain treble damages for antitrust injuries61 has been limited with regards to successful applicants for immunity. In an attempt to increase the effectiveness of its leniency program, the US ‘detrebled’ the potential civil liability of an immunity applicant,62 thus enhancing the effectiveness of the program. Provided for under the Clayton Act of 1914, see 15 USC § 15(a). Pub L 108–237, title II, § 213(a), June 22, 2004, 118 Stat. 666, 667 (‘[I]n any civil action alleging a violation of section 1 or 3 of the Sherman Act, or alleging a violation of any similar State law, based on conduct covered by a currently effective antitrust leniency agreement, the amount of damages recovered by or on behalf of a claimant from an antitrust leniency applicant […], shall not exceed that portion of the actual damages sustained by such claimant […]’). 61 62

Competition Compliance: Fines and Complementary Incentives 

77  

Since the detrebling of US civil damages for leniency applicants had the effect of removing the punitive element of civil claims, thereby reducing the scope of civil damages to compensation, it is not clear that this approach is relevant to the EU Member States which provide for compensatory, as opposed to punitive, civil claims. Nevertheless, the US example raises the question whether the Commission could or should grant immunity or reduction in respect of civil damages actions to whistle blowers in the EU. Unlike the US, where private enforcement is seen as another tool in the antitrust toolbox along with federal and state government enforcement, the EU has ordinarily tried to limit the role of private enforcement to achieving compensation for victims. Thus, Commissioner Almunia has stated, that, while fines through public enforcement should seek ‘to achieve a sufficient level of deterrence’, the goal of private enforcement is ‘[i]n contrast to this’ and should instead seek ‘full compensation for harm suffered’.63 This position is consistent with the Commission’s current Leniency Notice states that ‘[t]he fact that immunity or reduction in respect of fines is granted cannot protect an undertaking from the civil law consequences’ of its infringement.64 Although the Commission considered a rule that would have permitted double damages for civil litigation,65 the 2008 White Paper abandoned this proposal because the Commission believes that allowing punitive damages in private actions would ‘increase the risk of abusive litigation to an extent which is not compatible with the European legal tradition’.66 As a result, with private enforcement likely to remain compensatory rather than punitive, there is little to be gained from a deterrence perspective by eliminating compensation. Significant problems relating to thirdparty entitlement would likely arise from attempting to do so. Even in the US model, ‘de-trebling’ simply means eliminating the punitive portion of the private damages, while still preserving the compensatory, single damage function.

(b) Private securities litigation As noted above, in the US, the ‘34 Act gives purchasers or sellers of securities the right to bring private actions in federal court to recover damages that they have sustained as a result of securities fraud. In addition, state corporate law allows shareholders to sue directors and officers for violations of their fiduciary duties owed to the corporation or its shareholders. Shareholders in the US have made increasing use of these laws to pursue suits designed to impose individual sanctions on officers and directors of firms which violate the competition laws. These suits normally take the form of either shareholder suits for securities Almunia, cited above note 1, at p 1. Commission Notice on Immunity from fines and reduction of fines in cartel cases, 2006 OJ C298/17 at ¶ 39. 65 See 2005 Green Paper, § 2.3, Option 16 at p 7. 66 See Commission Staff Working Document, Public Consultation: Towards a Coherent European Approach to Collective Redress (4 February 2011), § 3.4, at ¶ 21. 63 64

78 

James S Venit and Andrew L Foster

fraud, which allow shareholders to recover the lost value of their shares from the corporation or directors or officers who committed fraud, or derivative suits on behalf of the corporation against the officers and directors, which allow the shareholders to impose new corporate governance mechanisms and recover damages for and on behalf of the company itself. (Although the damages awarded can be quite high, there is ordinarily no payout to the shareholders themselves.) Both of these approaches may increase the affirmative obligations on a firm and its managers and directors to avoid, detect and report anticompetitive activity.

(i) Securities fraud suits Under the ‘34 Act, joint and several individual liability for an executive or officer will inhere when he or she has ‘knowingly committed a violation of the securities laws’, and can include liability for costs incurred as a result of the unlawful activity. The securities fraud or wrongdoing alleged will ordinarily relate to securities Rule 10b-5,67 which prohibits manipulative or deceptive practices, including intentionally making material misstatements or omissions ‘in connection’ with the sale of any security (including, inter alia, signing and certifying mandatory SEC filings such as earnings reports). As it relates to antitrust, the wrongful conduct by an individual could include certifying a statement of accounts or earning statement as accurate despite knowing that the company’s revenues and profits had been artificially inflated by anticompetitive conduct such as a cartel.68 In particular, under the US securities laws, the failure to disclose the cartel activity as the ‘true’ source of the firm’s earnings may constitute a material omission, even if the financial statements themselves are otherwise accurate. Cases in which there have been securities fraud actions by shareholders in the aftermath of antitrust investigations include auction houses;69 rubber chemicals;70 parcel tankers;71 DRAM;72 thin film transistor liquid crystal display 17 C.F.R. § 240.10b-5. See, eg, In re Sotheby’s Holdings, Inc. Sec. Litig., 2000 US Dist LEXIS 12504, No 00 Civ 1041 (DLC);, at *4 (SDNY 31 August 2000); Menkes v Stolt-Nielsen S.A., 2006 US Dist LEXIS 42644, No. 3:03-cv-409, at *5-*6 (D. Conn. 19 July 2006) (examples of the types of statements potentially actionable included: ‘While pricing remains competitive in most markets […]’; ‘[…] pricing competition, weak utilization and empty repositioning costs negatively impacted the results’; ‘[…] we expect to continue to see strong price competition’). 69 In re Sotheby’s Holdings, Inc. Sec. Litig., No. 00-cv-1041 (S.D.N.Y) (securities fraud action). Some of the defendants prevailed on their motions to dismiss, but motions to dismiss were denied as to two other defendants, In re Sotheby’s Holdings, Inc Securities Litig, 2000 WL 1234601 (SDNY 31 August 2000), after which they settled. 70 In re Crompton Corp Sec Litig, No. 03-cv-1293 (D Conn) (securities fraud action which settled in 2010 for $11.4 million). 71 Menkes v Stolt-Nielsen SA, No. 3:03-cv-409 (D Conn). After the initial securities fraud action was dismissed, the parties amended their complaint and the district court denied the defendants’ motion to dismiss the amended complaint in June 2006, Menkes v Stolt-Nielsen S.A., 2006 WL 1699603 (D Conn 19 June 2006), after which the case settled. 72 In re Micron Techs, Inc Sec Litig, No 06-cv-085 (D Idaho) (securities fraud action, filed in 2006, which survived a motion to dismiss in February 2007, and a motion for judgment on the pleadings in February 2009). 67 68

Competition Compliance: Fines and Complementary Incentives 

79  

(TFT-LCD);73 and US domestic ocean shipping.74 Damages sought may include losses from the decline in stock price following revelation of the violation, any fine paid by the company to competition authorities, recovery of attorneys’ fees and costs incurred by the company in the course of the antitrust proceedings (even if the company received leniency), as well as damages paid by the company in treble-damage civil antitrust actions. In these actions, the plaintiffs seek to recover on behalf of a class of shareholders from the company itself or from individual officers or directors responsible for misleading statements made to investors. There is even one case indicating that an individual who has not him or herself directly made misleading statements may be held liable if it can be shown that he or she had concrete knowledge of the wrongdoing and could have prevented the company from making false statements but failed to do so.75 While other courts have not gone so far,76 and liability on this theory appears inconsistent with the recent decision of the U.S. Supreme Court in Janus Capital Group, Inc. v First Derivative Traders,77 this possibility is but another reason for diligent compliance efforts. The primary interest in these actions lies not in the monetary sanctions imposed on the offending corporations but, rather, in their ability to impose civil sanctions on responsible individuals. Even where an individual has not him or herself made misleading statements (ie, by signing a mandatory SEC filing), secondary individual liability may inhere if it can be shown that the director had concrete knowledge of the wrongdoing and nonetheless failed to prevent others within the company from making such statements. For example, in a securities litigation brought against Stolt-Nielsen following its admission of collusive behavior in the parcel tanker and intra-Europe inland barge operations sectors, a US district court concluded that individual corporate officers could be held personally liable for knowingly failing to disclose (or, worse, conspiring to ‘mask and conceal’) illegal cartel behavior from the Board of Directors of the Group parent.78 The case eventually settled for USD 2 million, which included attorney’s fees.79

73 Son et al v LG Display Co. Ltd. et al, No. 10-cv-4380 (SDNY) (securities fraud action by foreign investors); In re LG Philips LCD Co, Ltd Sec Litig, 1:07-cv-909 (SDNY) (securities fraud action). 74 City of Roseville Employees’ Retirement Sys v Horizon Lines, Inc et al, No 08-cv-969 (D Del), affirmed, Nos 10-2788 & 10-3815 (3rd Cir 2011). 75 Menkes v Stolt-Nielsen S.A., 2006 WL 1699603 (D Conn June 19, 2006). Skadden, Arps represented Stolt-Nielsen in the underlying antitrust proceedings but not in the subsequent securities proceedings. 76 United States v Schiff, 602 F.3d 152, 167 (3d Cir. 2010) (‘[T]he plain language of § 10(b) and corresponding Rule 10b-5 do not contemplate the general failure to rectify misstatements of others.’). 77 No 09-525, 2011 WL 2297762 (U.S. 13 June 2011). 78 Menkes v Stolt-Nielsen SA, 2006 WL 1699603 (D Conn 19 June 2006). 79 Menkes v Stolt-Nielsen SA, No 3:03-cv-409, 2011 US Dist. LEXIS 7066 (D Conn 25 January 2011).

80 

James S Venit and Andrew L Foster

(c) Shareholder derivative suits Shareholders may also institute derivative suits on behalf of the undertaking itself in order to recover damages caused by breaches of the fiduciary duties owed by the undertaking’s officers and directors. These breaches can include knowing violation of the antitrust laws or failure to detect, investigate and eliminate such offenses by other employees of the company. Recent competition antitrust cases in which there were follow-on derivative actions by shareholders include the graphite electrodes cartel80 and Intel’s alleged Article 102 TFEU misconduct.81 In a derivative suit, shareholders may sue for the institution of new or improved corporate governance mechanisms, and for restitution (to the company, not the shareholders) for harm caused by the fiduciary breaches, such as fines paid by the company to competition authorities and civil damages and settlements. For example, in the Intel Corp Derivative Litigation, the eventual settlement (which required approval by the court), included, inter alia, introduction of such measures as: creation of a Legal Compliance Group; creation of a Compliance Committee consisting of independent directors; annual review of antitrust compliance policies; records and audits of the program to review effectiveness; and revision of the Intel Code of Conduct to address antitrust compliance more specifically.82 In addition, Intel paid USD 2.65 million in attorney’s fees and stipulated that it had ‘expended millions of dollars to date in costs and resources to implement the [enhancements to its existing antitrust compliance program]’ and agreed ‘to commit the monetary and personnel resources necessary and appropriate to implement and enforce the Company’s future antitrust compliance programs, procedures, training and systems’.83 Such active monitoring and the required expenditure of millions of dollars would undoubtedly benefit a firm’s own internal compliance, but as derivative suits outside of the antitrust context have shown, companies and individual executives may also find themselves liable for hundreds of millions or even billions of dollars of restitution. For example, in September 2008, the UnitedHealth Group settled an options backdating derivative suit in which it paid USD 895 million, while the former CEO agreed to pay USD 30 million and the former General Counsel agreed to pay an additional $500,000. In September 2008, AIG settled a derivative suit against four former executives and its former managing general agent for USD 115 million. This was followed in August 2010 by settlement of an additional derivative action against former 80 Jaroslawicz v Krass et al, No. CV 980331117S (Conn.) (shareholder derivative action that was dismissed for failure to plead sufficiently that demand on board would have been futile, 1999 WL 436409 (Conn. Super. 16 June 1999)). 81 In re Intel Corp Derivative Litigation (MDL), Case 1:09-00867-JJF (D Del. 23 July 2010) (approving settlement of consolidated derivative claims). 82 In re Intel Corp Derivative Litigation, ‘Stipulation of Settlement’ Case 1:09-cv-00867-JJF (D Del. 25 May 2010) at pp 13–16. 83 Ibid, at p 12.

Competition Compliance: Fines and Complementary Incentives 

81  

AIG directors and officers, including its former CEO and former CFO for an additional USD 90 million. In September 2009, Broadcom Corp settled an options backdating-related derivative lawsuit for USD 118 million, while in December 2009, Comverse Technology settled an options backdating-related derivative lawsuit for USD 225 million (after the former CEO fled to Namibia to evade the charges), which included personal payments by the company’s former General Counsel of USD 1 million (in addition to USD 3 million he had previously paid to the SEC). These payments are all dwarfed, however, by the June 2009 USD 2.876 billion judgment against a former CEO in a derivative lawsuit by HealthSouth shareholders, resulting from systematically and fraudulently reported net income between 1996 and 2002.84 While director and officer liability insurance may cover the cost of an individual director or officer’s exposure in both a federal securities action and a shareholder derivative suit, such agreements usually include specific exceptions where a court finds that the individual concerned has engaged in wrongdoing (such as violation of the US criminal antitrust laws). Given the potential additional liability for individuals and exposure for undertakings which certify SEC filings without affirmatively investigating their internal competition compliance, these securities suits should only enhance competition deterrence in the US, both for individuals and undertakings.

6. Conclusion An antitrust enforcement system that puts all of its incentivizing eggs in one basket by relying only on corporate fines to deter firms from engaging in anticompetitive conduct can have unwanted side effects such as bankruptcy, stifling innovation, while placing substantial due process pressures on the administrative procedures through which these sanctions are imposed. Moreover, it is not clear that the systematic imposition of very high fines has resulted in the development of a militant culture of competition compliance. Complementary measures, such as civil sanctions for individuals and additional affirmative obligations derived from the anti-fraud provisions of securities laws and rules relating to corporate fiduciary duties may provide forceful additional avenues for significantly increasing incentives to comply. As concerns individual sanctions, there appears to be a growing consensus that individual sanctions constitute a legitimate part of an effective deterrence regime. Even if criminal custodial sanctions are not feasible at the EU level, other options such as civil 84 Tucker v Scrushy, CV-02-5212 AEH, ‘Memorandum Opinion Regarding Final Judgment Order Against Defendant Richard M. Scrushy Under Alabama Rules of Procedure Rule 54(b)’ (Ct. of Jeff. Cty. Ala. 18 June 2009).

82 

James S Venit and Andrew L Foster

fines or director disqualification may be open. The use of private enforcement may also have a beneficial deterrent effect when pursued under fiduciary duty and/or anti-fraud laws. Reliance on the latter two avenues could draw additional strength from the strong corporate compliance culture that characterizes these areas in practice.

Fred Louis*

Promoting Private Antitrust Enforcement: Remember Article 102

Private enforcement: what’s in a name? The present contribution is based on experience built over the years, starting long before the entry into force of Regulation 1/2003, of litigating (on behalf of the defendant or the plaintiff) a number of cases before Belgian and Dutch civil and administrative courts, where EU competition law arguments were raised, either solely or centrally, or as part of a wider dispute. These cases ran the gamut of competition law issues (pricing agreements, boycotts, parallel imports, refusal to supply, abusive patent strategies, exclusionary pricing practices) and were raised in very diverse settings: cease-and-desist orders linked to an alleged violation of Article 101 or Article 102 TFEU (or their equivalents in those days), competition law arguments incidental to a patent dispute, requests to set aside decisions of associations of undertakings allegedly violating Article 101 or Article 102, requests to reinstate an alleged abusively interrupted supply relationship, damages actions for alleged breach of Article 101 or Article 102, and even some cases raising the much maligned ‘Euro-defence’ in order to escape the enforcement of an otherwise valid contract.1 In some cases, these private litigation actions were initiated in parallel with a complaint to a competition authority (the Belgian NCA or the Commission). It is interesting to note that none of these cases were cartel damages actions. My experience with cartel damages actions is restricted to what is, to my mind, a wholly different type of litigation, ie follow-on actions initiated by putative victims of cartel behaviour, once they become aware of their status as a potential victim through a public announcement of an investigation by the public enforcer, or through the publication of that enforcer’s fining decision. Through a strange twist of events, ever since the first Ashurst report, public debate in recent years over ‘private enforcement’ has been – almost exclusively – dedicated to this latter type of follow-on actions. Yet in my view these are not * Partner, Wilmer Cutler Pickering Hale and Dorr LLP (‘WilmerHale’), Brussels. The views expressed in this contribution are personal and do not necessarily reflect those of WilmerHale. This contribution is meant as a brief outline, shorn of erudite references, to support the workshop discussions. 1 As it often happens, many of these cases did not proceed to a final judgment but were settled or solved in some other way, including through mediation by the judge.

86  Fred Louis instances of private enforcement. The respect of competition rules in the EU is in these cases wholly ensured by public enforcement actions, which give rise to hefty ‘administrative’ fines whose level is determined by the need to ensure deterrence. No competition violation that is not already uncovered by the public enforcer’s investigations comes to light through these private follow-on actions. Pursuant to the Court’s case law, citizens have the right to seek damages for cartel behaviour that causes them a loss, by virtue of the direct effect of Article 101. While follow-on damage actions thus enable citizens to enforce the rights they draw from the EU Treaties, these actions are purely about compensation; they do not serve public enforcement.2 The merits of encouraging follow-on cartel damage actions and how far the current favour these actions seem to enjoy should take us will be discussed by others. For present purposes, I would argue that the debate on these actions should not obscure the need to consider the ‘true’ private enforcement actions, ie actions which bring to light alleged infringements of competition law that have escaped the public enforcer’s scrutiny. In my view these are the actions that should primarily be encouraged by public enforcers insofar as they alleviate the enforcers’ burden while promoting competition in the EU. The rest of this contribution will reserve the term ‘private enforcement’ to this type of actions and will focus on Article 102 enforcement, where private actions have a central role to play.

Private enforcement of Article 102 is needed given the relative rarity and questionable effectiveness of public enforcement Article 102 has always lent itself to private enforcement. Contrary to Article 101 (in its entirety), all parts of Article 102 have always been directly effective. Contrary to secret cartels, most abusive practices are to a large extent ‘public’, and alleged victims do not have to wait for a public enforcer to start its investigation to realize they have been the victims of exclusionary or discriminatory practices.3 Article 102 damages actions are sometimes brought following an adverse decision by the public enforcer. But these are not necessarily pure follow-on 2 Unless one believes that damages awards are part of the punishment for illicit cartel behaviour. However, this is not how the law of damages works in most Member States, in particular in civil law countries where damages actions have a strict repair function and cannot include a punishment factor. With all due respect to a fascinating subject-matter that has kept most of us busy for a great many thrilling years, there is nothing so special about competition law that justifies turning the law of damages on its head. Furthermore, if private damages actions are to be encouraged because of their contribution to deterrence, then non bis in idem considerations have to be taken into account, and the cumulative effect of fines and damages awards needs to be carefully appraised. 3 This is not always the case. Non-transparent rebates practices are usually hidden by a web of confidentiality clauses and oral agreements which, if they are not always undetectable by their intended victims, are very often quite hard to prove to the satisfaction of a court of law.

Promoting Private Antitrust Enforcement: Remember Article 102 

87  

actions, since the intervention of the public enforcer may oftentimes have been prompted by a complaint from the very plaintiff seeking damages. The initiative for the enforcement rested with the complainant in such cases, not with the public enforcer. At the same time, there is relatively little public enforcement of Article 102.4 Competition agencies tend to pick dominance cases carefully. Selection criteria are not always clear or consistently applied but there does seem to be an attempt to pick cases where the potential impact on the EU economy is perceived as quite large, the precedent value is felt to be important and/or the wealth of evidence backing up the complaint greatly facilitates public enforcement. Somewhat paradoxically, this means that large, sophisticated complainants, who could easily afford private litigation and have the means and resources to produce relevant economic evidence, have a much greater chance of convincing the agencies to take their case, while smaller, weaker victims are mostly left to fend off for themselves in the courts. In any event, most abuse of dominance procedures before the public enforcer are fiercely defended. As a consequence, a lot of time and effort is spent to reach a final outcome. It is therefore sometimes doubted that public enforcement of Article 102 ever reaches its intended goal: by the time a decision is reached, the market has moved on.5 To address these difficulties, the Commission has increasingly resorted in recent times to ‘no-fault’ Article 9 commitment decisions. Although they are a useful tool in the Commission’s arsenal, the proliferation of such procedures does have a number of drawbacks. To name just a few, these include: questionable value as precedents because of the no-fault dimension; risk of developing remedies that are either ineffectual or too strict; doubts about the respect of the rights of all affected parties; and the inability to rely on these decisions as a foundation for a damages claim. However, even commitment decisions remain a poor option when speed is of the essence, as it often is when exclusionary practices are at stake. In the light of the restrictive enforcement practice of most competition authorities as regards interim measures, and the cumbersome procedural rules to which such proceedings are generally subject, national courts are often the only recourse whenever swift action needs to be taken. In considering the need to encourage private enforcement of Article 102, the importance of facilitating swift injunctive relief as much as possible cannot be overstated. Damages actions for exclusionary practices are extremely difficult to bring to fruition in a reasonable time frame, and they raise complex questions of economic analysis. It is far better for everyone involved if abusive behaviour can be quickly stopped rather than having to litigate over years, if not decades, on the 4 This is certainly so when one focuses on the Commission. By and large, NCAs tend to be relatively more active in pursuing abuse of dominance cases. Even at that level, however, the focus increasingly tends to be on secret cartels. 5 Admittedly, in a number of cases, this raises the question whether the alleged dominance or the alleged exclusionary nature of the practices under investigation were there in the first place.

88  Fred Louis consequences of having let these practices unfold. As is often the case, prevention is the best cure.6 Private enforcement thus has a key role to play for an appropriate level of enforcement of Article 102, arguably much more so than follow-on cartel damages actions in the field of anti-cartel enforcement.7

Specific challenges of Article 102 private enforcement Private enforcement of Article 102 raises a number of issues before one even gets to the quite complex proof-of-damages stage. To succeed in any attempt at enforcing respect of Article 102 through the courts, an applicant needs to provide proof of dominance, proof of the existence and substance of the allegedly abusive practices, and proof of their abusive nature. In damages cases, the applicant will also need to adduce proof regarding the extent of the damages and as to causation. None of this is easy. To assist him in his endeavours, an applicant can currently call upon the teachings of the European Courts’ jurisprudence and the European Commission’s case practice, the Commission’s 2009 Guidance Paper on its enforcement priorities relating to exclusionary conduct, various economic/legal reports prepared for Commission on the assessment of damages for antitrust violations and, when all else fails, he can call upon the principle of effectiveness. These are all useful tools. Yet they leave many questions unanswered, and sometimes they even induce some confusion, understandably so since most of these sources were not crafted with private enforcement actions in mind. The Courts’ case law and the Commission’s decision practice are by far the most significant sources. However, the number of cases remains fairly limited and, given that many judgments are very fact-specific, it is not always possible to deduce clear rules of general application. In particular, many Commission decisions seem to lend themselves relatively easily to a contrario reasoning, but experience shows that this apparent ease may be deceptive. Preliminary references via national courts to the ECJ may in that context prove quite useful.8 As to proving damages, Article 102 decisions have little to say on the actual effects of practices found to be abusive, and the Courts’ current jurisprudence is unlikely to move the Commission to do more in this respect. 6 However, even that simple principle does not enjoy unanimous approval. One need only think of the English Garden Cottage doctrine, which denies interim relief where damages are available as a remedy. 7 Note that private enforcement of Article 102 is generally pursued by a single plaintiff, or a limited number of them. It therefore does not typically raise the vexing question of whether and how to promote collective redress systems in the EU. 8 They may be particularly useful where the Court proves willing to consider in detail the questions put by the national court, as it was, for example, in Case C-52/09, Konkurrensverket v TeliaSonera Sverige AB [2011] ECR I-527.

Promoting Private Antitrust Enforcement: Remember Article 102 

89  

The Commission’s Guidance Paper is a much less straightforward document to use in private enforcement. It is well-known that, as a document setting out the Commission’s own enforcement priorities, its persuasive power for national courts is by definition quite limited. In any event, the focus of the document is on public enforcement, where fact-finding is not an insurmountable issue. The Guidance Paper is also a source of confusion because, on a number of points, it arguably deviates from the Courts’ case law, at least if one attempts to see in the offending passages an interpretative statement rather than a selection tool to prioritize enforcement action. The position adopted by the Commission in recent court cases where it had to defend Article 102 decisions is also unlikely to increase the standing of the Guidance Paper. Despite all this, the document can be helpful at times as a source of inspiration or confirmation. Until recently, reports prepared for the Commission had little to say of real significance for private enforcement of Article 102. While purporting to deal with private enforcement of antitrust rules in general, the real concrete focus was always on follow-on claims in cartel cases. The most recent Oxera report prepared with a team of jurists led by Professor Komninos is an exception. Although many of the economic techniques and methods listed in the report are not particularly suited to Article 102 damages assessment, some are certainly of interest. On the other hand, the report has little to say on causation. The report of course faced some inherent limitations, in that it was based, inter alia, on an analysis of national court decisions assessing damages for antitrust violations. The number of such decisions is however still quite limited. This is important to keep in mind, as the report’s commendable effort to try to classify and systematize damage assessment methods may obscure the fact that methods others than those described in the report, or variations on these, may be more suited to the facts (and the availability of data) in any given case. But litigants also face countless obstacles arising from national rules of procedure and evidence. As to these, there is little guidance to be found in the sources just discussed, beyond the general statements of the Court on the right to seek damages for antitrust violations. It is my experience and belief that even very conservative legal systems, such as those in the traditional civil law countries, offer more possibilities than is often assumed to be the case, in particular when combined with the principle of effectiveness.

The Commission’s proposals and guidance to facilitate private enforcement: A decisive step in the right direction The above lines were written in June 2011, two years before the Commission came out with its most ambitious intervention yet, in June 2013, to promote private antitrust enforcement. I will not here discuss the Commission’s

90  Fred Louis Recommendation on collective redress,9 but both the proposed Directive and the Communication and accompanying Staff Working Document on quantifying harm are worthy of note. The proposed Directive on certain rules governing actions for damages for infringements of both EU and national competition law rules is overall a very commendable attempt to facilitate private damages actions while seeking to maintain a balance between the right of victims to obtain redress, the rights of defence and the need to avoid deterring leniency applications, which are the main discovery and investigation tool in the antitrust authorities’ arsenal to fight cartels. The focus of the Directive is indisputably follow-on cartel damages actions. Where the proposal seeks to introduce new rules rather than repeating already established principles, over half of these new rules are of interest only for cartel damages actions. Yet the Directive’s other provisions will clearly also benefit victims of alleged abuses of dominance. Consider Article 9 of the draft Directive, which extends the Masterfoods doctrine as enshrined in Article 16 of Regulation 1/2003 to decisions by national competition authorities;10 or Article 10 on limitation periods; or Article 16(2) on quantification of harm, which emphasizes the importance of the principle of effectiveness for this topic and which seeks to confer on those national courts which do not have it already the power to estimate the amount of harm where it cannot be proven to the standard of certainty demanded by domestic tort liability rules. But by far the most important rules of the proposed Directive are contained in Chapter II on the disclosure of evidence. The system of court-ordered disclosure that the Directive would require Member States to introduce in their procedural rules constitutes a true revolution for many if not most Continental legal regimes.11 The possibility to force the allegedly dominant defendant (as well as third parties) to divulge information is crucial to establish some balance and equality of arms in damages actions for abuse of dominance. The dominant incumbent has – by definition – the most comprehensive and long-ranging information on the market it lords over; only the incumbent knows the true extent and scope of the practices uncovered and challenged by the plaintiffs; only it has the necessary cost information required to run many of the analyses described in the Commission’s Guidance Paper, not to mention the information required to run complex analyses 9 Other than to note with approval the fact that, by abandoning earlier ideas to seek to introduce some unconvincingly watered-down form of opt-out class actions in the European judicial landscape, the Commission has been able to concentrate its efforts in coming up with a set of balanced and well thought-out provisions to promote private enforcement which, if adopted in the form proposed by the Commission, will have a profound impact on antitrust damages claims in the EU. 10 This is potentially very important for damages actions for abuse of dominance, given that, as mentioned above, national authorities tend to focus more on dominance cases than does the Commission. 11 While the proposed Directive would thus aim to significantly increase plaintiffs’ right to evidence, the detailed rules it seeks to introduce are clearly intended to strike the right balance between the need for defendants to contribute to the establishment of relevant facts and the need to avoid the excessive costs and many drawbacks of US-style pre-trial discovery.

Promoting Private Antitrust Enforcement: Remember Article 102 

91  

to quantify harm. The ability to demand disclosure from third parties, and in particular from the dominant company’s customers, can also be extremely useful to check the accuracy of the dominant company’s factual submissions. Conversely, the dominant company will also have the possibility to demand information from the plaintiffs where it feels that they are understating their market performance. Most of this information may very well qualify as business secrets, normally immune from disclosure. While many years will be necessary to see how this plays out, Article 5(4) of the proposed Directive hints that business secrets will no longer constitute an absolute bar to disclosure.12 At this point, the main criticism one could make of the proposed Directive, from the perspective of what it fails to do to facilitate private enforcement in the abuse of dominance field, is that its scope is limited to ex post damages actions, and it does not extend to injunction or interim actions meant to stop harm from occurring. In the light of the duration, complexity and cost of an antitrust damages action, one might have hoped that the Commission would perceive the interest in promoting cheaper, more timely action. However, the Communication and Staff Working Document (SWD) on quantifying harm deserve praise for their contribution to the promotion of damages actions for abuse of dominance. Both documents stress the fact that the often complex analyses described in the SWD are not required or possible in many instances, and that it can be perfectly acceptable for a judge to base a ruling on simpler evidence. While the proposed Directive fails to introduce a presumption of harm similar to that envisaged for cartel damages claims, Article 16(2) of Regulation 1/2003 serves as a useful reminder to judges that the difficulties inherent in demonstrating the quantum of damages cannot result in the refusal to grant any redress. The SWD is also much more balanced than previous efforts by the Commission or its consultants, in terms of the discussion of quantification techniques in relation to both cartel cases and abuse of dominance cases. The document includes a specific chapter on techniques for assessing harm from exclusionary practices that could prove very useful to plaintiffs and courts, in particular where the use of the proposed new rules on disclosure of evidence may allow economic advisers and court appointed experts to gather sufficient data to run complex analyses.

The way ahead: Towards more practical guidance Compared with 2011, it is clear that the Commission has taken to heart the need not to forget about the private enforcement of Article 102. The Commission must 12 ‘Member States shall ensure that national courts have at their disposal effective measures to protect confidential information from improper use to the greatest extent possible whilst also ensuring that relevant evidence containing such information is available in the action for damages.’

92  Fred Louis be commended for this valiant attempt at a more balanced approach away from the near-exclusive attention devoted for many years to the promotion of private enforcement through follow-on cartel damages actions. Yet much remains to be done. The current package does not offer much help (except as concerns the crucial step of obtaining the necessary evidence) to plaintiffs seeking to demonstrate that the defendant is dominant and that it has committed an abuse of its dominance. It gives no help at all to plaintiffs seeking injunctive or interim relief, and it is still short on specifics and acceptable shortcuts where the evidence simply is not there to run complex econometrics. Rome wasn’t built in a day, however, and the time is not ripe, nor is experience sufficient to expect much more from the Commission in terms of a binding set of additional rules. Yet courts and litigants would certainly welcome further guidance and proactive assistance by the Commission, which could take the form of a guidance document, the sharing of best practices across Member States, and providing a discussion of how the principle of effectiveness can concretely help solve some of the most typical issues raised by every step in the private enforcement of Article 102 (showing dominance, demonstrating an abuse, and quantifying the resulting harm). Examples of what such a guidance document could contain might be: • A clarification as to the interaction between the case law and the Guidance Paper on enforcement priorities. • A confirmation that while the Commission, for the purposes of sorting out its own enforcement priorities, goes beyond market shares in its analysis of dominance, in private enforcement cases, once plaintiffs have shown that the allegedly dominant firm has very high market shares, the burden of proof is on the allegedly dominant firm to demonstrate specific circumstances to challenge the preliminary evidence of dominance provided by market shares. • A discussion of the role of legal certainty as a defence against significant fines as opposed to as a defence against damages actions, in particular in legal systems that do not necessarily consider that a violation of the antitrust rules automatically gives rise to a claim in damages. • A description of available techniques (use of independent experts, data rooms, etc.) to arrive at a sufficiently precise determination of facts where no reliable public sources exist, balancing the right to court-ordered disclosure of evidence proposed by the draft Directive with the protection of business secrets. • A more targeted list of the type of support that the Commission and NCAs could lend to courts faced with private litigation. Given the limitations of the preliminary reference procedure for providing effective assistance to national courts struggling with complex factual and economic assessments, a more proactive use of amicus curiae interventions by the Commission in Article 102 private enforcement would bring more clarity and harmonized enforcement, faster than any other Commission initiative could hope to achieve.

Promoting Private Antitrust Enforcement: Remember Article 102 

93  

• A discussion of the fact that, in many cases, competition in the EU would be better served by swift, injunctive relief rather than drawn-out costly damages litigation, including a discussion of the implications of the principle of effectiveness for legal systems which subject the availability of injunctive relief to the non-availability of damages or which request significant security deposits from the plaintiff. The list has no pretention whatsoever of being exhaustive. Not all of these examples are specific to Article 102 private enforcement, and the requested guidance could be part of a wider document. However, the more ambitious the scope of the document, the longer it risks being delayed and the higher the risk that essential messages become diluted. Given the draft Directive’s continued focus on follow-on cartel claims, it does not seem too much to ask of the Commission to devote more attention to Article 102 private enforcement in the coming months and years. At the Workshop in 2011, I argued that the energy of the regulator would be better spent in developing this kind of very concrete guidance than in searching to devise the best system for collective redress actions in the absence of significant experience of such actions in the Member States. The Commission seems to have understood this message. It is to be hoped that this new approach will pay off and that the proposed Directive will be adopted swiftly, without significant amendments to its current well thought-out and carefully balanced contents.

Jochen Burrichter* and Enno Ahlenstiel**

Integrating Public and Private Enforcement in Europe Legal and Jurisdictional Issues – The German Perspective

I. Introduction ‘Public enforcement’ is generally understood to refer to proceedings of competition authorities concerning violations of Article 101 and Article 102 TFEU. ‘Private enforcement’ with regard to Articles 101 and 102 concerns different types of actions such as actions for injunctions and damage claims, especially damage claims in the context of violations of Article 101 (‘cartel damage claims’). In Germany there have been several cartel damage claims in recent years. As these claims were closely connected to fining decisions of competition authorities, it will be analyzed hereafter how and to which extent public and private enforcement are integrated in Germany. Furthermore, jurisdictional issues arising in the context of cartel damage claims will be discussed. In 2005, the German law concerning cartel damage claims was significantly amended in favour of private enforcement. Sections 33(3) and 33(4) of the Act against Restraints of Competition1 (‘ARC’) now provide the following: (3) Whoever intentionally or negligently commits an infringement [of Article 101 or 102 TFEU] shall be liable for the damages arising therefrom. If a good or service is purchased at an excessive price, damages shall not be excluded on account of the resale of the good or service. The assessment of the size of the damage […] may take into account, in particular, the proportion of the profit which the undertaking has derived from the infringement. […] (4) Where damages are claimed for an infringement of a provision of [Article 101 or 102 TFEU], the court shall be bound by a finding that an infringement has occurred, to the extent such a finding was made in a final decision by the cartel authority, the Commission of the European [Union], or the competition authority – or court acting as such – in another Member State of the European Community. The same applies to such findings in final judgments resulting from appeals against decisions pursuant to sentence 1. […]’2 * Jochen Burrichter is a partner in the Düsseldorf and Brussels offices of Hengeler Mueller. E-mail: [email protected] ** Enno Ahlenstiel is a senior associate in the Düsseldorf office of Hengeler Mueller. E-mail: enno. [email protected] 1 Gesetz gegen Wettbewerbsbeschränkungen (GWB). 2 Federal Law Gazette (BGBl.), published on 20 July 2005, Part I. 2114, 2121.

96  Jochen Burrichter and Enno Ahlenstiel In the meantime, several important cartel damage claims have been filed. All of these claims are so-called ‘follow-on’ damage claims, i.e. they are based on or associated with public enforcement in cartel cases.

II. Damage claims under section 33 ARC 1. Types of damage claims under section 33 ARC Three types of cartel damage claims under section 33 ARC may be distinguished. Claims can be filed completely independent of proceedings by competition authorities. However, such claims have hardly been filed in the past and will therefore not be discussed in detail below. Practically all damage claims under section 33 ARC are follow-on damage claims. Follow-on cartel damage claims can be filed before there is a final decision by a competition authority or a court concerning the cartel, e.g. they can be based on a decision by the European Commission against which the alleged cartelists have appealed. In this case the plaintiff will have to prove an infringement by the cartelist, i.e. the defendant. However, the contested decision by a competition authority can serve as circumstantial evidence. Furthermore, follow-on cartel damage claims can be filed based on a final decision by a competition authority or a court. In this case the plaintiff does not have to prove an infringement of competition law by the defendant, but can rely on section 33(4) ARC. According to section 33(4) ARC the court before which a cartel damage claim is brought is bound by a finding that an infringement has occurred to the extent such a finding was made in a final decision by a competition authority or a court. In such cases the plaintiff does not have to provide any further evidence for the cartel. However, the binding effect of section 33(4) ARC is limited to the operative parts of the decision, i.e. the finding that a certain undertaking violated Article 101 TFEU and the facts covered by the operative parts. This usually concerns the following facts: (i) undertakings involved in the cartel; (ii) duration of the cartel; (iii) affected product; (iv) type of illegal conduct. The binding effect does not extend to the efficiency of the cartel (i.e. the effect on the market), causality between the infringement and a loss suffered by the plaintiff, damage caused, or fault. The exact extent of the binding effect has not yet been decided by the German Federal Court of Justice3 and is subject to debate in legal literature.4 Bundesgerichtshof (BGH). Rehbinder in Loewenheim/Meessen/Riesenkampff, Kartellrecht, 2nd edn 2009, § 33 GWB para 54; Emmerich in Immenga/Mestmäcker, GWB, 4th edn 2007, § 33 para 78; Bornkamm in Langen/ Bunte; Kartellrecht, 11th edn 2011, § 33 GWB para 136 and 144. 3 4

Public and Private Enforcement in Europe – German Perspective 

97  

2. Conditions of section 33 ARC For all types of cartel damage claims under section 33 ARC, the plaintiff has to prove intent or negligence on the side of the defendant concerning the infringement of competition law, causality between the infringement and damages and evidence allowing an assessment of the amount of damages suffered by the plaintiff (‘quantum’). However, section 33(3) ARC and its interpretation by the courts facilitate proving these conditions by the plaintiff.

a) Intent or negligence If there is a final decision by a competition authority or court, this decision will provide strong circumstantial evidence that the defendant acted intentionally or negligently. However, as already mentioned above the final decision is not binding in the strict legal sense of section 33(4) ARC in this regard. Generally the threshold for negligence is low in German law. The defendant will be found to have acted negligently if he failed to exercise reasonable care. This can already be the case if the cartelist was not sure whether his behaviour was legal but decided not to obtain a legal opinion.

b) Causal link With regard to the causal link between the infringement and the damage suffered by the plaintiff, the plaintiff can rely on prima facie evidence. This prima facie evidence is based on the fact that cartels serve the purpose to increase the gains of the cartelists.5 As a result customers, i.e. the plaintiff, suffer a corresponding loss. Only if the defendant is able to rebut the prima facie evidence will the plaintiff have to provide detailed evidence concerning the causal link.

c) Damage Finally, the calculation of the amount of damage suffered by the plaintiff (‘quantum’) is also facilitated. According to section 33(3) ARC and section 287 of the Code of Civil Procedure6 the amount of the damage can be estimated by the court. As a result the plaintiff does not have to prove the damage suffered in a strict sense and in detail. It is sufficient if the plaintiff provides evidence allowing an estimate of the damage suffered. Furthermore, section 33(3) ARC provides that when assessing the damage, the profit which the defendant has derived from the 5 BGH, judgment of 28 June 2005, WuW/E DE-R 1567, 1569 – Berliner Transportbeton I; BGH, judgment of 10 April 2008, WuW/E DE-R 2225, 2229 – Papiergroßhandel; Higher Regional Court of Berlin (KG), judgment of 1 October 2009, WuW/E DE-R 2773, 2777 – Berliner Transportbeton. 6 Zivilprozessordnung (ZPO).

98  Jochen Burrichter and Enno Ahlenstiel infringement can be taken into account. Evidence on or indication of the profit derived can sometimes be found in the (final) decision by a competition authority or a court. Another possibility is to provide the court with the opinion of an economist who may be able to calculate the profit (or even the damage suffered) by use of economic methods.

d) Passing-on defence Concerning the cartelist’s defence against a damage claim, section 33(3) ARC is, once again, an important provision. It expressly provides that if a good or service is purchased at an excessive price, damage shall not be excluded on account of the resale of the good or service. However the exact interpretation and meaning of section 33(3) ARC is highly disputed. (See section V below.)

3. Conclusion on damage claims under section 33(4) ARC Follow-on damage claims are facilitated both by sections 33(3) and 33(4) ARC, and by the application of the law by the courts. Therefore, the German rules on cartel damage claims serve to integrate public and private enforcement as the plaintiff can make use of the results of public enforcement for his private damage claim.

III. The leniency applicant in follow-on cases A crucial issue regarding the integration of public and private enforcement is whether the success of leniency programmes is endangered by private enforcement. The leniency programmes of both the European Commission and the German Federal Cartel Office (i.e., the Bundeskartellamt) provide for full immunity for the first undertaking cooperating with the authorities and a reduction of the fine by up to 50% for the other undertakings cooperating later. As part of their cooperation leniency applicants have to admit that their behaviour constituted a violation of competition law.

1. Cartel damage claims against leniency applicants Based on the admission and other information provided by the leniency applicant as well as the results of the competition authority’s investigation, the competition authority will issue a decision. This decision will also be addressed

Public and Private Enforcement in Europe – German Perspective 

99  

to leniency applicants qualifying for a reduction of up to 50% only. Furthermore the Commission also addresses its decisions to undertakings qualifying for full immunity,7 whereas the Federal Cartel Office does not issue a decision against an undertaking enjoying full immunity. Plaintiffs can use decisions addressed to leniency applicants in cartel damage claims. Especially final decisions within the meaning of section 33(4) ARC make leniency applicants ‘easy to sue’, as the finding that an infringement has occurred is binding for the civil court in a cartel damage case. Furthermore, with regard to whether the leniency applicant acted intentionally or negligently the plaintiff will probably also be able to refer to a corresponding admission by the leniency applicant. In many cases, the decision against the leniency applicant will become final much earlier than the decisions against the other undertakings appealing their decisions. Leniency applicants qualifying for full immunity do not have to pay a fine and therefore normally have no legitimate reason to appeal the decision unless they can argue that the facts stated in the decision are wrong and would therefore be detrimental in cartel damage claims against them. Leniency applicants who received a reduction of up to 50% will also find it difficult to appeal the decision as they have already admitted to their involvement in the cartel so that they can hardly dispute the facts of the case.

2. Isolated cartel damage claims against leniency applicants Based on a final decision against the leniency applicant, the plaintiff can single out the leniency applicant and sue him for the whole damage suffered including damages caused by the other cartelists. It is not necessary for the plaintiff to be a customer of the leniency applicant. A plaintiff could sue only the leniency applicant for the whole damage suffered (‘isolated cartel damage claim’) even if the plaintiff was only supplied by another cartelist, as all of the cartelists including the leniency applicant are jointly and severally liable. Currently, there is a pending case before the Regional Court in Düsseldorf8 where the plaintiff is suing only the leniency applicant. This isolated cartel damage claim concerns the carglass cartel against which the Commission imposed fines of over 1.3 billion euros in November 2008.9 In this case, the Commission fined Asahi, Pilkington, Saint-Gobain and Soliver for illegal market sharing and an exchange of commercially sensitive information regarding deliveries of carglass in the EEA. Between early 1998 and early 2003 these companies allegedly discussed target prices, market sharing and customer allocation in a series of meetings and 7 See Press Release IP/10/1487 of 9 November 2010 (Airfreight); Press Release IP/10/1297 of 30 June 2010 (Prestressing Steel) (corr. 6 October 2010). 8 Landgericht Düsseldorf (LG Düsseldorf). 9 Press Release IP/08/1685 of 12 November 2008 (Carglass).

100  Jochen Burrichter and Enno Ahlenstiel other illicit contacts. The fine against Asahi was reduced by 50% to a total of 113.5 million euros, as Asahi provided additional information to help the Commission to investigate and prove the infringement. While the decision against the leniency applicant Asahi is final, the three other undertakings appealed to the European General Court.10 Therefore, the only final decision within the meaning of section 33(4) ARC is the decision against Asahi. Based on this decision the German (car) insurance company HUK Coburg brought an isolated cartel damage claim only against Asahi. HUK Coburg filed its claim in December 2010. As of this writing, the case is still pending before the Regional Court in Düsseldorf.

3. Cartel damage claims against leniency applicants: A risk to leniency programmes? As outlined above, leniency applicants including those receiving full immunity are especially ‘easy to sue’ in Germany because decisions against them normally become final and thus binding for German courts much earlier than in other cases. This can result in a conflict between public and private enforcement. If a company only looks at the potential fines which can be imposed by competition authorities a leniency application will be favourable in many cases unless the risk that a violation of competition law will be uncovered is low or the violation will fall under the statute of limitations soon. However, due to section 33(4) ARC, companies will have to also consider the probability of damage claims. If there is a high risk that customers might claim substantial damages exceeding the reduction of the fine by way of a leniency programme, an undertaking might refrain from applying for leniency because of the risk of private enforcement. This endangers the effectiveness of public enforcement, as leniency programmes in the meantime are the most efficient tool to uncover illegal cartels. However, it has to be pointed out that, at least for companies qualifying for full leniency, the risk of damage claims is significantly lower in proceedings by the Federal Cartel Office, as the Federal Cartel Office does not issue a decision against such companies.

4. Settlements Both in proceedings before the Commission as well as before the Federal Cartel Office, undertakings (including leniency applicants qualifying for a reduction of up to 50%) can try to reach a settlement with the competition authority which would lead to a reduction of the fine by 10%. Settlement talks also offer the chance to discuss with the competition authority – at least with the German Bundeskartellamt 10 Case T-68/09, Soliver, 2009 OJ C90/32; Case T-72/09, Pilkington, 2009 OJ C102/24; Case T-73/09, Saint-Gobain, 2009 OJ C102/25.

Public and Private Enforcement in Europe – German Perspective 

101  

– whether the extent of the infringement (e.g. concerning products and duration) can be narrowed down. As a result the scope of cartel damage claims could also be limited. Furthermore, in case of a settlement the Commission and the Federal Cartel Office will only issue a curtailed decision which contains much less information which a plaintiff could use in a cartel damage claim as opposed to a full decision.

IV. Access to file in follow-on cases Another area concerning the integration of public and private enforcement is access to the file of a competition authority. In many cases cartel damage claims could be much facilitated if the plaintiffs had access to the file of a competition authority so that they can use evidence from the file.

1. German rules on access to file In Germany, section 406e, paragraph 1 of the Code of Criminal Procedure11 provides that an attorney may inspect, for the aggrieved person, the files that are available to the court or the files that would be required to be submitted to it and may inspect officially impounded pieces of evidence, if he can show legitimate interest in this regard. Inspection of the files shall be refused only if overriding interests worthy of protection, either of the accused or of third persons, constitute an obstacle thereto. It may be refused if the purpose of the investigation, also in another criminal proceeding, could be jeopardized. Notwithstanding the interpretation of these exceptions a potential plaintiff would generally have access to a file of the Federal Cartel Office under section 406e, paragraph 1 of the Code. However, according to the Federal Cartel Office’s leniency programme, where an application for immunity or reduction of a fine has been filed, the Federal Cartel Office shall use the statutory limits of its discretionary powers to refuse applications by third parties for file inspection or the supply of information, insofar as the leniency application and the evidence provided by the applicant are concerned.12 Therefore, it is unclear whether a victim of the cartel would be granted access to the file with regard to information provided by the leniency applicant under section 406e, paragraph 1 of the Code.

Strafprozessordnung (StPO). Notice No. 9/2006 of the Bundeskartellamt on the immunity from and reduction of fines in cartel cases – Leniency Programme – of 7 March 2006, para 22. 11

12

102  Jochen Burrichter and Enno Ahlenstiel

2. The Pfleiderer case The leading case with regard to access to file in Germany is the Pfleiderer case, which was decided by the European Court of Justice on 14 June 2011.13 This case concerned a reference for a preliminary ruling from the Local Court in Bonn.14 The Federal Cartel Office had imposed fines amounting in total to 62 million euros on three European manufacturers of decor paper in January 2008. After the conclusion of the procedure, Pfleiderer – a customer of decor paper – submitted an application to the Federal Cartel Office in February 2008 seeking full access to the file relating to the imposition of fines in the decor paper sector, in order to prepare civil actions for damages. In May 2008 the Federal Cartel Office replied to the application for access to the file by sending three decisions imposing fines, from which identifying information had been removed, and a list of the evidence recorded as having been obtained during the search. Pfleiderer then sent a second letter to the Federal Cartel Office expressly requesting complete access to the whole file, including documents relating to leniency applications including pieces of evidence. In October 2008 the Federal Cartel Office partly rejected the application and restricted access to the file to a version from which confidential business information, internal documents and documents covered by the Federal Cartel Office’s leniency programme had been removed, and refused access to the evidence which had been seized. Pfleiderer thereupon brought an action before the Local Court in Bonn challenging the decision. The Local Court in Bonn referred this case to the ECJ concerning the question whether the provisions of European competition law are to be interpreted in such a way that parties adversely affected by a cartel may not, for the purpose of bringing civil-law claims, be given access to leniency applications or to information and documents voluntarily submitted.15 The ECJ acknowledged that leniency programs are a useful tool to uncover and bring infringements of competition law to an end. Furthermore, the effectiveness of those programmes could be compromised if documents relating to a leniency procedure were disclosed to persons wishing to bring an action for damages as a person involved in an infringement, faced with the possibility of such disclosure, could be deterred from using the opportunity offered by such leniency programmes. However, the right to claim damages strengthens the effectiveness of the EU competition rules and discourages agreements or practices which restrict or distort competition. From that point of view, actions for damages before national courts can significantly contribute to maintaining effective competition in the EU.16 The ECJ decided that EU competition law does not preclude a person who has been adversely affected by an infringement of competition law and is seeking to Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161. Amtsgericht Bonn (AG Bonn). 15 Pfleiderer, cited above note 13, at paras 9 et seq. 16 Ibid, paras 25 et seq. 13 14

Public and Private Enforcement in Europe – German Perspective 

103  

obtain damages from being granted access to documents relating to a leniency procedure.17 It is, however, for the national courts to determine the conditions under which such access must be permitted or refused by weighing the conflicting interests.18 Subsequently, the Local Court in Bonn decided in 2012 that third parties do not have access to leniency applications if this would endanger the success of leniency programmes in the future. However, the Local Court in Bonn granted access to the Federal Cartel Office’s file including exhibits.19 This decision has not been appealed and thus is final. Similarly, in cartel proceedings concerning German coffee roasters, the Higher Regional Court in Düsseldorf found that third parties only had access to the fining decisions redacted for business secrets and personal data (as well as documents seized during the dawn raids), but not the leniency applications.20

3. Access to file in the future It can be expected that competition authorities will continue to protect their leniency programmes by arguing that their interests would be endangered if documents relating to a leniency procedure were provided to a person seeking damages. For example, the Commission has confirmed in the oral hearing in the CDC Hydrogene Peroxide v Commission case that the Commission is not willing to grant potential plaintiffs access to parts of its file concerning the leniency applicant.21 It can be expected that access to files of competition authorities will play a crucial part in the integration of public and private enforcement in the future. However, the effectiveness of leniency programmes could be endangered if potential plaintiffs are granted full access, as cartelists might refrain from cooperating with a competition authority. They may be wary of facilitating damage claims against them by submitting leniency applications and relating documents to competition authorities.

Ibid, paras 30 et seq. Advocate General Mazák had suggested that voluntary self-incriminating statements must not be disclosed to third parties as such a disclosure may prevent cartelists from applying for leniency. All other pre-existing documents which form part of the company’s submission should be disclosed, with the exception of business secrets and other confidential, internal information. Opinion of the Advocate General in Pfleiderer, cited above note 13. 19 Local Court Bonn, judgment of 18 January 2012, WuW/E DE-R 3499 – Pfleiderer II. 20 Higher Regional Court of Düsseldorf, judgment of 5 October 2012, WuW/E DE-R 3662 – Kaffeeröster. 21 Case T-437/08, CDC Hydrogene Peroxide/Commission, report of the oral hearing on 14 June 2011. See also Case T-437/08, CDC Hydrogene Peroxide v Commission, judgment of the General Court of 15 December 2011, not yet reported. 17 18

104  Jochen Burrichter and Enno Ahlenstiel

V. Passing-on defence The passing-on defence concerns the question whether the defendant can argue that the plaintiff suffered no damage because he was able to include the overcharge resulting from the cartel in his resale prices.

1. Case law on the passing-on defence The German Federal Court of Justice decided in the ORWI case that an indirect purchaser from the participants in a cartel is entitled to claim damages if he has suffered a loss as a result of the conduct in breach of cartel law. The burden of rendering evidence and proof lies with the indirect purchaser.22 The benefit that the injured party acquires by passing on the cartel-related price surcharge to his purchasers may have to be taken into account as a set-off of benefits. However, the burden of proof with regard to the passing on defence generally lies with the liable party. The case was decided under the law as it stood before Section 33(3) ARC entered in force.

2. Legal literature on the passing-on defence Section 33(3) ARC provides that damage shall not be excluded on account of the resale of goods or services if these have been purchased at an excessive price caused by a cartel. In legal literature that pre-dated ORWI, this was partly understood to exclude the passing-on defence. However, the extent to which the passing-on defence in fact is excluded by section 33(3) ARC is subject to controversial debate in legal literature.23 Some authors argue that the passing-on defence is completely excluded. Others argue that section 33(3) ARC simply shifts the burden of proof to the defendant who can still try to prove that the loss suffered by the plaintiff was reduced by the profit he gained from sales.24 Furthermore, the details with regard to the passing-on defence (e.g. the issue multiple cartel damage claims by direct and indirect customers) are still under discussion in legal literature. After the ORWI judgment, Section 33(3) must be understood consistently with the approach of the Federal Court of Justice. All in all, German law facilitates private enforcement and thus enhances the integration of public enforcement by at least making the passing-on defence difficult. BGH, judgment of 28 June 2011, WuW/E DE-R 3431 – ORWI (see also IIC 2012, 487–497). Jaeger in Loewenheim/Meessen/Riesenkampff, Kartellrecht, 2nd edition, 2009, § 33 GWB paras. 49 et seq. 24 Bechtold, GWB, 6th edition, 2010, § 33 para 28; Bornkamm in Langen/Bunte; Kartellrecht, 11th edn 2011, § 33 GWB paras 113 et seq and 120; Lübbig in Hirsch/Montag/Säcker, Kartellrecht – Band 2, 2008, § 33 GWB para 96 et seq. 22 23

Public and Private Enforcement in Europe – German Perspective 

105  

VI. Jurisdictional issues concerning private enforcement The most important jurisdictional issues concerning private enforcement are jurisdiction and forum shopping, respectively, and the bundling of claims. These issues are discussed below.

1. Jurisdiction In most cartel damage claims not only one jurisdiction is affected, but usually several jurisdictions. The reason for this is that either the plaintiff was affected by the cartel in different jurisdictions or the cartelists are located and/or have acted in different jurisdictions. If the plaintiff wants to file a damage claim against a defendant located and/or having acted in another Member State, Regulation 44/200125 applies. According to Article 5(3) of that Regulation, a person (where ‘person’ can mean an undertaking) based in one Member State may be sued in matters relating to tort in the courts competent for the place where the harmful event occurred. This refers both to the place where the person is established as well as the place where the damage was caused. However, if the plaintiff chooses the latter place, only the damage suffered in that jurisdiction can be claimed. Under Article 5(3) of the Regulation, the overall damage can only be claimed in the jurisdiction where the person is established.26 Cartel damage claims are matters relating to tort injury. Thus, Article 5(3) applies. Another possibility is to sue a cartelist for the whole damage on the basis of Article 6(1) of Regulation 44/2001. According to this provision a person may also be sued where he is one of several defendant cartelists, in the courts for the place where any one of them is domiciled, provided the claims are so closely connected that it is expedient to hear and decide them together in order to avoid the risk of irreconcilable judgments resulting from separate proceedings. In Germany cartelists are jointly and severally liable for a violation of competition law. Article 6(1) thus applies, in practice, to cartel damage claims. Consequently, if at least one of the cartelists is established in Germany, a plaintiff can sue all members of the cartelists for the whole damage, irrespective where they have their seat.

2. Forum shopping A cartel damage case concerning forum shopping was brought by CDC against the members of the hydrogen peroxide cartel before the Regional Court in Dortmund.27 25 Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters. 26 Case C-68/93, Fiona Shevill, Ixora Trading Inc, Chequepoint SARL and Chequepoint International Ltd v Presse Alliance SA [1995] ECR I-415. 27 Landgericht Dortmund, reference 13 O 23/09 (Kart).

106  Jochen Burrichter and Enno Ahlenstiel The case is based on a decision by the Commission. The Commission found that several producers of hydrogen peroxide participated in a cartel covering the whole territory of the EEA between 1994 and 2000.28 CDC bundled claims of 32 companies of the pulp and paper industry which purchased hydrogen peroxide for a total of 94 production sites located in 13 different European countries during the cartel period. The only defendant based in Germany was Evonik Degussa GmbH. However, by virtue of Article 6(1) of Regulation 44/2001, this ‘anchor defendant’ was sufficient to justify the inclusion of defendants based in other European jurisdictions in the same suit for damages. CDC later settled with the ‘anchor defendant’, Degussa Evonik GmbH, leaving no German-based defendant in the case. Nonetheless, the cartel damage claim against the other defendants is still admissible, as the rules on perpetuatio fori apply, meaning that a competent court does not lose its competence if the facts on which its competence was originally based change at a later stage.

3. Bundling of claims There are no special rules on collective redress in Germany. However, it is admissible under German law to bundle claims from several companies damaged by a cartel. One way to bundle claims is to convey them to a third party acting as a plaintiff in a cartel damage case. The leading case in Germany is a damage claim brought by CDC29 against numerous cement producers before the Regional Court in Düsseldorf.30 The case is based on a decision by the Federal Cartel Office which found that these cement producers had divided the German cement market between themselves, agreed on sales quotas, and fixed prices from the beginning of the 1990s until 2002. The cement producers appealed against this decision. The Higher Regional Court in Düsseldorf31 has generally confirmed the finding of a cartel by the Federal Cartel Office.32 An appeal against the decision of the Higher Regional Court in Düsseldorf is still pending before the Federal Court of Justice.33 CDC originally initiated its cartel damage claim in August 2005. The defendants tried to obtain a suspension of the damage proceedings until the end of their appeals against the Federal Cartel Office decision. This request was rejected by the Higher Regional Court in Düsseldorf in May 2006.34 Furthermore, the defendants claimed that the cartel damage claim by CDC was not admissible under German Commission Decision 2006/903/EC of 3 May 2006 in Case COMP/38620 – Hydrogen Peroxide. www.carteldamageclaims.com. 30 Landgericht Düsseldorf, reference 34 O (Kart) 147/05. 31 Oberlandesgericht Düsseldorf (OLG Düsseldorf). 32 Higher Regional Court of Düsseldorf, judgment of 26 June 2009, reference VI-2a Kart 2 – 6/08 OWi, BeckRS2010, 04805. 33 Bundesgerichtshof (BGH). 34 Higher Regional Court of Düsseldorf, judgment of 3 May 2006, WuW/E DE-R 1755. 28 29

Public and Private Enforcement in Europe – German Perspective 

107  

procedural law, inter alia because of the bundling of claims. However, the Federal Court of Justice confirmed that the cartel damage claim was admissible.35 As of this writing, the Regional Court in Düsseldorf has not yet decided on the merits of the case. It may therefore be concluded that conveying claims to a single plaintiff is a feasible and efficient way of bundling claims in Germany.

4. Conclusion on jurisdictional issues Even though there are no specific rules on collective redress in Germany, bundling damage claims of different customers against different cartelists based in different jurisdictions is admissible and the bundling of such claims is an efficient way of private enforcement. Therefore, there are no significant jurisdictional issues preventing an efficient integration of public and private enforcement in Germany.

VII. Conclusions The integration of public and private enforcement is a complex issue. However, German law provides for integration especially by way of the binding effect of a final decision of a competition authority or a court provided for in section 33(4) ARC. Furthermore the possibility to estimate the damage suffered and the difficulty of sustaining the passing-on defence facilitate cartel damage claims and thus strengthen private enforcement. The downside of an integration of public and private enforcement and generally the strengthening of private enforcement is the negative effect on the willingness of cartelist to cooperate with competition authorities in the framework of leniency programmes. Therefore, any further integration of public and private enforcement has to strike a judicious balance between making private enforcement more efficient and avoiding any danger of making public enforcement less efficient by making leniency programmes less attractive for cartelists.

35

BGH, judgment of 7 April 2009, GRUR-RR 2009, 319.

Luís Silva Morais*

Integrating Public and Private Enforcement of Competition Law in Europe: Legal Issues

1. Introduction and global overview 1.1. In order to discuss the idea of a possible integration of public and private enforcement of competition law in Europe (particularly as concerns the enforcement of EU competition law), it is useful on the one hand to perceive how the idea of private enforcement of EU competition law has developed over recent years and, on the other hand, to try to arrive at a systematic and analytic understanding of ‘private enforcement’ in terms of its relationship to public enforcement of competition law, the latter having clearly been the cornerstone of the EU system of competition law.1 * Professor of Lisbon Law Faculty (FDL)/Competition law and EU law, Jean Monnet Chair (Economic Regulation in the EU), Vice-President of the European Institute of FDL (IE), and of the Institute of Economic Financial and Tax law of FDL (IDEFF) and Chairman of CIRSF – Research Centre on Regulation and Supervision of the Financial Sector (IE and IDEFF in Scientific Partnership with the Bank of Portugal and the Institute of Insurance of Portugal). Founder and Partner of ‘Luis Silva Morais – Law Firm, rl’, based in Lisbon, Portugal. All the views expressed in this chapter are the sole responsibility of the author. A draft version of this chapter was presented and discussed at the 16th Annual EU Competition Law and Policy Workshop on 17 June 2011, held at the EUI in Fiesole and organized by Philip Lowe and Mel Marquis. The initial draft was updated in light of selected subsequent developments, such as, eg, the Donau Chemie judgment of the European Court of Justice. However, the review of the initial version of the chapter was completed by the end of May 2013. The revised version therefore did not cover the Commission’s legislative proposal and soft law documents of 11 June 2013. Nevertheless, given the importance of this package of measures (including, inter alia, a proposed Directive on private antitrust damages actions and a non-binding and horizontal recommendation on collective redress mechanisms), a slight further revision of the chapter was made in order to include very brief references to these new developments. These references do not purport in any manner to enter into an in-depth analysis of such developments. On the other hand, the approach followed in the chapter purports to anticipate, in prospective terms and for the purposes of theoretical and critical debate, the possible reach of the said measures – especially soft law initiatives, which may represent a more realistic way to move forward in this domain given the current state of affairs in the EU and the various Member States. 1 For a general perspective on public enforcement as the cornerstone of the EU system of enforcement of competition law see inter alia, Karl Hofstetter and Melanie Ludescher, ‘Fines against Parent Companies in EU Antitrust Law: Setting Incentives for ‘Best Practice Compliance’’, 33 World Competition 55 (2010); Wouter Wils, ‘Should Private Antitrust Enforcement Be Encouraged in Europe?’, 26 World Competition 473 (2003); Wils, The Optimal Enforcement of EC Antitrust Law, Essays in Law and Economics, Kluwer, 2003; Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for Damages’, 32 World Competition 3 (2009). For an overall review of the prospects of developing private enforcement, essentially understood as compensation for damages arising from infractions of competition law, see Veljko Milutinovic, The ‘Right to Damages’

110  Luís Silva Morais That introductory overview will in turn lead us to perceive and critically debate three contrasting views that may currently be espoused in the context of the gradual, albeit so far very limited emergence of forms of private enforcement of competition law in the EU. These three perspectives are as follows: (i) A view sustaining a fully integrated framework of enforcement of competition law, which ties public and private enforcement systems together in a distinctively European way (ie, distinct from the American way). This view calls for a drastic review of the theories and/or foundations underlying the enforcement of competition rules.2 (ii) A view advocating a fully autonomous or independent system of private enforcement of competition law, largely based on adequate mechanisms that allow for collective redress for claimants with small and dispersed losses to recover damages for losses they have suffered on account of anticompetitive conduct. While admitting that such private enforcement based on collective redress may, in some limited forms, complement public enforcement of competition law (particularly as regards follow-on cases, to which I refer below), according to this view private enforcement anchored in collective redress is basically independent of enforcement by public bodies and requires no coordination with public enforcement.3 (iii) According to a third possible view, public enforcement of competition law should remain a prevailing feature of the EU system of enforcement of competition rules, and private enforcement should play a strictly complementary and subsidiary role; this will require some forms of coordination between the two areas, but the dominant role of the public sphere should be maintained, and no fundamental shift of the theories and/ or foundations underlying enforcement of competition rules is necessary. For the sake of clarity, I will disclose from the start that I basically follow the third, alternative view, for reasons that will be put forward throughout this chapter. 1.2. As is widely known, the Treaty on the Functioning of the European Union (TFEU) does not provide specific remedies for the breach of Treaty obligations. As regards competition law, the possible right to damages arising from competition law infractions that citizens may derive from EU law has been established or under EU Competition Law – from Courage v Crehan to the White Paper and Beyond, Wolters Kluwer, 2010, especially pp 14–25 and pp 47 et seq. 2 This view is supported by Christopher Hodges, among others. See, eg, Hodges, ‘European Competition Enforcement Policy: Integrating Restitution and Behaviour Control. An Integrated Enforcement Policy, Involving Public and Private Enforcement with ADR’, in 34 World Competition 383 (2011). 3 This view was expressed, for example (and curiously, coming from a public enforcer), in very peremptory terms, by the Irish Competition Authority in the context of its submission to the European Commission Public Consultation: Towards a Coherent European Approach to Collective Redress, 4 February 2011, SEC (2011)173 final.

Competition Law in Europe – Legal Issues  111   recognized by the European Court of Justice in a number of landmark rulings, notably Courage v Crehan4 and Manfredi.5 In these rulings – particularly in the latter one – the ECJ clearly stated the principle that any individual can claim compensation for the harm suffered where there is causal relationship between that harm and an agreement or practice prohibited under Article 101 TFEU. Furthermore, it established a principle of effectiveness in this domain, according to which, in the absence of relevant EU rules, it was 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 directly from EU law, provided, inter alia, that they do not render practically impossible or excessively difficult the exercise of rights conferred by EU law.6 Building on this case law, the Commission subsequently published a 2005 Green Paper on Damages Actions for Breach of EC Antitrust Rules7 and a 2008 White Paper for Damages Actions for Breach of the EU Competition Rules.8 The White Paper, which contemplated the adoption of guidelines by the Commission and the formal harmonization of national legislation to foster private enforcement of EU competition rules, even led to an informal draft text of a Directive which was circulated among certain stakeholders in 2009 but never published. In parallel, the Commission also put forward ideas and proposals in order to foster some forms of collective redress for consumers in the context of the internal market for the purpose of safeguarding rights derived from EU legislation on consumer protection. This ultimately led to the adoption of a 2008 Green Paper on Consumer Collective Redress.9 This Green Paper of 2008 did not address collective redress for victims of antitrust law infringements because of the allegedly specific nature of antitrust law. However, several stakeholders argued (and rightly so, in my view) that there were important connections between consumer protection cases and antitrust cases and that the Commission should therefore postpone any specific proposals based on the 2008 White Paper for Damages Actions until a more thorough and global evaluation of collective redress in the field of consumer protection could be completed. The underlying rationale – to which I subscribe, although the same options need not be applied in the two fields, given their differences – was essentially the following. If, in the end, political or legal reasons posed an obstacle to the further introduction or fostering of collective redress measures on the basis of EU initiatives, those reasons would apply in both fields. See Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297. See Joined Cases C-295 to C-298/04, Manfredi v Lloyd Adriatico Assicurazioni [2006] ECR I-6619. 6 This is essentially the wording of the Manfredi judgment, updated with the terminology of the Treaty of Lisbon. 7 Commission, Green Paper on Damages Actions for Breach of the EC Antitrust Rules, COM (2005) 672 (19 December 2005). 8 Commission, White Paper on Damages Actions for Breach of the EC Antitrust Rules, COM (2008) 165 final (2 April 2008). 9 Commission, Green Paper on Consumer Collective Redress, COM (2008) 794 (27 November 2008). 4 5

112  Luís Silva Morais It is thus submitted that, if the Commission takes the introduction of collective redress further, any such initiative should be designed to govern collective actions in both consumer protection and antitrust areas while safeguarding their undeniable particularities. Ultimately, and possibly in conjunction with the institution of a new College of Commissioners, this led to a parenthesis in advancing the damages actions and collective redress proposals between 2009 and 2010. The parenthesis was closed in February 2011 with the Public Consultation ‘Towards a Coherent Approach to Collective Redress’, which did in fact envisage a broader approach encompassing both competition law and consumer protection law.10 Somewhat unexpectedly, although it would have been reasonable at that stage to foresee expedited action on the part of the Commission, it took another two years for the Commission to finally present its long-awaited proposals in this domain on 11 June 2013, which included: (i) a draft Directive on private actions for damages for infringements of competition law, and (ii) a non-binding Recommendation on collective redress mechanisms (these instruments being complemented by a Communication of the Commission entitled ‘Towards a European Horizontal Framework for Collective Redress’, a Communication of the Commission on quantifying harm in actions for damages for breaches of Articles 101 and 102 TFEU and a Commission Staff Working Document on quantifying harm in such actions).11 1.3. Very briefly, since I do not intend to comment ex professo on these successive policy documents, it may be noted that the approach followed by the Commission has become more limited and, at the same time, more realistic from the 2005 Green Paper to the 2008 White Paper and to the 2011 consultation document, in terms that are particularly relevant to the fundamental problem of the possible interplay between the spheres of public and private enforcement of competition law. In fact, although the 2005 Green Paper acknowledged that (a) private enforcement (basically understood as private actions for damages arising from competition law infringements) served the same deterrence objective as public Cited above note 3. I refer here to Commission, Proposal for a Directive of the European Parliament and of the Council 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, COM (2013) 404 final (11 June 2013); Commission Recommendation on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law, C(2013) 3539/3 (11 June 2013); Proposal of Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of Regions – Towards a European Horizontal Framework for Collective Redress, COM (2013) 401/2 (11 June 2013); Communication from the Commission on quantifying harm in actions for damages based on breaches of article 101 or 102 of the Treaty on the Functioning of the European Union, 2013 C167/19, accompanied by Commission Staff Working Document – Practical Guide – Quantifying Harm in Actions for Damages Based on Breaches of Article 101 of 102 of the Treaty on the Functioning of the European Union, SWD (2013) 205 (11 June 2013). 10 11

Competition Law in Europe – Legal Issues 

113  

antitrust enforcement,12 the 2008 White Paper (b) clearly rejected that approach, replacing it with the idea of fundamentally separate tasks for public enforcement of competition rules and for private enforcement through private actions for damages (with deterrence being the main task assigned to public enforcement and compensation of victims being the overriding goal of private enforcement). To some extent, the 2005 Green Paper implicitly assimilated the perspective underlying US antitrust law of private damages actions as a key instrument of deterrence, which in the US is essentially due to the origin and history of US federal antitrust enforcement, with limited means allocated to public enforcement and limited fines (leading to what authors such as Wouter Wils have suggestively designated as a ‘deterrence gap’ that had to be ‘filled by follow-on treble damages actions’).13 This assimilation, in turn, made room (at least hypothetically) for the idea of adopting in Europe (even if somehow adapting it or limiting its scope) American-modelled legal instruments or options of, eg, treble damages or the exclusion of the passing-on defence in actions brought by direct purchasers (just to mention some of the most noteworthy among the array of options contemplated in the US legal system). Having clearly – and more realistically – separated the tasks of deterrence and compensation of victims, and assigning them respectively to public enforcement and private enforcement, the 2008 White Paper understandably rejects those more radical features of the US legal system.14 In that process, however – and coming now to the February 2011 Public Consultation15 – there is a fundamental problem of legal competence to act on the part of the EU that has been continuously overlooked (since any action in this field taking the form of new and binding legal instruments will require a demonstration that there is a proper basis in the EU Treaties). Theoretically, new EU measures in this field may be based on: (i) Article 114 TFEU – that is, if the lack of harmonized measures in the field of collective redress in antitrust cases (which I take here autonomously in consideration) implies obstacles to trade and to the proper functioning of the internal market; or (ii) Article 81 TFEU – which relates to developing judicial cooperation in cross-border matters. The point here is that it is not justifiable to try to delineate an ambitious and overreaching legal architecture in the EU with the purpose of inducing a qualitative transition in the patterns and intensity of private enforcement of competition rules without ascertaining, from the start, a proper normative basis and legal foundation for possible EU harmonization measures (which are to a large extent debatable). This is also an argument in favour of a more gradualist and incremental approach that I would tend to favour in this domain, building 12 See the very straightforward statements produced in the 2005 Green Paper, including for example this one: ‘[P]ublic and private antitrust enforcement […] serve the same aims: to deter anti-competitive practices’. Green Paper, page 3 (emphasis added). 13 See Wils, ‘Relationship’, cited above note 1, pp 3 et seq. 14 For a short summary of the more limited instruments ultimately contemplated in the 2008 White Paper, see Assimakis Komninos, ‘The EU White Paper for Damages Actions: A First Appraisal’, in Concurrences No 2 - 2008, pp 84 et seq, especially p 85. 15 Commission, Towards a Coherent European Approach to Collective Redress, cited above note 3.

114  Luís Silva Morais on the consolidation of the decentralized public enforcement of EU competition rules and on multiple soft law steps in order to promote a soft approximation of legal instruments which may sustain private enforcement actions and which already exist in several Member States (in a context in which the lack of private enforcement actions does not derive entirely from a complete absence of adequate legal instruments, but from other factors, including issues pertaining to different legal cultures and lack of awareness about the potentialities offered by existing legal instruments.16 These issues have to be tackled in a more incremental manner, as I will contend below (see section 6). 1.4.1. At this point, and to put into perspective the fundamental issues related to the possible integration of public and private enforcement of EU competition rules in the context of the multiple evolutions and initiatives oriented towards the fostering of private enforcement, it is important to identify exactly what may be designated as private enforcement of competition law. Such precision is necessary since the term ‘private enforcement’ in itself is a rather loose, and not a particularly accurate notion. As a former Director-General for Competition of the Commission and founder of this Workshop, Claus-Dieter Ehlermann, has put it, ‘the conviction that competition law enforcement was based almost exclusively or even mainly on administrative action is not enough. Like other rules, competition law will only be efficiently applied if there is also private action’.17 (emphasis added) What ‘private action’ specifically means in this legal context has frequently been very loosely understood, as ‘private enforcement’ tends to be used to encompass very diverse or even heterogeneous legal actions. Trying to make the notion more accurate, we may identify two main pillars for private enforcement of competition rules (in a context in which the whole idea of private enforcement has sometimes come to be almost entirely identified with the first pillar and even with some sub-categories comprised in that first pillar): (i) the first pillar encompasses actions brought in national courts by private parties that have suffered harm caused by infringements of EU competition rules, in order to seek indemnity or compensation for that harm; On this starting point (within this first pillar), one may further identify several sub-categories of legal actions, which include, in particular: (i-1) – Actions brought by private parties for damages caused by infringements of competition rules; these, in turn, include actions for damages brought after a finding of infringement by public authorities and agencies (follow-on 16 On the relevance of these issues pertaining to different legal cultures influencing the operation of various judicial systems, which have to be duly perceived or ascertained through empirical analysis, see Barry Rodger, ‘Editorial – Private enforcement and collective redress: the benefits of empirical research and comparative approaches’, 8 Competition Law Review 1 (2012). 17 Claus-Dieter Ehlermann, ‘Deregulation and Enforcement of Competition Laws’, in Proceedings of the Second Seminar on European Union/Japan Competition Policy, Luxembourg, 1995, pp 22 et seq, at p 26.

Competition Law in Europe – Legal Issues 

115  

actions) or actions brought independently of such finding of infringement in the context of public enforcement (stand-alone actions); (i-2) – Actions brought in national courts by private parties harmed or that may be harmed or somehow affected by competition law infringements for injunctive relief. (ii) the second pillar encompasses various situations in which private parties defensively invoke competition law prohibitions in private contractual or intellectual property litigation, either in courts or before arbitrators (a domain in which this line of action by private parties in litigation concerning big business transactions has considerably expanded over recent years – something that may frequently be overlooked, since arbitral awards are generally not made public).18 Given this variety of situations, one may argue that the interplay or connection between public enforcement and private enforcement of competition rules – the way they integrate or not, to put it in terms of the ‘leitmotif’ of the 16th EU Competition Law and Policy Workshop at the EUI (where the remarks made in this chapter were presented) – will have qualitative features that are distinctively different depending upon the pillar or subcategory of private enforcement actions that may be at stake. Going one step further, one may even argue that it is possible to autonomously identify a more limited, stricto sensu area of private enforcement of competition rules. This area would only include actions brought by private parties claiming (for several reasons and envisaging different legal purposes, from compensation for harm or the determination or conditioning of the way in which some business transactions may be construed or executed) infringements of competition rules that have not been, as such, the object of actions by public enforcers. In that sense, the stricto sensu private enforcement of competition rules should comprise standalone actions (referred to above in (i-1), in fine) and the array of legal actions corresponding to what we have referred to above (in (ii)) as the second pillar of private enforcement. 1.4.2. Again, in this field I would emphasize the potential importance – too frequently overlooked – of arbitration. Turning again to Claus-Dieter Ehlermann, he rightly anticipated, in the context of the post-2003 modernization of EU competition law enforcement, that ‘in future the [European] Commission will have to take a more positive stand towards arbitration, as this is a pre-condition for the success of the modernisation exercise’19 (emphasis added). I refer here, in 18 There is no room within the limited remit of this paper to further elaborate on these systematic subdivisions of private enforcement of competition rules and on a possible legal categorization of these different realities. See generally Assimakis P Komninos, EC Private Antitrust Enforcement – Decentralised Application of EC Competition Law by National Courts, Hart Publishing, 2008; Clifford Jones, Private Enforcement of Antitrust Law – in the EU, UK and USA, Oxford University Press, 1999. 19 See Claus-Dieter Ehlermann and Isabela Atanasiu, eds, European Competition Law Annual 2001: Effective Private Enforcement of EC Antitrust Law, Hart Publishing, 2003, at p 303.

116  Luís Silva Morais particular, to a growing coverage of competition law issues – on several accounts and not necessarily induced by public enforcement actions – in the course of arbitral procedures that are too frequently ignored or underestimated given the non-public nature of those proceedings.20 It is true that, in certain cases, arbitration may be linked to the public enforcement of competition rules (e.g, where arbitral clauses are used as a way to monitor commitments in the context of commitment decisions adopted under Article 9 of Regulation 1/200321). However, in a growing number of cases, self-assessment by undertakings of cooperation agreements and practices in the post-2003 normative framework has been inducing such undertakings to use arbitration procedures, concerning those business cooperative practices, to evaluate competition law issues that may be raised in connection with these practices. Also, moving beyond cooperation practices, and considering that potentially abusive behaviours are more publicly and openly carried out by undertakings with market power (in contrast with cartels or other related practices that are bound to maintain some level of secrecy), the growing awareness of the legal issues at stake and the relative scarcity of abuse of dominance cases initiated by public authorities (ie, either by the Commission or by national competition authorities) tend to work as a positive incentive for the parties to raise those issues in the context of arbitral procedures. Jurisprudential developments in this field – in particular, the well-known Eco Swiss and Van Schijndel judgments22 – also contribute to that competition law awareness of private parties and to a corresponding use of competition law arguments and tools that may be used in the course of arbitral procedures23 (either to discuss alternative ways in which certain business practices may or may not be implemented, or even to claim compensation for harm produced in the context of those business practices, particularly when those practices put undertakings with considerable market power in contact with undertakings with lesser power).24 In short, I am considering here potential developments as regards arbitration chiefly connected with what I have tentatively designated as the stricto sensu 20 On the growing relevance of arbitration in the field of competition law, see the 2010 OECD Hearings on Arbitration and Competition (with two important studies by Luca Radicati di Brozzolo and Laurence Idot), http://www.oecd.org/competition/abuse/49294392.pdf. See also the special thematic issue on arbitration and competition law, with articles by Laurence Idot, Christopher Hodges and Assimakis Kominos, published in Competition & Regulation Review, edited by IDEFF of Lisbon Law School and by the Portuguese Competition Authority, Nos 11–12, December 2012, and the accompanying editorial, co-authored by Luís Silva Morais and João E. S. Noronha. 21 See, eg, Wouter Wils, ‘Settlements of EU Antitrust Investigations: Commitment Decisions under Article 9 of Regulation 1/2003’, 29 World Competition 345 (2006). 22 See Case C-126/97, Eco Swiss v Benetton [1999] ECR I-3055; Joined Cases C-430 and C-431/93, Jeroen van Schijndel and Johannes Nicolaas Cornelis van Veen v Stichting Pensioenfonds voor Fysiotherapeuten [1995] ECR I-4705. 23 On the possible impact of those judgments, see, eg, Laurence Idot, Union Européenne, Arbitrage et ADR, Study 293, in Guy Canivet, Laurence Idot and Denys Simon, eds, Lamy procédures communautaires, Wolters Kluwer, 2005. 24 See further Laurence Idot, ‘Arbitration, Competition Law and Public Order’, in Competition & Regulation Review, Nos 11-12, December 2012, cited above note 20; Assimakis Kominos, ‘Arbitration and EU Competition Law’, ibid.; Luís Silva Morais and João E. S. Noronha, Editorial, ibid.

Competition Law in Europe – Legal Issues 

117  

private enforcement of competition rules (meaning private actions oriented towards the application of competition rules that are neither dependent on public enforcement actions nor directly related with them). Furthermore, I believe that some incremental measures or steps may reinforce the important status (albeit largely unperceived as such) of arbitration procedures in order to foster stricto sensu private enforcement of competition rules. Such measures include, eg, the encouragement, through soft harmonization impulses, of legislative changes of national laws on arbitration that give more weight to public policy considerations, including the safeguarding of competition law and principles, as an element of possible annulment or non-enforceability of arbitral awards, as has happened to some extent in the legislative reform in the field of voluntary arbitration which recently took place in Portugal and which I will briefly mention below in section 5.25 Another possible measure could be the extension or adaptation of Articles 15(1) and 16(2) of Regulation 1/2003 in order to cover information and opinions to be addressed to arbitral courts and a systematic effort to identify and collect arbitral awards in which competition law issues have been discussed and considered as a basis for the final decision.26

2. The basic foundations of the interplay between public and private enforcement of competition rules 2.1. After this attempt to achieve a systematic and analytic understanding of the private enforcement vis-à-vis the public enforcement of competition law, and considering the status of the debate (ie, following the February 2011 Public Consultation and the ongoing and rather protracted discussion on possible EU initiatives in this field (arising not only from the output of that Public Consultation but also from issues raised in subsequent ECJ case law, to which I will refer in 25 On this 2011 reform of the Portuguese Act on Voluntary Arbitration, see the comments below in Section 5 (although the reform could still have been more ambitious as regards public order or public policy considerations to ensure that, for purposes of review of arbitral awards, arbitral courts take into account, on their own initiative, EU competition law issues. 26 On this latter aspect one has to acknowledge the practical hurdles that stand in the way of a systematic effort to identify and collect arbitral awards in which competition law issues have been discussed and considered as a basis for the final decision, since the private parties involved in arbitral disputes very often do not favour the disclosure of such arbitral awards. From my own empirical experience as arbitrator, in an appreciable number of arbitration cases in which competition law issues are raised these end with an agreement or some form of settlement between the parties who do not wish to further pursue the discussion of those competition law problems (the contents of which are not to be disclosed). In spite of those practical hurdles, a systematic effort of collecting relevant information should be launched in this domain. In Portugal, the Institute of Economic Financial and Tax Law and the European Institute of Lisbon Law University (IDEFF and IE of FDL) have, eg, approached the Portuguese Association of Arbitration (APA), which has begun to create a database of arbitral awards, so that the APA could also collect and disclose, whenever the parties grant their consent, arbitral awards in the context of which competition law issues have been pondered or discussed.

118  Luís Silva Morais section 4), one should try, at a high level basis, to ascertain what possible role may be played by the private enforcement of competition rules, and what kind of intersections one may reasonably consider between the public and private spheres of enforcement.27 A sobering way to start that analytical exercise is to recall a powerful statement made by Richard Posner about the paradigm of private enforcement in the US legal system. As he has stated, ‘students of antitrust laws have been appalled by the wild and woolly antitrust suits that the private bar has brought – and won. […] in short, the influence of private action in the development of antitrust doctrine has been on the whole a pernicious one’.28 While I may not entirely subscribe to the vehemence of that statement, it represents, nevertheless, a powerful illustration (even as regards the US legal context, whose history and evolution is very different from those of the EU competition law system) of the inherent limitation of private enforcement to pursue common tasks together with the sphere of public enforcement. Clearly, different overriding goals of antitrust law and policy cannot be pursued simultaneously or with the same intensity level through the public and private enforcement spheres (contrary to what was implicitly supposed in the 2005 Green Paper). High levels of deterrence and prevention of hardcore infringements of competition law (eg, in particular cartels), through public enforcement policies based on very high fines (with an allegedly strong deterrent effect)29 combined with strong recourse to leniency instruments, tend not to be compatible with a strong encouragement of private actions oriented towards the maximization of compensation for antitrust damages. Policy priorities have to be defined, and a more nuanced approach has to be delineated in order to reconcile or combine different policy objectives in public and private enforcement. The initial approach of the European Commission in the 2005 Green Paper, and the many analyses that followed and advocated a robust reinforcement of private enforcement, were not entirely realistic. Indeed, they sometimes reflected an unbalanced pursuit of contradictory objectives. Following the reforms introduced by Regulation 1/2003, the EU competition law system is still by and large going through a transitional period. In this transitional stage, priorities have to be reasserted as regards the development of EU competition policy. In that context, the first priorities should address public enforcement, which still has to be consolidated in the aftermath of the decentralization policy reform of 2003–2004. Enhancing the incentives of claimants to bring private claims should only come afterwards. Furthermore, any 27 I will not go into the details of the possible specific measures contemplated in the 2008 White Paper or of the particular questions raised in the February 2011 Public Consultation. 28 See Richard Posner, Antitrust Law, 2nd edition, University of Chicago Press, 2011, p 275. 29 The Commission’s policy of imposing high fines in recent years has been discussed by several authors. See, eg, Van Bael & Belis, Competition Law of the European Community, 5th edition, Wolters Kluwer, 2010, especially pp 341 et seq.

Competition Law in Europe – Legal Issues 

119  

moderate and well balanced initiatives designed to foster private enforcement should be based on a clear recognition of separate functions of public and private enforcement. EU competition policy has been anchored in two prevailing goals or functions. The first function has been to delineate the appropriate parameters of the prohibition of anticompetitive behaviours (parameters that sometimes assume an evolutionary nature, as seen in the shift from the initial overriding goal of fostering economic integration to the current emphasis on efficiency maximization, combined now, in the wake of the financial crisis, with considerations of indirect negative effects on consumers that affect the ‘well-being of the European Union’.30,31,32). The second function has been to deter and punish such unwanted behaviours. 2.2. By contrast, private enforcement should predominantly be oriented towards the distinct and autonomous goal of ensuring compensation for consumers injured by competition law infringements. That does not mean that private enforcement should be regarded, in the context of the pursuit of that relatively new goal, as independent of enforcement by public bodies and requiring no fundamental coordination or interactivity with public enforcement (as one of the three views on the relationship between public and private enforcement identified at the beginning of this chapter would have it). I believe that, while the two functions described above (delineating the parameters of the prohibition of anticompetitive behaviours and deterrence/ punishment) should remain the prevailing features of the EU system of enforcement of competition rules, in the context of a clearly leading role for public enforcement of those rules, private enforcement may perform an extremely important complementary and subsidiary role, which should gradually be consolidated. With regard to this subsidiary role, some type of coordination between the two areas will be necessary; but there will be no need for a fundamental shift of the theories or foundations underlying the enforcement of the competition rules. I do not consider, therefore, that the fundamental policy of public enforcement should be globally reconsidered in order to attenuate an allegedly excessive emphasis on deterrence as the almost exclusive enforcement theory, or to address an allegedly unjust focus on leniency policies, and with the purpose of adopting, 30 On this wide characterization of competitive harm that seems to go beyond what the Commission had been considering over the latest decade in the context of its new orthodoxy of economics- and efficiency-based analysis (frequently understood in a rather narrow manner), see Case C-52/09, TeliaSonera Sverige AB [2011] ECR I-527, paras 23–24. 31 The notion of efficiency to be predominantly taken into consideration is still subject to considerable debate. See, eg, Massimo Motta, Competition Policy – Theory and Practice, Cambridge University Press, 2004, especially chapter 1. 32 On the evolutionary nature of the core goals of EU competition law, and providing a comprehensive and critical overview of its more recent evolution, see Luís Silva Morais, Joint Ventures and EU Competition Law, Hart Publishing, 2013, chapter 4.

120  Luís Silva Morais instead, public powers that would not ‘privatize’ damages but would incorporate restitution of the injured parties.33 On the contrary, I admit that the functions of delineating the parameters of the antitrust prohibitions and deterring/punishing misconduct (for which leniency serves as an important instrument) should be consistently kept within the sphere of public enforcement. The relatively new functions of compensation or restitution of injured entities should be entrusted to private enforcement. However, that does not mean that public enforcement should be completely absent from that domain. As I say, some form of coordination with private enforcement is required. For example, as discussed below, such coordination may amount to taking into consideration, to a certain limited extent, forms of voluntary compensation provided by infringing entities when assessing appropriate levels of fines. However, on the whole, this represents a gradual evolution of public enforcement policy, and not a radical shift such as the one proposed by, among others, Christopher Hodges.34 Furthermore, without fundamentally calling into question what should remain the separate functions of public and private enforcement (ie, without pursuing a fallacy of a shared role of deterrence entrusted to both the public and the private spheres of enforcement), I admit that some positive interplay between public and private enforcement may also develop, not as regards the deterrence function proprio sensu but as regards relevant factors for a balanced competition policy, such as wider awareness and effective assimilation of competition parameters as elements actually conditioning market behaviour. In that respect, a particular contribution should be expected from the second pillar of private enforcement that I identified above in section 1.4.1 (at (ii)). In this context of essentially separate functions of public and private enforcement, but allowing some room for reciprocal positive interplay between the two spheres and to a certain extent requiring some types of coordination between those two spheres, one should acknowledge that, at least in the short term, it is easier to discern positive effects generated by public enforcement on private enforcement (namely in follow-on actions through the ‘Masterfoods rule’ of Article 16 of Regulation 1/2003) than it is to discern positive effects in the opposite direction (ie, positive effects flowing from private enforcement to the sphere of public enforcement). Given this overall picture, particular caution should be observed in order to prevent situations in which ill-designed attempts to foster private enforcement may adversely impact on the effectiveness and/or efficiency of public enforcement. 2.3. Considering the potentially positive interplay between the public and private enforcement of competition rules, on the one hand, and, on the other, the potential 33 This corresponds by and large to one of the three main views on the relationship between the spheres of public and private enforcement, supported by, inter alia, Christopher Hodges (see above section 1.1). 34 See Hodges, ‘Integrating Restitution and Behaviour Control’, cited above note 2.

Competition Law in Europe – Legal Issues 

121  

risks that private enforcement may raise in some cases for the effectiveness of public enforcement, one may for analytical purposes select three main areas in which the two spheres intersect and which therefore deserve further comment (albeit very succinctly given this chapter’s limited purview). The three areas are: (a) issues of access to information in the context of public enforcement and how that may interact with actions envisaged in the context of private enforcement; (b) issues related to leniency policies and instruments (and the impact private enforcement actions may have on them); and (c) issues concerning particular incentives or which otherwise focus on selectively encouraging, in the field of private enforcement, forms or domains of antitrust scrutiny that tend to be somehow elusive to public enforcement; in other words, these issues have to do with selectively encouraging private enforcement in domains that would be very difficult to cover, in the coming years and on the foreseeable horizon, by way of public enforcement actions, instead of developing an unrealistic agenda of simultaneously pursuing contradictory, irreconcilable goals, or at least goals that are difficult to combine, by way of both public and private enforcement.

3. Access to information in the context of public enforcement and interaction with private enforcement 3.1. It cannot be ignored that privileged access to information in the context of public enforcement actions may facilitate or even stimulate follow-on actions for damages. That facilitating effect may arise from two separate factors, namely, access to the Commission’s public enforcement file and access to the contents of Commission decisions on competition law infringements (subject to confidentiality issues in the disclosure of those decisions). There is a clear case of tension between conflicting elements and interests in cases where access to information related with public enforcement is needed by potential claimants in private enforcement follow-on actions. On the one hand, and considering the dynamic interplay between public and private enforcement and the potentially disruptive effects of the latter over the former, allowing claimants access to the Commission’s case file during the investigation phase of the public procedure would be apt to seriously affect the exercise of the Commission’s powers of investigation (even if we consider that potential claimants may also act as complainants in the public procedure, and that they may require, in that capacity, a certain degree of intervention in the public procedure, to the development of which they may have been instrumental). On the other hand, even outside the boundaries of leniency procedures (to which I will refer more extensively in section 4), the willingness of undertakings to cooperate

122  Luís Silva Morais with public investigations may be seriously undermined if those undertakings perceive a risk of public discovery powers being used as a tool for follow-on claims (through potentially abusive or distorted use of information rights conferred by Regulation 1049/200135). 3.2. Generally speaking, I would admit that this type of tensions must be resolved in favour of the effectiveness of public enforcement of competition rules. Accordingly, there are grounds to refuse potential claimants access to the Commission’s file while the investigation is taking place. The issue is controversial, however, and it will hopefully be settled or clarified by several judgments of the General Court (following its judgment in MyTravel Group36). It seems to be widely recognized that Regulation 1049/2001 is not a particularly adequate instrument by which to provide information to potential claimants, and the idea of creating a specific access regime under Regulation 1/2003 in connection with follow-on damages actions has already been put forward. Nevertheless, I would resist the idea of providing such claimants with that type of access while investigations are pending. But another scenario has to be considered: the initiation of private follow-on damages actions in the course of which national courts request access to the Commission file under the general duties of the Commission to cooperate with the courts (expressed in particular in Article 15 of Regulation 1/2003). In that case, it may be a matter of debate how the confidentiality of certain business secrets and information contained in the file are to be preserved (either through a confidentiality assessment by the Commission when delivering the file, or through a verification of confidentiality by the requesting court). Again, given the controversial and sensitive nature of such issues, a further jurisprudential clarification in this area would be a very positive development. As regards access to the content of Commission decisions verifying the existence of competition law infringements, and considering that the Commission is under an obligation to publish a non-confidential version of such decisions by virtue of Article 30 of Regulation 1/2003 (ie, a version purged of business secrets), it may also be debated to what extent a national court, in the course of a followon action for damages, can require comprehensive access to these decisions (if certain elements of the decisions are at the core of a certain claim). This is another area still in need of further jurisprudential clarification.

Regulation 1049/2001, 2001 OJ L145/43. Case T-403/05, MyTravel Group v Commission [2008] ECR II-2027, partially annulled, Case C-506/08, Sweden v Commission and MyTravel Group plc. [2011] ECR I-6237. 35 36

Competition Law in Europe – Legal Issues 

123  

4. Leniency policy and instruments, and reciprocal impact on private enforcement actions 4.1. In the context of the leniency regime, which has become a fundamental and highly effective part of the Commission’s arsenal for antitrust scrutiny of anticompetitive practices,37 undertakings that have taken part in cartel conduct may obtain fine reductions or even immunity from fines through their active cooperation with the competition authorities if it enables them to uncover the cartel or compile evidence against it. However, the grant of fine reductions or immunity does not protect a leniency applicant from the civil law consequences of its past participation in the cartel. On the one hand, while theoretically a reduction or mitigation in follow-on liability for damages could represent a further incentive to apply for leniency, that has not really been shown to be necessary, in terms of public interest, for the purposes of deterrence. Indeed, it is the high level of fines routinely imposed by the Commission that seems most apt to ensure that firms have sufficient incentives to apply for leniency. On the other hand, such a reduction of civil liability would amount to an unbalanced collision between the deterrence function (belonging to public enforcement) and the compensation function (belonging to private enforcement, although public enforcement may also contribute to that function). In other words, it would represent, on the part of the public enforcement sphere, an undue intrusion and conditioning of the possibilities conferred by private damages actions. 4.2. However, the disclosure of information provided in the context of a leniency application to potential follow-on damages claimants may be perceived as a risk by these undertakings; accordingly, it may undermine their willingness to apply for leniency – which would have serious adverse affects for the public interest inherent in maintaining an effective leniency program). To a certain extent, the incentives for undertakings to seek immunity under the leniency regime seem preserved in the proposals put forward by the 2008 White Paper. Indeed, in that document the Commission maintained that potential claimants should not have the right of access, at any stage, to ‘corporate statements’ (ie, statements through which an undertaking puts forward its knowledge of a cartel and admits its role in the cartel as part of an application for leniency).38 However, putting aside these corporate statements, a matter for further debate is the extent of 37 The same observations may be made, mutatis mutandis, about the correspondent or comparable leniency programs developed by the national competition authorities that are part of the European Competition Network. On the leniency programs developed at national level by the EU Member States, see Wouter Wils, ‘Leniency in Antitrust Enforcement. Theory and Practice’, in 30 World Competition 25 (2007); OECD Policy Roundtable – Leniency for Subsequent Applicants, 2012. 38 See the Staff Working Paper accompanying the 2008 White Paper, Chapter 10, Section B.1.

124  Luís Silva Morais complementary protection of other materials provided to the Commission by the leniency applicant concerning its participation in a cartel (and, as such, accessible to claimants in follow-on actions, including by way of intervention of the relevant national court). National legal systems may provide for such protection, but it would certainly be advisable to introduce clarifications in EU law concerning the extent and/or limits of such protection conferred on the leniency applicant. There are clearly conflicting interests at play here. 4.3.1. These issues – and especially the recognized need for legislative input at EU level in this domain – have become more prominent and more pressing, particularly in the wake of the ECJ’s landmark judgment of 14 June 2011 in the Pfleiderer case (the first preliminary reference specifically focusing on the interplay between private enforcement of competition law and leniency).39 Leaving aside the particulars of the underlying national court case, Pfleiderer concerned a situation in which the Bundeskartellamt, following a leniency application by some of the involved parties, imposed fines on three major European producers of decor paper and on individuals personally responsible for certain cartel practices (price-fixing and capacity-restriction agreements). Against that background, Pfleiderer brought a follow-on action of damages against the cartelists, from whom it had allegedly purchased certain goods. In this context, the company requested from the Bundeskartellamt extensive access to its case file. Having received from the Bundeskartellamt a version of the decisions imposing fines from which relevant information had been redacted, Pfleiderer further requested access to materials submitted voluntarily to the agency in connection with the defendants’ leniency applications. In response, the Bundeskartellamt provided only limited access to the file, excluding confidential business information and, above all, the corporate statements and documents provided by the leniency applicants. Following a judicial appeal, the referring German Court questioned the ECJ on a set of core issues relating fundamentally to the grant of access to leniency-related materials by claimants in follow-on civil proceedings. Advocate General Mazák summarized quite accurately the key legal problem with which the ECJ was confronted. As he explained, ‘the preliminary reference requires in particular the Court to weigh and balance the possibly diverging interests of ensuring the efficacy of leniency programs established for the purpose of detecting, punishing and ultimately deterring the formation of illegal cartels pursuant to Article 101 TFEU, with the right of any individual to claim damages for harm suffered as a result of such cartels’. (emphasis added)40 The ECJ acknowledged the general need for effective leniency programs as ‘useful tools’ that help to ‘uncover and bring to an end infringements of competition rules’, thereby serving ‘the objective of effective application of articles 101 TFEU 39 40

See Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161. See Opinion of Advocate General Mazák in Pfleiderer, cited previous footnote, at para 2.

Competition Law in Europe – Legal Issues 

125  

and 102 TFEU’.41 Their usefulness did not imply that they had to be protected at any cost, however, since, as the Court pointed out, neither the Commission’s Leniency Notice, nor the ECN Notice or the ECN Model Leniency Program were binding on the authorities or courts of the Member States, those instruments being merely of a ‘soft law’ nature.42 The Court did recognize that the effectiveness of leniency programs could be ‘compromised if documents relating to a leniency procedure were disclosed to persons wishing to bring an action for damages’.43 However, considering the importance of private enforcement of competition law in the EU, 44 it determined that national courts had to conduct a case-by-case balancing exercise in order to ascertain the conditions under which access to leniency information and documents must or must not be permitted. The ECJ did not follow the approach proposed by the Advocate General Mazák, who put forward specific parameters to determine what should and should not be discoverable based on a crucial distinction between corporate statements and other materials produced in the course of a leniency procedure. According to Mazák’s approach, (i) access to voluntary self-incriminating statements made by leniency applicants should not in principle be granted, while (ii) the opposite solution would apply (ie, access should be granted) as regards all other preexisting documents submitted by a leniency applicant.45 While the latter approach may be too schematic, and may not be adequate to deal with a wide variety of situations – which implies that the casuistic balancing approach of the ECJ is to be preferred – it has to be acknowledged that an ad hoc balancing approach inevitably leads to an appreciable degree of uncertainty. Indeed, the solution in Pfleiderer has already attracted a great deal of criticism.46 In my view, this criticism is largely unfair since the ECJ, given the current state of the law, could not really go much further than this. Nevertheless, the analysis of Advocate General Mazák provided abundant implicit signs as to the need for the EU to legislate in this domain. As the Advocate General suggestively put it: ‘in the absence of any EU legislative provision on the matter, […] different standards of disclosure to third parties by national competition authorities of information voluntarily communicated by leniency applicants could thus potentially affect the provisions on cooperation laid down by Regulation No 1/2003’.47 He further stated that ‘there is therefore an apparent tension between on the one hand, the effective operation of a leniency program by a national competition authority and thus the public enforcement of See Pfleiderer, cited above note 39, paras 25–27. See ibid, paras 21–23. 43 See ibid, paras 25–27. 44 See ibid, paras 28–30. Of course, when insisting on the need to ensure that victims of anticompetitive conduct must be able to enforce their rights in court, the ECJ made reference to its judgments in Courage and Manfredi. 45 See Opinion of Advocate General Mazák, cited above note 40, paras 46–48. 46 See, eg, Gianni Di Stefano, ‘Access of damage claimants to evidence arising out of EU cartel investigations: a fast evolving scenario’, [2012/3] Global Competition Litigation Review 95. 47 See Opinion of Advocate General Mazák, cited above note 40, para 27 (emphasis added). 41 42

126  Luís Silva Morais competition law and on the other hand, the grant of access to a third party, for the purposes of assisting it in bringing an action for damages pursuant to Article 101 TFEU, to information provided by a leniency applicant’. And, as regards this core legal question, the Advocate General emphasized that ‘Regulation No 1/2003 and the case-law of the Court have not established any de jure hierarchy or order of priority between public enforcement of EU competition law and private actions for damages. While no de jure hierarchy has been established, at present the role of the Commission and national competition authorities is, in my view, of far greater importance than private actions for damages in ensuring compliance with Articles 101 and 102 TFEU.’48 (emphasis added) Therefore, ultimately, an actual hierarchy should be retained in this domain in favour of public enforcement as enhanced by leniency procedures. This led the Advocate General to state that, in his view, the tension between effective operation of leniency programs and the grant of access to leniency information for purposes of actions for damages ‘is more apparent than real as in addition to the public interest in effective leniency programs in order to detect and punish secret cartels, such programs are also beneficial to private parties injured by such cartels […]. In the absence of effective leniency programs many cartels may never come to light and their negative effects on competition in general and on particular private parties could therefore persist unchecked.’49 I should add that it would in effect prevent multiple private actions for damages from being initiated. However, as there are overall grounds to maintain the hierarchical superiority of public enforcement, though it does not arise de iure condito from the current applicable rules, the need for terms and parameters under which the hierarchy can be effectively and consistently applied with a minimum level of certainty justify, de iure condendo, a clear solution established by formal legislative initiative at EU level. 4.3.2. Developments subsequent to the Pfleiderer case clearly illustrate the serious inconveniences of this persistent legal uncertainty. For example, in Donau Chemie, Advocate General Jääskinen produced his Opinion on 7 February 2013 and the ECJ issued its judgment on 6 June 2013.50 The case concerns the compatibility with EU law of an Austrian statute which bars third party access to court files in ‘public law competition proceedings’ if consent is not granted by the relevant parties. More specifically, the case involved a cartel investigation by the Austrian Competition Authority (ie, the BWB) which had been made possible by a leniency application and which led the Austrian Cartel Court, in the public enforcement action, to fine the cartelists for infringing Article 101 TFEU (since, in Austria, the BWB has to prosecute alleged infractions in court). See ibid, paras 39–40. See ibid, para 41. 50 See Case C-536/11, Bundeswettbewerbsbehörde v Donau Chemie et al., judgment of the ECJ of 6 June 2013, not yet reported. See also the Opinion of Advocate General Jääskinen, delivered on 7 February 2013. 48 49

Competition Law in Europe – Legal Issues 

127  

In that context, the plaintiffs in the civil action sought access to the relevant documents concerning the infraction at stake, but were confronted with the rule described above which precludes access to court documents in ‘public law competition’ cases. That, in turn, triggered the questions asked by the Austrian Cartel Court to the ECJ in the context of a preliminary ruling, concerning, inter alia, the compatibility of the Austrian rule with the principle of effectiveness of EU law. The ECJ was thus to some extent obliged to revisit key legal issues covered in its Pfleiderer ruling, although it did so in a manner that may be read as not entirely coincident with that ruling. The outcome seems to be further uncertainty in this crucial domain in which leniency applicants in various EU Member States need clearly predictable rules and treatment. Indeed, while Advocate General Jääskinen considered that Pfleiderer, with its ‘weighing exercise’ which ‘could only be conducted by the national courts on a case-by-case basis’, was ‘equally pertinent to the case to hand’, he also emphasized that there was a ‘different institutional context’ in Pfleiderer, which ‘concerned access to administrative rather than judicial documents’.51 As such, he regarded the dispute at stake ‘in some respects, closer to the problem considered by the Court in Courage and Crehan’,52 and, therefore, on balance, he ultimately considered that ‘the principle of effectiveness under European Union law […] precludes a provision of national competition law which makes the grant of access to documents of a national court, gathered within competition law proceedings involving the application of European Union competition law, to third persons who are not parties to those […] proceedings, but who wish to prepare actions for damages against participants in an agreement that has been the object of the […] proceedings […]. The answer will only be different if national law provides such alternative avenues for securing proof of breach of European Union competition law and the determination of damage […].’53 (emphasis added) This last part of the balancing test, concerning alternative means for the potential claimant to obtain adequate proof or information on competition law infringements, was not considered in Pfleiderer. This and other considerations of the Advocate General’s Opinion, which seem to send mixed messages, may be difficult to reconcile fully with the core considerations of the Pfleiderer judgment. Ultimately, the ECJ’s judgment of 6 June 2013 in Donau Chemie did not retain the coda of the balancing test put forward by the Advocate General Jääskinen. Instead, it adhered more closely to the reasoning followed in Pfleiderer.54 Nevertheless, the case highlights the uncertainty surrounding this highly sensitive See Opinion of Advocate General Jääskinen, cited previous footnote, paras 40–44. See ibid, para. 45. 53 See ibid, para. 71. 54 See especially the analysis of the ECJ in Donau Chemie, cited above note 50. According to the Court, although ‘a refusal to grant access to certain documents contained in the file of national competition proceedings’ may be justified in certain cases, that does not ‘necessarily mean that access may be systematically refused, since any request for access to the documents in question must be assessed on a case-by-case basis, taking into account all relevant factors in the case (see to that effect ‘Pfleiderer’, paragraph 31)’. (emphasis added) 51 52

128  Luís Silva Morais and crucial domain. At least until such time as the Union adopts a coherent framework in this field, the ECJ may well have to keep adding hermeneutical nuances as regards the relevant balancing test. This will likely preserve and indeed enhance the uncertainty facing leniency applicants, thus making leniency programs less attractive and undermining the effectiveness of competition law enforcement throughout the EU. Indeed, the ECJ’s judgment in Donau Chemie relied extensively on the balancing test established in Pfleideder and considered specifically, as grounds for refusal of access to leniency documents, the need to maintain the effectiveness of a leniency program and the existence of ‘overriding reasons relating to the protection of the interest relied on and applicable to the document to which access is refused’. However, this interest is only compelling ‘if there is a risk that a given document may actually undermine the public interest relating to the effectiveness of the national leniency programme’ in question. The ECJ also emphasizes, although in a much more mitigated manner compared to the Advocate General, the need to take account of the fact that ‘access may be the only opportunity [for persons adversely affected by the competition law infringement] […] to obtain the evidence needed on which to base their claim to compensation’). (emphasis added) The uncertainty described above creates intolerable risks, particularly for cartel enforcement, which relies heavily on leniency applications.55 The undesirable effects of this uncertainty can only be curbed by means of new legislative initiatives in this domain. Only in this way, it seems, can the EU create a level playing field across the Member States as regards the specific conditions of access or refusal thereof to leniency materials in the context of private damages actions. 4.3.3. This much has actually been acknowledged within the European Competition Network (ECN) in a Resolution of May 2012 by the Heads of the European Competition Authorities on the ‘protection of leniency material’.56 The Resolution states, on the one hand, that ‘civil damage claims and leniency programs are complementary tools to enforce competition law and deter further infringements’, but, duly recognizes, on the other hand, that ‘at present civil damage claims in cartel cases in the European Union mostly rely on public enforcement (follow-on actions) and public anti-cartel enforcement is nourished by leniency programs’. On that note, it concludes that ‘appropriate protection of leniency material is necessary to ensure the effectiveness of leniency programs and to enable authorities to uncover and terminate cartels’ and it also rightly emphasizes that, in the context of the decentralized enforcement of EU competition law and of the intense interplay between EU competition law and national competition 55 This reliance of cartel enforcement on leniency applications is widely acknowledged throughout the EU. See, eg, Mario Siragusa and Cesare Rizza, eds, EU Competition Law, Volume III: Cartel Law, Claeys & Casteels, 2007. 56 See the Resolution of the Heads of the European Competition Authorities (ECN) of 23 May 2012, ‘Protection of leniency material in the context of civil damages actions’.

Competition Law in Europe – Legal Issues 

129  

law enforcement, ‘an equivalent standard of appropriate protection of leniency material across the European Union’ should be adopted. (emphasis added) This is, in fact, tantamount to supporting a legislative initiative by the Commission at EU level in order to ensure the protection of leniency programs in a way that renders such protection compatible with some of the requirements of access to information for private actions. The ‘Commission Work Program 2012’ in fact contemplated such a legislative proposal in the field of competition law. According to this document, the objective of the initiative ‘would be to ensure effective damages actions before national courts for breaches of EU antitrust rules and to clarify the interrelation of such private actions with public enforcement by the Commission and the national competition authorities, notably as regards the protection of leniency programs in order to preserve the central role of public enforcement in the EU’.57 (emphasis added) Given that this option to legislate ex novo has been assumed in the postPfleiderer legal environment, it is of paramount importance to effectively initiate and conclude the legislative process as soon as possible, and to avoid a protracted period of uncertainty. On 11 June 2013, the legislative proposal finally emerged. It takes the form of a draft Directive 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. The draft Directive is a part of the long-awaited and much-delayed overall project for a comprehensive framework on private antitrust damage actions and collective actions.58 The draft text contemplates specific rules (Articles 5 to 8) on the disclosure of evidence concerning potential competition law infringements (with particular limits on access to some forms of leniency data provided for in Article 6). The purported goal of the system envisaged in the proposed Directive is to strike a balance between, on the one hand, ‘ensuring that in all Member States there is a minimum level of effective access to the evidence needed by claimants and/or defendants to prove their antitrust damage claim and/ or related defence’ and, on the other hand, ‘avoiding overly broad and costly disclosure obligations that could create undue burdens for the parties involved and risks of abuse’ and which would ‘jeopardiz[e] the public enforcement of competition rules by a competition authority’.59 As regards the protection of leniency information, Article 6 of the proposed Directive to some extent adjusts the Pfleiderer balancing test, providing that Member States must ensure that under their laws national courts are not entitled to order the disclosure or permit the use of corporate leniency statements or settlement submissions made to the Commission, and that the courts may only 57 See the ‘Commission Work Program 2012’ – Annex to the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, Brussels, COM (2011) 777 final (15 November 2011), particularly point 7 (‘Actions for damages for breaches of antitrust law’), under the heading ‘Competition’. 58 For references, see above note 11. 59 Commission, Explanatory Memorandum to the draft Directive, point 4.2 (emphasis added).

130  Luís Silva Morais order the disclosure or permit the use of other information prepared specifically for or by a competition authority after the authority has closed its proceedings or taken a decision. In my view, given the crucial importance of preserving the appropriate incentives to make leniency applications, in the context of this very delicate interplay between public and private enforcement of competition law discussed throughout this chapter, the precise extent and conditions for safeguarding leniency information should be further detailed in the final version of the Directive. In particular, the Directive should specify that protection against disclosure should apply not only to the first oral statement made by a party within a leniency procedure but also to subsequent and complementary submissions. By contrast, the draft Directive seems to imply that other documents prepared for investigations by the Commission or the national competition authorities or the parties concerned will not be protected from disclosure once the relevant investigation has been concluded. This should be carefully pondered in order to be sure that it achieves the right balance between guaranteeing adequate levels of information to prospective plaintiffs and maintaining the right incentives for leniency applications. On the other hand, the very terms of the balancing exercise required to provide access to leniency materials as contemplated in Pfleiderer and Donau Chemie could be decisively changed by the rules of the draft Directive on court-ordered disclosure, since prospective plaintiffs will be able to impose disclosure of other evidence by the defendants. Prospective plaintiffs should therefore not be as dependent on access to the files of competition authorities as was assumed in the case law that established possible access to leniency information on the basis of a casuistic assessment. In particular, Article 5 of the draft Directive allows prospective claimants to obtain court orders for the disclosure of evidence in the control of other parties or third parties, provided the claimant is able to show that the evidence is relevant to substantiate its claim. Implicit in the provision is the recognition that the national judge needs to ensure that disclosure orders are proportionate and that confidential information is duly protected.60 The protracted reform process on the part of the Commission has not only postponed a normative solution in light of the particular and highly sensitive problems linked to the Pfleiderer judgment but has also proceeded cautiously with other legislative and soft-law proposals on private enforcement of competition law arising from the 2011 Consultation document (and interlinked initiatives developed since the 2008 White Paper and other earlier analytical work by the Commission). The long gestation period has most probably been due to difficulties in finding the appropriate consensus for harmonizing different legislations of EU Member States, especially as regards admissible forms of collective redress in the 60 These rules on discovery are rather innovative in various continental legal systems (ie, in Romanistic-Germanic systems). Furthermore, account should be taken of the highly innovative and important element to be introduced that corresponds to the possibility for claimants to require disclosure of ‘categories of evidence’, which should be defined by them as precisely and narrowly as they can ‘on the basis of reasonably available facts’.

Competition Law in Europe – Legal Issues 

131  

context of different judicial systems and for purposes wider than the enforcement of competition law as such.61 4.3.4. As regards collective redress, the ‘Commission Work Program 2012’ contemplated an initiative that would establish ‘an EU framework for collective redress’, and ‘would follow up on the full range of previous Commission work on collective redress at EU level’. The project was conceived as involving the areas of ‘Justice, Consumer Affairs and Competition Policy’, and it admitted that the type of initiative (legislative, non-legislative) is still ‘to be determined’.62 This also has to be seen in the context of the positions adopted by the European Parliament in this field. In particular, the Parliament in its Resolution of 2 February 2012 on ‘the coherent approach to collective redress’ expressed a preference for a horizontal framework, including a common set of principles if the Commission were to propose EU action for collective redress. At the same time, the Parliament acknowledged the existence and importance of competition-specific issues which, supposedly and according to the Resolution, could be addressed either in a separate chapter of a horizontal instrument or in a separate legal instrument (but within the framework of a joint initiative).63,64 These complementary initiatives on collective redress and the legal and political hurdles they face will be briefly touched on in section 6. But I consider that the undeniably more controversial nature of those other initiatives has caused too much delay and should not have held back a more specific and limited legislative initiative to address the issues raised by the Pfleiderer jurisprudence (in the manner discussed above). In my view, those issues justified a more expedited ad hoc legislative solution independent of the overall Directive on private antitrust damages actions. 4.4. Another significant issue which may be raised in the context of the leniency program, and which is relevant for the interplay between public and private enforcement, is related to the possible reduction of fines if the infringing undertaking agrees to provide some form of voluntary compensation to entities 61 The lack of clear consensus persists notwithstanding some degree of gradual approximation of the judicial systems of these Member States. See, eg, Alan Dashwood and Angus Johnston, eds, The Future of the Judicial System of the European Union, Hart Publishing, 2011. 62 See point 110 of the Work Program (‘an EU framework for collective redress’, under the heading ‘Justice, Consumer Affairs and Competition Policy’). 63 See Resolution of the European Parliament of 2 February 2012, ‘Towards a coherent approach to collective redress’, 2011/2089 (INI). 64 To some extent this contrasts with the preferred approach of DG Competition and Commissioner Almunia. The Commissioner has stated that he acknowledged the position of the European Parliament on a horizontal initiative at EU level to stimulate collective redress, and that the Parliament’s position would have to be respected; his own view, however, was that it was preferable to develop a specific initiative in the field of competition law, without prejudice to initiatives in other areas. (Statement in reply to various questions in the Q&A session following Commissioner Almunia’s opening intervention in the annual Concurrences Conference – ‘Demain La Concurrence 2013’, Paris, 22 February 2013).

132  Luís Silva Morais that have suffered damages as a result of the competition law infraction. It should be recalled that, whereas under US leniency policy one of the conditions to obtain immunity is that, where possible, the undertaking must provide restitution to injured parties, no such condition exists under the European Commission’s leniency policy (or under comparable leniency policies adopted by the Member States). However, the European Competition Authorities’ ‘Principles for convergence on pecuniary sanctions imposed on undertakings for infringements of antitrust law’ adopted in May 2008 permit, as a possible mitigating circumstance (allowing a reduction of the fine) the fact that the offender takes active steps to provide voluntary compensation to those who have suffered damages due to the competition law infringement. As far as I am concerned, this may represent a promising way to gradually incorporate elements of compensation into the public enforcement of competition rules while acknowledging that such a contribution will not call into question the proposition that the compensation function remains properly within the private enforcement sphere alone. Nevertheless, the introduction of this new feature has taken place in a way that tries to strike a balance between the deterrence and punishment functions (which should be safeguarded entirely in the context of public enforcement and should not be jeopardized for compensation purposes) and the compensation function. In that regard, point 18 of the European Competition Authorities’ ‘Principles for convergence on pecuniary sanctions’ of 2008, already mentioned above, which provides for the possible reduction of fines in exchange for voluntary compensation, is accompanied by the proviso that, ‘where compensation is taken into account as a mitigating circumstance, this reduction should not in any case be such as to undermine the deterrent effect of the fine’ (emphasis added). However, the qualitative level upon which a certain mitigation of the fine would seriously undermine the deterrent effect of the same fine is by no means clear.65 The issue should be further clarified in the interest of promoting a gradual, albeit limited, integration of compensation considerations in the field of public enforcement of competition rules – without affecting its other functions.

5. Particular incentives that may selectively encourage domains of antitrust scrutiny more elusive to public enforcement As mentioned earlier, and considering the crosscurrents and intersection points I have been exploring between private enforcement and public enforcement 65 On mitigation of fines for competition law infractions, see Luís D S Morais and Francisco Costa Cabral, ‘Mitigation of fine and competition law: An overview of EU and national case law’, (Foreword to the Concurrences/e-Competitions database of leading national and EU cases), 19 July 2012, e-Competitions, No. 48098, www.concurrences.com.

Competition Law in Europe – Legal Issues 

133  

– the latter undoubtedly corresponding to the prevailing dimension of the EU competition law system – it would make sense, whenever possible, to encourage, on a selective basis, private enforcement in domains that tend to be more difficult to address through public enforcement actions. In a very broad manner, I would ‘prima facie’ consider the following to be illustrative scenarios ripe for increased recourse to private enforcement: (a) potentially abusive practices under Article 102 TFEU (where cases are seldom initiated by the Commission or even by national competition authorities, at least in relative terms, bearing in mind the potential range of abuse of dominance cases that may exist, in different moments, in the market); (b) joint ventures and possible anticompetitive effects that may be inherent to minority shareholdings of a significant dimension but short of conferring control (chiefly, crossed minority shareholdings in two or more undertakings) in certain undertakings;66 and (c) multiple relationships relating to the exercise and/or management of IP rights. As regards abuse of dominance cases, these are almost inevitably complex. Scrutiny is bound to be time-consuming and frequently involves remedies that are extremely difficult to identify and calibrate.67 Therefore, competition authorities have to carefully select a limited number of cases or situations they will monitor. It may be easier for undertakings and/or consumers with lesser economic power to react timely to these situations through multiple private actions (including injunctive relief and not necessarily large collective redress mechanisms).68 And, as I will try to emphasize below in section 6, there are multiple incremental steps that may be inserted in a soft law/soft harmonization approach at the EU level in order to increase the awareness of, efficiency and corresponding use of private action mechanisms. In the fields of joint ventures, (crossed) minority shareholdings of a significant dimension and IP-related relationships, significant market situations may be kept 66 See Luis Silva Morais, Joint Ventures and EU Competition Law, cited above note 32, Chapter 3 (discussing, in addition to joint ventures, possible anticompetitive effects inherent to certain minority shareholdings that do not confer control); Robin A. Struijlaart, ‘Minority share acquisitions below the control threshold of the EC Merger Control Regulation: An economic and legal analysis’, in 25 World Competition 173 (2002); Daniel O’Brien and Steven Salop, ‘Competitive effects of partial ownership: financial interest and corporate control’, 67 Antitrust Law Journal 559 (2000); Enrique González-Diaz, ‘Minority Shareholdings and Creeping Acquisitions: The European Union Approach’, in Barry Hawk, ed, International Antitrust Law & Policy: Fordham Competition Law Institute 2011, Juris Publishing, 2012. 67 On these difficulties, which seriously constrain the intervention of competition authorities, see Harry First, ‘Netscape is Dead – Remedy Lessons from the Microsoft Litigation’, in Review of Competition and Regulation (C&R), No 1, 2010, special thematic issue on the abuse of dominance. 68 As a particularly significant example which in a sense confirms or illustrates the potential relevance of abuse of dominance cases in private enforcement actions (maxime, in stand-alone cases), see, eg, Purple Parking Limited and Meteor Parking Limited v Heathrow Airport Limited, ruling of the English High Court of 15 April 2011, in which the court found that Heathrow Airport had abused its dominant position. Significantly, what was essentially at stake for the claimants was the granting of access to the forecourts of Heathrow Airport Terminals 1 and 3 to enable them to carry out their valet parking activities. In Portugal, another relevant (stand-alone) case was decided by an arbitrator. Although the content of the arbitral award cannot be fully disclosed, this very significant case concerned an abuse of dominance in the pharmaceutical sector.

134  Luís Silva Morais secret or relatively undisclosed and the type of cross-interests involved may render less probable potential complaints by some of the parties at the core of those situations. It may well happen that the very business dynamic of those situations will lead the parties – either parties directly involved or parties somehow affected by those situations – to multiple forms of private litigation in which they may introduce issues of compliance or lack of compliance with competition rules. Of course, that may also happen in the course of arbitration procedures normally employed in the biggest business transactions and in the context of the functioning or managing of big corporate structures and of joint ventures. This leads me to underline once again the particular importance that arbitration may have for the development of private enforcement of competition rules (especially in the domains described above). And, considering the potential importance of arbitration – in light of the Eco Swiss and Van Schijndel jurisprudence – it would make sense to encourage (through orientations and/or recommendations) legislative reforms in the Member States adopting solutions that establish public order principles or criteria as a basis for the control of arbitral awards, as has been done to some extent (albeit with a lot of criticism from several stakeholders heavily involved in arbitration) in the Portuguese Law on Voluntary Arbitration, Law No. 63/2011 of 14 December. (This reform could have been more ambitious, however, and the reference in Article 56(1)(b)(ii) of the Law to awards contrary to principles of international public order could have been extended to public order in general, even if it may be somehow contemplated that such reference should already include what Idot calls a kind of third concept of ‘European Union public order’).69 What is at stake, ultimately, is the possibility of annulment of arbitral awards due to the infringement of public order principles, a normative solution which, whenever phrased with enough latitude will tend to include the principles of competition law.

6. Possible gradual and incremental approach to foster private enforcement without adversely affecting public enforcement 6.1. In light of the preceding comments and analysis, I admit that, following the Public Consultation of February 2011, a ‘soft harmonization’, phased approach (through a Communication of the Commission or, more specifically, a set of Recommendations addressed to the Member States) will be a good option to foster 69 See especially Laurence Idot, ‘Arbitration, Competition Law and Public Order’, cited above note 24, pp 213 et seq. As noted by Idot, another noteworthy aspect of the new Portuguese Law on Voluntary Arbitration (Law No. 63/2011), more modern in this regard than other arbitration laws found in other EU Member States, concerns the latitude it allows parties to use arbitration (which creates a wider legal window for the arbitrability of competition law issues). In fact, according to Article 1 of the Law, the parties may submit to arbitration any dispute involving economic interests (of a ‘patrimonial’ nature, ie, any dispute involving assets).

Competition Law in Europe – Legal Issues 

135  

private enforcement of competition rules without unduly or adversely affecting public enforcement, which must remain the cornerstone of the system. As noted above in section 4.3.3, the Commission included in its Work Program 2012, besides a more specific legislative proposal to clarify issues pertaining to the interplay of access of relevant information by private claimants with leniency programs and the information these generate about competition law infringements (following the Pfleiderer ruling), an initiative to develop an ‘EU framework for collective redress’, the form of which, legislative or non-legislative, was still to be determined. At this stage, a combination of a proposed harmonization through a Directive on damages and a soft-law initiative (ie, a Recommendation) on collective redress was probably to be expected in this domain. To some extent it is a political acknowledgement of the difficulties of pursuing a more extensive, proactive or vigorous strategy of formal EU harmonization in this domain, given the persistent and rather intractable differences of judicial systems and related legal cultures of the various EU Member States. This also must have influenced the current emphasis of the Commission on the establishment of ambitious frameworks for alternative dispute resolution (ADR) throughout the EU,70 where the goals of consumer redress, including in the area of competition law, may be more easily attained, however indirectly, rather than through formal harmonization of national procedural laws and of certain features of national judicial systems. 6.2. That mixed approach has indeed materialized in the June 2013 Commission Proposals, which include, as already mentioned, a draft Directive on private actions for damages for infringements of competition law provisions and a nonbinding Recommendation. As regards the draft Directive, and as stated in the Commission’s Explanatory Memorandum, its two key goals are, on the one hand, to ‘optimise the interaction between the public and private enforcement of competition law’ (assuming its ‘complementary’ character), and on the other hand to ‘ensure the effective exercise of the victims’ right to full compensation’ (assuming that, ‘while the right to full compensation is guaranteed by the Treaty itself and is part of the acquis communautaire, the practical exercise of this right is often rendered difficult or almost impossible because of the applicable rules and procedures’). I have already observed (above in section 4.3.3) that in the domain concerning the first goal (ie, optimizing the interaction of public and private enforcement) the outcome of the draft Directive may be far from optimal; accordingly, certain adjustments should be considered in the course of the EU’s legislative procedure. 70 On the implications and basic features of this comprehensive ADR approach, see Christopher Hodges, ‘New Modes of Redress for Consumers and Competition Law’, in Competition & Regulation Review (C&R), December 2012, pp 225 et seq. See also Christopher Hodges, Iris Benöhr and Naomi Creutzfeld-Banda, Beck/Hart, Consumer ADR in Europe (Civil Justice Systems), Hart Publishing, 2012.

136  Luís Silva Morais These include, above all, adjustments oriented toward a more effective protection of the incentives to submit leniency applications in the field of public enforcement. With regard to the exercise of the right of victims of competition law infringements to obtain full compensation, the draft Directive puts forward some important solutions. In addition to the aspects I have already covered above on disclosure of documents, reference should also be made to proposed rules on: (a) rebuttable presumptions for cartel infringements; (b) the relevance of decisions taken by national competition authorities and national courts; (c) the recognition of a passing-on defence in damages actions; (d) the definition of joint and several liability for competition law infringements; and (e) limitation periods for damages actions. As regards (a), rebuttable presumptions for cartel infringements, the draft Directive establishes a distinction between ‘competition law infringements’ and ‘cartel infringements’ (defined very broadly), and sets forth a rebuttable presumption that a cartel infringement has caused harm (although that presumption is not extended to the level of harm to be ascertained in any given case). Interestingly, this type of presumption was abandoned in the UK in the context of the recent proposals of the UK Government for the reform of private enforcement of competition law,71 and it may raise some level of controversy in other legal systems of EU Member States as well. With respect to (b), decisions of national competition authorities, the draft Directive provides that these are to be treated as full proof before civil courts that competition law infringements have taken place. (The draft Directive also proposes that a civil court is to be precluded from taking decisions in damages actions arising from such infringements if such decisions would run counter to the findings of review courts in other Member States. While the objective of contributing to a level playing field in this domain within the EU is understandable, this proposed rule may lead to rather intractable legal problems given the potentially different approaches of national competition authorities in various Member States despite their integration within the European Competition Network. Concerning (c), the passing-on defence in damages actions, the draft Directive recognizes that defendants can generally avoid or diminish liability by showing that the claimants passed a part of the overcharge arising from the infringement (or even the whole overcharge at stake) to their own customers. However, in an action brought by an indirect purchaser, the claimant will have to prove the scope and existence of the pass-on. As regards (d), the definition of joint and several liability for competition law infringements, the draft Directive provides that competition law infringers 71 See the ‘Draft Consumer Rights Bill, presented to Parliament by the Secretary of State for Business, Innovation and Skills by Command of Her Majesty’ – curiously presented by the UK Government on 12 June 2013, one day after the Commission unveiled its long-awaited proposal on private enforcement. This Draft Consumer Rights Bill includes the UK Government’s proposals to facilitate private enforcement of competition law rights through litigation, including via collective actions. It is subject to consultation until mid-September, 2013.

Competition Law in Europe – Legal Issues 

137  

are jointly and severally liable for the entire harm caused by anticompetitive behaviour; but it provides for a more favourable regime for leniency applicants, which I regard as very positive (in contrast with my view about the proposed rules on disclosure). According to the draft text, provided they have obtained full immunity, leniency applicants will merely be liable for harm they have caused to their own direct or indirect customers. (This solution is subject to the principle of ensuring full compensation for the injured parties; hence, leniency applicants remain fully liable if the injured parties are unable to obtain full compensation from the other infringers.) This approach adds to the attractiveness of leniency programs, although it has to be acknowledged that leniency applicants may be left in a considerable state of legal uncertainty until proceedings against the other infringers are completed. Finally, the draft Directive seeks to clarify decisively (e) the applicable rules on limitation periods for damages actions. In particular, it establishes that claimants must have at least five years in which to bring an action after becoming aware of the infringement. Further, the period of limitations would have to be suspended when a competition authority opens an investigation, thus giving claimants the possibility to wait until public proceedings are complete, thereby enhancing the potential of follow-on cases. As regards the Recommendation on collective redress, the Commission, taking a much more moderate stance than it did in the 2008 White Paper, does not require Member States to introduce collective redress mechanisms within their legal systems. Following a rather low-key approach, the Recommendation merely puts forward non-binding principles to guide the introduction of some form of collective redress in all EU Member States. Although it does require the introduction of collective actions as such, the Recommendation favours compensatory and injunctive relief based on the so-called ‘opt-in model’: in order to be part of a group initiating any form of collective action lato sensu, any potential claimants must give their express consent. This low-key approach may be compensated by the somewhat implicit possibility of future EU binding instruments in this area if the Commission ascertains, in its review of the implementation of the Recommendation (to be conducted within two years after its adoption), that Member States have not actually adopted collective redress mechanisms along the lines of the recommended principles. 6.3. On the whole, and in spite of the various limitations of the June 2013 Commission proposals on which I have commented above, the gradualist approach taken by the EU, based on a very limited degree of formal harmonization and largely relying on soft law, may be a valuable strategy and is most probably a more realistic one, in the current legal and political context, to encourage private enforcement in antitrust and collective redress across a broader range of areas. One should also recall the extent of impressive soft harmonization of national competition laws of the EU Member States, largely achieved through soft law

138  Luís Silva Morais instruments,72 which has in effect sustained or accompanied the process of decentralization of enforcement of EU competition law. This process shows that important results may also be achieved in the field of private enforcement through a carefully calibrated and gradual soft law approach. In fact, apparently small and incremental steps may have a large impact and may promote a smooth transition to more intensive recourse to multiple forms of private enforcement, particularly in those areas which may escape the attention of public authorities. I therefore consider that there is room for the gradual but consistent promotion of private enforcement in a manner that does not limit it to follow-on procedures, as the Commission sometimes seems to fundamentally envisage it. In the domain of follow-on actions, and given the high level of fines imposed, private enforcement is not a priority for purposes of deterrence, and even for purposes of compensation such actions may be gradually integrated, as I have observed (section 4.3), in the public measures adopted (without prejudice to the undeniable importance of a consistent expansion of private claims for damages). However, even through the lens of this gradualist and moderate approach in this domain, I consider the June 2013 Commission proposals rather disappointing, particularly in light of the range of creative solutions that could be considered, albeit through mere low-key recommendations. Indeed, I would endorse other solutions including – with no intention to be in any way exhaustive – incremental steps that would mainly ensue from changes of the Member States’ national competition laws which may be recommended at EU level through an extensive soft-harmonization approach, as in fact happened, on the whole, with the gradual approximation of national competition rules in accordance with the EU competition law paradigm. (A proviso is that some of the solutions that I put forward below are, albeit in variable forms, already contemplated in the national legislation of certain Member States.) These incremental changes, which on various points go further than the solutions contemplated in the June 2013 proposals, could include, inter alia: • Enabling the national competition authority (or giving it further incentives) to intervene as amicus curiae in any court proceedings, at the request of national courts, or on the competition authority’s own initiative or upon request by the parties, whenever Articles 101 or 102 TFEU or equivalent rules of national law are to be applied or are relevant for the case; • Mandatory communication by national courts to their respective national competition authority of all the rulings and appeals that involve application of the competition rules. This apparently limited and low-key measure may, in my view, have a significant incremental effect on the recourse to private enforcement in the medium term, creating awareness about the potentialities of these kinds of private actions and gradually influencing a change of legal 72 In this regard, and on the impressive convergence of national competition laws in the absence of any formal harmonization methods, see, eg, Michaela Drahos, Convergence of Competition Laws and Policies in the European Community, Kluwer Law International, 2001.

Competition Law in Europe – Legal Issues 

139  

culture in this domain in various Member States (and indeed while in some States such as the UK there is easy and widespread access on a systematic basis to court rulings touching on competition law issues,73 in other Member States such access or knowledge is not easy to obtain in a systematic manner, leading to the fact that empirical research in this domain, and covering relevant judicial precedents of private enforcement, tends to produce data that frequently come as a relative surprise to the legal community since the level of private enforcement is more significant than expected by most stakeholders74); • The possibility of staying court cases in which the application of the competition rules may be at stake, whenever the court is aware – or is made aware – of proceedings initiated by the Commission or by the national competition authority, and considering that it may be convenient to wait for the adoption of a decision by those authorities; • Recognition (ie, formal recognition) as legitimate parties in judicial proceedings concerning compensation or indemnity for infringement of competition rules of competing undertakings or of any other market players directly or indirectly affected by those possible competition law infringements (extended to associations that may be deemed fit under national law for that purpose, eg, trade or professional or consumer associations; indeed, a curious aspect in the current context is the scarce intervention in many EU Member States of consumer associations in procedures of this kind); • Mandatory information to be collected and published by the national competition authority (on its website), and in particular the full text of all judicial rulings covering competition law issues, including private actions concerning the application of competition rules (competition authorities may be fundamental agents, however indirectly, that foster awareness about the potentialities of private enforcement of competition law, and this may correspond to a rather sui generis but extremely relevant interplay between public and private enforcement throughout the EU; this type of approach may also translate into some form of peer pressure within the ECN which indirectly enhances incentives and conditions for private enforcement and which promotes the influence of authorities in the context of their growing competition advocacy role,75 whereby they may nurture a process of soft 73 This is illustrated and supported by evidence in the research coordinated by Barry Rodger in a project entitled Comparative Private Enforcement and Collective Redress in the EU. The project covers the EU-27 and is due to be published in the course of 2013. I have had access to parts of the study, for which I gratefully acknowledge its Coordinator. 74 This has been verified, eg, in Portugal in the context of empirical research on judicial precedents involving some level of private enforcement of competition law in Portugal. The research was conducted as part of a study coordinated by Barry Rodger and referenced in the previous footnote. A first version of the Report on Portugal is being published as Leonor Rossi and Miguel Ferro, ‘Private Enforcement of Competition Law in Portugal (1) An Overview of Case Law’, Competition & Regulation Review (C&R), April/June 2012, pp 91 et seq. 75 On the growing role of competition advocacy on the part of competition authorities, including advocacy vis-à-vis the State and other public entities, see, eg, Bruno Lasserre, ‘Towards the ECN’s

140  Luís Silva Morais harmonization of core elements of procedural and judicial systems of Member States, thus augmenting the potentiality of private enforcement); and • the adoption of measures concerning arbitration (referred to throughout this chapter). Accordingly, on the whole, I admit that the soft law approach currently on the EU horizon may allow significant steps on a gradual way for building a new and enhanced basis of private enforcement of competition law throughout the EU. However, I also consider that such an approach should have a much broader scope (as illustrated above), namely putting forward creative elements of interaction between public and private enforcement of competition rules (as contemplated, eg, in parts of the 2012 Study commissioned by the European Parliament, ‘Collective Redress in Antitrust’76 and in other analytical documents originated in the EU Member States such as, eg, the UK 2013 ‘Consultation on options for reform – government response, Private Actions in Competition Law’77) or extending, more decisively, to the areas that I discussed in section 5 (more elusive, as I explained, to public enforcement proceedings). Hopefully, an extensive critical discussion of the June 2013 proposals – which is beyond the purview of this chapter – should pave the way for a much-needed fine-tuning of those proposals, rendering them more ambitious (albeit in terms of very practical and apparently limited incremental steps, which may be highly relevant to the gradual promotion of private enforcement of competition law within the EU).

Second Decade’, in Barry Hawk, ed., International Antitrust Law & Policy Fordham Competition Law 2011, Juris Publishing, 2012, pp 19 et seq. 76 See the Study commissioned by the European Parliament (Committee on Economic and Monetary Affairs) and entitled ‘Collective Redress in Antitrust’, by Paolo Buccirossi, Michele Carpagnano, Lorenzo Chiari, Massimo Tognoni and Cristiana Vitale (June 2012). 77 See Department for Business Innovation & Skills, ‘Private Actions in Competition Law – A Consultation on Options for Reform – Government Response’ (January 2013) (which launched a discussion leading to the above-mentioned proposals of the Government of 12 June 2013 to reform private enforcement of competition law in the UK).

Assimakis P Komninos*

The Relationship between Public and Private Enforcement: quod Dei Deo, quod Caesaris Caesari

I. Public and private antitrust enforcement and the objectives of EU competition enforcement The concurrent public and private enforcement of legal rules is not unique to the antitrust laws. It certainly predates those laws and expresses more fundamental ideas about the relationship between the State and the citizens and their respective roles in the implementation of the law as such. Private enforcement has been particularly strong in the United States, because, for constitutional reasons, there are no specialised agencies to the same extent as in Europe and, thus, private enforcement tends to compensate for this lack of statutory control. In contrast, if there are statutory agencies entrusted with the enforcement of the law, there is less need to grant equal powers to individuals.1 From a purely competition law perspective, antitrust enforcement pursues three systematically different, yet substantively interconnected, objectives.2 The first one is injunctive, ie to bring the infringement of the law to an end, which may entail not only negative measures, in the sense of an order to abstain from the delinquent conduct, but also positive ones to ensure that such conduct cease in the future. The second objective is restorative or compensatory, ie to remedy the injury caused by the specific anticompetitive conduct. The third one is punitive,3 ie * At the time of writing, Commissioner and Member of the Board of the Hellenic Competition Commission; Visiting Professor, IREA – University Paul Cézanne Aix – Marseille III; Visiting Research Fellow, University College London. The author expresses his personal views. The chapter covers developments through to the end of August 2011. 1 See Hans Micklitz, ‘Transborder Law Enforcement – Does it Exist?’, in Stephen Weatherill and Ulf Bernitz, eds., The Regulation of Unfair Commercial Practices under EC Directive 2005/29, New Rules and New Techniques, Hart Publishing, 2007, p 252, with further references; Ilya Segal and Michael Whinston, ‘Public vs. Private Enforcement of Antitrust Law: A Survey’, University of Chicago, John M Olin Program in Law and Economics, Working Paper No 335 (December, 2006), p 1. 2 See Christopher Harding and Julian Joshua, Regulating Cartels in Europe, A Study of Legal Control of Economic Delinquency, OUP, 2003, pp 229 et seq; Assimakis P Komninos, EC Private Antitrust Enforcement, Decentralised Application of EC Competition Law by National Courts, Hart Publishing, 2008, pp 7 et seq For a slightly different classification of tasks-objectives of antitrust enforcement, compare Wouter Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for Damages’, 32 World Competition 3 (2009), who speaks of three tasks: (a) clarifying and developing the content of the antitrust prohibitions, (b) preventing the violation of these prohibitions through deterrence and punishment, and (c) pursuing corrective justice through compensation. 3 The term ‘punitive’ is used here in its generic sense and does not necessarily correspond to criminal law.

142  Assimakis P Komninos to punish the infringer and also to deter him and others from future transgressions. Ideally, these three basic objectives can be pursued inside an enforcement system that combines both public and private elements. Both public and private enforcement may – directly or indirectly – pursue all three objectives: • The injunctive objective is served with cease and desist orders and negative or positive injunctions ordered both by competition authorities, in the course of public proceedings, and by the courts, in the course of civil proceedings. Indeed, the latter may go even further than public enforcement. For example, it may be easier to obtain a preliminary injunction from a national judge within the EU than from the European Commission, while the latter, unlike the former, cannot issue orders imposing positive measures to undertakings in Article 101 TFEU cases.4 • Private enforcement primarily serves the restorative-compensatory objective, since private actions ensure compensation for those harmed by anticompetitive conduct. However, even in such cases, the role of public enforcement is not inexistent. For example, a competition authority’s action may in effect amount to redress in specific cases. Then, the competition authority may impose on the wrongdoer or accept commitments from him to put in place a compensatory scheme.5 In addition, some competition regimes give powers to certain public authorities to claim damages, acting on behalf of the victims. For example, this is the case in France6 and in the United States, where State Attorneys General can bring parens patriae actions on behalf of victims located within their States.7 • Finally, as for the punitive objective, public enforcement is undoubtedly predominant. This objective is pursued through the imposition of fines, which punish the wrongdoers and deter them from breaching the law in the future (specific deterrence) but also deter other persons from entering into or continuing to engage in behaviour that is contrary to the competition rules 4 Case T-24/90, Automec Srl v Commission (II) [1992] ECR II-2223, para 51. According to the General Court, freedom of contract is the rule, so the Commission cannot order a party to enter into a contractual relationship ‘where as a general rule the Commission has suitable means at its disposal for compelling an enterprise to end an infringement’. In the Commission’s view such purely positive measures may be more justifiable in Article 102 TFEU cases (see the Commission’s arguments in para 43). 5 This has happened in very exceptional cases. See, for example, the OFT’s case on the independent schools’ cartel, where the OFT settled the case and accepted a commitment by the schools to make an ex-gratia payment totalling £3 million into an educational charitable trust to benefit the pupils who attended the schools during the academic years to which the cartel related. 6 Article L442-6(3) Code de commerce. France also has another interesting specificity: it is possible for aggrieved parties to take part as partie civile in a criminal proceeding against the wrongdoer and, thus, be awarded damages, apart from the criminal conviction of the (natural) persons involved in anticompetitive conduct. 7 At the same time, laws that provide for the possibility of the competition authority to skim off economic benefits achieved from certain conduct (eg sections 34 and 34a of the German Competition Act) cannot be said to pursue compensatory but rather punitive objectives.

The Relationship between Public and Private Enforcement 

143  

(general deterrence).8 However, here again, private actions may supplement the retributive and deterrent effect of the public sanctions by attaching punitive elements to the civil nature of remedies, this being the case of legal systems that provide for punitive antitrust damages. More importantly, the existence of private enforcement furthers the overall deterrent effect of the law, by adding a supplementary system of sanctions and risks for the wrongdoer.9 In my view, there is some misunderstanding as to the interests protected by competition law in the contexts of public and private enforcement. Some authors distinguish between public enforcement, which pursues the public interest of protecting the competition norms through administrative or criminal sanctions, and private enforcement, which pursues the private interest of competitors and consumers.10 The European Commission has also, in the past, followed a similar approach, with statements which seem to ignore the public interest nature of civil claims.11 Such distinctions, however, do not do justice to the role of civil courts when they enforce competition law in the context of private disputes between economic operators. Although the courts certainly decide disputes inter partes, at the same time, they cannot simply confine themselves to considering the interests of the litigants, but must also have regard to the general interest.12 This explains why courts must raise the competition law question even ex proprio motu and may not allow the performance of an anticompetitive agreement, even if the parties have not raised the issue of its legality.13 Courts are also not bound by judicial or in-court 8 See Commission, Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003, 2006 OJ C210/2, para 4; Case 41/69, ACF Chemiefarma v Commission [1971] ECR 397, para 173; Case C-308/04 Ρ, SGL Carbon AG v Commission [2006] ECR Ι-5977, para 37; Case C-76/06 P, Britannia Alloys & Chemicals Ltd v Commission [2007] ECR Ι-4405, para 22. 9 See further Commission, Green Paper on Damages Actions for Breach of the EC Antitrust Rules, COM(2005) 672 final, under 1.1; Commission Staff Working Paper Accompanying the White Paper on Damages Actions for Breach of the EC Antitrust Rules, SEC(2008) 404, paras 15 and 20; Komninos, cited above note 2, pp 7–10. 10 See, eg, August Braakman, ‘The Application of the Modernised Rules Implementing Articles 81 and 82 EC Treaty in Injunction Proceedings: Problems and Possible Solutions’, in Barry Hawk, ed, International Antitrust Law and Policy 1999, Annual Proceedings of the Fordham Corporate Law Institute, Juris Publishing, 2000, chapter 1; Harding and Joshua, cited above note 2, p 239; Barry Rodger, ‘The Big Chill for National Courts: Reflections on Market Foreclosure and Freezer Exclusivity under Article 81’, 11 IJEL 77 (2004), p 107. Compare also Jo Shaw, ‘Decentralization and Law Enforcement in EC Competition Law’, 15 Legal Studies 128, 158–159 (1995); Wils, cited above note 2, p 9. 11 See Ernst-Joachim Mestmäcker, ‘The Modernisation of EC Antitrust Policy: Constitutional Challenge or Administrative Convenience?’, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds, European Competition Law Annual 2000: The Modernisation of EC Antitrust Policy, Hart Publishing, 2001, pp 233–234; Komninos, cited above note 2, p 12. 12 See Jacques Bourgeois, ‘EC Competition Law and Member State Courts’, in Barry Hawk, ed, Antitrust in a Global Economy 1993, Annual Proceedings of the Fordham Corporate Law Institute, Transnational/Juris Publications, 1994, chapter 21, at pp 485–486; Claude Lucas de Leyssac and Gilbert Parleani, Droit du marché (Paris, 2002), p 971. 13 See Guy Canivet, ‘The Responsibility of the Judiciary in the Implementation of Competition Policy’, in OECD, Judicial Enforcement of Competition Law, 1997, p 24.

144  Assimakis P Komninos settlements, when the latter infringe the competition rules; indeed, they are under a duty stemming from Article 4(3) TEU to consider the above as null and void, since they are against public policy.14 Likewise, the possibility for public competition authorities in the EU and in some national competition systems to intervene in civil proceedings and submit amicus curiae observations is partly due to the public policy / interest nature of competition law-related litigation.15 Finally, laws that attach a punitive element to civil claims for damages precisely prove that there is something more at stake than just the pursuit of the private interest. It is sometimes argued that the private interest that permeates civil litigation is harmful for the public interest that the competition laws should only be about. However, such criticism is unfounded. The protection of private rights cannot by itself set in motion the mechanisms. The protection of free competition, since the law is indifferent to harm caused to a specific person, unless that harm is the consequence of a certain practice whose object or effect is the distortionprevention-restriction of effective competition on the market. To give some examples, an agreement or practice might cause harm to certain persons but still not be considered anticompetitive, because it may not affect appreciably competition in the market (de minimis). Then, certain unilateral conduct may foreclose competitors and actually harm them, but may not be anticompetitive, in the first place. Conversely, there may be anticompetitive conduct that does not harm any specific person, or no person has standing to sue. The existence of private actions and, in particular, the availability of damages to the victim of anticompetitive practices, is perfectly consistent with the public interest that is inherent in competition norms. Indeed, the Court of Justice has recognised that private actions strengthen the working of the EU competition rules and discourage practices that are liable to restrict or distort competition, thus making a significant contribution to maintaining effective competition in the Union.16 In other words, this is a case where the private interest contributes to the safeguarding of the public interest, so no antinomy should exist. 14 Judicial settlements can have in some jurisdictions res judicata effect between the parties and may even constitute an enforceable act or be so declared by a court. The Court of Justice has stressed that such settlements, even if constituting judicial acts, are capable of falling within the prohibition of Article 101 TFEU, since that provision ‘makes no distinction between agreements whose purpose is to put an end to litigation and those concluded with other aims in mind’ (Case 65/86, Bayer AG and Maschinenfabrik Hennecke GmbH v Heinz Süllhöfer [1988] ECR 5249, paras 14–15). See also Case 258/78, L.C. Nungesser KG and Kurt Eisele v Commission (Maize Seed) [1982] ECR 2015, paras 80–89; Cass.com., 26-3-79, Ateliers de Construction de Compiègne v Fabry, where the French Supreme Court quashed a lower judgment that had refused to examine the Article 101(2) TFEU nullity of a patent licensing agreement, because the parties had concluded a settlement having the force of res judicata. The Supreme Court stressed the ordre public nature of the Treaty competition rules and considered, therefore, that the nullity of Article 101(2) TFEU could not a matter to be settled. 15 See Philippe Rincazaux, ‘Les autorités de la concurrence doivent-elles être autorisées à intervenir dans les procédures relatives à des problèmes de concurrence, plus particulièrement lorsqu’elles sont menées devant les juridictions ordinaires ? Dans l’affirmative, quel devrait être le fondement de leur pouvoir d’intervention ? Rapport international’, LIDC Questions 2001/2002, p 1. 16 Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297, para 27; Joined Cases C-295/04 to C-298/04, Vincenzo Manfredi et al. v Lloyd Adriatico Assicurazioni SpA et al. [2006] ECR I-6619, para 91.

The Relationship between Public and Private Enforcement 

145  

Even if we suppose that in a given case a civil litigant’s private interest might not be compatible with the public interest, as may be the case, for example, if inefficient competitors allege the ‘anticompetitive nature’ of certain practices that in reality enhance effective competition, such a private action would fail, because the alleged harm would not have been caused by conduct prohibited by Articles 101 and 102.17 Consequently the private interest simply cannot be seen in contradiction with the public interest, in this context. Hence the complementarity and the ‘private attorney-general’ function of the civil litigant. In Pfleiderer, the Court of Justice dispelled any doubts about the role and function of courts and considered both the competition authorities and the former as servants of the general interest and did not make any differentiation between them as to the interest they pursue when applying the Treaty competition rules. In the words of the Court, the competition authorities of the Member States and their courts and tribunals are required to apply Articles 101 TFEU and 102 TFEU, where the facts come within the scope of European Union law, and to ensure that those articles are applied effectively in the general interest.18

In sum, an effective system of private enforcement does not alter the basic goal of the competition rules, which is to safeguard the public interest in maintaining a free and undistorted competition, and should by no means be thought of as antagonistic to the public enforcement model. Ideally the two models can work to complement each other.19 Encouragingly, the Commission has, of late, realized this by speaking of the two limbs of antitrust enforcement as complementary and serving the same aim ‘to create and sustain a competitive economy’20 and by also stressing the public interest element in private actions for damages.21 17 Indeed, in such cases, there is no need to have recourse to an ‘antitrust injury’ doctrine in Europe, since there would be no infringement of EU competition law in the first place. 18 Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161, para 19, emphasis added. 19 See eg Recital 7 of Regulation 1/2003: ‘The role of the national courts here complements that of the competition authorities of the Member States’. Such a complementary function was advocated by the majority of the participants at the 2001 EU Competition Workshop in Florence, which dealt with private enforcement. See the individual contributions and discussions in Claus-Ehlermann and Isabela Atanasiu, eds, European Competition Law Annual 2001: Effective Private Enforcement of EC Antitrust Law, Hart Publishing, 2003. 20 See Green Paper, section 1.1: ‘The antitrust rules in Articles 81 and 82 of the Treaty are enforced both by public and private enforcement. Both forms are part of a common enforcement system and serve the same aims: to deter anticompetitive practices forbidden by antitrust law and to protect firms and consumers from these practices and any damages caused by them. Private as well as public enforcement of antitrust law is an important tool to create and sustain a competitive economy.’ 21 See Commission Staff Working Paper Accompanying the White Paper on Damages Actions for Breach of the EC Antitrust Rules, SEC(2008) 404, para 322: ‘The right of victims of a competition law infringement to bring an action for damages must thus be seen as being also in the public interest. Guaranteeing that right should therefore be considered of the utmost importance.’ See also Joined Cases T-22/02 and T-23/02, Sumitomo Chemical Co. Ltd and Sumika Fine Chemicals Co. Ltd v Commission, [2005] ECR II-4065, para 128, where the Commission pleaded that ‘the existence of remedies for civil damages may also fulfil a public interest, in so far as they are likely to dissuade infringement of the competition rules’.

146  Assimakis P Komninos

II. The independence of private antitrust enforcement Notwithstanding their substantive complementarity, private and public enforcement remain institutionally independent of each other.22 The independence of the two models means that in principle there is no hierarchical relationship between them, ie between the public authority and the ‘private attorney-general’. Indeed, Advocate General Mazák’s Opinion in Pfleiderer makes this clear.23 Introducing a rule of primacy would be problematic because it would undermine the role of courts as enforcers of equal standing. In the US, primacy of public over private antitrust enforcement and deference to the public enforcer has never been accepted by the courts as a valid principle.24 On the other hand, in Europe, this remains a question because of the unique features of EU law. However, the fact that the Court of Justice appears to have entrusted the Commission with a certain primacy over national proceedings and courts, does not contradict the analysis here.25 This ‘primacy’ is not one of the Commission, as a competition authority, over civil courts, but rather of the Commission, as a supranational EU organ, over national courts.26 Then, this principle of supremacy does not mean that a national court is widely prevented from employing reasoning which could be inconsistent with that used by the European Commission in a decision, a fortiori a decision which concerns different facts. While the ECJ has not yet been required to provide an answer to this question, the UK House of Lords, in its judgment in Crehan,27 held that a national court is not required to accord full deference to the reasoning of past Commission decisions but is merely precluded from reaching a different result than the Commission in the operative part of a specific decision which essentially deals with the same case.28 22 See further Assimakis P Komninos, ‘Public and Private Antitrust Enforcement in Europe: Complement? Overlap?’, 3(1) Competition Law Review 5 (2006). 23 Opinion of 16 December 2010, para 40: ‘I consider that Regulation No 1/2003 and the case-law of the Court have not established any de jure hierarchy or order of priority between public enforcement of EU competition law and private actions for damages.’ 24 See Clifford Jones, ‘A New Dawn for Private Competition Law Remedies in Europe? Reflections from the US’, in Ehlermann and Atanasiu, eds., Effective Private Enforcement, cited above note 19, p 99. Note that the so-called Chevron doctrine, under which courts often defer to specialised administrative agencies (Chevron USA v Natural Resources Defense Council, 467 U.S. 837 (1984)), has not been applied in the area of antitrust law and, in any event, courts are allowed to overrule the agencies if need be. 25 Case C-344/98, Masterfoods Ltd v HB Ice Cream Ltd [2000] ECR I-11369. Article 16(1) of Regulation 1/2003 has adopted verbatim the Masterfoods case law and states that national courts ‘cannot take decisions running counter to the decision adopted by the Commission’ or which ‘would conflict with a decision contemplated by the Commission in proceedings it has initiated’. 26 See also Emil Paulis and Céline Gauer, ‘La réforme des règles d’application des articles 81 et 82 du Traité’, 11 Journal des tribunaux – Droit européen 65 (2003), p 69. 27 Inntrepreneur Pub Company (CPC) et al. v Crehan, [2006] UKHL 38. 28 See Notice on the cooperation between the Commission and the courts of the EU Member States in the application of Articles 81 and 82 EC, 2004 OJ C101/54, para 8, which states that ‘the application

The Relationship between Public and Private Enforcement 

147  

Thus, according to this view, a real conflict between a Commission decision and a national court judgment would happen only if the latter were to prevent compliance by the addressee of a Commission decision with the operative part of that decision. The 2008 FIDE Community report expresses this rationale in the following terms: The Commission’s reasoning leading it to a particular decision, including its interpretation of Article 81 or Article 82 and its findings of fact, are clearly not ‘binding’ as such. The addition to the Community legal order that Commission decisions represent is not a particular interpretation of Article 81 or Article 82, or its findings of fact. It is in the operative part of the decision that specific provisions are found, creating legal effects: the obligation to pay a fine, the duty to conform to an order to cease certain behaviour or to take certain positive action. This is the part of the decision that becomes part of Community law and is vested with supremacy as long as the decision stands.29

My view is that a nuanced principle of supremacy does not call into question the principle of independence of private vis-à-vis public antitrust enforcement, precisely because of the specific characteristics of EU law and of its supranational nature. In the past, I have criticized the proposals to confer a ‘binding effect’ on all infringement decisions taken by competition authorities for the civil litigation that takes place before the civil courts,30 and I continue to do so, although I am no longer as absolute as I was. I certainly realize that some national laws have introduced specific provisions on the binding effect of national (or EU) infringement decisions on civil follow-on damages litigation.31 I also understand of Articles [101] and [102 TFEU] by the Commission in a specific case binds the national courts when they apply [EU] competition rules in the same case in parallel with or subsequent to the Commission’ (emphasis added). The same view is expressed in the Commission’s article-by-article Explanatory Memorandum to the 2000 draft Regulation, which states under Article 16 that ‘the potential for conflict depends on the operative part of the Commission decision and the facts on which it is based’ (Commission Proposal for a Council Regulation on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty and amending Regulations (EEC) No 1017/68, (EEC) No 2988/74, (EEC) No 4056/86 and (EEC) No 3975/87 (‘Regulation implementing Articles 81 and 82 of the Treaty’), COM(2000) 582 final, 2000 OJ C365E/284). See also Dorothe Dalheimer, in Dorothe Dalheimer, Christoph Feddersen and Gerald Miersch, eds., EU-Kartellverfahrensverordnung, Kommentar zur VO 1/2003 (Munich, 2005), p 120; Andrea Klees, Europäisches Kartellverfahrensrecht mit Fusionskontrollverfahren (Cologne/Berlin/Munich, 2005), pp 300–301. 29 Eric Gippini-Fournier, ‘Institutional Report: The Modernisation of European Competition Law: First Experiences with Regulation 1/2003’, in Franz Koeck and Margit Maria Karollus, eds, The Modernisation of European Competition Law, Initial Experiences with Regulation 1/2003 (Vienna, 2008), p 471. Compare the approach of Assimakis P Komninos, ‘Effect of Commission Decisions on Private Antitrust Litigation: Setting the Story Straight’, 44 Common Market Law Review 1387 (2007), pp 1397 et seq. 30 See, eg, Assimakis P Komninos, ‘The EU White Paper for Damages Actions: A First Appraisal’, 2/2008 Concurrences 84, p 91. 31 Section 58A of the UK Competition Act aims at facilitating follow-on civil actions for damages brought before the ordinary civil courts and provides that findings of infringement of UK or EU competition law by the OFT (or by the Competition Appeal Tribunal on appeal) bind the courts deciding on follow-on civil claims for damages. Section 47A of the Act, for its part, extends the binding effect of infringement findings to decisions of the European Commission. In addition, section 58A refers to follow-on civil proceedings for damages before the ordinary civil courts (the Chancery Division of the

148  Assimakis P Komninos the cost and time-related benefits for potential claimants, if they did not have to reargue the competition law infringement before a court. If these provisions are seen as a proactive mechanism to incentivize more follow-on private enforcement, the detriment to the principle of independence is acceptable. If a competition authority has decided in a judicially-reviewable final decision that there has been an infringement, I can see the rationale for disallowing the defendant to challenge anew this cardinal finding. If, however, these provisions are interpreted in such a way that courts would not be able to discuss and even contradict any of the findings contained within the administrative decision and would be turned to mere assessors of damages, then this brings into question the independence of private enforcement. For this reason, a more nuanced approach should be followed in interpreting these rules and the meagre national case law so far supports this.32 In any event, in real life, a competition agency’s infringement decision de facto is highly persuasive for a civil court. I would caution, therefore, against changing the legal reality across Europe for little – if any – real practical benefit. Conferring a general binding effect on all administrative agency findings would unsettle the relationship and balance between private and public enforcement. My concern is not so much the principles of separation of powers and judicial independence. Rather, it has to do with general principles of law and policy. An unqualified binding effect would essentially subjugate private to public enforcement; it would also withdraw from the ambit of national civil courts a substantial part of competition law disputes, in particular those referring to the infringement of the antitrust norm, thus undermining the role of courts as enforcers of equal standing. Allowing the courts to be fully involved in the application of substantive competition law would enrich national litigation and national courts would remain active players in the application and enforcement of the Treaty competition rules and would not be turned into mere assessors of damages. It may also make life a bit more difficult for competition authorities, which might feel restrained in their action, if they knew that they decide not only for themselves and the administrative proceedings, but also for the judge and the civil proceeding. For example, an authority might shy away from including broad statements in its decisions about harm to the economy caused by a specific cartel, High Court), while section 47A refers to follow-on claims brought before or transferred to the CAT. Section 33(4) of the German Act against Restraints on Competition (ie, the GWB) confers a binding effect follow-on civil litigation not only on European Commission and the Bundeskartellamt, but also on all other EU Member States competition authorities’ infringement decisions. See also Article 88/B(6) of the Hungarian Competition Act. Compare also Article 35(1) of the Greek Competition Act (Law 3959/2011), which provides that the judgments of the administrative courts reviewing the Hellenic Competition Commission’s decisions – but not the decisions themselves – have the force of erga omnes res judicata before the civil courts. 32 See, for example, Enron Coal Services Ltd v English Welsh & Scottish Railway Ltd, [2011] EWCA Civ 2. In that case, the UK Competition Appeal Tribunal, in a follow-on action, held that there was no liability in damages for lack of causation, notwithstanding the existence of an infringement decision by the sectoral regulator, which acted as antitrust agency. On appeal, the English Court of Appeal ruled that tribunals overseeing damage claims are bound by the facts contained in an antitrust decision, but stressed that these have to be clear statements and not ‘stray phrases’.

The Relationship between Public and Private Enforcement 

149  

which may be a good practice in term of advocacy and sensitizing of consumers, for fear that such statements might be ‘interpreted’ as a binding calculation of the damage caused to specific claimants in a follow-on civil proceeding. Therefore, I conclude, on this topic, by stressing that it is preferable – if at all – to introduce a rebuttable33 presumption of antitrust liability.

III. The specific issue of leniency The principle of independence of private antitrust enforcement has many serious practical consequences. The most important question concerns its impact on agencies’ leniency programmes. There are two main issues: (a) The question of civil liability itself of successful leniency or immunity applicants and (b) the extent to which leniency-related material is discoverable by litigants in a civil proceeding for damages. The particular question of reducing the civil liability of successful leniency or immunity applicants is quite complex and goes to the core of the relationship between public and private enforcement. It also touches upon the fundamental issue of the objective of private antitrust damages actions: if the paramount objective is compensation, then limitations of an immunity beneficiary’s liability will be difficult; if, on the other hand, deterrence and effectiveness of enforcement is also an objective, then it will be easier to protect the integrity and attractiveness of the leniency programme by limiting civil liability. In general, courts are not bound by administrative leniency schemes. Immunity from administrative fines is totally unconnected with civil litigation claims. The de-trebling of antitrust damages for corporate amnesty applicants in the US does not call this into question, because de-trebling does not mean protection from single damages liability, and furthermore this takes place only if the amnesty beneficiary assists the claimant in his private action.34 Thus, the specific US rule is not one-sided, but rather aims at protecting the effectiveness of both the leniency programme and private enforcement. In Europe, the 2005 Green Paper on damages had examined two options. One was to grant a successful immunity applicant the option to claim a rebate on any damages claim facing him, in return for helping claimants bring damages claims This is the case, for example, of the Cypriot Competition Act of 2008 (section 40(1)). Section 213(b) of the 2004 Antitrust Criminal Penalty Enhancement and Reform Act only limits the damages recoverable from a corporate amnesty applicant to the harm actually inflicted by the applicant’s conduct, ie to single and not treble damages, if that person also co-operates with private claimants in their damage actions against the remaining cartel members. An appropriate level of cooperation as defined by the Act involves: (a) providing a full account of all facts relevant to the civil action; (b) furnishing all documents relevant to the civil action; and (c) making oneself available for interviews, depositions and testimonies in connection with the civil action. The Act also limits the recovery of damages from amnesty applicants to damages attributable to the defendant, ie it eliminates joint and several liability for successful amnesty applicants. 33 34

150  Assimakis P Komninos against all cartel members.35 The claims against the other infringers, jointly and severally liable for the entire harm, would have remained unchanged.36 The other option was to remove joint and several liability for the successful immunity applicant and limit his liability to the share of the harm corresponding to his share in the cartelised market.37 The 2008 White Paper, however, no longer pursues any of these options38 and makes, instead, a proposal that was never discussed in the Green Paper: limiting the civil liability of successful immunity recipients39 to claims by their ‘direct and indirect contractual partners’. In other words, the immunity recipient would be liable only to persons that bought directly from him the products or services in question (direct contractual partners) or to those down the supply chain who bought these products or services from the direct contractual partners themselves.40 Thus, a victim that did not buy cartelised products or services41 directly or indirectly from him and, more interestingly, a harmed competitor, will not be able to claim damages. At the same time, this rule would in effect remove the immunity recipient’s joint liability,42 since, as the Commission explains in an example, ‘where 30% of a victim’s total purchases of cartelised products originate from the immunity recipient, the latter would only be liable for 30% of the total harm suffered by this victim due to the overcharge of the cartelised products’.43 This is certainly a novel and interesting proposal. The question is whether the limitation of the right of competitors and others not falling under the Commission’s definition of ‘direct and indirect contractual partners’ is at odds with primary EU law, ie with the Treaty itself and the ECJ’s judgments in Courage and Manfredi, which stress that the right to damages should be open to ‘any individual’. In principle, while some compromise between the principle of effective judicial protection (compensation of all harmed individuals) and effectiveness of competition law enforcement (safeguarding the effectiveness and attractiveness of the leniency programme) is desirable and can easily be accommodated by EU law, Option 29 of the Green Paper. If there had been a system of double damages for horizontal cartels, this rebate would have dedoubled the award for the immunity recipient, thus restoring single damages. 37 Option 30 of the Green Paper. 38 While the White Paper stays clear of any proposal to introduce punitive damages at EU level, such damages may be available under national law (eg in Irish law), so it is desirable to include some protection for immunity recipients from national punitive damages awards. 39 This proposal does not cover the other leniency applicants that did not receive full immunity. 40 Commission Staff Working Paper Accompanying the White Paper on Damages Actions for Breach of the EC Antitrust Rules, SEC(2008) 404, para 305. 41 A remaining issue is what happens with cartels that do not involve sale of goods or services to contractual partners (eg a cartel not to sell in a particular market ot to a particular client). 42 The White Paper considers that removal of joint liability by itself is not sufficient to effectively limit the immunity recipient’s liability (Commission Staff Working Paper Accompanying the White Paper on Damages Actions for Breach of the EC Antitrust Rules, SEC(2008) 404, para 304). Compare, however, para 322 of the Staff Working Paper, where removal of joint liability is surprisingly mentioned as a separate proposed measure. Perhaps the reference in para 322 was left in from a previous draft by mistake. 43 Commission Staff Working Paper Accompanying the White Paper on Damages Actions for Breach of the EC Antitrust Rules, SEC(2008) 404, footnote 160. 35 36

The Relationship between Public and Private Enforcement 

151  

a secondary EU legislative measure cannot result in the total exclusion of the right to damages for certain categories of persons, as guaranteed by primary Union law. For example, introducing limitations to the joint civil liability of the immunity recipient and, thus, protecting him from the risk of the other cartel members’ insolvency, does represent a sensible solution and primary EU law would not stand in the way. Similarly, protecting this undertaking from potential punitive damages at national level, or at EU level, were the EU ever to proceed to the introduction of such a system in the long term, would not raise concerns, because the Union right to damages could still be exercised by all harmed individuals with regard to the immunity recipient’s liability to pay single damages for the part of the damage that is attributable to him. The problem here is that the White Paper proposes totally to bar certain classes of individuals from claiming damages against an undertaking that has infringed Article 101 TFEU. In particular, it proposes that competitors and other potential claimants, such as customers of other cartel members or shareholders in derivative actions (assuming there is a causal link between the infringement and their harm) should not be able to claim damages from that undertaking. If we look closer, however, at the proposed solution, we see that in reality the White Paper does not propose to affect the exercise of those persons’ right to damages against the other cartel members that did not receive full immunity from fines. Indeed, joint and several liability of these cartel members continues to be the rule, so they would still be jointly and severally liable to pay damages to a harmed person for the whole of the harm. Thus, in reality what the White Paper proposes is not totally to bar some persons from suing for damages, but rather it would make those persons only slightly worse off by increasing their risk in case of the insolvency of all or some of the other cartel members. This is a rather low risk. In fact, irrespective of this White Paper proposal and of what primary Union law dictates, all claimants always bear the risk of the insolvency of all cartel members. So, it seems to me that the proposed solution would most likely not seriously affect the existence of the Union right to damages, while at the same time it would undoubtedly strengthen the effectiveness of one aspect of the leniency programme, the race to the authority to be the first undertaking that self-reports, thus ensuring full immunity.44 Being second or third would not only mean the loss of full immunity but also the exposure to damages liability for the whole of the harm.45 An even better solution46 would be to completely exclude the immunity recipient’s liability also for claims by his direct and indirect contractual partners. Again, as above, this would not dramatically affect the exercise of the right to 44 See also p 521 of the Impact Assessment Study attached to the 2008 White Paper (Making Antitrust Damages Actions More Effective in the EU: Welfare Impact and Potential Scenarios, Final Report for the European Commission, 30 March 2008). 45 Note, however, that the Commission does not propose to disallow contribution among the (nonimmunity recipient) cartel members. 46 See Assimakis P Komninos, ‘The Road to the Commission’s White Paper for Damages Actions: Where We Came From’, 4(2) Competition Policy International 80 (2008), p 104.

152  Assimakis P Komninos damages by these persons, since they could still claim compensation for the whole of their damage against the other cartel members, who would remain jointly and severally liable. As a safety valve, the law could provide that this total exclusion of liability would not apply to the exceptional case of insolvency by one or more of the jointly and severally liable (other) cartel members.47 While not affecting the right of compensation, such a solution would enhance the effectiveness of the leniency programme.48 Indeed, this has now been adopted in the Hungarian Competition Act, which, with the intention to increase the attractiveness of the leniency policy, provides that a leniency applicant receiving full immunity from fines will not be liable to pay damages to third parties so long as such damages can be collected from other cartel members (ie, from those which did not receive full immunity under the leniency policy). Moreover, ensuring that the leniency programme remains attractive and thus effective is quite beneficial for private enforcement and potential claimants. First, the claimants become aware of the cartel infringement, which is more effectively exposed to the public authority by the leniency applicants; second, the facts are established during the administrative proceedings; third, courts or claimants could, under certain circumstances, ask for documentary evidence in the hands of the public enforcer, in order to prove the damage; and fourth, a final public decision, depending on the applicable rules, may have a binding effect on the follow-on civil proceeding49 or may constitute prima facie evidence of the cartel violation. As far as the second question is concerned, discoverability of corporate statements made or submitted by leniency applicants, the ECJ’s judgment in Pfleiderer50 has in my view offered the Commission and national competition authorities support in their approach to resist or limit access to such evidence by civil claimants when the effectiveness of their leniency programmes is at stake. It is not by chance that the first preliminary reference on the relationship between private enforcement and leniency came to Luxembourg from Germany, since the Bundeskartellamt generally publishes much reduced versions of the final decision in cartel cases pursuant to a successful immunity application. In the specific case, the Bundeskartellamt had imposed fines on the three largest European producers of decor paper and on five individuals personally responsible for price-fixing agreements and agreements on capacity closure. 47 In such a case, the claimants would have to sue first the other cartel members and, in case of insolvency of the latter, they could then bring a new action against the immunity recipient for the part of harm that is attributable to him (in other words, removal of joint liability for him should here be the rule). 48 Of course, a debate is still possible, if one views Courage and Manfredi not only as authority for a Union right to damages available to victims but also as authority for a Union obligation imposed on wrongdoers (to compensate the victims). In that case, indeed, any exclusion of a wrongdoer’s liability to certain classes of victims would be contrary to primary EU law. In my view, however, the language in Courage and Manfredi (paras 26–27 and paras 89–91, respectively), which is more rights- than obligations-centred, and the powerful rationale of effectiveness would allow for a compromise in order to safeguard public enforcement and thus by implication the effet utile of Article 101 TFEU. 49 This is certainly the case under German, UK and Hungarian law. 50 Cited above note 18.

The Relationship between Public and Private Enforcement 

153  

Pfleiderer, a purchaser of decor paper, brought an action for damages against the producers of decor paper, from which it stated it had purchased goods. In order to prepare for the civil proceedings, it applied to the Bundeskartellamt for comprehensive access to the files relating to the cartel proceedings. After Pfleiderer received a version of the three decisions imposing fines, from which identifying information had been removed, and a list indicating the evidence collected in a search, it expressly requested, by way of a second application, access also to the leniency applications, the documents voluntarily transmitted by the immunity recipients and the evidence collected. The authority informed the claimant that it intended to accede to that request only in part and to limit access to the file to a version from which confidential business information, internal documents, the corporate statements themselves and documents provided by the applicants had been removed. On appeal, the referring court sent a number of questions to the Court of Justice, having essentially to do with the broader question of providing access to leniency-related documents by claimants in follow-on civil proceedings. The Court of Justice recalled first of all that neither the EU Leniency Notice, nor the ECN Notice, nor the ECN Model Leniency Programme bind the authorities and courts of the Member States, due to their soft law nature.51 Then, the Court preferred to give a judgment of principle and not to offer a list of discoverable and non-discoverable evidence, as Advocate General Mazák had proposed in his Opinion.52 Since this was a matter for which no Union legislation was in existence, it fell under the competence of the Member States. Therefore, the Court probably felt that it was impossible or – at least – inelegant for itself to construct a specific rule of discoverability. Its reliance, however, on the general principle of effectiveness53 leaves no doubts: [L]eniency programmes are useful tools if efforts to uncover and bring to an end infringements of competition rules are to be effective and serve, therefore, the objective of effective application of Articles 101 TFEU and 102 TFEU. The effectiveness of those programmes could, however, be compromised if documents relating to a leniency procedure were disclosed to persons wishing to bring an action for damages ... [Therefore,] [t]he view can reasonably be taken that a person involved in an infringement of competition law, faced with the possibility of such disclosure, would be deterred from taking the opportunity offered by such leniency programmes, particularly when, pursuant to Articles 11 and 12 of Regulation No 1/2003, the Commission and the national competition authorities might exchange information which that person has voluntarily provided.54

Ibid, paras 21–23. AG Mazák proposed that a distinction be made between the corporate statements themselves, which are put together for the specific purpose of the leniency application and should not be discoverable, and ‘all other pre-existing documents submitted by a leniency applicant in the course of a leniency procedure’, which should be discoverable (Opinion, paras 43–47). 53 Pfleiderer, cited above note 18, para 24. 54 Ibid, paras 25–27. 51 52

154  Assimakis P Komninos Then, after repeating the Courage and Manfredi mantras as to the important role of private antitrust enforcement in Europe,55 the Court ruled that a balancing exercise is here necessary that can be conducted by the national courts only on an ad hoc basis, taking into account the above principles and the overall circumstances of each case. In my view, the Court’s solution is wise and allows for more flexibility. A very specific rule on what should and what should not be discoverable, as the Advocate General had proposed, would have been too stringent and might lead to problems in specific cases. There are cases where even providing access to corporate statements may not affect the effectiveness of the leniency programme, for example when the leniency applicant itself has disclosed the contents of its corporate statement to a third party and the issue is whether other co-claimants should have access to it. More importantly, there are cases where granting access to ‘all other pre-existing documents’, as the Advocate General had proposed, may be totally unsatisfactory and a more case-by-case approach is necessary. In that sense, I think the ECJ’s judgment in Pfleiderer should be welcomed.

IV. Should there be a role for competition authorities in calculating damages (harm)? We have seen that it is inaccurate to say that public antitrust enforcement has no restorative or compensatory elements. In exceptional cases, the public authorities do enter into this field. Of course, quantification of harm resulting from anticompetitive behaviour is typically related to private antitrust litigation where the victims of such behaviour put forward claims for damages. This is a process that will be put in effect either by the court itself, after the court has established that the conditions of both antitrust and civil liability are fulfilled, or even by the parties in the course of a settlement. The question, however, is whether competition authorities should, as a matter of course, consider it their task to quantify the damage suffered by specific persons or, at least, the harm caused by anticompetitive conduct. My answer is a clear no. First, with regard to whether competition authorities should themselves calculate the damage suffered by victims of anticompetitive conduct, there have, indeed, been voices that support this, relying on the general protective duties of the public authorities but also on the antitrust agencies’ expertise in such highly complicated matters.56 It is also the case that in some legal system the competition authority Ibid, paras 28–29. For a proposal to confer powers to antitrust authorities to award civil damages to victims of anticompetitive behaviour see Igartua Arregui, ‘Should the Competition Authorities Be Authorized to Intervene in Competition-related Problems, when they Are Handled in Court? If so, what Should Form the Basis of their Powers of Intervention? National Report from Spain’, LIDC Questions 2001/2002, in: http://www.ligue.org, p 5. 55 56

The Relationship between Public and Private Enforcement 

155  

has been entrusted with specific duties in this area. For example, Article 25(c) of the Spanish Competition Act of 2007 provides that civil courts can request the competition authority (the Comisión Nacional de Competencia) to issue an opinion on the quantification of damages to be awarded.57 In Finland, Article 18a(4) of the Competition Act provides that, during the judicial proceedings regarding an action for damages, courts may request a statement from the competition authority. Then, Article 84 of the Japanese Antimonopoly Act provides that, if an action for damages has been filed, the court may ask for an opinion of the competition authority with respect to the amount of damages. I think that to impose an additional task on competition authorities, as a matter of law, and require them to assume a role in compensating victims, may be a well-intentioned but flawed idea. Competition authorities should continue to be primarily focussed on putting the infringement to an end (injunctive task) and on punishing and deterring anticompetitive conduct (punitive task).58 These specific tasks should remain central. Getting involved in the personalization and quantification of antitrust harm (ie, in identifying specific victims and awarding them specific compensatory sums) would result in the authorities’ loss of their neutral status and would inevitably mean that they will be found serving specific private interests instead of the general public interest. Therefore, the task to compensate the victims of illegal behaviour should stay in the hands of courts. Second, as far as calculation of the anticompetitive harm in general is concerned, it is fair to say that competition authorities usually avoid engaging in this process, unless they have specific reasons to do so. Agencies, when enforcing the competition rules, do not usually reason in terms of ‘quantifiable harm’, in the same way as courts do in private actions. Instead, authorities reason in terms of ‘anticompetitive object’ and ‘anticompetitive effect’. The underlying rationale behind the first concept is precisely that economic evidence generally shows that certain conduct is detrimental to competition because it causes or is likely to cause anticompetitive harm. At the same time, the fact that certain conduct is considered anticompetitive by object or even by actual or potential effect, does not necessarily imply the actual existence of quantifiable and compensable harm, in the private tort sense.59 In T-Mobile, the Court of Justice referred to this question in the following terms:

57 Article 13(3) of the old Competition Act of 1989 provided that civil courts could request the then Tribunal de defensa de la competencia to issue an opinion on the appropriateness and amount of the damages to be paid in a case of antitrust violation. 58 See Wils, cited above note 2, p 21. That is not to say that I am opposed to the practice of the European Commission and of some national competition authorities to give a fine rebate or generally reduce fines when the undertakings in question have compensated the victims of their conduct (see Wils, p 20, with examples). This, however, is connected with the degree of culpability of antitrust offenders and mitigation factors in the course of fining and does not represent a more general stance vis-à-vis the relationship between public and private antitrust enforcement. 59 See, eg, Case C-534/07 P, William Prym GmbH & Co. KG and Prym Consumer GmbH & Co KG v Commission [2009] ECR I-7415, paras 79–82.

156  Assimakis P Komninos With regard to the assessment as to whether a concerted practice ... pursues an anti‑competitive object, it should be noted, first, ... that in order for a concerted practice to be regarded as having an anti‑competitive object, it is sufficient that it has the potential to have a negative impact on competition. In other words, the concerted practice must simply be capable in an individual case, having regard to the specific legal and economic context, of resulting in the prevention, restriction or distortion of competition within the common market. Whether and to what extent, in fact, such anti-competitive effects result can only be of relevance for determining the amount of any fine and assessing any claim for damages.60

Thus, the more general concept of anticompetitive harm, to which competition authorities are better attuned, is seen in an all-encompassing and abstract manner, unlike the concept of ‘harm’ employed in the context of civil litigation, which is direct, causal, more concrete and personalized. In other words, the former is more representative of harm done to the economy than to specific harm to certain persons. Especially in the cases where a competition authority finds an infringement ‘by object’, its reluctance to quantify the resulting harm is reasonable. The authority is satisfied that such infringements, pursuant to basic principles of economics, harm competition and prefer to leave the task of quantifying the specific harm, if any, to the courts in the context of private actions. Embarking on a quantification of harm in cases of infringement ‘by object’, even if not required by the applicable legal standards, might result in the weakening of the moral and legal repulsion of the hard-core violation. As for cases of infringements ‘by effect’, most such cases refer to exclusionary conduct and it is notoriously difficult for a competition authority to quantify the resulting harm. At the same time, the fact that the authority has proved an anticompetitive effect is sometimes seen as ‘sufficient’ and the authority may see no advocacy benefit in quantifying the harm that a specific competitor has suffered. The above, however, does not suggest that competition authorities may never have to quantify antitrust-related harm in general, when prosecuting conduct. They may be obliged to do so under their law, in order to prove an infringement, or in the process of imposing fines.61 Alternatively, from an advocacy point of view, they may find it simply useful, in terms of raising awareness and enhancing the visibility and popular legitimacy of their actions. Doing so might be easy, because it may result from the case file, or difficult, if the harm can only be quantified following certain complicated economic methodologies. 60 Case C-8/08, T-Mobile Netherlands BV et al v Raad van bestuur van de Nederlandse Mededingingsautoriteit [2009] ECR I-4529, para 31 (emphasis added). 61 For example, the new Greek Competition Act provides that among the other factors (severity, duration, geographical coverage, degree of liability) relevant for the determination of the level of the fine, the financial benefit obtained by the defendant shall be taken into account (Article 25(2)(a) of Law 3959/2011). If calculation of such financial benefit is possible, the level of the fine may not be lower than that. Compare the previous version of the German Competition Act, which provided that fines in cartel cases could amount to three times the additional illegal earnings. Both provisions necessitate a degree of quantification of harm by the competition authority.

The Relationship between Public and Private Enforcement 

157  

There are cases where the public agency enforcing the competition rules may even take into account the injury to specific victims of an anticompetitive practice and impose on the wrongdoer the obligation to compensate those persons. Indeed, the public agency may pursue this informally, for example through a settlement.62 In addition, some competition regimes also provide for a role for the State in claiming damages, acting on behalf of the victims.63 If, in the course of the public proceedings, the authority finds it easy to quantify the resulting harm (eg, because it has discovered relevant evidence), there is no reason why it should not proceed to do so, but this task should not be primary. Generally, competition authorities may sometimes find it expedient to quantify harm based on calculations made by the antitrust offenders themselves. If the latter aimed at specific effects on the market and calculated such effects, there is no reason why the competition authority should not refer to such calculations and presume that they represent the actual harm. The situation of courts is slightly different: they can only draw inferences from such evidence but cannot consider it as conclusive, since they have to resolve a civil dispute and arrive at a concrete award in damages, by proving and quantifying the harm in question. In all such cases, the authority or the court should be open to arguments and evidence showing that the offenders’ calculations were misplaced. Finally, it may be possible for competition authorities to make certain statements about the harm in question, based on sound factual and economic evidence. Such statements should fall short of actually quantifying the harm, but can be extremely useful to courts and victims, which can then avail themselves of a strong basis and starting point in their efforts to prove and quantify the specific harm they suffered. A final interesting question is whether courts and authorities can rely on certain proxies or presumptions about the existence and (possibly) the quantification of harm. Empirical studies show that certain types of anticompetitive behaviour, such as cartels, result in harm and such studies have actually referred to averages of harm per type of infringement (eg, cartels).64 It may be that courts, in particular, may find such proxies prima facie useful, but their use by competition authorities is unlikely, since the proxies usually refer to types of conduct that are patently anticompetitive: in such cases the authorities are either unlikely to research the effects of the practice in question or bound to arrive at a specific figure of harm (for reasons related to calculation of fines).

See above note 5 on the OFT’s independent schools case. See above notes 6 and 7 and accompanying text. 64 See Oxera, Assimakis P Komninos et al, Quantifying Antitrust Damages: Towards Non-binding Guidance for Courts, Study Prepared for the European Commission, December 2009, pp 88 et seq, with further references. 62 63

Barry E Hawk and Yolaine Seaton*

US Antitrust Arbitration

I. Introduction A. 1985 as watershed year It is now the general rule in the United States that both international and domestic antitrust claims are arbitrable so long as the arbitration agreement is valid. Historically, US courts were reluctant to enforce agreements to arbitrate antitrust claims. The argument against the arbitrability of antitrust claims was first outlined in American Safety Equipment Corp v J P Maguire & Co1 where the Second Circuit Court of Appeals reasoned that antitrust enforcement was a matter of public policy.2 The Second Circuit also expressed concerns about the potential bias of arbitrators and stressed that the scope and complexity of antitrust claims made them inappropriate for arbitral tribunals.3 Federal courts subsequently adhered to the American Safety doctrine even though the Federal Arbitration Act of 1925 (‘FAA’)4 clearly enunciated a federal policy favoring arbitration of private legal claims. In 1985 the Supreme Court rejected the American Safety approach in Mitsubishi Motors Corp v Soler Chrysler-Plymouth, Inc.5 A Japanese car manufacturer and its Puerto Rican dealer agreed to arbitrate in Japan all disputes arising under a sales and distribution agreement. The manufacturer sued the dealer in the federal district court in Puerto Rico, seeking to compel the arbitration of alleged breaches of the sales procedure agreement.6 The dealer counterclaimed for violations of, inter alia, the Sherman Act.7 Plaintiff manufacturer then sought to compel the arbitration of the dealer’s antitrust counterclaims. The Supreme Court upheld the district court’s decision to compel arbitration of the antitrust counterclaims. The Court concluded that the rationale behind * Barry Hawk is the Director of the Fordham Competition Law Institute and Senior Counsel at Labruna Mazziotti & Segni. Yolaine Seaton is a Dean’s Fellow at Fordham Law School. 1 American Safety Equipment Corp v J P Maguire & Co, 391 F.2d 821 (2d Cir 1968). 2 Ibid at 826–827. 3 Ibid. 4 9 USC § 1 et seq. 5 Mitsubishi Motors Corp v Soler Chrysler-Plymouth, Inc, 473 US 614 (1985). 6 Alleged breaches included nonpayment of stored vehicles, contractual storage penalties, damage to manufacturer’s warranties and good will and expiration of distributorship. 7 15 USCA §§ 1 to 7.

160  Barry E Hawk and Yolaine Seaton American Safety did not justify the assumption that arbitration is an inadequate mechanism for resolving antitrust disputes.8 In the Court’s view, arbitrating antitrust claims would not violate US public policy because the antitrust statute ‘seeks primarily to enable an injured competitor to gain compensation’9 and ‘by agreeing to arbitrate a statutory claim, a party does not forgo the substantive rights afforded by the statute; it only submits to their resolution in an arbitral, rather than a judicial forum.’10 As for the complexity of antitrust matters, the Court noted the efficacy of arbitral procedures because ‘adaptability and access to expertise are hallmarks of arbitration.’11 Finally, the Court dismissed concerns about unfairness in arbitral proceedings given that parties could select ‘competent, conscientious, and impartial arbitrators’12 and if necessary can ‘attack directly the validity of the agreement to arbitrate.’13 The Court’s decision in Mitsubishi was driven by ‘concerns of international comity, respect for the capacities of foreign and transnational tribunals, and sensitivity to the need of the international commercial system for predictability in the resolution of disputes.’14 The holding was restricted to international disputes and did not directly address whether domestic antitrust claims could be arbitrated. But in 1991 the Supreme Court extended its reasoning on antitrust arbitrability for the first time to the domestic context in Gilmer v Interstate/Johnson Lane Corp.15 The Court declared that: While arbitration focuses on specific disputes between the parties involved, so does judicial resolution of claims, yet both can further broader social purposes. Various other laws, including antitrust and securities laws and the civil provisions of the Racketeer Influenced and Corrupt Organizations Act (RICO), are designed to advance important public policies, but claims under them are appropriate for arbitration.16

In the 25 years since the Mitsubishi decision in 1985 favoring if not encouraging arbitration of antitrust disputes, the total number of arbitrations involving federal or state law antitrust issues or claims is difficult to obtain. The ICC reports a total of at least 55 antitrust disputes to date.17 Since Mitsubishi there have been more than 40 federal court decisions concerning antitrust arbitration: one in the Mitsubishi, 473 US at 632–637. Ibid at 635. 10 Ibid at 628. 11 Ibid at 633. 12 Ibid at 634. 13 Ibid at 632. 14 Ibid at 629. 15 Gilmer v Interstate/Johnson Lane Corp, 500 US 20 (1991). 16 Ibid at 20–21(emphasis added). Even before Gilmer, lower federal courts had already begun to enforce arbitration agreements in purely domestic cases. 17 To date, the ICC has administered more than 15,000 arbitrations with 3465 arbitrations from 2006–2010. It should be noted that in comparison to other leading arbitral tribunals the ICC has handled an impressive number of antitrust arbitrations and the Tribunal’s experience and practice of antitrust arbitration is not necessarily reflective of the treatment of antitrust arbitration internationally. See Gordon Blanke and Phillip Landolt, eds., EU and US Antitrust Arbitration, Wolters Kluwer, 2011, at 1789–1790. 8 9

US Antitrust Arbitration 

161  

Supreme Court (Stolt-Nielson),18 17 in the Circuit Courts of Appeals and 26 in the district courts.19 Thus the following calculations may be based on an overly narrow subset of arbitrations, ie those subject to judicial intervention or review.

B. Kinds of antitrust claims 1. Horizontal per se restrictions Approximately 17 federal decisions concerned per se horizontal antitrust violations: two in the Supreme Court (Mitsubishi and Stolt-Nielson), eight in the Circuit Courts of Appeals and seven in the district courts.

2. Horizontal non-per se restrictions Approximately 10 federal decisions concerned non-per se horizontal antitrust violations: five in the Circuit Courts of Appeals and five in the district courts.

3. Vertical restrictions Approximately 13 federal decisions concerned vertical restraints: one in the Supreme Court (Mitsubishi), two in the Circuit Courts of Appeals and 10 in the district courts.

4. Monopolization/unilateral conduct Approximately 10 federal decisions concerned alleged monopolization under Section 2 of the Sherman Act: two in the Circuit Courts of Appeals and eight in the district courts.

C. Procedural postures 1. Antitrust claim raised as a defense There have been two federal decisions in which the parties raised an antitrust claim as a defense: Mitsubishi and Baxter Intern., Inc v Abbott Laboratories.20

18 19 20

Stolt-Nielsen SA v AnimalFeeds International Corp, 559 US 662 (2010). There also have been at least ten decisions concerning state antitrust laws in the state courts. Baxter Intern., Inc v Abbott Laboratories, 315 F.3d 829 (2003).

162  Barry E Hawk and Yolaine Seaton

2. Antitrust claim raised as a counterclaim There have been three federal decisions in which the parties raised an antitrust counterclaim: Mitsubishi, Swensen’s Ice Cream Co v Corsair Corp21 and Western Intern. Media Corp v Johnson.22

3. Antitrust claim raised as plaintiff claim There have been approximately 23 federal decisions in which the plaintiffs affirmatively raised antitrust claims: two in the Circuit Courts of Appeals and 21 in the district courts.

4. Class actions There have been approximately 18 federal decisions in which the plaintiffs brought antitrust class actions: one in the Supreme Court (Stolt-Nielsen), eight in the Circuit Courts of Appeals and nine in the district courts.

II. Scope of arbitrability Pursuant to the general federal court stance ‘to favor compulsory arbitration, even of antitrust disputes,’23 federal courts have generously interpreted the scope of arbitration clauses. For example, Simula, Inc v Autoliv, Inc24 concerned a development and licensing agreement between an inventor of automotive air bag systems and a supplier of automotive components. The inventor brought a federal court action alleging, inter alia, violations of Sections 1 and 2 of the Sherman Act. The inventor sought preliminary and permanent injunctive relief, treble damages for antitrust violations, punitive damages, compensatory damages, attorney’s fees and costs. Defendant licensee moved to compel arbitration pursuant to Sections 3–4 of the FAA and either to dismiss the complaint or stay the action pending resolution of the arbitration proceedings. The district court granted defendant’s motion to compel arbitration, and plaintiff appealed. The Court of Appeals for the Ninth Circuit affirmed the district court. Plaintiff argued that its antitrust claims did not fall within the scope of the arbitration clause, and that to hold otherwise would fail to protect the public policy interests promoted Swensen’s Ice Cream Co v Corsair Corp, 942 F.2d 1307 (1991). Western Intern Media Corp v Johnson, 754 F. Supp. 871 (S.D.Fla. 1991). 23 Philip Areeda and Herbert Hovenkamp, Antitrust Law, 3rd edn, Wolters Kluwer, 2007, at 217– 218. For general discussion of arbitration clauses, see Blanke and Landholt, eds, Arbitration, cited above note 17, at 80–88. 24 Simula, Inc v Autoliv, Inc, 175 F.3d 716 (9th Cir 1999). 21 22

US Antitrust Arbitration 

163  

by US antitrust law.25 The relevant provision of the licensing agreement provided that: ‘All disputes arising in connection with this Agreement shall be finally settled under the Rules of Conciliation and Arbitration of the International Chamber of Commerce by three arbitrators appointed in accordance with the said rules.’26 The court found that arbitration clauses containing the phrase ‘any and all disputes arising under the arrangements contemplated hereunder’ must be interpreted liberally so as not to ‘conflict with the strong federal policy favoring arbitral dispute resolution.’27 Once the arbitration clause was construed to encompass all disputes between the parties, the court held that the antitrust claims were arbitrable. A second example of the courts’ willingness to compel arbitration under broad contractual language is JLM Indus., Inc v Stolt-Nielsen SA28 Charterers of ocean tankers for the transport of liquid chemicals brought a class action alleging price fixing by the tanker owners.29 The transactions at issue were governed by 80 identical, industry standard, charter shipping contracts between the plaintiffs and each of the defendants.30 Each contract contained a broad arbitration clause that covered ‘any and all differences and disputes of whatsoever nature arising’ from the contracts.31 On the basis of these arbitration clauses, the defendants sought to compel arbitration of the price fixing claims but the district court denied these motions on the ground that the price fixing allegations fell outside of the scope of the arbitration clause. The Second Circuit Court of Appeals agreed with defendants and reversed. The Court of Appeals examined the standard arbitration clause in the contracts between the parties and concluded that the clause was a ‘broad’ one which covered matters ‘collateral’ to the contract itself. The court then examined the price fixing allegations and held that, because they affected the actual price terms contained in the charter agreements, the claims arose out of the underlying contract and were covered by the arbitration clause. In addition, the Court rejected plaintiffs’ argument that tanker owners could not invoke the arbitration clause ‘since the conspiratorial conduct giving rise to the claims in this case was conduct of the parent companies rather than the contracting subsidiaries.’ 32 The Court, under the principle of estoppel, declared: A nonsignatory to an arbitration agreement may compel a signatory to that agreement to arbitrate a dispute where a careful review of the relationship among the parties, the contracts they signed ..., and the issues that had arisen among them discloses that the issues the nonsignatory is seeking to resolve in arbitration are intertwined with the agreement that the estopped party has signed.33 Ibid at 721. Ibid at 720. 27 Ibid. 28 JLM Indus., Inc v Stolt-Nielsen SA, 387 F.3d 163 (2d Cir. 2004). 29 Ibid at 167. 30 Ibid. 31 Ibid. 32 Ibid at 177. 33 Ibid. 25 26

164  Barry E Hawk and Yolaine Seaton The Circuit Court concluded that the issues that defendants wanted to arbitrate were undeniably intertwined with the charter agreements, because plaintiffs entered into the charters containing allegedly inflated price terms that gave rise to the claim.

III. Unfair and unconscionable arbitration provisions Courts will not force a party to an antitrust dispute to arbitrate if the arbitration clause at issue is adjudged unfair, inappropriate or unconscionable, particularly with regard to available remedies.34 Parties must be able to effectively vindicate their ‘statutory rights under the terms of an arbitration agreement.’35 As seen below, however, the Supreme Court recently restricted state law ‘unconscionability’ limitations on class arbitrations waivers in federal courts.

IV. Antitrust class action claims A. Arbitrability of class action claims The arbitrability of class action claims in federal courts has been subject to a roller coaster of judicial views. In Green Tree Financial Corp v Bazzle36 the US Supreme Court implicitly endorsed the arbitration of antitrust class actions by not prohibiting clauses that authorized the use of such procedures. The Court held that the question of whether an arbitration agreement allowed class action claims was a matter of contract interpretation to be decided by the arbitrator, not the courts. Bazzle’s tolerant acceptance of arbitrable class action claims recently was modified by the Supreme Court in Stolt-Nielsen SA v AnimalFeeds International Corp.37 The case involved a private class action against Stolt-Nielsen after it had been found to have engaged in illegal price fixing in a prior criminal action by the DOJ. After being served with a demand for arbitration by AnimalFeeds pursuant to standard maritime arbitration clauses, Stolt-Nielson argued that it had not consented to arbitration. After defendants were served with a demand for class arbitration, the parties subsequently entered into a supplemental agreement Areeda and Hovenkamp, cited above note 23, at 217–218. In re Cotton Yarn Antitrust Litigation, 505 F.3d 274, 283 (4th Cir. 2007) (citing Green Tree Fin. Corp-Ala. v Randolph, 531 US 79, 90–91 (2000); Booker v Robert Half Int’l, Inc, 413 F.3d 77, 81 (D.C. Cir. 2005)). 36 Green Tree Fin. Corp v Bazzle, 539 US 444 (2003). 37 Stolt-Nielsen SA, cited above note 18. 34 35

US Antitrust Arbitration 

165  

that stipulated that the standard agreement between them did not expressly contemplate or otherwise address the issue of class action arbitration. This was submitted to an arbitral panel to decide the issue. The panel concluded that class arbitration was appropriate and certified the class in a partial award. Ultimately, the Supreme Court disagreed. In reversing the Second Circuit’s decision finding no manifest disregard for the law in the arbitrators’ determination that class arbitration was appropriate, the Supreme Court acknowledged that defendants faced a ‘high hurdle’ under Section 10(a)(4) of the FAA, which required them to show not merely that the panel’s decision was erroneous, but that the arbitration panel had in fact ‘exceeded [its] powers,’ straying so far from contract interpretation that its ruling effectively imposed its own public policy. The Court then assessed whether the panel exceeded its powers and concluded that the panel’s faulty determination resulted from its misunderstanding of the Court’s reasoning in Bazzle as establishing the standard for determining whether a contract may be interpreted as allowing class arbitration; that is, the arbitral panel wrongly read Bazzle as requiring it to consider the terms of the parties’ agreement to ascertain whether they intended to permit, or preclude, class arbitration.38 But because the parties had stipulated that the arbitration clause was silent as to the class action mechanism, there was no need to ascertain party intent given the parties’ ‘complete agreement’ that they did not desire it.39 The Court reasoned that the arbitral panel’s only task was to identify the ‘default rule’ governing the class arbitration issue, and that the panel had erroneously failed to inquire whether the FAA, maritime or New York law ‘contained a ‘default rule’ permitting an arbitration clause to allow class arbitration absent express consent.’ 40 Instead, the panel had determined on its own that class arbitration had its merits and was permissible as a matter of public policy.41 The majority decision placed great emphasis on the importance of party intent, holding that ‘differences between bilateral and class-action arbitration are too great for arbitrators to presume, consistent with their limited powers under the FAA, that the parties’ mere silence on the issue of class action arbitration constitutes consent to resolve their disputes in class proceedings.’42 After Stolt-Nielsen, it appears that an arbitration clause that does not clearly address the question of class arbitration risks being interpreted as ambiguous and as evidence that the parties did not agree to the class mechanism. In re TFT-LCD (Flat Panel) Antitrust Litigation43 involved multidistrict litigation stemming from allegations of a global price fixing conspiracy in the market for Thin Film Transistor Liquid Crystal Display (‘TFT-LCD’) panels. Ibid at 1772. Ibid at 1770. 40 Ibid at 1769. 41 Ibid at 1767–1770. 42 Ibid at 1776. 43 In re TFT– LCD (Flat Panel) Antitrust Litigation, 2011 WL 1753784 (ND Cal 2011). 38 39

166  Barry E Hawk and Yolaine Seaton Defendants were manufacturers who collectively had market shares between 82% and 95% for TFT-LCD panels between 1996 and 2006. During this same period the TFT-LCD industry experienced significant consolidation and a number of cross-licensing agreements, joint ventures, and other cooperative arrangements which plaintiffs alleged facilitated collusion. Direct purchaser plaintiffs (‘DPPs’) filed a consolidated class action complaint on behalf of all purchasers of TFT-LCD panels in the United States from the defendants. Only after plaintiffs’ third amended complaint did certain defendants assert general affirmative defenses involving arbitration agreements and forum selection clauses but none moved to stay litigation at that time. It was not until after the class was certified that defendants moved to stay the litigation pursuant to arbitration agreements with the DPPs. Plaintiffs argued, inter alia, that the antitrust claims were outside the scope of the arbitration clauses and that plaintiffs were not required to arbitrate claims against defendants with whom they did not sign an arbitration clause, and that defendants could not compel arbitration based on contracts with absent class members. The court agreed with plaintiffs that ‘defendants demonstrated their intent to litigate in court rather than seek arbitration by, among other things, filing two motions to dismiss, opposing class certification without raising arbitration and engaging in extensive discovery.’44 However, the court limited this finding to named members of the DPP class and concluded that defendants’ actions did not waive their right to arbitrate with unnamed members of the DPP class: ‘defendants could not have moved to compel arbitration against such entities prior to the certification of a class in this case because, as defendants point out, putative class members are not parties to an action prior to class certification.’45 The court also stated that it could not ‘determine whether an as yet unidentified DPP class member should or should not be required to arbitrate its claims against all defendants, as opposed to only the defendant with which it entered into an arbitration agreement.’46 The court reviewed other case law and reiterated that ‘a nonsignatory defendant faces a heavy burden to show that the signatory plaintiff intended to submit to arbitration, notwithstanding the absence of a formal agreement.’47 If defendants intended to seek arbitration against absent DPP class members, the court ordered that they must present each contract along with evidence sufficient to demonstrate such intent.

B. Enforceability of class action waivers Since Bazzle in 2003 opened the doors for the arbitration of antitrust class actions, companies doing business in the United States have increasingly drafted arbitration Ibid at *3. Ibid at *4. 46 Ibid at *5. 47 Ibid. 44 45

US Antitrust Arbitration 

167  

clauses that expressly prohibit arbitration on a class basis. Companies now have a general preference for combining class arbitration waivers and agreements to submit disputes to arbitration because this effectively forces plaintiffs to proceed individually in an arbitration forum. In the US challenges to class action waiver provisions of arbitration agreements are decided by the courts rather than arbitrators.48 The courts have differed on whether class action waivers are enforceable and under what conditions. In In re Am. Express Merchants’ Litig.49 and Kristian v Comcast Corp,50 both the Second and First Circuits found class arbitration waivers to be unenforceable. The courts in these cases viewed class action waivers as effectively preventing injured plaintiffs from pursuing antitrust claims because the relatively small amounts in controversy made the costs of pursuing the claims individually economically infeasible.51 In In re American Express Merchants’ Litigation,52 merchants filed a class action antitrust suit under Section 1 of the Sherman Act53 against American Express and its affiliates (hereinafter ‘Amex’). The Card Acceptance Agreement between merchants and Amex contained an arbitration provision in addition to a provision stating that ‘if arbitration is chosen by any party with respect to a claim, neither you nor we will have the right to litigate that claim in court or have a jury trial on that claim….’54 The district court granted Amex’s motion to compel arbitration pursuant to the terms of the Card Acceptance Agreement and dismissed plaintiff’s action against Amex. The district court held that the arbitration clause was paradigmatically broad enough to encompass all the plaintiff’s substantive antitrust claims and that the collective action waiver was enforceable. The Second Circuit reversed, finding that the mandatory class action waiver provision contained in the arbitration clause in card acceptance agreement was unenforceable. Amex petitioned for writ of certiorari. The Supreme Court granted writ, 559 US 1103, 130 S.Ct. 2401, 176 L.Ed.2d 920, and vacated and remanded for reconsideration in light of its decision in Stolt-Nielsen. On remand, Amex argued that the Court of Appeal’s original decision ‘cannot stand in the wake of Stolt-Nielsen, reading the decision as repeatedly emphasiz[ing] courts’ obligation to faithfully enforce (not just construe) the parties’ arbitration Blanke and Londolt, eds., Arbitration, cited above note 17, at 1357. In re Am. Express Merchants’ Litig., 634 F.3d 187 (2d Cir 2011), reversed by the Supreme Court following the completion of this chapter. See American Express v Italian Colors, 570 US (2013). 50 Kristian v Comcast Corp, 446 F.3d 25, 61 (1st Cir. 2006). 51 See Ibid at 59; In re Am. Express Merchants’ Litig., 634 F.3d 187 (2d Cir 2011). 52 In re Am. Express Merchants’ Litig., 634 F.3d 187 (2d Cir 2011). 53 Plaintiffs alleged that American Express charged a supra-competitive 3% merchant discount fee that greatly exceeded the fee charged by Visa and MasterCard. Plaintiffs contended that this supra-competitive fee could be sustained only through enforcement of an Honor All Cards (‘HAC’) provision, included in every merchant contract, that unlawfully tied American Express’s charge and credit card services in violation of § 1 of the Sherman Act. Merchants who chose to accept the charge card had to agree to accept all American Express cards. In re Am. Express Merchs. Litig., 2006 WL 662341, at *1. 54 In re Am. Express Merchants’ Litig., 634 F.3d 187 (2d Cir 2011). 48 49

168  Barry E Hawk and Yolaine Seaton agreement’ and that ‘courts [cannot] invalidate the parties’ agreement ...based on the absence of class procedures.’55 The Court of Appeals disagreed, concluding that it did not follow from Stolt-Nielsen that ‘a contractual clause barring class arbitration is per se enforceable.’56 Instead of examining the enforceability of class action waivers under state unconscionability law, the court, like other federal circuits, chose to focus its analysis on ‘whether the class action waiver is enforceable when it would effectively strip plaintiffs of their ability to prosecute alleged antitrust violations’ pursuant to Section 2 of the FAA.57 The circuit court emphasized that there is a strong ‘public interest in the vigilant enforcement of antitrust laws through the instrumentality of the private trebledamage action,’ class actions included.58 Thus, ‘any agreement which confers even a partial immunity from civil liability for future violations of the antitrust laws is inconsistent with the public interest. Class actions are well-recognized as an economically rational alternative [for vindicating statutory rights] when a large group of individuals or entities has suffered an alleged wrong, but the damages due to any single individual or entity are too small to justify bringing an individual action.’59 The court found that plaintiff merchants met their burden of showing that the cost of individually arbitrating their dispute with Amex would ‘ensure[ ] that no small merchant may challenge American Express’s tying arrangements under the federal antitrust laws.’60 The court declared further that Stolt–Nielsen should not be read to ‘imply that an arbitration is valid and enforceable where, as a demonstrated factual matter, it prevents the effective vindication of federal rights.’61 Kristian v Comcast Corp62 involved customers who brought state and federal court actions against Comcast, a cable television provider, for antitrust violations alleging that Comcast traded customers with competitors through ‘swapping agreements’ thereby effectively eliminating competition in a given geographical area. The state case was removed to federal court. Comcast moved to compel arbitration. On appeal, the First Circuit first noted that, under Mitsubishi, arbitration was an acceptable alternative to litigation, ‘so long as the prospective litigant effectively may vindicate its statutory cause of action in the arbitral forum.’63 It then inquired whether the class arbitration waiver precluded the plaintiffs from obtaining a recovery because their individual claims were too small to pursue separately.64 The court found this to be the case on the facts before it, severed the class action Ibid at 193. Ibid. 57 Ibid at 194. 58 Ibid at 197. 59 Ibid (internal quotation marks omitted). 60 Ibid at 197–199. 61 Ibid at 199. 62 Kristian v Comcast Corp, 446 F.3d 25 (1st Cir. 2006). 63 Ibid at 54 (quoting Mitsubishi, 473 US at 637). 64 Ibid at 55. 55 56

US Antitrust Arbitration 

169  

ban from the arbitration agreement and ordered the arbitration to go forward under the arbitration clause as modified.65 The enforceability of class waivers in these cases was examined under the ‘vindication of statutory rights’ standard.66 Another legal test to assess enforceability of class action waiver is unconscionability67 which is grounded in state law rather than federal law. Courts in some states, particularly California, have condemned the use of class action waivers whereas others, such as New York, have accepted them as neither unconscionable nor contrary to public policy. However, in April of this year the Supreme Court in AT&T Mobility LLC v Concepcion,68 (by a five-to-four vote) held that states cannot prohibit the inclusion of class action waiver provisions in arbitration agreements. Thus, the unconscionability defense to compulsory arbitration is no longer available as a basis for invalidating class waiver provisions in arbitration agreements governed by the FAA. The case arose under a cell phone service contract whose arbitration clause provided that all disputes be arbitrated and that all claims be brought in an ‘individual capacity, and not as a plaintiff or class member in any purported class or representative proceeding.’69 Plaintiffs were disgruntled with a sales tax imposed on advertised ‘free phones’ under the agreement and sued AT&T in federal district court in California. The district court deemed the class arbitration waiver unconscionable and unenforceable under the California Discover Bank rule.70 The Ninth Circuit agreed with the district court’s reasoning and in addition held that the FAA, which makes arbitration agreements ‘valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract,’71 did not preempt its ruling. A majority of the Supreme Court reversed, holding that the FAA preempts California’s state law rule. The Court struck down the Discover Bank rule on the ground that a rule ‘requiring the availability of classwide arbitration interferes with fundamental attributes of arbitration and thus creates a scheme inconsistent with the FAA.’72 In reversing the Ninth Circuit’s decision, the Supreme Court explained that the FAA’s primary purpose is to ensure that parties have discretion in designing arbitration processes and that arbitration agreements are enforced according to their terms. For the majority, arbitration should be an informal, efficient and 65 Ibid at 64. In reaching this conclusion, the court examined, and distinguished, the Truth in Lending Act (‘TILA’) cases in which other circuits had upheld class arbitration waivers. In particular, the court held that TILA claims were far different and far simpler than antitrust claims, and that a class action waiver in the TILA context might not deprive a plaintiff of his statutory rights, whereas in the antitrust case before the court, it would. Ibid at 58–59. 66 Blanke and Landolt, eds., Arbitration, cited above note 17, at 1137. 67 Ibid at 1357. 68 AT&T Mobility LLC v Concepcion, 2011 WL 1561956 (US). 69 Ibid at *1. 70 Discover Bank v Superior Court, 36 Cal. 4th 148 (2005). The criteria to determine the unconscionability of a waiver provision are (i) it is in an adhesion contract; (ii) it governs disputes over small sums of money; and (iii) it is alleged to be part of a scheme to deliberately cheat consumers out of individually small sums of money. 71 9 USC § 2. 72 AT&T Mobility, 2011 WL 1561956 at *8.

170  Barry E Hawk and Yolaine Seaton expedient dispute resolution tool.73 Rules that dictate the procedures for arbitration can undermine its informality to such an extent that the rules effectively hinder the resolution of disputes. The Court illustrated this point by restating its previous holdings that state laws or rules invalidating arbitration agreements that failed to require judicially supervised discovery during arbitration or the use of Federal Rules of Evidence would prevent the ‘streamlined proceedings’ that arbitration agreements are supposed to facilitate.74 Rules ‘aimed at destroying arbitration’ or ‘demanding procedures incompatible with arbitration’ are preempted by the FAA regardless of whether they arise from a generally applicable rule of contract law such as the unconscionability doctrine. That is, ‘the grounds available under FAA Section 2’s saving clause, should not be construed to include a State’s mere preference for procedures that are incompatible with arbitration and would wholly eviscerate arbitration agreements.’75 Applying this rationale to class arbitrations, the Court offered three reasons why the California Discover Bank rule was ‘inconsistent with the FAA.’ First, ‘the switch from bilateral to class arbitration sacrifices the principal advantage of arbitration—its informality—and makes the process slower, more costly, and more likely to generate procedural morass than final judgment.’76 Second, class arbitration requires a great deal of procedural formality to bind absent class members to the arbitral awards and the Court reasoned that it was highly unlikely that, in passing the FAA, Congress had meant to entrust an arbitrator ‘with ensuring that third parties’ due process rights are satisfied.’77 Third, the Court was particularly concerned that there would be an increased risk to defendants because arbitration was ill-suited for the high stakes of class proceedings and errors made in the course of rendering large judgments would not be subject to adequate judicial review. Thus, according to the majority, class arbitration is an ill-conceived procedure that makes dispute resolution lengthy, complicated and less confidential, with no resemblance to traditional arbitration. In his dissent, Justice Breyer argued that class arbitration is consistent with the use of arbitration – a position supported by the American Arbitration Association, which described class arbitration as ‘a fair, balanced, and efficient means of resolving class disputes.’78 Justice Breyer also provided numerous counterexamples in which parties willingly submitted high-stakes disputes to arbitration79 and emphasized that class proceedings had their advantages, such as greater simplicity and speed compared to full-blown class litigation.80 Ibid. Ibid at *8–9. 75 Ibid at *7 (internal quotation marks omitted). 76 Ibid at *10. 77 Ibid at *11. 78 Ibid at *19. 79 Ibid at *20. 80 The dissent also accused the majority of contravening basic federalism principles by forcing on states its substantive understanding of what arbitration is. States should be allowed to strike down arbitration agreements as long as there is no discrimination against arbitration. Ibid at *22. Justice 73 74

US Antitrust Arbitration 

171  

In sum, AT&T Mobility LLC v Concepcion strongly suggests that although the FAA can be regarded as a statute favoring arbitration, the FAA disfavors class arbitrations. Postscript: On 20 June 2013, the US Supreme Court reconfirmed AT&T Mobility and rejected the 2nd Circuit’s position in Amex (described above).

V. Claims concerning horizontal price fixing/per se violations As discussed above, the court in JLM Indus., Inc v Stolt-Nielsen SA81 concluded that an arbitration agreement can extend to antitrust claims alleging a price fixing conspiracy where plaintiffs seek joint and several liability against their contracting counterparts, even though other nonsignatory co-conspirators are involved. The court declined to accept the argument that horizontal price fixing claims are not arbitrable.82 Similarly, in Ross v American Express Co83 individual credit cardholders and credit card issuers sought interlocutory appeals from a district court decision compelling cardholders to arbitrate, even though defendant issuer was not party to arbitration clauses binding cardholders. Plaintiff credit card holders (MasterCard, VISA and Diners Club credit cards) alleged that American Express (‘Amex’) had conspired with other issuers to fix foreign currency transaction fees. Plaintiffs were not holders of Amex credit cards and did not seek relief as purchasers of Amex products. Each plaintiff had entered into a standard cardholder agreement with its Issuing Bank. The cardholder agreements included a broad arbitration clause which provided that ‘[a]ny dispute, claim, or controversy ... arising out of or relating to this Agreement’ be settled in an arbitral, not a judicial, forum. The district court granted Amex’s motion to compel arbitration, quoting JLM Indus., Inc v Stolt-Nielsen SA: ‘a nonsignatory to an arbitration agreement may compel a signatory to that agreement to arbitrate a dispute where a careful review ... discloses that the issues the non-signatory is seeking to resolve in arbitration are intertwined with the agreement that the estopped party has signed.’ The district court also cited the reasoning of Denney v BDO Seidman, LLP, that allegations of concerted misconduct between a signatory and a nonsignatory could establish a close relationship between them sufficient to invoke the doctrine of equitable estoppel and found that plaintiffs and Amex possessed the requisite close relationship to compel arbitration of their dispute. Breyer suggested that the majority’s preoccupation with the ‘merits and demerits of class actions’ was the driving force behind its decision, which consequently encroaches on California’s federalism interests by forcing upon California its view that class arbitration has none of the benefits of arbitration due to its procedural deficiencies. 81 82 83

JLM Indus., Inc, cited above note 28. Ibid at 182. Ross v American Express Co, 547 F.3d 137 (2d Cir. 2008).

172  Barry E Hawk and Yolaine Seaton The Court of Appeals for the Second Circuit reversed, holding that Amex could not, via equitable estoppel, compel cardholders to arbitrate based on arbitration clauses not signed by Amex. The Court of Appeals began its analysis with a reminder that arbitration ‘is a matter of consent not coercion’ and that ‘[w]hile the FAA expresses a strong federal policy in favor arbitration, the purpose of congress in enacting the FAA was to make arbitration agreements as enforceable as other contracts, but not more so.’84 The Court reasoned that the district court improperly extended the principle of equitable estoppel by compelling arbitration in a situation in which the parties did not have some sort of ‘corporate relationship to a signatory party; that is, th[e] Court has applied estoppel in cases involving subsidiaries, affiliates, agents, and other related business entities.’ 85

VI. Remedies in antitrust arbitration The statutory remedies under the federal antitrust laws generally are available in arbitration to plaintiffs who have suffered antitrust injury and these remedies are not contractually waivable.86 Accordingly, treble damages are available in antitrust arbitrations and it is mandatory for arbitrators to award them.87 Treble damages are ‘not punitive for the purposes of interpreting the scope of an arbitration clause’ but are ‘a mere mathematical expansion of the actual damages calculated by the arbitrator.’88 Arbitrators also ‘have the power to fashion equitable relief.’89 For example, the Fifth Circuit Court of Appeals confirmed an arbitration award that granted a mandatory injunction which compelled the defendant to offer the plaintiff a new contract that contained the reduced price terms that defendant had offered to a competitor of the plaintiff. The court emphasized that injunctive remedies are available under Section 16 of the Clayton Act and no distinction should be made between the courts and arbitral fora.90 In another case, parties asserted that the arbitrator had insufficient authority to arbitrate equitable claims under an arbitration agreement specifying that all equitable claims were the exclusive jurisdiction of state and federal courts. The Ibid, 547 F.3d at 140. Ibid, 547 F.3d at 144. Blanke and Landolt, eds., Arbitration, cited above note 17, at 1400. See Mitsubishi, 473 US at 637, noting that the claimant must be able to ‘effectively … vindicate its statutory cause of action in the arbitrable forum’; Kristian v Comcast Corp, 446 F.3d 25, 37, 46 (1st Cir. 2006), noting that ‘the remedies provided by the antitrust statute cannot be contractually waived.’ 87 Ibid. 88 Investment Partners, LP v Glamour Shots Licensing, Inc, 298 F.3d 314 (5th Cir. 2002). 89 Ibid. 90 Am. Cent. Eastern Tex. Gas. Co v Union Pacific Res. Group, Inc, 93 Fed Appx 1, 11 (5th Cir. 2004). 84 85 86

US Antitrust Arbitration 

173  

Ninth Circuit Court of Appeals held that the arbitrator ‘had authority over all disputes, equitable and legal and…he did not exceed his authority by arbitrating equitable claims.’91 The court upheld a permanent injunction against the plaintiff and its principals.92 Finally, the Fourth Circuit Court of Appeals has stated that arbitrators ‘enjoy broad equitable powers’ and they ‘may grant whatever remedy is necessary to right the wrongs within their jurisdiction.’93

VII. Arbitration in doj and ftc enforcement actions A. Overview The Administrative Dispute Resolution Act (‘ADRA’) permits federal agencies, including the Department of Justice (‘DOJ’) and the Federal Trade Commission (‘FTC’), to use arbitration and other alternative dispute resolution (‘ADR’) proceedings where the parties consent.94 The statute also directs the agencies to ‘consider not using’ ADR in six situations.95 An agency must consider whether an ADR proceeding is appropriate if a definitive or authoritative resolution of the matter is required for precedential value. Agencies must also consider not using ADR in a matter that involves or may bear upon significant questions of Government policy that require additional procedures before a final resolution may be made, if the ADR proceeding likely would not serve to develop a recommended policy for the agency. Maintaining established policies is of special importance and so the ADRA cautions against ADR proceedings if they would not likely reach consistent results among individual decisions. The ADRA discourages using arbitration or ADR if a full public record of the proceeding is important or if the matter significantly affects persons or organizations who are not parties to the proceeding. Finally, agencies should refrain from using ADR if the proceeding would interfere with the agency maintaining continuing jurisdiction over the matter. While the DOJ and FTC are statutorily authorized to ‘induce the target of an investigation to agree prospectively to compulsory, binding arbitration of identified areas of sensitivity or potential misconduct,’96 they apparently rarely if ever have used arbitration to resolve disputes between themselves and targets. Instead, over the past several years, both agencies have sparingly required arbitration to resolve Comedy Club, Inc v Improv West Associates, 553 F.3d 1277, 1284–1286 (9th Cir. 2009). Ibid. Gilmer v Interstate/Johnson Lane Corp, 895 F.2d 195, 199 (4th Cir. 1990), affirmed, 500 US 20 (1991) (citation omitted). 94 5 USC §§ 571, et seq. 95 5 USC § 572(b). 96 Blanke and Landolt, eds., Arbitration, cited above note 17, at 1299. 91 92 93

174  Barry E Hawk and Yolaine Seaton disputes that may arise between targets and third parties when implementing remedies pursuant to consent decrees or administrative enforcement orders. The arbitration provisions that have been incorporated into remedial decrees and orders have ‘varied greatly with respect to the applicable rules and other conditions, as well as the agencies’ control and oversight of and participation in the arbitration.’97 The provisions usually state that the arbitration is binding and, except in United States v Imetal SA, 98 do not generally provide for or authorize agency participation in the arbitration.99 Agency arbitral mechanisms have mostly been used in merger cases brought under section 7 of the Clayton Act. Arbitration provisions apparently have been used only once in a matter brought under Section 1 of the Sherman Act.

B. Arbitration in DOJ consent decrees The DOJ has included arbitration in six consent final judgments.100 The arbitration provisions do not address the resolution of liability issues between the DOJ and defendants, but rather focus on the defendants’ remedial obligations to third parties. The only final consent judgment pertaining to a Section 1 Sherman Act violation was in United States v El Paso Natural Gas Co.101 The DOJ alleged that El Paso, one of the largest gas pipeline operators in New Mexico, was unlawfully requiring gas well owners in New Mexico looking to connect to the pipeline to purchase metering equipment from El Paso in addition to its natural gas collection services. As a result of this tying arrangement the price for meter installation was raised. The consent decree required El Paso to arbitrate any future dispute that could not be resolved under an agreement between El Paso and a well operator in which the well operator was permitted to perform meter installation in accordance with El Paso’s allowable ‘specifications, standards and procedures.’102 Arbitration pursuant to this provision was to be binding and the arbitrator would apportion the costs.103 However no reference was made to applicable law or rules for the arbitration etc. The parties did not have to notify the DOJ of any arbitral proceedings initiated under the judgment. Ibid at 1746. United States v Imetal SA, 2000–1 Trade Cases (CCH) (D.D.C. 25 May 2000). 99 Blanke and Landolt, eds., Arbitration, cited above note 17, at 1758. 100 See United States v Cargill, Inc, 2000–2 Trade Cases (CCH) (D.D.C. 30 June 2000); United States v Imetal SA, 2000–1 Trade Cases (CCH) (D.D.C. 25 May 2000); United States v El Paso Natural Gas Co, 1995–2 Trade Cases (CCH) (DDC, 4 August 1995) (final judgment); United States v Morton Plant Health Sys., 1994–2 Trade Cases (CCH) (MD Fla, 28 June 1994) (final consent judgment); United States v The Dow Chemical Co, 1987–2 Trade Cases (CCH) (ND M, 12 August 1987) (final judgment); United States v Data Card Corp, 1987–1 Trade Cases (CCH) (DDC, 29 January 1987) (final judgment). 101 El Paso Natural Gas Co, 1995–2 Trade Cases at 65. 102 Ibid at para. 75,364 – 75,365. 103 Ibid. 97 98

US Antitrust Arbitration 

175  

It is common for arbitration provisions in merger consent decrees to involve remedial divestiture to a specific buyer or requirements relating to supply agreements concerning quantities of critical inputs to be sold for their fair market value. This is generally mandated to ‘ensure that the purchaser of the divested product line would be ‘a viable, on-going business’ with the ‘managerial, operational and financial capacity to compete effectively’ in the relevant market.’104 If either the seller or the purchaser cannot agree on a fair market price, they can elect to settle the issue through arbitration after ‘entering an agreement that is acceptable’ to the DOJ, and abide by rules set out in the consent judgment.105

C. Arbitration in FTC administrative orders The FTC has had similar infrequent use of arbitration proceedings106 due to four constraints unique to its role as a governmental entity that make arbitration unsuitable for resolving substantive antitrust matters. First, most if not all of the six situations highlighted in Section 572(b) of the ADRA prompting agencies to consider not using such proceedings are commonly present in the majority of FTC antitrust enforcement matters. Second, the FTC risks unlawful subdelegation of its decision-making authority should the agency resort to arbitration proceedings to resolve antitrust cases or significant issues within those cases. The FTC may not, without congressional authorization, subdelegate to an outside party its responsibilty to decide the substantive antitrust dispute.107 The concern is that arbitration is binding or may undermine the agency’s independent deliberations and decisions. If the FTC 104 United States v Data Card Corp, 1987–1 Trade Cases(CCH) (DDC, 29 January 1987), at para. 59,822. 105 For example, the arbitration provision in the consent decree in United States v Imetal, et al., Final Judgment dated 26 May 2000, http://www.justice.gov/atr/cases/f223200/223278.htm provides as follows: 1. Any controversy to be settled by arbitration shall be submitted to the American Arbitration Association; 2. The arbitrator appointed shall be one acceptable to the United States in its sole discretion; 3. The United States shall provide its assistance to the arbitrator and may submit evidence; 4. Rules and procedures shall be adopted to ensure that the controversy shall be completed within four months from the appointment of the arbitrator and any award made pursuant to any arbitration shall be final and binding on the parties to the arbitration. 106 See, eg, In re Evanston Northwestern Healthcare Corp, No. 9315, 2008 FTC LEXIS 47 (FTC, 28 April 2008) (final order); In re DTE Energy Co, 131 FTC 962 (2001) (decision and order); In re America Online, Inc, 131 FTC 829 (2001) (decision and order); In re El Paso Energy Corp, No.C3915, 2000 FTC LEXIS 7 (FTC, 6 January 2000) (decision and order); In re Ciba-Geigy Ltd. 123 FTC 842 (1997) (decision and order). 107 Perot v Federal Election Commission, 97 F.3d 553, 559 (DC Cir. 1996), cert. denied (1997). See also R.H. Johnson & Co v SEC, 198 F.2d 690, 695 (2d Cir), cert denied 344 US 855 (1952) (holding that, although it is unlawful for an agency to subdelegate its decision-making responsibilities to an outside party without Congressional authorization, the subdelegation is not unlawful when the agency retains the ultimate decision-making authority).

176  Barry E Hawk and Yolaine Seaton wishes to utilize binding arbitration, the ADRA imposes special prerequisites requiring the involvement of both the FTC Chairman and the US Attorney General.108 Not surprisingly, these requirements ensure that the FTC seldom engages in binding arbitration. Unlawful subdelegation is of less concern for the DOJ, which ‘initiates all of its antitrust cases, merger and non-merger matters alike, in federal district court, where it then appears as a litigant.’109 Third, the FTC’s use of arbitration could, if not handled properly, involve an unauthorized augmentation of the agency’s budget. ‘An agency may not augment its appropriations from an outside source without specific statutory authority.’110 If another party reimburses the FTC for the costs of using an arbitrator, the payment might be viewed as an unlawful augmentation. Finally, rules protecting the confidentiality of information could prevent the FTC from sharing even pertinent information with neutrals and third parties.111 The FTC Act Sections 6 and 21b–2(b)(3)(C) provide that the relevant information that the FTC receives from third parties during the course of its investigation must be treated with confidentiality and cannot be given to an independent neutral. As a result the FTC staff probably could not share its investgatory material with an arbitrator unless that arbitrator were an officer or employee of the Commission, which would create a conflict of interest, or unless the arbitrator were the officer or employee of another law enforcement agency, in which case he still cannot provide information to the other party in the arbitration proceedings.112 Arbitration provisions in FTC administrative orders ordinarily do not pertain to violations of the orders themselves, but are aimed at commercial disputes between the party under order and third parties. For instance, in In re Evanston Northwestern Healthcare Corp113 the FTC Order required that persons paying for hospital services have the option to resolve any price term disputes first via mediation and if no resolution was found then arbitration by a single mutually agreed upon arbitrator. In In re America Online, Inc114 the FTC mandated that any disputes, with third non-merging parties in connection with compliance with any of the rates, terms and conditions in the agreement ordered between the parties, be submitted to arbitration. The arbitrator had no authority to resolve issues concerning Respondents’ compliance with the order. Lastly, in In re Exxon Corporation and Mobil Corporation, where parties failed to agree on various commercial terms, the FTC ordered binding arbitration.115 5 USC 575(c). Blanke and Landolt, eds., Arbitration, cited above note 17, at 1515. Ibid at 1520. 111 Ibid at 1519–1520. 112 Ibid. 113 In re Evanston Northwestern Healthcare Corp, FTC No. 9315, 2008 FTC LEXIS 47 (28 April 2008) (final order). 114 In re America Online, Inc, 131 FTC 829 (2001) (decision and order). 115 In re Exxon Corporation and Mobil Corporation, FTC No. C-3907 (30 November 1999) (Agreement Containing Consent Orders) ¶¶ XIV.B.5, XV.A, and XV.C.2, http://www.ftc.gov/ os/1999/11/exxonmobilagr.pdf. 108 109 110

J Thomas Rosch

Designing a Private Remedies System for Antitrust Cases– Lessons Learned from the US Experience

Thank you for inviting me to participate in your workshop. Today I want to share with you my own perspective—as a public enforcer and previously as a trial lawyer for private litigants—on the function and structure of a private remedies system for antitrust cases in the European Union. Consistent with the goals of this workshop, I hope that my remarks set down here on paper and what I discuss orally will stimulate debate, critique and brainstorming among the workshop participants regarding the topics of private remedies and collective redress in the EU.

I Let me start by summarizing what I understand to be the current state of play in the EU regarding private remedies for antitrust cases. As I see it, there are three elements. The first is that Vice-President and Commissioner for Competition Joaquín Almunia (and Commissioner Neelie Kroes before him) have been determined to establish a system of private remedies in the EU in order to supplement public enforcement. One reason for having private enforcement as well as public enforcement is the fact that the EU has continued to grow through the accretion of Member States, which obviously enlarges the territorial scope of application of EU laws and hence increases the number of potential enforcement cases.1 As a result, and as pointed out in an October 5, 2010 information note signed by VicePresident Almunia and his fellow Commissioners for Justice and Consumer Policy and submitted to the EU College of Commissioners, there needs to be a more decentralized enforcement of EU law, which logically raises the question of ‘whether further mechanisms of private enforcement should be added to the current system of EU remedies in order to strengthen the enforcement of EU law’.2 * At the time of writing, Commissioner, US Federal Trade Commission. Currently, Of Counsel, Latham and Watkins. The views stated here are my own and do not necessarily reflect the views of the Federal Trade Commission or other Commissioners. I am grateful to my attorney advisor, Henry Su, for his invaluable assistance in preparing this paper. 1 Vice-President Viviane Reding, Vice-President Joaquín Almunia and Commissioner John Dalli, European Commission, Towards a Coherent European Approach to Collective Redress, Joint Information Note (5 October 2010), at 3, ¶ 3, http://ec.europa.eu/transparency/regdoc/rep/2/2010/ EN/2-2010-1192-EN-1-0.Pdf. 2 Ibid. Former Commissioner Mario Monti has made the same observation: ‘For a future, enlarged Community with 27 or 28 Member States, it is not a desirable—or even a viable—concept that the

180  J Thomas Rosch Another reason for having private enforcement as well as public enforcement is the European Court of Justice’s 2001 pronouncement in Courage Ltd. v Crehan that a private right of action ‘strengthens the working of the [EU] competition rules and discourages agreements or practices, which are frequently covert, which are liable to restrict or distort competition’.3 Private actions brought in the national courts ‘can [therefore] make a significant contribution to the maintenance of effective competition in the [EU]’.4 Commissioner Kroes was presumably referring to the Courage decision when she said ‘that the European Court of Justice has been very clear when it pronounced that the right to damages is a necessary element to guarantee the full effectiveness of the [EU] competition rules’.5 In her view, ‘[t]he Commission, as the guardian of the Treaty, is therefore required to take any action that is necessary to make that right a reality’.6 The second element of the current state of play is that a system of ‘collective redress’ is envisioned, which the October 5, 2010 information note describes as ‘a broad concept encompassing any mechanism that may accomplish the cessation or prevention of unlawful business practices which affect a multitude of claimants or the compensation for the harm caused by such practices’.7 This concept emerges from the recognition that certain violations of EU law, such as infringements of EU competition law, may harm a large group of citizens and businesses in the EU, so as to make individual redress an ineffective means of stopping the violations or obtaining compensation for the resulting harms.8 Moreover, individual redress may produce different and inconsistent results in the national courts of the Member States. If a claim arises from a breach of EU law that affects all EU citizens in the same way, then the right to compensatory or injunctive relief available to each EU citizen for that breach should be the same, regardless of where the action is instituted.9 A couple of observations should be stressed at the outset about collective redress. One observation is that the concept of collective redress is not new in the EU.10 Rather, what is—or would be—new is a coherent approach, which means getting the Commission as a whole, the national authorities of the Member States, and the application of the EC competition rules should largely be limited to administrations acting as public enforcers’. Monti, ‘Effective Private Enforcement of EC Antitrust Law’, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2001: Effective Private Enforcement of EC Antitrust Law, Hart Publishing, 2003, at p 3. 3 Case C-453/99, Courage Ltd. v Crehan [2001] ECR I-6297, para 27; accord Joined Cases C-295/04 to C-298/04, Manfredi v Lloyd Adriatico Assicurazioni SpA [2006] ECR I-6619, para 91. 4 Ibid. 5 Neelie Kroes, ‘Collective Redress – Delivering Justice for Victims’, Address at the ALDE Conference (4 March 2009), at 4, http://europa.eu/rapid/pressReleasesAction.do?reference=SPEECH /09/88&format=PDF&aged=1&language=EN&guiLanguage=en. 6 Ibid. 7 Joint Information Note, cited above note 1, at 4, ¶ 7. 8 Ibid at 3, ¶ 4. See also Joaquín Almunia, ‘Common Standards for Group Claims Across the EU’, speech, Valladolid, 15 October 2010, at 4, http://europa.eu/rapid/pressReleasesAction.do?reference=S PEECH/10/554&format=PDF&aged=1&language=EN&guiLanguage=en. 9 Almunia, ibid, at 4. 10 Joint Information Note, cited above note 1, at 4–5, ¶¶ 8–9.

Designing a Private Remedies System for Antitrust Cases – the US Experience  181   various stakeholders to agree on a common EU-wide approach.11 To this end, the October 5, 2010 information note sketches out some core principles underlying an EU-wide approach but fundamental questions relating to a collective redress mechanism still loom on the horizon, such as who would be allowed to bring a representative action and whether participants would be allowed to opt in or to opt out.12 The other observation is that the concept of collective redress is not confined to infringements of EU competition law. It would also apply to other violations of EU law, such as environmental and consumer protection law. As a result, the process of developing private enforcement for antitrust cases likely involves at least two steps: the Commission first must reach agreement on a set of common principles for collective redress, and after that, Vice-President Almunia will be able to present a specific proposal for antitrust damages actions that sets common standards and minimum requirements for the national authorities to implement in their courts.13 The third element of the current state of play is that both Vice-President Almunia and Commissioner Kroes have promised not to import perceived flaws in the US system of private enforcement into the EU model.14 It is this promise that frames the discussion and debate today. So what I want to do is to give you a first-hand, insider account of what exactly are the flaws in the US system, and what fixes or safeguards I would recommend be put into place in the EU to avoid the same problems and to allay concerns among constituencies and stakeholders about any type of collective redress mechanism.

II I would like to acknowledge that the current system in the US is flawed in at least four respects, based on my experience in defending firms in many antitrust trebledamages class actions during the forty-plus years that I have practiced. A first flaw is the development of treble-damages antitrust class actions. This flaw was exacerbated by the proliferation of indirect purchaser class actions in a number of states. When I started practising antitrust law in 1965, antitrust treble-damages cases were rarely brought by classes of plaintiffs. They were instead brought by single plaintiffs. The trebling of damages (and an award of attorney’s fees) 11 Ibid at 5, ¶¶ 10–11. The fact that there have been prior experiences in the EU with collective redress may arguably make it more difficult to adopt a coherent approach because constituencies and stakeholders may prefer a particular approach based on their own prior experiences. 12 Almunia, cited above note 8, at 6. 13 Ibid at 5. 14 See ibid at 6 (‘Second, collective action in Europe has not led to abuse – and this is something we should be proud of. We must identify safeguards that will prevent importing a US-style litigation culture and ensure balance in our European approach for collective redress’.); Kroes, cited above note 5, at 4 (‘Let me also explain the safeguards our proposal puts in place. I would like to assure you we are not proposing anything like the US system. Not at all’.).

182  J Thomas Rosch under Section 4 of the Clayton Act15 was considered incentive enough to initiate a private antitrust lawsuit.16 According to court statistics, antitrust class actions gained some popularity in the 1970s but then declined dramatically in the 1980s.17 By the mid-1990s, however, class actions had again become the rule, not the exception.18 To quote Professor Steve Calkins, who was General Counsel to the Federal Trade Commission at the time he wrote these remarks, ‘Antitrust class actions are once more in vogue’.19 Not only was there a resurgence in antitrust class actions in the federal courts but many actions were being filed in the state courts as well, as a result of the so-called ‘Illinois Brick repealers’ enacted by some states. On this point a bit of history is in order. In 1968, the Supreme Court held in Hanover Shoe, Inc v United Machinery Corp20 that a defendant in a treble-damages antitrust case could not assert a ‘passing-on’ defense, which involved arguing that the plaintiff had passed on any illegal overcharge associated with its lease of shoe machinery from the defendant to consumers in the form of higher shoe prices, and therefore suffered no injury cognizable under the antitrust laws.21 In so holding, the Court expressed its concern that the passing-on defense, if carried to its logical conclusion, would shift the burden of private enforcement to the ultimate consumers (in this case, buyers of single pairs of shoes), who ‘would have only a tiny stake in a lawsuit and little interest in attempting a class action’.22 ‘In consequence’, the Court reasoned, ‘those who violate the antitrust laws by price fixing or monopolizing would retain the fruits of their illegality because no one was available who would bring suit against them’.23 The effectiveness of private treble-damages actions as a mechanism for enforcing the antitrust laws would thereby be substantially reduced.24 15 15 USC § 15(a) (2009) (providing that ‘any person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws … shall recover threefold the damages by him sustained, and the cost of suit, including a reasonable attorney’s fee’). 16 See, eg, Hawaii v Standard Oil Co, 405 US 251, 262, 266 (1972) (‘By offering potential litigants the prospect of a recovery in three times the amount of their damages, Congress encouraged these persons to serve as ‘private attorneys general’’.) (‘The fact that a successful antitrust suit for damages recovers not only the costs of the litigation, but also attorney’s fees, should provide no scarcity of members of the Bar to aid prospective plaintiffs in bringing these suits’.); Perma Life Mufflers, Inc v Int’l Parts Corp, 392 US 134, 139 (1968) (recognizing ‘that the purposes of the antitrust laws are best served by insuring that the private action will be an ever-present threat to deter anyone contemplating business behavior in violation of the antitrust laws’); Byram Concretanks, Inc v Warren Concrete Prods. Co, 374 F.2d 649, 651 (3d Cir. 1967) (‘It is well known that a primary objective of the private treble-damage suit is to provide a means for enforcement of the anti-trust laws in addition to Government prosecutions’.). 17 See Stephen Calkins, ‘An Enforcement Official’s Reflections on Antitrust Class Actions’, 39 Arizona Law Review 413, 416–417 (1997) (reporting some statistics on the number of federal court antitrust class actions filed from 1972 to 1995). 18 Ibid (showing a marked uptick in the number of antitrust class action filings in 1994 and 1995, not only in absolute numbers but also as a percentage of all federal class actions filed). 19 Ibid at 419. 20 392 US 481 (1968). 21 Ibid at 487–488. 22 Ibid at 494. 23 Ibid. 24 Ibid.

Designing a Private Remedies System for Antitrust Cases – the US Experience  183   Although the Supreme Court has subscribed to the view that private trebledamages actions are an important means of enforcing the antitrust laws, it has also expressed concern with duplicative recoveries. Thus, in Hawaii v Standard Oil Co,25 the Court held that the State of Hawaii could not bring a parens patriae action under Section 4 of the Clayton Act for injury to its economy from the defendants’ alleged conspiracy to restrain trade and to monopolize the market for refined petroleum products.26 The Court explained that any alleged harm to Hawaii’s economy is in large part a reflection of the injuries to the ‘business or property’ of Hawaiian consumers,27 who have a right to sue on their own behalf under Section 4, and indeed, may join together in a class action to do so.28 Accordingly, allowing Hawaii to sue separately in parens patriae would create a problem of duplicative recoveries,29 which class actions avoid with specific rules for identifying an appropriate plaintiff-class and determining who is bound by the outcome of the action.30 This takes us to 1977, when the Supreme Court held in Illinois Brick Co v Illinois31 that only the direct purchasers of a product or service have standing to sue for injury to their business or property flowing from allegedly illegal overcharges.32 The Court thus rejected the use of a ‘passing-on’ theory by plaintiffs as well as by defendants; indirect purchasers could not come to court claiming that they too had been injured under the federal antitrust laws because part or all of the overcharge had been passed on to them by the direct purchasers.33 Once again, the Court was concerned with the problem of multiple liability and duplicative recoveries,34 which ‘would add whole new dimensions of complexity to treble-damages suits and seriously undermine their effectiveness’.35 In response to Illinois Brick, some states have passed laws known as ‘Illinois Brick repealers’ that extend standing to indirect purchasers, including consumers.36 Such laws have had the effect, of course, of subjecting an antitrust defendant to liability from multiple lawsuits by direct purchasers and indirect purchasers 405 US 251 (1972). Ibid at 252–253. 27 Ibid at 264. 28 Ibid at 266 (‘Rule 23 of the Federal Rules of Civil Procedure provides for class actions that may enhance the efficacy of private actions by permitting citizens to combine their limited resources to achieve a more powerful litigation posture’.). 29 Ibid at 263–264. 30 Ibid at 266. It should be noted that the Congress, in 1976 and in response to Hawaii v Standard Oil, amended the Clayton Act to permit states to bring parens patriae actions on behalf of their citizens. 15 USC § 15c(a)(1) (2009) (added to the Clayton Act as part of the Hart-Scott-Rodino Antitrust Improvements Act of 1976). But the Court’s fundamental concern with duplicative recoveries did not go away. 31 431 US 720 (1977). 32 Ibid at 746. 33 Ibid at 735. 34 Ibid at 730–731. 35 Ibid at 737; see also ibid at 741, 746–747. 36 For a recent survey of the different state responses to lllinois Brick, see Sullivan v DB Investments, Inc, 613 F.3d 134, 146–148 and footnotes 10–13 (3d Cir.), rehearing granted and vacated, 2010 US App. LEXIS 18088 (3d Cir. 27 August 2010). 25 26

184  J Thomas Rosch and duplicative recoveries—just as the Supreme Court had feared. An antitrust defendant’s exposure thus has sometimes far exceeded treble damages, which can put tremendous pressure on the defendant to settle a case regardless of its merit, and can lead to extortionate settlements. A second flaw is the judicial rule that antitrust defendants who do not settle are subject to joint and several liability with no right to contribution from settling defendants.37 As the Supreme Court explained, the fact that antitrust defendants are jointly and severally liable ‘simply ensures that the plaintiffs will be able to recover the full amount of damages from some, if not all, participants’; it does not imply that the courts have the power to order contribution when the Congress has not provided for such a remedy in the antitrust statutes.38 While the Court may be right in its interpretation of the Sherman and Clayton Acts, without a right to contribution, a non-settling defendant could end up paying the full amount of a price-fixing judgment (less the pre-trebled amounts of prior settlements). Such a result has also led to the prevalent view that class action outcomes, both litigated and settled, are excessive and unjustified. A third flaw is an uneven ‘discovery’ playing field for antitrust plaintiffs and defendants. Even in the beginning, a treble-damages plaintiff generally had an advantage over, say, a defendant firm like General Motors, because of the cost and burden that the latter bore in searching for, segregating and producing relevant documents.39 A defendant also generally bore the larger share of the cost and burden in producing witnesses, because a large company would typically have multiple, knowledgeable managers40 whereas a plaintiff, being an individual, would have relatively few, if any, knowledgeable witnesses. Moreover, the cost and burden of document discovery increased exponentially for treble-damages defendants as compared to plaintiffs when ‘electronic 37 Texas Indus., Inc v Radcliff Materials, Inc, 451 US 630, 640, 646 (1981) (‘We therefore conclude that Congress neither expressly nor implicitly intended to create a right to contribution’.) (‘We are satisfied that neither the Sherman Act nor the Clayton Act confers on federal courts the broad power to formulate the right to contribution sought here’.). 38 Ibid at 646. 39 Judge Richard Posner of the United States Court of Appeals for the Seventh Circuit refers to this problem as one of ‘asymmetric discovery burdens’. See Thorogood v Sears, Roebuck & Co, 624 F.3d 842, 849–850 (7th Cir.) (‘An additional asymmetry, also adverse to defendants, involves the cost of pretrial discovery in class actions […]. In most class action suits, including this one, there is far more evidence that plaintiffs may be able to discover in defendants’ records (including emails, the vast and ever-expanding volume of which has made the cost of discovery soar) than vice versa’.), rehearing denied, 627 F.3d 289 (7th Cir. 2010), petition for cert. filed, No. 10-1087 (US 2 March 2011); Swanson v Citibank, N.A., 614 F.3d 400, 411 (7th Cir. 2010) (Posner, J., dissenting) (‘Behind both Twombly and Iqbal lurks a concern with asymmetric discovery burdens and the potential for extortionate litigation […] that such an asymmetry creates […]. In most suits against corporations or other institutions, and in both Twombly and Iqbal—but also in the present case—the plaintiff wants or needs more discovery of the defendant than the defendant wants or needs of the plaintiff, because the plaintiff has to search the defendant’s records (and, through depositions, the minds of the defendant’s employees) to obtain evidence of wrongdoing’.) (citations omitted). I will have more to say about Twombly and Iqbal in Part III below. 40 Rule 30(b)(6) of the Federal Rules of Civil Procedure provides that a corporate defendant is to be deposed through one or more officers, directors, managing agents, or other persons designated to testify about specified matters based on information known or reasonably available to the corporation.

Designing a Private Remedies System for Antitrust Cases – the US Experience  185   discovery’ became de rigueur in the federal courts. That meant that a large defendant had to search and segregate all of its on-line, as well as its off-line documents, including e-mails.41 The disproportionality in burden and expense between plaintiffs and defendants was exacerbated when the courts generally held that discovery directed at individual, unnamed members of a class was truncated or prohibited.42 A fourth flaw has been the emergence of numerous ‘opt-outs’ in class action litigation. Rules 23(c)(2) and (c)(3) of the Federal Rules of Civil Procedure require that unnamed class members be given notice and an opportunity to opt out of a Rule 23(b)(3) class action for monetary relief before any judgment entered in such an action, favorable or unfavorable, can bind them as class members.43 The reasons for opt-outs are many and varied. However, as far as corporate plaintiffs are concerned, a major factor is that large companies that would otherwise be class members sometimes decide they could do better if they ‘opted out’ of the class and let their own attorneys represent them. Such a decision would lead to the possibility that a defendant in a class action might secure a defense verdict after a lengthy trial, only to be confronted by the prospect of multiple subsequent trials by plaintiffs who had ‘opted out’ before the trial or judgment. Here is the conundrum posed by opt-outs. This mechanism was preferred over an affirmative, ‘opt-in’ requirement (which is used in some statutory, collective actions like claims under the Fair Labor Standards Act of 1938, for example)44 for class actions because of the concern that ‘[r]equiring a plaintiff to affirmatively request inclusion would probably impede the prosecution of those class actions involving an aggregation of small individual claims, where a large number of 41 See, eg, Swanson, 614 F.3d at 411 (Posner, J., dissenting) (‘With the electronic archives of large corporations or other large organizations holding millions of emails and other electronic communications, the cost of discovery to a defendant has become in many cases astronomical. And the cost is not only monetary; it can include, as well, the disruption of the defendant’s operations. If no similar costs are borne by the plaintiff in complying with the defendant’s discovery demands, the costs to the defendant may induce it to agree early in the litigation to a settlement favorable to the plaintiff’.). 42 See, eg, Cox v Am. Cast Iron Pipe Co, 784 F.2d 1546, 1556–1557 (11th Cir. 1986) (disapproving the service of interrogatories on unnamed class members as a strategy to reduce class size and regarding the imposition of discovery sanctions for failure to respond as an impermissible, affirmative ‘opt-in’ device); Dellums v Powell, 566 F.2d 167, 187 (D.C. Cir. 1977) (acknowledging that discovery against absentee class members is not available as a matter of course, although exceptions can be made ‘at least when the information requested is relevant to the decision of common questions, when the interrogatories or document requests are tendered in good faith and are not unduly burdensome, and when the information is not available from the representative parties’); Blackie v Kushner, 524 F.2d 891, 906 footnote 22 (9th Cir. 1975) (‘A defendant does not have unlimited rights to discovery against unnamed class members; the suit remains a representative one’.); Clark v Universal Builders, Inc, 501 F.2d 324, 341 (7th Cir. 1974) (‘The taking of depositions of absent class members is—as is true of written interrogatories—appropriate in special circumstances. And, not unlike the use of interrogatories, the party seeking the depositions has the burden of showing necessity and absence of any motive to take undue advantage of the class members’.); 43 Fed R Civ P 23(c)(2) & (c)(3). 44 See, eg, Ervin v OS Rest. Servs., Inc, 632 F.3d 971, 976 (7th Cir. 2011) (concluding that ‘[i]t does not seem like too much [in a combined action involving federal FLSA claims and ancillary state law class claims] to require potential participants to make two binary choices: (1) decide whether to opt in and participate in the federal action; (2) decide whether to opt out and not participate in the state-law claims’).

186  J Thomas Rosch claims are required to make it economical to bring suit’.45 Specifically, the concern, as recounted by the Supreme Court in Phillips Petroleum Co v Shutts, was that ‘[t]he plaintiff’s claim may be so small, or the plaintiff so unfamiliar with the law, that he would not file suit individually, nor would he affirmatively request inclusion in the class if such a request were required by the Constitution’.46 Based on this reasoning, the Court in Phillips Petroleum declined to impose an affirmative, opt-in requirement as one of the protections guaranteed by the Due Process Clause on out-of-state plaintiff class members who had no minimum contacts with the forum State but are nonetheless being subjected to the personal jurisdiction of the courts in that State.47 By contrast, plaintiffs who have sufficiently large or important claims, and have the wherewithal to hire their own attorneys, do not need the protections afforded by the opt-out mechanism. They are perfectly capable of filing suit on their own, and as the Court noted in Phillips Petroleum, they can therefore opt out of the class action if they wish.48 In antitrust treble-damages class actions, direct purchasers who have standing to sue are more likely to fall into this camp; they are more likely to be sophisticated corporate plaintiffs who may well opt out of the class action and pursue relief on their own. (Contrast that with indirect purchasers who might sue under state law; these plaintiffs are more likely to be individual consumers whom the opt-out mechanism was meant to protect.) Because the opt-out mechanism does not distinguish between the needs of individual consumer plaintiffs and corporate plaintiffs, the class action vehicle fails to achieve its purpose of minimizing an antitrust defendant’s exposure to multiple lawsuits and multiple liabilities. A verdict exonerating a defendant in an antitrust treble-damages class action does not mean that there will not be other lawsuits brought by plaintiffs—especially corporate plaintiffs—who have opted out.

III I now would like to describe the Supreme Court antitrust jurisprudence that these flaws in antitrust treble-damages class actions have spawned. The first Phillips Petroleum Co v Shutts, 472 US 797, 812–813 (1985). Ibid at 813. See also ibid at 813 footnote 4 (‘[Requiring] the individuals affirmatively to request inclusion in the lawsuit would result in freezing out the claims of people—especially small claims held by small people—who for one reason or another, ignorance, timidity, unfamiliarity with business or legal matters, will simply not take the affirmative step.’) (quoting Benjamin Kaplan, ‘Continuing Work of the Civil Committee: 1966 Amendments of the Federal Rules of Civil Procedure (I)’, 81 Harvard Law Review 356, 397–398 (1967)). 47 Ibid at 813–814 (‘Petitioner’s ‘opt in’ requirement would require the invalidation of scores of state statutes and of the class-action provision of the Federal Rules of Civil Procedure, and for the reasons stated we do not think that the Constitution requires the State to sacrifice the obvious advantages in judicial efficiency resulting from the ‘opt out’ approach for the protection of the rara avis portrayed by petitioner’.). I will come back to the constitutionality of opt-in versus opt-out in Part III below. 48 Ibid at 813. 45 46

Designing a Private Remedies System for Antitrust Cases – the US Experience  187   consequence has been the fact that the Supreme Court has moved up the time that a federal court can appropriately dismiss a treble-damages action, including an antitrust class action, from the summary judgment stage to the pleadings stage. In 1986, the Supreme Court handed down a trio of important cases addressing the standard for granting summary judgment under Rule 56 of the Federal Rules of Civil Procedure.49 One of the cases, Matsushita Electric Industrial Co, Ltd. v Zenith Radio Corp, involved an alleged conspiracy ‘to raise, fix and maintain artificially high prices for television receivers sold by [petitioners] in Japan and, at the same time, to fix and maintain low prices for television receivers exported to and sold in the United States[.]’50 Essentially, the plaintiffs were claiming that the defendants had engaged in a predatory pricing conspiracy in the United States market—the only market in respect of which the US antitrust laws could apply.51 The defendants moved for summary judgment, which the district court granted but the Third Circuit reversed. The Supreme Court sided with the district court, holding that the plaintiffs had the burden, in order to defeat a summary judgment motion, to come forward with evidence raising a genuine issue of material fact concerning the existence of a predatory pricing conspiracy.52 Noting that a predatory pricing conspiracy is ‘by nature speculative’ because it requires the alleged conspirators, presumed to be rational economic actors, ‘to forgo profits that free competition would offer them’,53 the Court held that ‘if the factual context renders respondents’ claim implausible—if the claim is one that simply makes no economic sense— respondents must come forward with more persuasive evidence to support their claim than would otherwise be necessary’.54 Such evidence must tend to exclude the possibility that the defendants underpriced the plaintiffs in order to compete for business ‘rather than to implement an economically senseless conspiracy’.55 Matsushita therefore represented a relaxation of an earlier-held view ‘that summary procedures should be used sparingly in complex antitrust litigation where motive and intent play leading roles, the proof is largely in the hands of the alleged conspirators, and hostile witnesses thicken the plot’.56 Although motive and intent can and do indeed play ‘leading roles’ in many antitrust cases, summary judgment is nonetheless appropriate if the evidence fails to reveal any plausible, rational motive to engage in the alleged conduct, and 49 Celotex Corp v Catrett, 477 US 317 (1986); Anderson v Liberty Lobby, Inc, 477 US 242 (1986); Matsushita Elec. Indus. Co, Ltd. v Zenith Radio Corp, 475 US 574 (1986). 50 475 US at 578 (quoting In re Japanese Elec. Prods. Antitrust Litig., 723 F.2d 238, 251 (3d Cir. 1983)) (emphases in the original). 51 Ibid at 582 (‘We begin by emphasizing what respondents’ claim is not. Respondents cannot recover antitrust damages based solely on an alleged cartelization of the Japanese market, because American antitrust laws do not regulate the competitive conditions of other nations’ economies’.) (emphasis in the original). 52 Ibid at 586–587. 53 Ibid at 588. 54 Ibid at 587. 55 Ibid at 597–98. 56 Poller v Columbia Broadcasting Sys., Inc, 368 US 464, 473 (1962).

188  J Thomas Rosch if the defendants’ conduct is ‘consistent with other, equally plausible [and benign] explanations’.57 In response to the flaws in antitrust treble-damages class actions, the Supreme Court has subsequently relaxed the standard for summary dismissal even further. In Bell Atlantic Corp v Twombly,58 the Court was persuaded to apply the Matsushita plausibility standard to test the sufficiency of a complaint claiming that a group of incumbent local exchange carriers (ILECs) had conspired under Section 1 of the Sherman Act to charge inflated prices for local telephone and high-speed Internet services.59 Specifically, the Court held that even under a liberal, notice-pleading standard, ‘[f]actual allegations must be enough to raise a right to relief above the speculative level’,60 that is, they must ‘state a claim to relief that is plausible on its face’.61 The Court did not limit this plausibility standard to antitrust cases either, as Ashcroft v Iqbal62 later made clear. The reasons for the Court’s application of the plausibility standard to the pleading stage were laid out in Twombly and Credit Suisse Securities (USA) LLC v Billing,63 the latter case involving a pair of antitrust class actions challenging underwriting practices regulated by the federal securities laws. In Twombly, the Court’s majority rejected Justice Stevens’s suggestion in dissent that groundless claims can be weeded out early in the discovery process through ‘careful case management’; the majority expressed the concern that ‘the threat of discovery expense will push cost-conscious defendants to settle even anemic cases before reaching [summary judgment or trial]’.64 This concern, of course, relates to the uneven ‘discovery’ playing field that I have already mentioned. In Credit Suisse, the Court made the additional observation that ‘antitrust plaintiffs may bring lawsuits throughout the Nation in dozens of different courts with different nonexpert judges and different nonexpert juries’.65 In the Court’s view, there was an ‘unusually high risk that different courts will evaluate similar factual circumstances differently’ because the task of evaluating the evidence in order to distinguish permissible conduct from impermissible conduct in antitrust cases is necessarily nuanced and fact-bound.66 The threat of serious ‘antitrust mistakes’ (what we frequently call ‘false positives’) would cause antitrust defendants to avoid not only conduct that is legally forbidden but also conduct that is permitted or encouraged, out of fear of triggering an antitrust lawsuit and the risk of treble damages.67 Once again, the Court was reacting to concerns posed by the prevalence of antitrust treble-damages class actions. Matsushita, 475 US at 596 (bracketed language added). 550 US 544 (2007). Ibid at 550–551. 60 Ibid at 555. 61 Ibid at 570. 62 556 US 662 (2009). 63 551 US 264 (2007). 64 550 US at 559. 65 551 US at 281. 66 Ibid at 281–282. 67 Ibid at 282. 57 58 59

Designing a Private Remedies System for Antitrust Cases – the US Experience  189   The second consequence is that the Supreme Court jurisprudence has threatened to ‘spill over’ into and ‘infect’ public antitrust enforcement proceedings. For one thing, Matsushita and Twombly articulated procedural standards under Rule 56 and Rule 12 of the Federal Rules of Civil Procedure, respectively, which apply across the board to all civil cases, including cases brought by the federal antitrust agencies, regardless of subject matter or claim. As I noted earlier, Matsushita was handed down as part of a trio of cases dealing with the summary judgment standard—along with Celotex v Catrett and Anderson v Liberty Lobby, both nonantitrust cases. Similarly, the teachings in Twombly carried over to Iqbal, a case involving the alleged deprivation of the constitutional rights of a detainee. Furthermore, with the exception of the Federal Trade Commission’s Section 5 authority,68 public and private plaintiffs enforce the same substantive antitrust laws (the Sherman and Clayton Acts). Thus, whenever the Supreme Court has reined in substantive law out of concerns about the negative impact of private antitrust litigation, those changes have affected public enforcement proceedings as well. Credit Suisse is a good example of that; there the Court essentially made a policy choice in favor of ‘the efficient functioning of the securities markets’ and against ‘any enforcement-related need for an antitrust lawsuit’, which it viewed to be ‘unusually small’.69 Similarly, in Verizon Communications Inc v Trinko,70 the Court concluded that the existence of a regulatory structure governing telecommunications service providers means ‘that the additional benefit to competition provided by antitrust enforcement will tend to be small, and it will be less plausible that the antitrust laws contemplate such additional scrutiny’.71 Moreover, the Trinko Court was concerned, as in Credit Suisse, with the risk that antitrust enforcement (in this case, of Section 2) in the courts would produce costly false positives.72 Lastly, the Supreme Court thus far has declined to say whether there is anything in the Constitution that requires ‘opt-out plaintiffs’ to be able to re-litigate a case that the class has lost or settled. After holding in Phillips Petroleum Co v Shutts that an opt-out mechanism provides sufficient protection under the Due Process Clause—from the standpoint of in personam jurisdiction—to out-of-state class members who have no minimum contacts with the forum State, the Court had the opportunity in Ticor Title Insurance Co v Brown73 to consider ‘whether a federal 15 USC § 45 (2009). 551 US at 283. 70 540 US 398 (2004). 71 Ibid at 412. 72 Ibid at 414 (‘Against the slight benefits of antitrust intervention here, we must weigh a realistic assessment of its costs…. Mistaken inferences and the resulting false condemnations ‘are especially costly, because they chill the very conduct the antitrust laws are designed to protect’…. Judicial oversight under the Sherman Act would seem destined to distort investment and lead to a new layer of interminable litigation, atop the variety of litigation routes already available to and actively pursued by competitive LECs’.) (citations omitted). 73 511 US 117 (1994) (per curiam). As the Court’s opinion notes, this case related to 12 different, ‘tag-along’ antitrust class actions that came on the heels of the Commission’s 1985 enforcement proceedings against six title insurance companies for allegedly conspiring to fix fees for title searches 68 69

190  J Thomas Rosch court may refuse to enforce a prior federal class action judgment, properly certified under Rule 23, on grounds that absent class members have a constitutional due process right to opt out of any class action which asserts monetary claims on their behalf’.74 After taking briefs and hearing oral argument, the Court voted 6 to 3, however, to dismiss the writ of certiorari as improvidently granted.75 In dissent, Justice O’Connor (joined by Chief Justice Rehnquist, who wrote the opinion of the Court as an Associate Justice in Phillips Petroleum, and Justice Kennedy) pointed out that ‘[t]he [Ninth Circuit’s] decision below rests exclusively on a constitutional right to opt out of class actions asserting claims for monetary relief’, and not on whether the classes in MDL No. 633 had been properly certified under Rule 23(b)(1) and (b)(2) despite the presence of monetary claims.76 (As I mentioned earlier in these remarks, claims for monetary relief are typically certified under Rule 23(b)(3), which requires that proposed class members have an opt-out right.) In Justice O’Connor’s view, the Court had to assume that MDL No. 633 had been properly certified under Rule 23(b)(1) and (b)(2), and therefore should not have dodged the question of whether an opt-out right is constitutionally required by the Due Process Clause.77

IV Lastly, I’d like to describe how I think the EU can fashion a private antitrust system that will work but is not flawed in the ways I have described. As a threshold matter, I want to emphasize that most plaintiffs’ antitrust lawyers are, first and foremost, and title examinations in 13 states, including Arizona and Wisconsin. Ibid at 118. See FTC v Ticor Title Ins. Co, 504 US 621 (1992). Those class actions, which included ones filed on behalf of title insurance consumers in Arizona and Wisconsin, were consolidated for pretrial purposes under the multidistrict litigation statute, 28 USC § 1407, and transferred to the Eastern District of Pennsylvania as MDL Case No. 633. The class representatives in these actions eventually settled their claims with the defendant title insurance companies, and the district court certified the classes under Rule 23(b) (1)(A) and (b)(2) and entered final judgment, overruling objections from the States of Arizona and Wisconsin that due process required that proposed class members have an opportunity to opt out— even for classes certified under Rule 23(b)(1) and (b)(2). See In re Real Estate Title & Settlement Servs. Antitrust Litig., 1986-1 Trade Cases ¶ 67,149 (E.D. Pa. 1986), affirmed without opinion, 815 F.2d 695 (3d Cir. 1987), cert. denied, 485 US 909 (1988). 74 Ticor Title, 511 US at 120–121. After judgment had been entered in MDL No. 633, plaintiff Walter Brown filed a new class action on behalf of Arizona and Wisconsin title insurance consumers in the District of Arizona. The district court granted summary judgment to the defendant title insurance companies, holding that Brown and the class members, as parties to the prior class actions, were bound by the settlement and judgment in MDL No. 633 under the doctrine of res judicata. The Ninth Circuit reversed, holding that the application of res judicata to the money damage claims of class members who did not have an opportunity to opt out of the class in MDL No. 633 would violate due process. Brown v Ticor Title Ins. Co, 982 F.2d 386, 392 (9th Cir. 1992). 75 Ticor Title, 511 US at 118. The Court subsequently had another opportunity to consider this constitutional question but again dismissed the writ as improvidently granted. See Adams v Robertson, 520 US 83 (1997). 76 Ticor Title, 511 US at 126 (O’Connor, J., dissenting). 77 Ibid at 123–125.

Designing a Private Remedies System for Antitrust Cases – the US Experience  191   investors. That was true even in the ‘good old days’ when private treble-damages actions were generally brought as individual actions instead of as class actions. So the trick is to incentivize them to invest (and thereby ensure a healthy private enforcement regime) while at the same time avoiding the features of trebledamages class action litigation that I (and more importantly, our Supreme Court) have considered to be pernicious. Here I give you seven suggestions. The first four suggestions have to do with avoiding the problems we have experienced in the US The first suggestion is to avoid ‘opt-out’ class actions. Requiring that individuals or firms must instead ‘opt in’ to a class (or other collective action lawsuit) will minimize extortionate settlements and eliminate follow-on ‘opt-out’ actions for damages. In my view, requiring ‘opt in’ instead of ‘opt out’ does not raise the same due process concerns that the Supreme Court has avoided resolving. The constitutional question before the Court in Ticor Title was whether a proposed class member has to be given a choice—in that case, to opt out of the class—before he or she can be bound by any resulting settlement or judgment. What I am proposing here is that proposed class members should indeed be given a choice—which should satisfy any due process concerns, but that the choice would involve affirmatively opting in to the class. This is not a new concept; as I observed earlier, there are some forms of collective action in the US that use an opt-in mechanism (like claims under the Fair Labor Standards Act).78 The EU could require that potential plaintiffs opt in to a collective action established to pursue claims for relief for violations of EU law. If a potential plaintiff were to choose not to opt in, then he or she would not have a remedy for the alleged violation of any EU law (although he or she may still be able to pursue individual relief for any related violations of the national laws of a Member State). In this regard, a basic observation I would offer is that the EU’s pursuit of a collective action model of redress does not mean that the EU also has to embrace a representative model, which is what an opt-out class action embodies. Recall that the October 5, 2010 information note describes ‘collective redress’ as ‘a broad concept encompassing any mechanism that may accomplish the cessation or prevention of unlawful business practices which affect a multitude of claimants or the compensation for the harm caused by such practices’.79 The primary reason for an opt-out mechanism is that, under a representative model, somebody one usually does not know (ie, a plaintiffs’ antitrust attorney and his or her client, the class representative) has decided to bring an action on one’s behalf. The opt-out is there to ensure that if a person has no interest in the litigation, or wants to pursue the claims individually, he or she can choose not to be part of the class and to be represented by others. By contrast, if collective action requires an individual 78 See note 44 above. Congress’s stated rationale for adding an opt-in mechanism to claims under FLSA was to ‘[limit] private FLSA plaintiffs to employees who asserted claims in their own right and [to free] employers of the burden of representative actions’, which had become excessive and were often brought on behalf of plaintiffs lacking a personal interest in the outcome. Hoffman-La Roche, Inc v Sperling, 493 US 165, 173 (1989). 79 Joint Information Note, cited above note 1, at 4, ¶ 7.

192  J Thomas Rosch decision on the part of each potential plaintiff to participate, then there is no need for an opt-out mechanism. Each participating plaintiff still benefits from the efficiencies and economies associated with a collective action model of redress, but the difference is that the participation is active, not passive. A second suggestion is to ban ‘indirect purchaser’ claims whenever they might result in multiple damage recoveries for the same injury. To be sure, overcharges arising from antitrust violations may sometimes be ‘passed on’ to ‘indirect purchasers’. Moreover, it sometimes may be hard to determine whether that has happened without a trial. However, these concerns are outweighed by the prospect of multiple damage recoveries for the same antitrust injury, and the complexity of conducting a trial or numerous mini-trials—assuming the question of indirect purchaser status does not stand as an obstacle to class certification80—to make that determination. This was the view of the Supreme Court in Illinois Brick, as I have recounted above.81 A third suggestion is to abandon the rule of joint and several liability with no right of contribution because it creates an unnecessary downside with no countervailing benefits. Like indirect purchaser claims, this rule gives rise to the risk that defendants will end up paying more than their fair share of the harm caused by their alleged anticompetitive conduct. At the same time, this rule does not provide a necessary incentive to plaintiffs’ antitrust attorneys to invest in private enforcement litigation; even without the rule, they should still be able to recoup their investment from a combination of settlements with some defendants and judgments against others. A fourth suggestion is to permit discovery only upon a showing of good cause by the party seeking the discovery but provide, in the law or legislative history, that good cause will be conclusively presumed to exist if that party cannot obtain the discovery sought from public law enforcement authorities (which may be prevented by law from sharing the fruits of their investigations).82 This approach yields two benefits. First, it would minimize the asymmetric discovery burden that currently prevails in the US, which the Supreme Court has sought to alleviate in Twombly 80 See, eg, Sullivan v DB Investments, Inc, 613 F.3d 134, 153 footnote 17 (3d Cir.) (‘We briefly note that the indirect purchasers face factual obstacles to class certification notwithstanding the legal defects discussed above because competition in the market for rough gem diamonds waxed and waned during the class period […]. Thus, the class, as currently defined, includes members that acquired diamonds in a market controlled by the CSO, members that purchased diamonds from competitors that were not participating in the CSO’s price-fixing activities, and members that have no antitrust injury whatsoever’.), rehearing granted and vacated, 2010 US App. LEXIS 18088 (3d Cir. 27 August 2010). 81 See notes 31–35 above, and accompanying text. 82 For example, in the US we have a statutory provision that entitles a state attorney general, who may decide to bring a follow-on civil action in parens patriae, to request that the Attorney General of the United States ‘make available to him, to the extent permitted by law, any investigative files or other materials which are or may be relevant or material to the actual or potential cause of action under [the Clayton] Act’. 15 USC § 15f(b) (2009). The phrase ‘to the extent permitted by law’ operates to exclude certain materials, such as grand jury materials, from disclosure except upon a showing of particularized need. In re Ill. Petition to Inspect & Copy Grand Jury Materials, 659 F.2d 800, 804 (7th Cir. 1981) (interpreting Fed. R. Crim. P 6(e)(3)(C)(i)).

Designing a Private Remedies System for Antitrust Cases – the US Experience  193   by allowing implausible claims for relief to be dismissed at the pleading stage.83 In accordance with my suggestion, plaintiffs would not have a right to virtually unlimited discovery. Instead, they would have to make some threshold showing as to the discovery they need and why they need it.84 Second, my approach would encourage plaintiffs’ attorneys to hitch private antitrust claims as much as possible to the results of government antitrust investigations. Their first source of discovery should be from the relevant public enforcement authorities, but if they cannot obtain the needed discovery through this channel, then they would have a good argument for having a national court in a Member State order that the discovery be provided by the defendants. And the defendants should have less reason to complain from the standpoint of a disclosure burden because presumably they have already turned over most of the same information to the enforcement authorities. My remaining three suggestions relate to how the EU can incentivize private enforcement so that it best complements and supports public enforcement efforts. First, I would limit private actions to those that follow public enforcement actions in the EU, including those public enforcement actions that are based on amnesty or leniency applications and grants. This would help ensure that someone else besides plaintiffs’ counsel has done the investigative and trial work that counsel must ordinarily do in the US. Having someone else do the work (including gathering the relevant evidence—see my previous suggestion about discovery) should be a powerful incentive for plaintiffs’ antitrust attorneys to invest in private collective action litigation. Second, I would recommend that the EU specifically provide that prejudgment interest dating back to the beginning of the conduct challenged be awarded to prevailing plaintiffs. In the US, the Clayton Act permits courts to award prejudgment interest at their discretion,85 but this is honored more in the breach than in the observance. By making an award of prejudgment interest mandatory, the EU (and its Member States) could help ensure that most damage awards equal or exceed treble-damages awards in the United States, which would promote convergence on the private enforcement front. Judging by the incentives that 83 In the US, some district courts have used the pendency of motions to dismiss brought under Twombly as a basis for staying discovery, so as to minimize the expense and burden to the defendants. See, eg, Dowdy & Dowdy Partnership v Arbitron Inc, No. 2:09-cv-253 KS-MTP, 2010 US Dist. LEXIS 108798, at *4 (S.D. Miss. 30 September 2010). 84 In the US, some district courts have more aggressively applied the ‘proportionality principle’ under Rule 26(b)(2)(C)(iii) of the Federal Rules of Civil Procedure (which takes into account whether ‘the burden or expense of the proposed discovery outweighs its likely benefit, considering the needs of the case, the amount in controversy, the parties’ resources, the importance of the issues at stake in the action, and the importance of the discovery in resolving the issues’) to ensure that discovery is proportionate to the specific circumstances of an antitrust case. See, eg, Tamburo v Dworkin, No. 04 C 3317, 2010 US Dist. LEXIS 121510, at *7–8 (N.D. Ill. 17 November 2010). 85 15 USC § 15(a) (2009) (‘The court may award under this section, pursuant to a motion by such person promptly made, simple interest on actual damages for the period beginning on the date of service of such person’s pleading setting forth a claim under the antitrust laws and ending on the date of judgment, or for any shorter period therein, if the court finds that the award of such interest for such period is just in the circumstances’.).

194  J Thomas Rosch existed in the United States before class actions came into vogue, this too should incentivize investment in private actions, notwithstanding the limiting measures described above. Third, I would disallow ‘loser pays’ rules. Avoiding such rules would mean that the most an investor in an antitrust lawsuit could lose would be his or her investment, and this too would help to preserve proper incentives. As you may know, in the US the prevailing ‘American rule’ provides that, win or lose, litigants bear their own litigation expenses, with the exception of only certain limited categories of costs that can be taxed against the losing party.86 The American rule applies to civil actions brought under Section 4 of the Clayton Act; the ‘cost of suit’ that may be awarded under Section 4 has been interpreted by courts as limited to taxable costs.87 * * * These are my thoughts and suggestions as a public enforcer and previously a trial lawyer for private litigants. I look forward to our discussion and debate.

See Fed R Civ P 54(d); 28 USC § 1920 (2009). See, eg, Reazin v Blue Cross & Blue Shield, Inc, 899 F.2d 951, 981–982 (10th Cir. 1990); Illinois v Sangamo Constr. Co, 657 F.2d 855, 865–866 (7th Cir. 1981). 86 87

James Keyte*

Collective Redress: Perspectives from the US Experience

For the past decade, the European Union has been seriously considering the idea of implementing a collective redress mechanism. Currently, around half the Member States have some form of collective redress, but citizens across EU Member States do not have any means of collectively bringing suit for similar claims. An EUwide collective redress system would afford citizens across Member States with the opportunity to bring suits that are similar to the United States’ class action suit. The United States model allows plaintiffs to bring concurrent class action suits under both state and federal laws to address the same alleged misconduct. Thus, it provides a perfect natural experiment for the EU since individual Member States – much like individual States in the US – often have overlapping yet distinct laws that target the same conduct as the EU laws. The US experience, however, is far from simple. For example, some critics have questioned whether the US class action system is in ‘crisis,’ especially given that class action plaintiffs’ lawyers often profit as much as (or more than) the actual plaintiffs in these suits.1 This situation potentially leads to excessive litigiousness and strains already depleted judicial resources. Looking at the US experience from the EU’s perspective, two fundamental questions arise: first, given the United States example, why does Europe want to construct a collective redress mechanism at all, and second, can the EU create a collective redress system that takes the ‘good’ part of US class actions, but not the ‘bad’? The answers are not simple. Yes, the EU appears to want a system of collective redress, but, of course, it wants to create such a system without interfering too much with Member States. And, yes, the EU appears to want the ‘good’ parts of the US class action system, but may not fully realize that the perceived bad parts – eg, contingency fees, opt-out procedures, etc. – are what provide the economic motivation to keep the system running. And, absent an economic incentive to bring private collective redress actions (however designed), there will be very few collective redress suits filed. * James A Keyte is a Partner in the Antitrust and Competition group in the New York office of Skadden, Arps, Slate, Meagher & Flom LLP. He would like to thank Marissa E Troiano, an associate of Skadden, Arps, who assisted in the preparation of this chapter. The views expressed are solely those of the author and do not necessarily reflect the views of other individuals at Skadden. 1 See Anne Bloom, ‘From Justice to Global Peace: A (Brief) Genealogy of the Class Action Crisis’, 39 Loyola of Los Angeles Law Review 719 (2006); Richard Epstein, ‘Class Actions: The Need for a Hard Second Look’, Civil Justice Report No. 4 (March 2002), http://www.manhattan-institute.org/ html/cjr_4.htm.

196  James Keyte This paper first evaluates the US system under Rule 23, weighing both the positive benefits of the class action system and the problems it has created in the US, particularly in the antitrust context. It then considers how such a system might play out in the European Union. Finally, we look at a number of variations on the US class action model being considered by the EU, and ask whether the EU can have its cake and eat it too (the answer, by the way, is probably not).

I. The US system under Rule 23 and its drawbacks The US class action model is probably one of the most widely recognized and criticized systems in the world. In the United States, plaintiffs can aggregate similar claims if they meet the procedural requirements of Federal Rule of Civil Procedure 23. Rule 23 became a more widely used procedural tool after it was amended in 1966. The Advisory Committee for the Federal Rules of Civil Procedure described the amendments as addressing two main concerns: foremost, the amendments were meant to promote judicial economy by minimizing duplicative suits, and second, the amendments were also meant to ‘provide [a] means of vindicating the rights of groups of people who individually would be without effective strength to bring their opponents into court at all.’2 Thus, the Committee wanted to increase plaintiffs’ access to justice ‘even at the expense of increasing litigation.’3 While the European Union shares some of these goals, it expressly has noted that ‘[a]ny European approach [must] avoid from the outset the risk of abusive litigation.’4 Given this focus, it is important to view the United States class action model through the lens of the litigation culture it has created. Foremost, it is important to concede that this ‘litigation culture’ was certainly not a goal of the class action mechanism, even if the Committee was aware that such a culture could emerge. Rather, the class action mechanism was meant to address a serious problem of equity and justice: when each plaintiff’s claims are worth a relatively small amount of money, the costs of bringing suit individually will deter plaintiffs from bringing potentially meritorious claims. Moreover, even if each of these plaintiffs were to bring their claims individually, most of these cases would present similar evidence and require decisions on the same questions of law and fact. Accordingly, if each case was brought individually, these cases would clog the judicial docket and waste judicial resources. Therefore, the class action was ‘mothered by the practical necessity of providing a procedural device so that mere numbers would not disable large groups of individuals, united in 2 See Bloom, cited previous footnote, quoting Benjamin Kaplan, ‘A Prefatory Note’, 10 Boston College Industrial and Commercial Law Review 497, 497 (1968–1969). 3 Ibid. 4 Commission, Staff Working Document, Public Consultation: Toward a Coherent European Approach to Collective Redress, SEC (2011) 173 final (4 February 2011) (‘2011 Consultation Paper’).

Collective Redress: Perspectives from the US Experience 

197  

interest, from enforcing their equitable rights nor grant them immunity from their equitable wrongs.’5 However, class actions can create serious issues for both defendants and plaintiffs. The class action suit makes it easier to bring frivolous suits on behalf of an unmanageably large group of plaintiffs against defendants with deep pockets. Fearing time-consuming and costly litigation and possibly enormous damages, defendants could be pressured into settling baseless claims and spending large sums on litigation defense. From the plaintiff’s perspective, there may be thousands of absent class members who are bound by a class action judgment, despite having little to no involvement in the class action legal proceedings. Presumably to address the aforementioned concerns, Rule 23 imposes certain ‘threshold’ conditions which must be met to bring claims as a class: (1) the class must be so numerous that joinder of all the members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties must be typical of the claims or defenses of the class; and (4) the representative parties must fairly and adequately protect the interests of the class.6 Rule 23 also imposes a series of notice requirements to make sure that absent class members are aware of their legal rights and are afforded the opportunity to opt-out of any legal decisions that will bind them in the future.7 By now, class action suits are ubiquitous – and we are left to question whether the 1966 amendments got it right. Plaintiffs have undoubtedly received greater ‘access’ to collective justice, but the costs to the judicial system and defendants have been high. While many unmeritorious class actions are stopped at the class certification stage of litigation, other suits undoubtedly ‘sneak’ by – and defendants are still forced to settle these frivolous claims. Thus, it is doubtful that these procedural thresholds are sufficient to protect against the perils of overlitigiousness through the class mechanism. Ultimately, the United States example evidences that the class action system both enables and incentives plaintiffs’ lawyers to exploit this procedural device to bring weak, unmeritorious and overbroad suits on behalf of a class if they can find one willing representative of the class (and this task is almost always an easy one). Because of the high costs of litigation at the motion to dismiss, summary judgment, and class certification stage, defendants may settle these claims long before the procedural issues related to the class even come to the forefront. Some of the major problems under the US system include: 1. there are weak limits on the standing of plaintiffs to bring class-action lawsuits; 2. the opt-out mechanism subjects defendants to massive potential damages exposure by the mere filing of a putative class-action lawsuit; 5 Montgomery Ward & Co v Langer, 168 F.2d 182, 187 (8th Cir. 1948), citing United Mine Workers of America v Coronado Coal Co, 259 US 344, 387–389 (1922). 6 Federal Rule of Civil Procedure (FRCP) 23 (a)(1–4). 7 FRCP 23 (c)(2). See also Edward F Sherman, ‘American Class Actions: Significant Features and Developing Alternatives in Foreign Legal Systems’, 215 FRD 130, 133–136 (2003).

198  James Keyte 3. treble damages in the antitrust context result in overcompensation (similar to punitive damages); 4. joint and several liability with no right of contribution increases the stakes for defendants exponentially; 5. contingency fees and the lack of a ‘loser-pays’ rule makes it financially rewarding to bring class-action antitrust lawsuits without the concomitant risk of bringing unmeritorious, weak, or overbroad actions; 6. wide-ranging discovery raises defendants’ costs, diverts management time and attention, and risks ‘fishing expeditions’; and 7. jury trials often make the risk of proceeding to a verdict too high, especially in light of the automatic trebling of any damages award in the antitrust context. These problems are not theoretical. In fact, they are nearly unavoidable byproducts of a system that encourages plaintiffs’ lawyers to specialize in complex class action litigation.8 For example, class action lawyers understand the potential benefits of an opt-out system. By finding one or a few willing representative class plaintiffs, lawyers can use the class action as a vehicle to leverage large settlements from defendants who do not want to (or who cannot afford to) risk going to trial. These lawyers understand that the case will stay large because most unnamed class members will choose not to (or will forget to) opt out of the case. Indeed, many cases are now routinely settled before a class has even been certified, and parties agree to a settlement class without even considering the procedural thresholds of numerosity, commonality, typicality, and adequacy imposed by Rule 23. Moreover, many of these defendants do not want to face the cost, risk or bad publicity of being involved in a protracted class action suit. This is particularly true in the antitrust context, where plaintiffs often allege an industry-wide conspiracy, while only alleging misconduct by a few of the larger industry players. Smaller competitors in the same industry may be lumped into a class action suit, despite the fact that they are not implicated in the complaint, they do not have the deep pockets of their larger competitors, and/or they cannot afford the reputation costs of being associated with this type of suit. The costs of this bad publicity cannot be measured in any real way, but this by-product of the American class-action system nonetheless remains a real motivator at the backdrop of settlements of class-action litigation. Yet another issue created by the class action suit is the disparity in legal rights (both procedural and substantive) across US states. For example, in the antitrust context, the Sherman Act governs federal claims brought against potential antitrust violators. Under Illinois Brick Co v Illinois,9 the Supreme Court held that indirect purchasers are not allowed to bring suit against antitrust violators, in part because 8 See, eg, John C Coffee, Jr, ‘The Regulation of Entrepreneurial Litigation: Balancing Fairness and Efficiency in the Large Class Action’, 54 University of Chicago Law Review 877, 882–83 (1987). 9 Illinois Brick Co v Illinois, 431 US 720 (1977).

Collective Redress: Perspectives from the US Experience 

199  

of the difficulty in proving pass-on costs and damages. Other principles guide who has standing to bring claims, and what must be pleaded to show antitrust injury. At the backdrop of the federal antitrust scheme are parallel state laws for antitrust and unfair competition. Many states have adopted so-called ‘Illinois Brick repealer’ laws to grant standing to indirect purchasers.10 Additionally, some state laws offer different remedies or allow state attorneys general to bring class action lawsuits on behalf of their citizens. The conflict between state and federal law often complicates class action lawsuits that include both state and federal claims – when deciding these cases, courts often have to issue separate rulings for sub-classes pertaining to particular jurisdictions or have to parse out which claims can go forward and which should be dismissed based on state law. In these cases, the nation-wide class action arguably complicates the litigation, rather than streamlining common questions of fact and law.11 In the United States, these aforementioned problems are most acutely felt by antitrust defendants who face economic hardship if they are forced to litigate frivolous claims – especially given that most juries do not understand the complex legal and economic issues at the backdrop of most antitrust cases and defendants risk a treble damages penalty if found guilty. This hardship is exacerbated by how easy it is to bring class action suits against antitrust defendants for ‘parallel’ behavior that may be completely legal. Interestingly, rather than attacking the class action mechanism itself, in Bell Atlantic v Twombly,12 the Supreme Court targeted FRCP 8. Twombly leaves wide discretion to trial courts deciding a motion to dismiss, thereby allowing lower courts to get rid of frivolous class action suits at the very onset of the case. The Twombly court focused primarily on the high litigation costs and the enormous risk that comes even with potentially frivolous class action suits, and accordingly, held that antitrust plaintiffs should have to do more than merely recite the elements of the Sherman Act (and point to parallel conduct) to survive a motion to dismiss. But the question remains – should the US courts target other procedural rules, or is it the class action model itself that needs reform? And, is any meaningful reform possible that would target all of the aforementioned problems while still providing ‘small’ plaintiffs a significant and efficient means of redress when their individual claims represent lesser monetary damages?

10 Robert Lande, ‘New Options for State Indirect Purchaser Legislation: Protecting the Real Victims of Antitrust Violations’, 61 Alabama Law Review 447 (2010). 11 See, eg, Edward Sherman, ‘Complex Litigation: Plagued By Concerns Over Federalism, Jurisdiction and Fairness’, 37 Akron Law Review 589, 599 (2004) (noting that ‘[w]hat to do about complex litigation is not a neutral question of modernizing procedure. Because a significant portion of these cases – for instance class actions based on products liability – depend on state law, problematizing complex litigation problematizes the question of which sector of government, state or federal, has the authority to resolve the policy questions that arise.’). 12 Bell Atl. Corp v Twombly, 550 US 544 (2007).

200  James Keyte

II. How are these issues playing out in the EU? These are precisely the questions that appear to be at the forefront of the collective redress discussion in the EU. Commentators understand the risks of the US system, and have launched a public consultation to determine whether to implement a collective redress mechanism and what such a system should look like. This debate spans the past decade and takes note of many of the potential issues raised in the previous section of this paper. In initially raising the potential of an EU-wide collective redress mechanism, the European Commission issued both a Green Paper in 2005 on antitrust damages actions13 and a White Paper on the same subject in 2008.14 These Papers identified potential limitations and benefits of the collective redress system, and thus proposed four distinct options for collective redress. The Commission solicited comments on these options to determine how practitioners in both the EU and US felt about these options and to test which of these options posed the least issues. 1. No action: The first of these options was no action, which could lead to ‘distortions of competition and give consumers across the EU a different level of redress.’15 Obviously, this option was not satisfying to a large portion of the commentators – while it imports none of the problems posed by the US model, it also provides none of the benefits of judicial economy and increased justice for individual plaintiffs. Accordingly, this is not an option pushed by many of the commentators, who recognize that citizens across Member States should have some way of addressing their injuries collectively. 2. Just ‘cooperation’ among Member States: The second option calls for cooperation between Member States. However, the Commission noted that some problems may arise from this option because Member States with collective redress mechanisms do not have the resources to deal with actions on behalf of consumers from Member States that do not have collective redress mechanisms. Considering the US perspective, this option could raise many issues. US states have very different individual state laws. Procedural rules based on standing, indirect purchasers, damages, and other crucial issues also differ based on location. Because the federal law provides one standard and therefore harmonizes these issues, plaintiffs have less incentive to forum shop based on where they expect to get the highest damages. (Of course, plaintiffs lawyers will – and should – always bring suit strategically based on where their claims are strongest, but judicial systems should provide equal justice to both plaintiffs and defendants regardless 13 Commission, Green Paper on Economic Damages Actions for Breach of the EC Antitrust Rules, COM (2005) 672 final (19 December 2005). 14 Commission, White Paper on Damages Actions for Breach of the EC Antitrust Rules, COM (2008) 165 final (2 April 2008). 15 2005 Green Paper.

Collective Redress: Perspectives from the US Experience 

201  

of where a suit is commenced). If the EU only provides redress at the Member State level, plaintiffs may engage in forum shopping and this could in turn push Member States to shape their laws around attempting to minimize or maximize litigation in their courts. 3. Expanding Existing Means: A third proposed option was a mix of policy instruments, which could include improving alternative dispute resolution mechanisms, extending the scope of national small claims procedures to mass claims, extending the scope of the Consumer Protection Cooperation Regulation, encouraging businesses to improve their complaint handling schemes and/ or taking actions to raise consumers’ awareness of existing means of redress. Again, this mechanism avoids some problems of an EU-wide collective redress mechanism, but it still raises many of the same issues as option two. Another potential limitation of this option is that it depends on private parties to proactively target issues related to collective redress. While many consumer protection agencies are quite active, it is unclear whether such groups would have the economic resources or incentives to act on behalf of plaintiffs. Moreover, the EU may be able to improve conditions by encouraging businesses to improve their own internal systems of dealing with complaints, but it is unclear how the EU would police those businesses that fail to implement such measures. 4. Private, Collective Redress Across the EU: Finally, the Commission proposed judicial collective redress. The Commission identified several issues at the onset, including: financing of the procedure; how to prevent unmeritorious claims; standing in court; the question of an opt-in or opt-out procedure; the distribution of compensation; and how to avoid elements which are said to encourage a litigation culture such as is said to exist in some non-European countries, such as punitive damages, contingency fees and other elements. Since this option most closely mirrors the US system, it also imports the most problems present in the US system. At the backdrop of these four options is the 2008 Evaluation Study prepared by Civic Consulting and Oxford Economics. The study considers the efficiency and effectiveness of the national-level collective redress mechanism in 13 member states, emphasizing that these mechanisms differ vastly by Member State. As a whole, the Study concluded that the various means of collective redress led to a decrease in costs for both plaintiffs and defendants, and concluded that in general, the collective redress mechanisms work effectively. However, the Survey also concluded that there is no evidence that the creation of a uniform EUwide approach to collective redress would be an improvement over the current individual systems. These findings push strongly in favor of option two mentioned above, but do not address the cross-border situation that precipitated much of the current debate on cross-border claims. Indeed, the 2008 Evaluation Study predicted an upswing in cross-border cases, including travel cases and other similar situations where plaintiffs are not citizens of the Member States in which their injury occurred.

202  James Keyte In 2011, the Commission issued a Consultation Paper that attempts to consolidate the comments received in response to the 2008 White Paper. The 2011 Consultation Paper suggests that some of the downsides of collective redress (like the costs, unmeritorious lawsuits, and or choice of law issues) can be mitigated.

III. Can the EU have its cake and eat it too? Experience in the US suggests that the position taken in the 2011 Consultation Paper may be an overly optimistic viewpoint – even after Twombly, many unmeritorious lawsuits make it to settlement and/or trial phase, and plaintiffs are not deterred from bringing overboard or baseless suits. Moreover, even with one national law that governs a particular claim, different courts apply the same rules in slightly different fashions, thus still opening the door for forum shopping. More importantly, if the EU truly wants to promote private, collective redress – either though Member State cooperation or at an EU level – it is going to have to make it financially attractive for cases to be brought. To be sure, as more and more plaintiffs (and plaintiffs’ lawyers) realize that they can bring suits across Member States, they are more likely to increase the number of suits filed. Indeed, this is inherent in a system that tries to promote collective redress as the best method for approaching complex cases – to save judicial resources, the system must be successful enough that plaintiffs reduce the number of duplicative individual suits that are brought. But only if the EU settles on a system that in some manner sufficiently rewards those who identify, prepare and litigate these cases will the EU’s policy objectives be met. And therein lies the dilemma: can the EU meet its lofty goals of collective redress without providing the incentives that have led to perceived abuses of the US system? The United States experience is not helpful on this score. There has been little in the way of effort or results in removing some of the abuses that appear to be motivated by the financing of class action, contingency fees, opt outs and the like. The real question is how much of the bad the EU will have to take in order to reap the good that collective redress can bring. The answer to this question will depend in part upon whether the EU focuses on deterrence or compensation as the goal of collective redress. The 2011 Consultation Paper suggests that both of these goals exist, but the conflict between these goals may make it more difficult to minimize costs and/or reduce meritless claims. If deterrence becomes the primary goal of the collective redress system, the rules will likely make it easier to bring cases and will make penalties steeper. Such rules would inevitably incentivize the plaintiffs’ bar to bring more cases and push for early settlement of baseless claims, and defendants in the EU would therefore face the same problems that class action defendants in the US face. If, however, the rules focus more on compensation, some of the incentives to bring baseless claims in hopes of a big payoff will be quelled.

Collective Redress: Perspectives from the US Experience 

203  

Some proposed answers to these problems include tinkering with the opt-out/ opt-in system or having the ‘loser’ in a case pay for the costs of litigation. Neither of these options offers a real solution. For instance, France considers an opt-out procedure (like the one adopted in the United States) to be contrary to public policy because such cases may affect the rights of parties who are not involved in the litigation and who may never even be aware of the litigation. However, an opt-in procedure creates other issues – for instance, that the class may still be too small to lower costs and avoid the unnecessary expenditure of judicial resources, or that damages may be more difficult to apportion.16 Similarly, a pay-if-you-lose system will shift economic incentives – if plaintiffs may have to pay for the costs of litigation if they lose, they will be strongly incentivized not to litigate even potentially meritorious claims, and the EU will end up with a collective redress system that no one uses. Finally, rule-makers in the EU should be cognizant that antitrust cases present a particular problem. Because enforcement agencies already impose fines, many consumer suits would be ‘follow-on’ suits that impose fines on companies for a second time. Because fines are often as much or more than compensatory damages, collective redress actions would be unlikely to improve deterrence efforts, unless it reverted to the US treble damages scheme. Accordingly, courts would have to reconsider other procedural rules (as the US Supreme Court did in Twombly) to address the potential problem of the ‘blackmail’ suit for large settlements. While the EU may not be able to have its cake and eat it too, it can shape rules to minimize the perceived abuses of the US system yet still provide enough incentives to have these actions brought. It will be a delicate balance. Rule-makers should look to the US deterrence model and how it has led to litigiousness when deciding their collective redress goals. Perhaps it may be as simple as reducing damages to single or double; or in some manner capping contingency fees; or precluding certain financing arrangements that support a plaintiffs’ bar. The EU may be able to adopt these changes faster than the US has adopted them, but they will need to retain enough of an economic incentive for plaintiffs to bring claims. If they do not, at some point, the incentives (unattractive as they may be) could become so small that all the EU is left with are lofty goals and good intentions.

16 Andrew Jeffries, ‘EU Proposals on Collective Redress – Lost in Consultation?’, CPI Antitrust Chronicle (April 2011).

Brian A Facey and David Rosner*

Collective Redress for Cartel Damages in Canada

I. Introduction A civil right to damages arising from anticompetitive conduct was first introduced in Canada in 1976.1 However, for several decades, civil enforcement of antitrust claims was rare.2 The reasons for the dearth of civil proceedings included uncertainty regarding the constitutionality of the civil right of action (which the Supreme Court resolved in 1989),3 the unavailability of treble damages, and the rarity of jury trials in Canada. Furthermore, until 1992, Quebec was the only province that permitted class proceedings, and most Canadian jurisdictions did not allow contingency fee arrangements. Now, every province and territory in Canada has either class action legislation or representative action rules that allow the commencement of class proceedings.4 Contingency fee arrangements are also now permitted.

* Brian Facey is Co-Chair of the Competition, Antitrust and Foreign Investment Group at Blake, Cassels & Graydon LLP, Toronto, Canada. David Rosner is an associate at Blakes in its Competition, Antitrust and Foreign Investment Group. This chapter draws on a 2006 paper by Cal Goldman (CoChair of Blakes Competition, Antitrust and Foreign Investment Group and former head of the Canadian Competition Bureau). The authors thank Charles Dobson, an associate in the Blakes Litigation Group, for his assistance in preparing the final version of the chapter. The views expressed herein do not necessarily reflect those of Blakes or any of its clients. 1 Under the predecessor legislation of the Competition Act, ie, the Combines Investigation Act, RSC 1970, c. C-23. 2 Calvin S Goldman, Robert E Kwinter, Jeff Galway et al, ‘Private Access to Antitrust Remedies: The Canadian Experience’ presented to the American Bar Association, Section of Antitrust Law, 2003 Spring Meeting (2–4 April 2003) at 5. 3 General Motors of Canada Ltd v City National Leasing, [1989] 1 SCR 641 [General Motors Canada]. 4 The following provinces have enacted comprehensive class proceeding legislation: Quebec (1978), Ontario (1992), British Columbia (1996), Newfoundland and Labrador (2001), Saskatchewan (2001), Manitoba (2002), Alberta (2003), New Brunswick (2006), and Nova Scotia (2007). The province of Prince Edward Island does not have comprehensive class proceeding legislation. Nevertheless, in Western Canadian Shopping Centres Inc v Dutton, cited below note 66, the Supreme Court of Canada held that where a legislative framework governing class actions does not exist but a province has a representative action rule (as they all do), courts can adopt class action type procedures on an ad hoc basis. The Supreme Court noted that the certification test in such a province would be analogous to the current test in the provinces of Ontario and British Columbia. The Federal Court of Canada also has comprehensive class action rules. As discussed below, it has limited subject matter jurisdiction, but such jurisdiction does include claims made under the federal Competition Act.

206  Brian A Facey and David Rosner A plaintiff class action bar has emerged, and it is increasingly advancing claims based, in whole or in part, on breaches of the Competition Act (the ‘Act’),5 particularly cartel activity. Indeed, over the last few years, there has been a noticeable increase in the number of antitrust class actions filed in Canada, both ‘copycat’ litigation based on actions initiated in the US and follow-on litigation arising from the Competition Bureau’s enforcement of the Act.6 There is also growing coordination between plaintiffs’ counsel in the US and Canada. A number of factors suggest that, going forward, there will be an increasing number of civil actions under the Act. One of the principal reasons for this is the high priority that the Canadian Parliament has placed on cartel enforcement. Recent amendments to the Act create a ‘US style’ per se criminal cartel offence, with penalties as high as $25 million per count, and 14 years in prison for individuals (the same punishment as for treason).7 The interplay between public and private enforcement has been identified by the Canadian Commissioner of Competition (the head of the public enforcement agency, the ‘Competition Bureau’) as an area that the Bureau is actively monitoring, particularly in relation to the effectiveness of its immunity and leniency program. The Commissioner recently noted: We expect that more issues may arise as a result of class actions, given the move to a per se standard, and recent jurisprudence. We are monitoring major developments in class actions to assess what steps we may need to take to press the Crown to protect the public interest, including the integrity of our immunity and leniency programs. (emphasis added)8

A further development is a recent confirmation in Canadian jurisprudence that indirect purchasers have no cause of action to recover an ‘overcharge’ resulting from anticompetitive conduct that has been passed on to them. Instead, direct purchasers have a claim to the entirety of the overcharge, and the defendant cannot avoid that claim by arguing that the direct purchaser passed the overcharge on to its customers. The same principles apply in US federal law, under the famous case of Illinois Brick Co v Illinois.9 RSC, 1985, c. C-34 (‘Competition Act’ or ‘Act’). These cases include follow-on litigation regarding cartels involving graphite electrodes, chlorine, and vitamins. 7 Competition Act, s 45(2). Prior to these amendments, a conspiracy was only an offence if the agreement unduly restricted competition: see R v Nova Scotia Pharmaceutical Society, [1992] 2 SCR 606. 8 See Melanie Aitken, ‘Remarks’, Cambridge, Ontario, 24 February 2011, http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03377.html (date accessed: 18 August 2011). See also the remarks of Sheridan Scott, Canada’s former Commissioner of Competition (2004–2009): ‘[T]he announcement [on January 9, 2006] ... of a major cartel decision involving the paper industry made headlines across the country. Cascades Fine Papers Group Inc, Domtar Inc and Unisource Canada, Inc each pled guilty in the Superior Court of Justice in Toronto to two counts of conspiring to lessen competition unduly contrary to section 45 of the Competition Act. Each company was sentenced to record fines of $12.5 million for their part in the domestic conspiracy of carbonless sheets. A prohibition order was issued against the companies ... This decision was important in another way – it demonstrates that we [the Competition Bureau] are going to take domestic cartels very seriously’: Winter Meeting of the American Bar Association (23 January 2006), http://www.competitionbureau. gc.ca/eic/site/cb-bc.nsf/eng/02021.html (date accessed: 17 August 2011) (our emphasis). 9 Illinois Brick Co v Illinois, 431 US 720 (1977). 5 6

Collective Redress for Cartel Damages in Canada  207   At one point, it was unclear whether the Illinois Brick rule would be adopted in Canada. In 2003, the Court of Appeal for Ontario (Canada’s most populous province) released its decision in Chadha v Bayer,10 which, coincidentally, also involved bricks. Chadha related to an alleged cartel involving iron oxide pigment, used to colour bricks. A motions judge had certified a class of homeowners (indirect purchasers) whom the cartel allegedly affected. The Ontario Court of Appeal overturned certification. In doing so, the Court expressed approval of much of the reasoning of the US Supreme Court in Illinois Brick, including the risk of double recovery if both direct and indirect purchaser claims were allowed, and the complexity of determining who had suffered what proportion of the price overcharge. The Court did not immediately close the door on indirect purchaser claims completely, holding that the question of whether they could obtain relief was still open in Canada.11 The Court nevertheless made it clear that certification could only be granted in an indirect purchaser case if pass-through of the price increase to the indirect purchasers could be proven on a class-wide basis. If this were not the case, liability to the class would not be a common issue, and class proceedings would not be the preferable procedure for addressing claims, as the individual trials needed to determine loss and therefore liability would be unmanageable.12 Following Chadha, certification hearings in price-fixing class actions revolved largely around the question of whether the expert evidence sufficiently demonstrated that pass-through could be determined on a class-wide basis. Two later decisions of the Supreme Court of Canada, Canfor (2004)13 and Kingstreet (2007),14 concluded that a defendant has no pass-on defence against a direct purchaser in either tort or restitution. On April 15, 2011, the British Columbia Court of Appeal released two decisions applying Canfor and Kingstreet in the antitrust class action context: Pro-Sys Consultants Ltd v Microsoft Corporation15 and Sun-Rype Products Ltd v Archer Daniels Midland Company.16 As discussed in further detail below, the B.C. Court of Appeal held that indirect purchasers do not have a cause of action for cartel damages under the Act. The Court of Appeal noted that Canadian law on this point is consistent with American federal law, as established by the Supreme Court of the United States in Illinois Brick. (2003), 63 O.R. (3d) 22 (CA) [Chadha], leave to appeal to SCC refused, [2003] SCCA No 106. Ibid at para 65. 12 Ibid at para 56. Chadha has been distinguished ‘where the class includes a discrete and readily identifiable number of direct purchasers from a small group of identified distributors’: Axiom Plastics Inc v E.I. DuPont Canada Co, 2008 CanLII 23490 (Ont. Div. Ct.) at para 52. 13 British Columbia v Canadian Forest Products Ltd, [2004] 2 SCR 74 [Canfor]. 14 Kingstreet Investments Ltd v New Brunswick (Finance), [2007] 1 SCR 3 [Kingstreet]. 15 2011 BCCA 186 [Pro-Sys]. 16 2011 BCCA 187 [Sun-Rype]. Both Pro-Sys and Sun-Rype were appealed to the Supreme Court of Canada. Following the completion of this chapter, the Supreme Court in Pro-Sys reversed the Court of Appeal, opening the door to offensive use of passing on despite the ban on its defensive use. The Sun-Rype appeal was dismissed on the facts. 10 11

208  Brian A Facey and David Rosner These developments have the potential to streamline certification of pricefixing cases. The adoption of the Illinois Brick approach in Canada will reduce the evidentiary burden on plaintiffs, as they now must only demonstrate harm to direct purchasers. Plaintiffs no longer need to prove pass-on of any overcharge through layers of direct and indirect purchasers, as was previously required by Chadha. In the remainder of this chapter, we examine the availability of private antitrust remedies in Canada. We focus on situations where multiple plaintiffs are alleged to have suffered damages arising from the same defendant(s)’ anticompetitive conduct. First, we will discuss the scope of private remedies under the Act. This will include an analysis of section 36 of the Act, and related evidentiary issues. Next, we will review the procedural considerations that commonly arise in class proceedings, including the complexities raised by parallel US and Canadian class actions. For simplicity’s sake, we principally focus on procedural rules in Ontario, although similar rules exist in most other provinces.

II. Scope of private action under the Competition Act (a) Section 36 of the Competition Act, limitation periods and damages (i) Section 36 Although Canada’s original antitrust statute dates from 1910,17 the government did not introduce a right to bring private damages actions for criminal antitrust violations until 1976. Subsection 36(1) of the Act18 provides a right of action to any person who has suffered loss or damage as a result of a violation of the criminal provisions of the Act.19 To succeed in a private action under the Act, an injured party must establish that: (i) the impugned conduct is contrary to any provision of Part VI of the Act;20 and (ii) such conduct has caused loss or damage to the plaintiff. An action can be brought at any time prior to the expiration of the statutory limitation period.21 Where criminal proceedings22 have been commenced, See Combines Investigation Act, SC 1952 sup C 314. Section 36 of the Act is reproduced at Appendix A. 19 A claim can also be raised under section 36 of the Act for a breach of an order of the Competition Tribunal or a court of competent jurisdiction. 20 Criminal offences under Part VI of the Act include: conspiracy (section 45), bidrigging (section 47), misleading advertising (section 52), and deceptive telemarketing (section 52.1). 21 See below for further discussion of the limitation period. The limitation period does not apply to the Crown’s ability to prosecute indictable criminal offences, including, for example, conspiracies under section 45 of the Act. 22 The term ‘criminal proceedings’ has been interpreted in the case law. In R v Roddy (1877) 41 UCQB 291 at 297, 1 Cart BNA 709 (Ont. CA), the court stated that: ‘what is a criminal proceeding …‘depends on the manner in which the Legislature have [sic] treated the cause of complaint, and for this purpose the scope and object of the statute, as well as the language of its particular enactments, 17 18

Collective Redress for Cartel Damages in Canada  209   prospective plaintiffs may choose to wait until the proceedings have ended, because the limitation period becomes two years from the date the criminal proceedings are finally disposed of.23 In such cases, subsection 36(2) of the Act establishes a rebuttable presumption permitting a record of the criminal proceedings to be entered as evidence of the defendant’s alleged anticompetitive conduct in the civil proceedings. It now appears that only direct purchasers can hope to advance antitrust actions successfully. As mentioned above, the B.C. Court of Appeal’s recent decisions in Sun-Rype and Pro-Sys have confirmed that indirect purchasers have no cause of action.24 In Sun-Rype, the plaintiffs alleged that the defendants had fixed the price of high-fructose corn syrup, a sweetener used in various food products. Claims were asserted on behalf of both direct and indirect purchasers. The lower court certified a single class action on behalf of both types of purchasers. However, a majority of the Court of Appeal found that the indirect purchasers had no cause of action and remitted the application for certification on behalf of direct purchasers back to the trial court for further consideration. The Supreme Court of Canada had previously held in Kingstreet25 and Canfor26 that there is no passing on defence in Canadian law, in either tort or restitution; that is, it is no defence to a claim relating to an overcharge that the plaintiff passed its loss on to its customers. The majority of the Court of Appeal concluded that it must follow that any passing on of an overcharge is not recognized at law. Accordingly, an indirect purchaser could not have a cause of action for an overcharge passed through to it. To hold otherwise would result in double recovery. If the cartel has no pass-on defence against direct purchasers (as held in Kingstreet and Canfor), those purchasers have a cause of action for the entirety of the overcharge. Also recognizing a cause of action for indirect purchasers would either result in: (a) the defendants being liable for the same overcharge twice; or (b) a reduction of the direct purchasers’ recovery by the amount of the indirect purchasers’ claim, effectively reviving the pass-on defence and denying direct purchasers part of their claim. In Pro-Sys, decided by the Court of Appeal at the same time as Sun-Rype, the plaintiffs alleged that Microsoft had engaged in various kinds of anticompetitive behaviour that purportedly allowed it to maintain a dominant position in the relevant markets and charge higher prices for its operating system and applications software. The class members were all indirect purchaser end users of the software. should be considered … as a general rule … every proceeding before a magistrate where he has power to convict, in contradistinction to the power of making an order, is a criminal proceeding, whether the magistrate be authorized, in the first instance, to direct payment of a sum of money as a penalty, or at once to adjudge the defendant to be imprisoned […]’ Subsection 36(4)(a)(ii). See above notes 15 and 16. 25 See above note 14. 26 See above note 13. 23 24

210  Brian A Facey and David Rosner The plaintiffs claimed that the alleged overcharges at the direct-purchaser level were passed through to all indirect purchasers and ultimately to end users, and they claimed redress in tort and restitution. The motions judge certified the proposed class action, but Microsoft appealed. The majority of the British Columbia Court of Appeal set aside the certification order. As in Sun-Rype, the Court held that indirect purchasers did not have a cause of action that could be maintained at law. Given that the proceeding had been brought solely on behalf of indirect purchasers, the Court of Appeal dismissed the action. Assuming that the courts of other provinces will follow the B.C. Court of Appeal’s decisions in Sun-Rype and Pro-Sys, which is likely given that they are based on binding Supreme Court precedent, the Canadian approach to indirect purchasers will now be consistent with federal law in the United States.27 The US Supreme Court has concluded that there is no pass-on defence (in Hanover Shoe, Inc v United Machinery Corp.)28 and thus no cause of action for indirect purchasers (in Illinois Brick).29

(ii) Limitation periods The limitation period for commencing a private antitrust action in Canada pursuant to subsection 36(4) of the Act is the later of two years from: ‘(i) a day on which the conduct was engaged in, or (ii) the day on which any criminal proceedings relating thereto were finally disposed of, whichever is the later ...’30 A number of caveats must be considered regarding the wording of subsection 36(4). First, the limitation period does not apply to the Crown’s ability to commence criminal proceedings. Potential defendants may face criminal liability at any time and, consequently, follow-on civil proceedings thereafter. It is thus possible for a defendant to be faced with a civil lawsuit more than two years after the infringing conduct ceased, if the Crown has initiated criminal proceedings. This situation can create economic and business uncertainty. In practice, however, private antitrust class actions are being commenced at increasingly early stages of related criminal proceedings. This is likely aimed at pre-empting competing suits and improving the position of early-filing plaintiffs’ 27 State law varies in the US ‘Following Illinois Brick, nineteen states and the District of Columbia passed legislation, so-called ‘Illinois Brick repealers,’ expressly allowing indirect purchasers to recover damages under their antitrust statutes. The thirty-one states that do not have Illinois Brick repealer statutes have, more or less, left it to their courts to decide whether their respective antitrust statutes should be interpreted consistently with Illinois Brick or not’: Jonathan T Tomlin and Dale J Giali, ‘Federalism and the Indirect Purchaser Mess’, 11 George Mason Law Review 157, 161 (2002). 28 392 US 481 (1968). 29 See above note 9. 30 Act, subsection 36(4)(a). It should be noted that any economic tort or equitable causes of action pleaded on the basis of the same facts as the Competition Act allegations may have longer limitation periods, which are determined on the basis of the law of the relevant province(s). For example, in Sun-Rype Products Ltd v Archer Daniels Midland Company, 2008 BCCA 278, leave to appeal refused, [2008] SCCA No 416, the British Columbia Court of Appeal held that a ten-year limitation applies to a claim for a remedial constructive trust in that province.

Collective Redress for Cartel Damages in Canada  211   counsel if there is a carriage motion to decide which of a number of plaintiffs’ counsel will represent the class. Second, the Federal Court has recently held that the limitation period in subsection 36(4) of the Act is not subject to the doctrine of discoverability. The applicability of the judge-made discoverability doctrine to a statutory limitation provision is a matter of statutory construction.31 Where it applies, time will not run until the plaintiff has discovered, or ought with reasonable diligence to have discovered, the material facts upon which the cause of action is based.32 In Garford Pty Ltd v Dywidag,33 the Federal Court held that discoverability does not apply to actions brought under the Competition Act. Given that section 36(4) of the Act is expressed as running from a particular date (the date of the impugned conduct or of the disposition of criminal proceedings), the Court concluded that the limitation period commences on that date, regardless of the plaintiff’s knowledge of a potential claim.34 The Court granted the defendant a motion for summary judgment dismissing claims under sections 36 and 45 of the Competition Act, on the basis that the limitation period had expired. Third, where a cause of action is based on conduct that constitutes a ‘continuing’ practice, it may be difficult to determine when the limitation period actually ends. In Garford, the Federal Court held that a continuing offence under Part VI of the Act would require ongoing acts that, in themselves, are an offence under Part VI.35 Continued lessening of competition due to acts that are no longer occurring would not be sufficient to extend the limitation period.36 Similarly, continued damages to the plaintiff would not do so.37 31 Peixeiro v Haberman, [1997] 3 SCR 549 at para 37. See also Ryan v Moore, [2005] 2 SCR 53 at paras 23, 24 and 31. 32 Kamloops (City) v Nielson, [1984] 2 SCR 2; Central Trust Co v Rafuse, [1986] 2 SCR 147; M(K) v M(H), [1992] 3 SCR 6; Peixeiro, ibid; Novak v Bond, [1999] 1 SCR 808. Where the discoverability principle applies, it can create difficult issues in class actions, as different members of the class may have different dates upon which they knew or should have known of a cause of action. In Smith v Inco, 2010 ONSC 3790 at para 118, an environmental class action relating to soil contamination reducing property values, the Ontario Superior Court applied discoverability as of the date the ‘overwhelming majority’ of the class ought to have known of the claim, even though some members of the class knew or ought to have known earlier. 33 2010 FC 996 [Garford]. 34 See Garford, ibid at paras 31–33, following the same conclusion expressed in obiter in Laboratoires Servier v Apotex Inc, 2008 FC 825 at para 488, affirmed, 2009 FCA 222, leave to appeal to SCC refused, [2009] SCCA No 403. The British Columbia Court of Appeal allowed postponement of a Competition Act claim in Sun-Rype Products Ltd v Archer Daniels Midland Company, note 30 above, but explicitly did so ‘[b]ased on the agreement of the parties’ (para 137). The Court of Appeal permitted postponement with respect to the consumer representative plaintiff’s claims, but not those of the corporate representative plaintiff, as it purchased large quantities and had ‘the means of knowledge’ to know about the price-fixing conspiracy within the limitation period: ibid at para 129. 35 Garford, cited above note 33, at paras 39–45. See also Eli Lilly and Co v Apotex Inc, 2009 FC 991 at para 736 [Eli Lilly], affirmed, 2010 FCA 240, leave to appeal to SCC refused, [2010] SCCA No 434: ‘in the Court’s view, ongoing conduct can only be qualified as ongoing for the purposes of subs. 36(4) so long as it continues to constitute an offence under Part VI of the Competition Act.’ 36 See also Eli Lilly, ibid at para 743: ‘Effects may be examined for the purposes of determining whether or not this agreement was likely to unduly lessen competition, but it does not extend the period during which such conduct occurred’ (emphasis added). 37 Garford, cited above note 33, at para 46.

212  Brian A Facey and David Rosner

(iii) Damages Unlike US law, Canadian antitrust law does not provide for treble damages. Section 36 also excludes punitive damages.38 However, the costs of any ‘investigation in connection with the matter and of proceedings’ under section 36 can be recovered. Under subsection 36(1) of the Act, the plaintiff must have ‘suffered loss or damage’ to obtain recovery. Plaintiffs in class actions frequently claim that despite the wording of subsection 36(1), liability and damages can be assessed on an aggregate basis for the entire class. The courts of British Columbia and Ontario have recently taken different approaches to this question. The leading British Columbia case is Pro-Sys Consultants Ltd v Infineon Technologies AG, which involves allegations of a price-fixing conspiracy among producers of computer memory chips (known as DRAM).39 At first instance, the judge denied certification. He found that loss or deprivation was an element of each class member’s claim, essential to a finding of liability, and could not be established on a class-wide basis. The B.C. Court of Appeal unanimously overturned the lower court’s decision and granted certification. Without deciding the issue, the Court of Appeal stated that damage may not be an element of certain restitutionary claims, such that these claims may be made out on the basis of proof of wrongful conduct and resulting gain without any proof of loss.40 The Court of Appeal placed significant weight on the fact that some of the defendants had pleaded guilty in the United States. Even though the pleas were in a different jurisdiction, under different law, the Court found that the defendants had admitted that they had engaged in the wrongful conduct and that they had obtained an unlawful benefit from that conduct. In light of these admissions, the Court of Appeal found that there was a sufficient basis to conclude, as a preliminary matter at the certification stage, that aggregate damages could be assessed and that the plaintiffs could use statistical evidence to establish that the ‘admitted’ unlawful benefit was attributable to the class, as well as the amount of that benefit.41 In contrast, in Chadha, the Ontario Court of Appeal concluded that the aggregate damages provisions of the Ontario class proceedings statute could only be used 38 Wong v Sony of Canada, [2001] OJ No 1707 (SCJ) at paras 16–18. However, Canadian law does provide for the recovery of pre-judgment interest, which can be substantial when the time period between the date of injury and final judgment is extensive. 39 Pro-Sys Consultants Ltd v Infineon Technologies AG, 2009 BCCA 503, leave to appeal to SCC refused, [2010] SCCA No 32 [DRAM]. 40 DRAM, ibid at paras 31–33, and 70. As discussed below, at the certification stage, the Court only considers whether it is ‘plain and obvious’ that an asserted cause of action has no chance of success; accordingly, in DRAM, the Court of Appeal did not rule on whether a restitionary claim could be established without proof of loss. This is an area of controversy in Canada: see, eg, Serhan Estate v Johnson and Johnson, (2006), 269 D.L.R. (4th) 279 (Div. Ct.), leave to appeal to SCC refused, [2006] SCCA No 494 [Serhan Estate]. The issue is likely to be decided in an ongoing products liability trial against St. Jude Medical Inc that is expected to conclude in September 2011. 41 See also Knight v Imperial Tobacco Canada Limited, 2006 BCCA 235.

Collective Redress for Cartel Damages in Canada  213   to determine the quantum of damages once liability to the entire class had been proven; they could not be used to establish the fact of damage.42 In Irving Paper v Atofina Chemicals Inc, Justice Leitch of the Ontario Superior Court of Justice recently reaffirmed that, pursuant to Chadha, aggregate damages cannot be used to avoid proof of class-wide injury.43 A further controversial question in Canadian antitrust law is whether codefendants in a section 36 case have a right of contribution and indemnity against each other, if found liable to the plaintiffs. In the US, the Supreme Court has held that co-conspirators in a price-fixing action have no such right.44 The Ontario courts have recently noted that this issue has not yet been settled in Canada.45

(iv) Evidentiary standards / burden of proof The traditional evidentiary standard in civil matters is the ‘balance of probabilities,’ otherwise referred to as the ‘preponderance of evidence’ standard. Differing degrees of proof may apply depending on the subject-matter of the case. In R v Oakes,46 the Supreme Court of Canada held: The case may be proved by a preponderance of probability, but there may be degrees of probability within that standard. The degree depends on the subject-matter. A civil court, when considering a charge of fraud, will naturally require a higher degree of probability than that which it would require if considering whether negligence were established. It does not adopt so high a degree as a criminal court, even when it is considering a charge of a criminal nature, but still it does require a degree of probability which is commensurate with the occasion. (emphasis added)47

Since section 36 of the Act provides a civil remedy for damages resulting from specific criminal acts, Oakes would require plaintiffs to prove that criminal conduct with a higher degree of probability.48

Chadha, cited above note 10 at paras 49 and 61. 2010 ONSC 2705; see also Quizno’s Canada Restaurant Corporation v 2038724 Ontario Ltd, 2010 ONCA 466 at para 55, leave to appeal to SCC refused, 2011 CanLII 4632 (the aggregate damages provision of the Ontario class proceedings statute ‘is procedural and cannot be used in proving liability’). 44 Texas Industries v Radcliff Materials, 451 US 630 (1981). 45 Crosslink Technology, Inc v BASF Canada (November 30, 2007), London, 5030CP (Ont. SCJ) at para 47 (unreported); Osmun v Cadbury Adams Canada Inc, 2010 ONSC 2643 at paras 48, 60–65, affirmed, 2010 ONCA 841, leave to appeal to SCC refused, 2011 CanLII 40927; Main v Cadbury Schweppes plc, 2010 BCSC 816 at para 19, affirmed, 2011 BCCA 21, leave to appeal to SCC refused, 2011 CanLII 40927. 46 [1986] 1 SCR 103 [Oakes] (quoting Bater v Bater, [1950] 2 All ER 458 (CA)). 47 Ibid at 459. 48 Oakes, cited above note 46 at 137–38. See, eg, Maritime Travel Inc v Go Travel Direct.Com Inc, 2008 NSSC 163 at para 92, affirmed, 2009 NSCA 42 (holding that where an allegedly misleading advertisement ‘has two possible meanings, one of which would attract criminal and heavy civil sanctions and the other which would not, I conclude the heavier burden of proof on the balance of probabilities is not met.’) 42 43

214  Brian A Facey and David Rosner

III. Procedural considerations (a) Jurisdiction Canada is a federal state, which means that it has two main levels of government: the federal and provincial governments.49 Canada’s Constitution divides all legislative powers between these two levels of government.50 The federal government’s heads of power include ‘trade and commerce,’ and the Competition Act, including section 36 thereof, have been found to validly fall within the scope of that power.51 Actions under section 36 of the Competition Act can be brought in either the Federal Court or a provincial superior court. However, the Federal Court lacks jurisdiction over common law torts, such as intentional interference with economic interests, which frequently have a longer limitation period than subsection 36(4) of the Act and may (unlike section 36 claims) result in punitive damages. As a result, Competition Act claims, particularly class actions, are frequently brought in provincial superior courts, where the plaintiffs can advance both the federal statutory cause of action and common law claims that lie outside the Federal Court’s jurisdiction. This creates significant complexity. The procedural rules in the different provinces are not always the same, and one province, Quebec, maintains civil law traditions whereas the rest of Canada has adopted the common law. The situation in Canada is thus not dissimilar to that which prevails in the European Union – a single competition law must be applied in the courts of different jurisdictions, where different legal traditions and legal procedures may apply. A Canadian provincial superior court will assume jurisdiction over a claim where there is a ‘real and substantial connection’ between the province and the subject matter of the action.52 In Vitapharm Canada Ltd v F Hoffmann-La Roche Ltd,53 defendants from outside of Canada challenged the jurisdiction of the Ontario Superior Court of Justice over class actions related to a conspiracy to fix the prices of bulk vitamins. The Court applied the real and substantial connection test and held that it had jurisdiction over the foreign defendants. The Court found 49 Canada has ten provinces: Newfoundland and Labrador, Prince Edward Island, Nova Scotia, New Brunswick, Quebec, Ontario, Manitoba, Saskatchewan, Alberta, and British Columbia. It also has three northern territories, each of which exercises powers delegated by the federal government: Nunavut, Northwest Territories, and the Yukon. 50 See Constitution Act, 1867, 30 & 31 Victoria, c. 3 (U.K.), sections 91–92 [Constitution Act]. 51 General Motors Canada, cited above note 3. 52 Morguard Investments Ltd v De Savoye, [1990] 3 SCR 1077 [Morguard]. In March 2011, the Supreme Court of Canada heard and reserved on an appeal from the Ontario Court of Appeal’s decision in Van Breda v Village Resorts Ltd, 2010 ONCA 84, which will clarify the factors to be considered in applying the real and substantial connection test. Currently, the Van Breda test sets out a long list of criteria for examining the connection between the forum and the plaintiff’s claim, and the connection between the forum and the defendant. 53 [2002] OJ No 298 (SCJ).

Collective Redress for Cartel Damages in Canada  215   that it was reasonable to believe that, if the international price-fixing scheme were proven, Canadian consumers would have been detrimentally impacted by paying unnecessarily high prices. This was found to be a sufficient connection to Ontario. In asserting jurisdiction, the Court also considered the consequences of refusing to certify; namely, a multiplicity of proceedings in multiple jurisdictions, which would lead to substantially higher litigation costs for the plaintiffs, raise the possibility of inconsistent results, and likely lead to the inability of some plaintiffs to pursue their claims. In theory, a defendant can be faced with claims (individual and/or class) in all thirteen Canadian provinces and territories. There is no formal process similar to the multi-district litigation procedure available in the US for coordinating discovery or trials among the provinces.54 However, the Canadian Bar Association’s Task Force on Class Actions has recently circulated for comment a draft judicial protocol that would allow courts in different provinces to work together to coordinate class actions as they move towards a hearing or settlement.55 A number of Canadian courts have also found that they can certify a ‘national class,’ even though such a class includes persons who reside outside of the province where the class proceeding was commenced.56 The ‘national class’ is discussed below.

(b) Class actions (i) Overview Class actions may now be the most commonly-used procedural tool to enforce private antitrust remedies in Canada. Class proceeding legislation is said to have three policy objectives: judicial economy, access to justice and behaviour modification. These objectives inform all aspects of class proceedings.57 All but the smallest of the Canadian provinces (Prince Edward Island) have class action legislation. If a plaintiff wishes to bring proceedings on a class basis, a court must approve or ‘certify’ a class proceeding. The criteria for certification in Ontario are typical of most provinces (except for Quebec, which is discussed below). The plaintiff must establish that: a) the claim discloses a viable cause of action; 54 See, generally, Uniform Law Conference of Canada, Civil Law Section, ‘Report of the Uniform Law Conference of Canada’s Committee on the National Class and Related Interjurisdictional Issues: Background, Analysis, and Recommendations’ (March 9, 2005). 55 See ‘Judicial Protocol Consultation’, http://www.cba.org/CBA/ClassActionsTaskForce/Main/ judicial_protocol_consultation.aspx (date accessed: 18 August 2011). See also: http://www.cba.org/ CBA/ClassActionsTaskForce/Main/default.aspx (date accessed: 23 September 2011). 56 See, eg, S. Gordon McKee and Jeff Galway, ‘Constitutional Considerations Concerning National Class Actions,’ Law Society of Upper Canada Special Lectures 2001 [McKee & Galway]. 57 Ontario Law Reform Commission, Report on Class Actions (Toronto: Ministry of the Attorney General, 1982) at 117 [OLRC Report].

216  Brian A Facey and David Rosner b) there is an identifiable class of two or more people who are willing to be represented by a representative plaintiff; c) the claims of the class members raise common issues; d) a class action would be the preferable procedure for resolution of the action; and e) there is an appropriate representative plaintiff.58 Whether a court certifies a proposed class action is based on these five criteria, not on the merits of the claim itself. The standard applied is that the plaintiffs must show ‘some basis in fact’ for each of the five certification criteria, other than the requirement that the pleading disclose a cause of action, which is assessed on the basis that the facts alleged in the claim are presumed true.59 Further, the fact that the action may require individual assessments of damages will not bar certification.60 The threshold for class certification in Canada is generally considered to be lower than in the United States, in large part because there is no requirement that the common issues ‘predominate’ over the individual issues. In Canada, predominance is only one of many factors to be considered in determining whether a class action is the ‘preferable procedure.’ Certification motions typically involve affidavit evidence, including expert reports.61 In most provinces, cross-examination is usually conducted on these affidavits, although it is limited to matters that are relevant to certification and not solely to the merits of the case.62 Unlike the United States, there is no broad pre-certification discovery. Plaintiffs frequently cite this lack of discovery as a reason that the certification test should be relatively easy for them to meet, as they claim that without significant disclosure of the defendant’s documents and other information, they could not meet a rigorous standard at the preliminary certification stage.63 On the other hand, the ‘some basis in fact’ test is already a relatively low evidentiary threshold, and exposure of defendants to the additional expense and risk of a class proceeding is an important decision that should not merely be rubber-stamped. Each of the criteria for certification is discussed briefly below. o Disclosure of a Cause of Action. To succeed on a motion for certification, a plaintiff must demonstrate that the claim discloses a cause of action. In an antitrust case, plaintiffs typically plead the statutory cause of action under section 36 of the Competition Act, the common law torts of conspiracy and Class Proceedings Act, 1992, SO 1992, c 6, section 5 [Ontario Class Proceedings Act]. Hollick v Toronto (City), [2001] 3 SCR 158 at para 25. 60 Ontario Class Proceedings Act, cited above note 58, section 6. 61 Recent Ontario and BC decisions have held that expert opinions adduced at certification hearings should not be subjected to the same exacting scrutiny as required at trial. See, eg, DRAM, cited above note 39 at para 66; Quizno’s, cited above note 43, at para 45. But see Dumoulin v Ontario, [2005] OJ No 3961 (SCJ) at para 27 (where one of the criteria for certification depends on a disputed fact, that fact must be decided by the motions judge on a balance of probabilities). 62 Pro-Sys, cited above note 15. 63 See, eg, DRAM, cited above note 39, at para 54. 58 59

Collective Redress for Cartel Damages in Canada  217   intentional interference with economic interests, and restitutionary claims for disgorgement of profits. Unlike the other criteria for certification, evidence is not admissible on the question of whether the plaintiff has pleaded a cause of action. The allegations made in the pleadings are presumed to be true for the purposes of the motion. The test to be applied is whether it is ‘plain and obvious’ that the claim has no chance of success.64 Even where a defendant has a good defence on the merits, it is difficult to defeat certification on the basis that there is no cause of action. Courts have held that the fact that a plaintiff’s case involves complex issues of fact and law or a novel legal proposition will not preclude the action from proceeding as a class action. Further, it has been held that actions based on uncertain areas of the law should not be resolved at the pleadings stage without the advantage of a factual foundation.65 o Identifiable Class. To succeed on the certification motion, the plaintiff must show that the proposed class is capable of clear definition. Given that the class definition is used to determine who is entitled to any remedy awarded and who is bound by the class action judgment, the courts have held that it is necessary that any particular person’s claim to membership in the class be determinable by stated, objective criteria.66 However, it is not necessary that every class member be named, or be known from the outset of the class proceeding. There must also be a ‘rational relationship’ between the class and the common issues.67 The scope of the class should not be arbitrary. It should be limited to individuals who share the same interest in the resolution of the common issues.68 o Common Issues. ‘Common issues’ are: (a) common but not necessarily identical issues of fact, or (b) common but not necessarily identical issues of law that arise from common but not necessarily identical facts.69 The underlying question is whether a determination of the common issues will avoid the duplication of fact-finding and/or legal analysis. However, resolution of the common issues need not resolve the class members’ claims or determine issues of liability with respect to all class members. It is assumed that individual issues will remain to be determined through individual trials following a determination of the common issues at the common issues trial. The common issues must be both: (a) necessary to the resolution of each class member’s claim; and (b) a ‘substantial ingredient’ of each of the class members’ claims.70 Accordingly, the question on a motion for certification Hollick, cited above note 59, at para 25; Hunt v Carey Canada Inc, [1990] 2 SCR 959 at 980. See, eg, Serhan Estate, cited above note 40 (refusing to strike out a novel pleading of ‘waiver of tort’). 66 Western Canadian Shopping Centres Inc v Dutton, [2001] 2 SCR 534 at para 38. 67 Hollick, cited above note 59, at para 20. 68 Ibid at para 21. 69 Class Proceedings Act, cited above note 58, section 1. 70 Hollick, cited above note 59 at para 18; Dutton, cited above note 66 at paras 39–40. 64 65

218  Brian A Facey and David Rosner is not simply whether the claims advanced raise common issues. It is also whether the resolution of the proposed common issues will move the litigation forward to a degree sufficient to justify certification of the action as a class proceeding. o Preferable Procedure. For a class action to be the preferable procedure, the plaintiff must satisfy the court that it is more desirable or suitable than any available alternatives, such as a test case, consolidation of individual actions or proceedings before a regulatory body with jurisdiction over the subject matter of the claims.71 Preferability is assessed in the context of the three objectives of class proceedings: judicial economy, access to justice and behaviour modification.72 Class action legislation frequently describes this branch of the test as being whether a class proceeding is the preferable procedure for determining the common issues. That said, the Supreme Court has held that the test is whether a class proceeding is the preferable procedure for determining the entire action, not only the common issues.73 The question of preferability must, therefore, take into account the importance of the proposed common issues in relation to the claims as a whole, including the individual issues involved in the claims. However, unlike in the US, the common issues need not predominate over the individual issues for a class proceeding to be the preferable procedure. The court must also be satisfied that a class proceeding would be fair, efficient and manageable in the circumstances.74 o Representative Plaintiff. In this stage of the analysis, the court will consider whether there is a representative plaintiff who: (i) would fairly and adequately represent the interests of the class; (ii) has provided a plan for the proceeding that sets out a workable method of advancing the action on behalf of the class and notifying the class of the proceeding; and (iii) does not have an interest in conflict with the interests of other class members on the common issues.75 Compared to the rest of Canada, the certification threshold in Quebec is substantially lower.76 Quebec courts will certify class actions where: (1) the claims of the members raise identical, similar or related questions of law or fact; (2) the facts alleged seem to justify the conclusions sought; (3) the composition of the group makes the application of article 59 (representative actions) or 67 (joint actions) of the Quebec Code of Civil Procedure difficult or impracticable; and (4) the member to whom the court intends to ascribe the status of representative is in a position to represent the members adequately.77 Changes to procedural rules in Quebec have also made it very difficult to challenge the veracity of the plaintiffs’ Hollick, ibid at paras 28 and 31. Ibid at para 27. 73 Ibid at para 30. 74 Ibid at para 28; Rumley v British Columbia, [2001] 3 SCR 184 at para 35. 75 Class Proceedings Act, cited above note 58, section 5(1)(e). 76 See below the discussion on Option Consommateurs v Novopharm Ltd, [2006] QCCS 118 [Option Consommateurs]. 77 Quebec Code of Civil Procedure, art 1003. 71 72

Collective Redress for Cartel Damages in Canada  219   factual submissions at the certification hearing.78 Defendants no longer have a right to cross-examine a representative plaintiff on a motion for authorization to institute a class action, and must obtain leave of the court to adduce oral evidence at this stage of the proceedings.79 Consequently, defence counsel in Quebec often focus on building a case for decertification later in the process. Despite Quebec’s lower threshold for certification, Option Consommateurs v Novopharm Ltd,80 a decision of the Quebec Superior Court, demonstrates that a motion for authorization to institute class proceedings is more than a mere formality. In 2003, a Quebec newspaper published an article claiming that certain unnamed generic manufacturers of pharmaceutical products may have provided pharmacists in Quebec and other provinces with illegal rebates, discounts and kickbacks in return for prescribing their drugs. The newspaper article also noted that a Quebec regulatory body was investigating the matter. Option Consommateurs, a consumer rights not-for-profit association, launched a motion for leave to institute a class proceeding against nine pharmaceutical manufacturers shortly after the newspaper article was published, seeking damages of approximately $4 billion. Its motion material included the newspaper article. Stating that a motion to authorize the institution of class proceedings is not a mere formality, Justice Roy refused to certify the class on the basis of speculative, vague and general allegations.

(ii) Discovery (including cross-border discovery) As discussed above, in general, pre-certification discovery is quite limited in Canada. If the class action is certified, oral examination for discovery and production of documents on the common issues follows.81 Recently, the Ontario courts have been bifurcating the common issues trial, with quantification of damages and entitlement to punitive damages being deferred.82 Bifurcation 78 The amendments to the Quebec Code of Civil Procedure came into force on 1 January 2003. These rules provide that a motion for authorization of a class action no longer requires the support of sworn statements, that it can only be contested orally, and that the court has the discretion to authorize the presentation of evidence at the relevant hearing. The amendments have the effect of depriving respondents with an opportunity to cross-examine applicants on the facts alleged in their certification motion, and provide courts with the right to reject evidence relevant to the proceedings on the basis that the evidence is not appropriate. The Quebec Court of Appeal confirmed the constitutional validity of these amendments in Pharmascience Inc v Option Consommateurs, [2005] QCCA 437. 79 Young v Noranda Inc, [2005] JQ No 16113 (Sup. Ct.) (permitting deposition of the representative plaintiff prior to the certification hearing, but limiting it to the certification criteria, and requiring that if the defendant wished to introduce the transcript of the deposition in evidence at the certification hearing, permission of the certification judge would have to be obtained, and the judge would have to be convinced that the evidence was relevant to the certification.); see also Mazzona c. Daimler Chrysler, 2008 JQ No 13987 (SCJ) at para 33. 80 Cited above note 76. 81 Axiom Plastics Inc v EI DuPont Canada Co, 2011 ONSC 4510 at paras 38–45 (scope of discovery in class proceedings). 82 See, eg, Robinson v Medtronic, Inc, [2009] OJ No 4366 (SCJ), affirmed, 2010 ONSC 3777 (Div. Ct.); Peter v Medtronic, Inc, [2009] OJ No 4364 (SCJ), affirmed 2010 ONSC 3777 (Div. Ct.); Markson v MBNA Canada (21 October 2009) (Ont. SCJ) (unreported).

220  Brian A Facey and David Rosner avoids expensive discovery on issues that may never be reached. Discovery on individual issues is deferred until after all phases of the common issues trial have been completed. Typically (other than in Quebec),83 discovery rules require the parties to produce all relevant documents that are in their possession, power or control without the need for any specific interrogatory or request from opposing counsel.84 Oral discoveries in Canada are more limited than in the US, often being restricted to one corporate representative of each party, rather than every person with knowledge of the matters in issue.85 Discovery of a non-party requires leave of the court.86 In most provinces, there is an implied or deemed undertaking that documents and information produced by a party through the discovery process will not be used for any purpose other than the conduct of that litigation.87 The existence of this implied undertaking can make it more difficult to obtain confidentiality/ protective orders in Canada. Such orders are generally limited to documents that contain trade secrets or other proprietary information the disclosure of which could reasonably be expected to cause financial or other harm to the company.88 Given the limits on discovery in Canada, particularly pre-certification, Canadian plaintiffs are increasingly seeking access to evidence given in discovery in related US cases. The scope of permissible cross-border discovery was discussed in Vitapharm v F Hoffmann-LaRoche Ltd.89 As noted earlier, this case related to a global price-fixing conspiracy, where manufacturers of certain vitamins were alleged to have conspired to fix prices and allocate markets, including in Canada. Litigation was brought in Canada, the United States and elsewhere. Millions of documents were produced in the US litigation under US rules, subject to a protective order. The Ontario plaintiffs moved to intervene in the US litigation to modify the protective order to allow them access to the defendants’ documents and to US depositions. The defendants in the Ontario action brought a motion in Ontario for an order preventing the plaintiffs from gaining access to US discovery. The Ontario Superior Court of Justice dismissed the defendants’ motion, noting that the motion in the US was necessary only because of the protective order. If there had been no protective order, and the plaintiffs were simply given access 83 In Quebec, the parties are only required to produce those documents that they intend to rely upon or that are specifically requested by opposing counsel during oral discovery. 84 See, eg, Ontario Rules of Civil Procedure, RRO 1990, Reg. 194 [Ontario Rules of Civil Procedure], R 30.03. 85 See, eg, ibid, r 31.03. 86 See, eg, ibid, r 31.10. 87 See, eg, ibid, r 30.1.01. There are certain exceptions; for instance, under r 30.1.01(6) of the Ontario Rules of Civil Procedure, the deemed undertaking does not preclude use of evidence from one proceeding for the purpose of impeaching a witness in a different case. Where there are multiple class actions in Canada, the parties may agree that discovery of the defendants be conducted simultaneously for purposes of all of the certified proceedings. 88 Sierra Club of Canada v Canada (Minister of Finance), 2002 SCC 41. 89 Cited above note 53.

Collective Redress for Cartel Damages in Canada  221   to the US discovery, the defendants could not object (as there was no deemed or implied undertaking in the US proceedings). Justice Cumming held: ‘A Canadian court generally will be reluctant to prevent someone from gathering evidence extraterritorially, as its ultimate admissibility in a Canadian proceeding will be determined by the Canadian courts.’90 He found that the plaintiffs’ motion before the US court was not unfair, as access to the US discovery would be consistent with the three policy objectives underlying class proceedings – facilitating access to justice, judicial efficiency and behaviour modification.91 In particular, there would be significant savings in litigation costs. This decision was upheld by both the Ontario Divisional Court and Court of Appeal.92 The Court of Appeal noted that the US judge could make any amendment to his protective order conditional upon similar protective orders being issued in the Canadian proceedings (to protect confidential documents).93 Canadian plaintiffs have sometimes sought a more active role in US proceedings. For example, the plaintiffs in Coleman v Bayer Inc94 sought to intervene in a US class action for the purpose of modifying a protective order to allow them access to documents produced, but also sought to conduct oral discovery. The US court held that ‘allowing the Canadian plaintiffs access to discovery materials already produced will not offend the Canadian tribunal, nor will it prejudice defendants by further intruding into their privacy or subjecting them to further questioning or document production.’95 However, the Canadian plaintiffs were not permitted to actively conduct depositions in the US case. In In Re: Urethane Antitrust Litigation, the United States District Court for the District of Kansas modified its protective order to grant access to US discovery materials, which it described as ‘a rather passive act that simply lifts a restriction on who may access confidential material.’96 However, the Court denied the Canadian plaintiff an order granting it permission to obtain and/or use such discovery in the Canadian litigation, finding that this was an issue to be determined by the Canadian courts.

90 Ibid at para 42. See, eg, Pro-Sys Consultants Ltd v Microsoft Corporation, 2008 BCSC 1263, where the plaintiffs had obtained access to US discovery. The Court found that the plaintiffs’ expert could attach reports she had authored for the US cases to her Canadian affidavit, and be cross-examined on them, and could also rely on reports filed by the defendants in the US. However, US reports of other plaintiffs’ experts, who had not filed an affidavit in Canada, were not admissible. 91 OLRC Report, above note 57 at 117. 92 (2002), 212 DLR (4th) 563 (Ont. Div. Ct.); (2003), 223 D.L.R. (4th) 445 (Ont. CA); leave to appeal refused, [2003] SCCA No 245. 93 Ibid (CA) at para 6. The United States District Court, District of Columbia, declined to rule on the Canadian plaintiffs’ motion until the appeals in Ontario had run their course: In re Vitamins Antitrust Litigation, MDL No 1285, Misc. No 99–197 (TFH). The Ontario litigation subsequently settled. 94 [2004] OJ No 1974 (SCJ) [Coleman]. 95 In re Baycol Products Litigation, MDL No 1431, pretrial order No 77 (US Dist. Minn. 2003) at 9; see also In re Linerboard Antitrust Litigation, 333 F. Supp. 2d 33 (East. Penn. Dist. Ct., 2004). 96 Case No 04-MD-1616-JWL (10 September 2010).

222  Brian A Facey and David Rosner

(iii) Opt-out and opt-in regimes In Ontario, an ‘opt-out’ regime exists whereby persons who fall within the class definition are deemed to be members of the class and are bound by any judgment or settlement unless they take the required steps to opt-out of the proceeding. Alberta, Manitoba, Saskatchewan, and Nova Scotia have similar ‘opt-out’ regimes. British Columbia and Newfoundland and Labrador have hybrid ‘optout’ and ‘opt-in’ regimes, whereby residents of the province must opt-out of class membership, but non-residents are required to opt-in if they wish to be part of the class. In Quebec, courts have traditionally been averse to including non-residents in the class, except where a non-resident demonstrates that his or her individual claim has a real and substantial connection to Quebec.97 That said, Quebec courts have recently certified national classes.98 Although provincial courts have certified national or multi-jurisdictional classes, issues nonetheless remain regarding the enforceability and constitutionality of class action judgments/settlements that include non-residents as class members. No Canadian decision has yet resolved these issues unequivocally.99 This issue is discussed in further detail below in the section on national classes.

(iv) Costs There are significant differences between the provinces regarding liability for legal costs in a class proceeding. In British Columbia and Manitoba, class action legislation provides that, absent exceptional circumstances, costs and disbursements are not to be awarded against either the plaintiff or defendant for any aspect of a class proceeding. In Alberta, Saskatchewan and Newfoundland, the court has the discretion to award costs. In Ontario and Quebec, costs are generally awarded to the successful party.100 Where costs are available, the normal rule is that the successful party is entitled to a portion of its costs of the proceeding, but any award is subject to the discretion of the court to reduce or deny costs. In Ontario, if an unsuccessful party is the plaintiff, only the representative plaintiff, and not the other class members, is responsible for any adverse costs award.101 A representative plaintiff can, however, insulate himself or herself from 97 An example of a non-resident’s real and substantial connection to Quebec would be where the injurious conduct occurred in the province but the class member left Quebec prior to the litigation being commenced. 98 See, eg, Brito v Pfizer Inc, [2008] JQ No 4642 (Sup. Ct.). 99 See the section on ‘National Class’ herein. See also McKee & Galway, cited above note 56; Hocking v Haziza, 2008 QCCA 800 [Hocking]; Société canadienne des postes v Lépine, [2007] JQ No 8498 (SCJ), affirmed, [2009] 1 SCR 549 (refusing to enforce Ontario national class judgments); and Punit v Wawanesa Mutual Insurance Co, [2005] OJ No 1928 (SCJ). 100 The Supreme Court of Canada has held: ‘it should not be assumed that class proceedings invariably engage access to justice concerns to an extent sufficient to justify withholding costs from the successful party’: Kerr v Danier Leather Inc, 2007 SCC 44 at para 69. 101 Ontario Class Proceedings Act, cited above note 58 at s. 31(2).

Collective Redress for Cartel Damages in Canada  223   a costs award by obtaining funding from the Ontario Class Proceedings Fund or an indemnification from class counsel’s firm.102 Ontario courts have, on occasion, denied costs to a successful defendant where the class proceeding is considered a test case, raises a novel point of law or involves a matter of public interest.103 In Caputo v Imperial Tobacco Limited,104 an Ontario court denied the defendants’ request for costs from the representative plaintiffs (and their counsel) following an unsuccessful certification motion. The plaintiffs had sought to certify a class that was broadly defined to include all residents of Ontario, whether living or deceased, who had ever smoked cigarette products manufactured, marketed or sold by the defendants. In denying certification, the Court concluded that ‘[t]he only apparent common element in this action is that all of the proposed class members allegedly smoked cigarettes at one time or another.’105 The defendants wanted costs in excess of $1.2 million on a joint and several basis against both the representative plaintiffs and their counsel. They argued that plaintiffs’ counsel were the ‘real plaintiffs’ in the case and had been motivated by a potentially large contingency fee to pursue a case that the representative plaintiffs would not have pursued on their own out of fear of liability for costs. The defendants blamed plaintiffs’ counsel for driving up the costs of the litigation by, among other things, advancing nine causes of action, conducting excessive cross-examination of the defendants’ affiants, and continuously tailoring the list of proposed common issues, which together resulted in the certification motion taking nine years to reach a hearing. The Court refused the defendants’ request for costs in its entirety on two grounds. First, the Court noted that the action raised novel points of law, and second, that it raised a matter of public interest, namely health issues related to smoking.106 Accordingly, the Court found that the case was a proper one in which to deviate from Ontario’s usual loser pays costs rule. The Court also rejected the defendants’ argument that the greater involvement of plaintiffs’ counsel in class action litigation is grounds for elevating them to the status of a party liable for costs, finding that this would run entirely counter to the ‘access to justice’ purpose of class proceedings.107 This access to justice rationale has also been used as a reason to approve relatively generous contingency fees for plaintiffs’ counsel.108 However, the See also the discussion below regarding third party funding. See, eg, Ontario Class Proceedings Act, cited above note 58 at s. 31(1); Caputo v Imperial Tobacco, [2005] OJ No 842 (SCJ) (costs) [Caputo]; (2004), 74 OR (3d) 728 (SCJ) (certification). 104 Caputo, ibid. 105 Ibid. 106 Ibid at para 33. It may be that the court looked harder for a way to avoid awarding any costs because of a view that the defendant was overreaching by asking for such a large award, and that it be paid by plaintiffs’ counsel. 107 Ibid at para 42. But see Attis v Ontario Minister of Health, 2010 ONSC 4508 in which costs were ordered against the plaintiffs’ solicitors as they had failed to adequately apprise the putative representative plaintiffs of their potential liability for costs awarded to the defendant. 108 The Ontario Class Proceedings Act, cited above note 58 at subsections 31(3)–(9), allows class counsel to enter agreements with their clients providing for a ‘multiplier’ of their usual fees, if successful at the common issues trial or by settlement, and if approved by the courts. 102 103

224  Brian A Facey and David Rosner Ontario Court of Appeal has recently raised concerns about the uncontested nature of class counsel fee requests where a settlement is being approved.109 In such circumstances, class counsel are seeking fees (usually including a contingency fee premium) that will reduce the funds available to class members. The settling defendant usually takes no position on fees being paid by the class. The Court of Appeal found that an amicus or monitor could be appointed in appropriate cases to ensure fairness to the class.

(v) Third-party funding Recent jurisprudential developments may have opened up new and somewhat alarming possibilities for the funding of class action litigation. In Dugal v Manulife Financial Corporation,110 Justice Strathy approved a funding arrangement wherein an Irish company would indemnify the plaintiffs against any costs order in exchange for 7% of any recovery in the class proceeding. This company engages exclusively in funding litigation and is one of a burgeoning crop of litigation-finance companies, some of which have been formed in Canada. The funding arrangement, prima facie, smacks of historically unlawful champerty. However, Justice Strathy rejected an argument that the arrangement was champertous, pointing out that there was no evidence of the funder inciting the litigation.111 The decision is worrisome for a number of reasons. First, taking the financial risk out of the hands of parties initiating litigation shifts motives and could incentivize frivolous and vexatious litigation, clogging the civil justice system needlessly. Second, bringing in a ‘silent partner’ who is not a party to the action raises questions regarding confidentiality and the degree to which the funder is entitled to information about the progress of the case (Justice Strathy noted that some degree of information disclosure between the plaintiff and funder was acceptable). Third, generally, the plaintiff class already receives a relatively small share of any amounts recovered in a successful lawsuit or settlement, so it is unclear if permitting third-party funding benefits the actual class members at all, or merely increases the profits of class counsel.

(vi) Standing As previously noted, a representative plaintiff must be accepted as an appropriate representative at the certification hearing. In Ontario, a would-be representative plaintiff must have a cause of action against a named defendant, or the case may Smith Estate v National Money Mart Company, 2011 ONCA 233. [2011] OJ No 1239 (SCJ) [Dugal]. See also Hobshawn v Atco Gas and Pipelines Ltd (14 May 2009), Action 0101-04999 (Alta. Q.B.) (unreported); MacQueen v Sydney Steel Corporation (19 October 2010), Action 218010 (NSSC) (unreported); see also Sandra Rubin, ‘Enter the Silent Partner’ in Lexpert, vol. 12, No 9 (Carswell: Toronto, 2011) at 58 [Silent Partner]. 111 Dugal, ibid at para 33. 109 110

Collective Redress for Cartel Damages in Canada  225   be dismissed for ‘no cause of action’ prior to the certification motion.112 In some provinces, where multiple defendants are named in an action, at least one of the named plaintiffs must have a cause of action against each named defendant.113 Certain Canadian jurisdictions have residency requirements that only permit a resident of the jurisdiction to institute class proceedings.114 Other provinces, notably Quebec, Alberta, and Ontario, allow non-residents to bring and be the named plaintiff in class proceedings. The Nova Scotia Court of Appeal has recently held that non-residents whose claims have no real and substantial connection to the province cannot be representative plaintiffs, as the court lacks constitutional jurisdiction over them.115

(vii) National classes Anticompetitive conduct can have effects in more than one province or territory. As a result, antitrust class actions can be, and often are, instituted in several provinces contemporaneously (usually Ontario, Quebec and B.C., which have the largest populations and longest history of class actions). Plaintiffs’ counsel may attempt to coordinate their actions across Canada if they consider it strategically advantageous to do so. Typically, they will commence provincial class actions in Quebec and British Columbia and an opt-out class for the rest of Canada in Ontario. While there is precedent for the certification of multi-jurisdictional or ‘national’ classes in Canada, this is only appropriate where there is a real and substantial connection (the Canadian test for jurisdiction) between the forum and the claims of the non-residents.116 There have been several cases in which Canadian courts have held that a ‘real and substantial connection’ will exist so long as the residents and non-residents in a national class share the same common issue.117 However, this appears to Such a motion is brought under the Ontario Rules of Civil Procedure, rule 21. This is the case in Alberta and Ontario, but not in British Columbia. But see Lupsor Estate v Middlesex Mutual Insurance Co, [2003] OJ No 1038 (SCJ), leave to appeal granted, [2003] OJ No 3745 (Div. Ct.), allowing the named plaintiff to seek certification against multiple defendants even where the named plaintiff did not have a cause of action against each of them (leave to appeal granted on the basis that the decision conflicted with the case law, and there was good reason to doubt its correctness). 114 This is the case in the provinces of British Columbia and Newfoundland and Labrador. 115 Bellefontaine v Purdue Frederick Inc, 2010 NSCA 58 [Bellefontaine Appeal], affirming unreported oral decision (26 February 2009) (NSSC) [Bellefontaine]. 116 A province has no competence to legislate extra-territorially. The Constitution Act, 1867 imposes a territorial limitation on provincial legislative power. In particular, the opening words of section 92 of the Constitution Act, 1867 provide: ‘In each Province the Legislature may exclusively make Laws in relation to Matters coming within the Classes of Subjects next hereinafter enumerated’ [emphasis added]. The enumerated powers include: ‘Property and Civil Rights in the Province’ (section 92(13)), ‘The Administration of Justice in the Province’ (section 92(14)), and ‘Generally all Matters of a merely local or private Nature in the Province’ (s 92(16)) (emphasis added). See also: Unifund Assurance Co v Insurance Corp. of British Columbia, 2003 SCC 40 at paras 50–56 (one province cannot regulate civil rights in another province). 117 See, eg, Carom v Bre-X Minerals Ltd (1999), 43 OR (3d) 441 (Gen. Div.) [Bre-X]; Wilson v Servier Canada Inc (2002), 59 OR (3d) 656 (Sup. Ct.) and Wilson v Servier Canada Inc (2000), 50 OR 112

113

226  Brian A Facey and David Rosner be inconsistent with the Supreme Court of Canada’s decision in Bisaillon v Concordia University. In Bisaillon, the Supreme Court held that the class action procedure cannot have the effect of conferring jurisdiction on a court over a claim that would not otherwise fall within its jurisdiction.118 Accordingly, a ‘common issue,’ which is simply a class action procedural categorization of an issue that can be determined on a class-wide basis, cannot give a province jurisdiction over the claims of individuals who could not have brought the same action on an individual basis in that province. If there is no real and substantial connection between the non-residents’ claims and the forum, then their cases do not fall within the subjectmatter jurisdiction of the court. In light of Bisaillon, in Bellefontaine v Purdue Frederick Inc, a Nova Scotia court dismissed, pre-certification, the claims of non-resident plaintiffs in a class action relating to the prescription medicine OxyContin, as their claims had no real and substantial connection to Nova Scotia.119 In particular, the Court found that plaintiffs who had not lived in Nova Scotia, been treated there, or obtained OxyContin there, lacked the requisite connection to the jurisdiction. The Court also rejected the plaintiffs’ submissions that sharing common issues, as between the resident and non-resident plaintiffs, was sufficient to establish a real and substantial connection to Nova Scotia.120 That decision was affirmed on appeal.121 Canada’s foremost constitutional scholar, Prof. Peter Hogg, collaborating with one of Canada’s leading class action practitioners, recently published an article examining the constitutionality of national classes. They concluded: We see no escape from the conclusion that the jurisdiction of provincial courts over national class actions is restricted to a plaintiff class that includes only persons whose claims, judged individually, all have a real and substantial connection to the forum. This is by no means an unusual situation. Where the action has been brought in the province where the defendant resides and where the wrongdoing took place, it will be satisfied. And, even where the action has been brought in a province where the defendant does not reside, the material facts (a catastrophic accident for example) may still provide a real and substantial connection to the forum province for the claims of everyone in the plaintiff class. But, where the claims of some members of the class do not have a close connection to the forum, the convenience of a national class action cannot overcome the constitutionally prescribed territorial restriction on the jurisdiction of provincial courts. (3d) 219 (Sup. Ct.), leave to appeal to Div. Ct. refused, 52 OR (3d) 20, leave to appeal to SCC refused, [2001] SCCA No 88; Harrington v Dow Corning Corp. (1997), 29 BCLR (3d) 88 (SC), affirmed, [2000] 11 WWR 201 (CA), leave to appeal to SCC refused, [2001] SCCA No 21 (but note that British Columbia only takes jurisdiction over non-resident class members who ‘opt-in’ and thus attorn to jurisdiction in British Columbia). 2006 SCC 19 at paras 17–22. Bellefontaine, cited above note 115. Bellefontaine Appeal, ibid at para 12. See also Hocking, cited above note 99 at paras 150–151, in which the Quebec Court of Appeal confirmed (in obiter) that the Canadian constitution requires a real and substantial connection between each class member’s claim and the forum province. See also Goyette c. GlaxoSmithKline Inc, 2009 QCCS 3745 at para 124 (in obiter), affirmed, 2010 QCCA 2054. 121 Bellefontaine Appeal, cited above note 115. 118 119

120

Collective Redress for Cartel Damages in Canada  227   In our opinion, a national class action in the superior court of a province cannot include anyone in the plaintiff class whose claim, if brought individually, would not have a real and substantial connection to the forum province. (emphasis added)122

(viii) International classes There is also an emerging trend of certifying international classes. For example, in McKenna v Gammon Gold,123 an Ontario court certified a ‘global’ class consisting of all purchasers worldwide of Gammon Gold’s securities during a specified period. To justify asserting jurisdiction over a global class, the Court found that a ‘real and substantial connection’ existed between non-resident class members and Ontario.124 The feasibility of granting adequate notice to the class is a major concern in international class actions. Class action litigation proceeds on the assumption that it is in the interests of justice and efficacy to authorize a plaintiff to act on behalf of a larger group of individuals who share common legal claims. However, such an approach is only appropriate where the fruits of the litigation may be shared among all members of the class and where each class member may assert his or her rights. Where the proposed class extends globally, it is difficult to achieve a workable means of ensuring that non-resident class members have been made aware of the foreign proceeding, such that they could be bound by the foreign judgment. This problem of adequate notice has also been an issue in enforcing foreign judgments or settlements relating to international classes in Canada, as discussed below.

(ix) Certification for the purposes of and standards for settlement In most provinces, an action commenced as a class action cannot be settled or discontinued without court approval, even prior to certification. The settlement agreement will generally require certification, if it has not yet occurred, to ensure that all class members are bound by the settlement. Where certification is sought solely for settlement purposes, the test for certification tends, in application at least, to be easier to meet than in an opposed certification motion. In particular, a class action that would not be the preferable procedure for active litigation may be the appropriate procedure for administering a settlement.125 122 Peter W Hogg and S Gordon McKee, ‘Are Class Actions Constitutional?’ (2010) NJCL 279, 292. For a response to this article, see Janet W Walker, ‘Are National Class Actions Constitutional? – A Reply to Hogg and McKee’ 48 OHLJ 95 (2010). 123 [2010] OJ No 1057 (SCJ) [McKenna], leave to appeal granted, 2010 ONSC 4068 (Div. Ct.), varied (but not on this point), 2011 ONSC 3782 (Div. Ct.). Other examples of courts certifying global classes include Ford v F Hoffman-LaRoche Ltd (2005), 74 OR (3d) 758 (SCJ) and Mandeville v Manufacturers Life Insurance Co, [2002] OJ No 5387 (SCJ). 124 McKenna, ibid at paras 114–116. 125 See, eg, Fischer v IG Investment Management Ltd, 2010 ONSC 5132, in which Justice Perell of the Ontario Superior Court certified a class action for settlement purposes even though he had previously denied certification on the basis that the preferable procedure criterion had not been satisfied.

228  Brian A Facey and David Rosner For example, certification was granted for the purposes of settlement in Bona Foods Ltd v Ajinomoto USA Inc,126 a class action alleging that the defendants had illegally fixed the price of monosodium glutamate and nucleotides in the Canadian market. The motions judge questioned whether the losses alleged could be proved on a common basis, but concluded that it was reasonable not to give weight to such considerations when certification was requested only for purposes of a settlement that would resolve and terminate the litigation. A settlement of a certified class proceeding is not binding unless it is approved by a court. The court assesses whether the proposed settlement is fair, reasonable and in the best interests of the class.127 However, there is a ‘strong initial presumption of fairness when a proposed class settlement, which was negotiated at arms-length by counsel for the class, is presented for court approval.’128

(x) Enforcement of settlements or judgment The Supreme Court’s 1990 decision in Morguard Investments Ltd v De Savoye has dramatically changed Canadian law regarding the recognition and enforcement of foreign judgments.129 Companies doing business in Canada that ignore foreign proceedings do so at their peril. Courts in Canada now commonly enforce foreign judgments if the court issuing the order had a real and substantial connection to the litigation.130 The Ontario Court of Appeal confirmed in Currie v McDonald’s Restaurants of Canada that it is willing, in proper circumstances, to enforce a settlement in a US class action that also includes Canadian class members.131 To do so, the Canadian court must confirm: (a) that the foreign court had a proper basis for asserting jurisdiction (the ‘real and substantial connection’ test); and (b) that the interests of the Canadian class members were adequately protected (such that the requirements of ‘order and fairness’ are met). [2004] OJ No 908 (SCJ). Vitapharm Canada Ltd v F Hoffmann-La Roche Ltd, [2005] OJ No 1118 (SCJ) at para 110 [Vitapharm Settlement]. See also Dabbs v Sun Life Assurance Co of Canada (1998), 40 OR (3d) 429 (Gen. Div) and Bre-X, cited above note 117. In Vitapharm Settlement (at para 117), Justice Cumming noted that in determining whether to approve a settlement, a court will take into account factors such as: ‘(a) the likelihood of recovery or likelihood of success; (b) the amount and nature of discovery, evidence or investigation; (c) the proposed settlement terms and conditions; (d) the recommendation and experience of counsel; (e) the future expense and likely duration of litigation; (f) the recommendation of neutral parties, if any; (g) the number of objectors and nature of objections; (h) the presence of arms-length bargaining and the absence of collusion; (i) the information conveying to the court the dynamics of, and the possible taken by (j) [sic] the parties during the negotiations; and (k) the degree and nature of communications by counsel and the representative (l) [sic] plaintiff with class members during the litigation.’ 128 Vitapharm Settlement, ibid at para 113. 129 See above note 52. 130 See, eg, Beals v Saldanha, [2003] 3 SCR 416; Pro Swing v Elta Golf Inc, 2006 SCC 52. 131 (2005), 74 OR (3d) 321 (CA) [Currie], affirming Parsons v McDonald’s Restaurants of Canada Ltd (2004), 45 CPC (5th) 304 (Ont. SCJ) and Currie v McDonald’s Restaurants of Canada Ltd (2004), 4 C.P.C. (6th) 299 (Ont. SCJ). The two actions were brought together on appeal. 126 127

Collective Redress for Cartel Damages in Canada  229   In Currie, the plaintiffs instituted class proceedings in Canada against McDonald’s Restaurants of Canada Ltd on the basis of alleged wrongdoing regarding promotional games offered to customers. A parallel Illinois class action that included US, Canadian and other international class members was settled. The Illinois court approved the settlement and ordered that notice of the settlement be made to Canadians via an advertisement in three Quebec newspapers for French Canadians and two advertisements in a national monthly magazine for English Canadians. The Ontario Court of Appeal was asked to consider whether the Illinois judgment/settlement was binding in Ontario. If so, the plaintiffs’ proposed class action in Ontario could not proceed, because the subject matter of the litigation would already have been determined. The Court recognized that Ontario residents frequently engage in cross-border activities that may become the subject matter of class litigation in Ontario, in another province or in a foreign jurisdiction. The Court found that there were strong policy reasons that favoured the fair and efficient resolution of inter-provincial and international class action litigation and that conflict of law rules should recognize the advantages of having cases finally resolved in one jurisdiction in appropriate cases. The Court held that the principle of comity will often require that foreign class action judgments be given full effect. Applying the real and substantial connection test, the Court determined that Illinois was a proper jurisdiction for the original class action. On the other hand, the Court of Appeal also found that the principles of ‘order and fairness’ required that careful attention be paid to the situation of ordinary Canadian McDonald’s customers whose rights were at stake. These non-resident class members would have no reason to expect that any legal claim they were asserting against McDonald’s Canada as a result of visiting an Ontario restaurant would be adjudicated in the United States. The Court thus determined that it was required to consider the procedural rights that were given to customers of McDonald’s Canada, particularly the right of unnamed class members to ‘opt out’ of the US class action. The Court found that this right to ‘opt out’ was of vital importance, and was only meaningful if the unnamed class members were given adequate notice of this right. The motions judge at first instance found that the notice of the right to ‘opt out’ of the Illinois proceeding was written in ‘wall to wall legalese’ and its dissemination in Canada was woefully inadequate. It was not reasonable to consider publication in two issues of one magazine as adequate notice to English-speaking Canadians, or to French-speaking Canadians outside of Quebec. As a result, the Ontario Court of Appeal refused to enforce the US judgment in Ontario.132 Nevertheless, it is clear from the Court’s reasons that a US settlement will be enforced against Canadian class members if the requirements of jurisdiction and procedural fairness are met. 132 The Illinois judgment was found to be enforceable against specific Canadians who appeared before the Illinois court to object to the settlement and thus attorned to the jurisdiction of that court: Currie, ibid.

230  Brian A Facey and David Rosner Difficulties obtaining Canadian approval of an international class action settlement are also a feature of Kelman v Goodyear Tire and Rubber Company.133 Goodyear was alleged to have negligently designed, manufactured and sold a heating product. Class actions were commenced throughout Canada and internationally. A single settlement fund was ultimately constituted, from which class members in the US and Canada could make claims. The settlement provided that claim administration would be headquartered in the US, and the US court appointed a special master to hear appeals of denied claims. A Canadian class was certified for purposes of settlement in Ontario (rather than seeking enforcement of the US decision as in Currie). The Ontario court found that it could not approve the settlement as initially presented, as doing so would require it to take jurisdiction, but then cede that jurisdiction to the US court. Regional Senior Justice Winkler (now Chief Justice of Ontario) concluded that, where an Ontario court is involved in approving a settlement, it must retain jurisdiction over implementation of the settlement. The settlement was ultimately approved with a supervisory role for the Ontario court.134 In light of these decisions, it is important for defendants who are considering an international settlement to ensure that they consult with Canadian counsel so that all of the necessary steps, such as proper notice, are taken to minimize the chances of similar claims proceeding in Canada, and to maximize the chances that the settlement will be enforced by a Canadian court. Lack of proper notice can even be an issue where enforcement of a settlement approval in one province is sought in another province. In Canada Post Corp. v Lépine, the Supreme Court affirmed a Quebec decision refusing to enforce an Ontario certification of a national class and settlement approval, due in part to the fact that the Ontario notice was likely to be confusing to Quebec class members.135

(xi) Mandatory arbitration/mediation clauses Companies commonly insert mandatory arbitration/mediation clauses into their standard-form consumer agreements. In the face of a class action, defendants have tried to use these clauses as a basis to defeat certification. Given recent developments in the law, this strategy is now less promising. There is consumer protection legislation in every Canadian province and territory.136 Ontario and Quebec have now specifically modified their [2005] OJ No 175 (SCJ). Ibid at para 25. 135 2009 SCC 16. 136 Alberta Fair Trading Act, RSA 2000, c. F-2; Manitoba Consumer Protection Act, CCSM c C200; Newfoundland and Labrador Consumer Protection Act, CNLR 773/96; Northwest Territories Consumer Protection Act, RRNWT 1990, c C-16; Nova Scotia Consumer Protection Act, RSNS 1989, c 92; New Brunswick Consumer Product Warranty and Liability Act, SNB 1978, c C-18; Ontario Consumer Protection Act, 2002, SO 2002, c 30, Sched. A; Prince Edward Island Consumer 133 134

Collective Redress for Cartel Damages in Canada  231   consumer protection statutes to override arbitration clauses in consumer contracts.137 A recent Supreme Court decision suggests that other provincial consumer legislation may be interpreted in a similar manner, even in the absence of such specific amendments. In Seidel v TELUS Communications Inc,138 the Supreme Court held, in a 5–4 majority, that a putative class action could proceed notwithstanding a contractual mediation/arbitration clause. The appeal concerned a customer of a telecommunications company whose contract provided that any disputes would first be referred to mediation and, failing a successful resolution, to arbitration. The contract also purported to waive any right to participate in or commence a class action. The Court held that, absent legislative intervention, mediation/arbitration clauses are to be enforced, even when they are contained in standard form contracts of adhesion. Canadian courts are generally quite receptive to alternative dispute resolution. However, they are currently even more receptive to class actions, in the belief that they promote access to justice. The Supreme Court found that the British Columbia Business Practices and Consumer Protection Act139 should be interpreted generously in favour of consumers. Even though the statute did not explicitly foreclose use of arbitration to resolve consumer disputes, the majority concluded that general language voiding any waiver or release of rights or protection conferred by the Act rendered both the arbitration clause and the class action waiver void. A similar class action-friendly approach is evident in the Ontario Court of Appeal’s decision in Griffin v Dell.140 The proposed class action related to allegedly defective computers. Approximately 70% of the putative class members were consumers, rather than business purchasers. After deciding the Ontario Consumer Protection Act precluded the arbitration clause from applying to the consumer claims, the Court of Appeal also exercised its discretion to stay arbitration of the non-consumer claims, so they could also proceed as part of the class action. The Court held that ‘it would not be reasonable to separate the consumer from the non-consumer claims’ and that ‘granting a stay of the non-consumer claims would lead to inefficiency, a potential multiplicity of proceedings, and added cost and delay.’141

Protection Act, PEI Reg. EC578/83; Quebec Consumer Protection Act, RSQ, c P-40.1; Saskatchewan Consumer Protection Act, RRS c C-30.1, Reg. 1; Yukon Consumer Protection Act, Y.C.O. 1972/400; Nunavut Consumer Protection Act, RSNWT (Nu) 1988, c C-17; and British Columbia Business Practices and Consumer Protection Act, SBC 2004, Chapter 2 [Business Practices and Consumer Protection Act]. 137 Ontario Consumer Protection Act, ibid, sections 7–8, Quebec Consumer Protection Act, ibid, section 11.1. 138 2011 SCC 15. 139 See above note 136. 140 [2010] OJ No 177 (CA). 141 Ibid at paras 46–47.

232  Brian A Facey and David Rosner

(xii) Coordination with foreign counsel Many of the antitrust class actions instituted in Canada are ‘copycat’ cases based on similar proceedings commenced in the US. Because many class actions have facts that are common to the US and Canada, plaintiffs’ law firms have been forging formal and informal relationships across the border. In practice, the majority of cases first arise in the US, followed by proceedings in Canada. US counsel often assist their Canadian counterparts in various aspects of a case, and Canadian courts have approved counsel fees that include the fees of US counsel.142 Interestingly, on a recent carriage motion in a securities class action, one Ontario firm championed its relationship with a US firm (Milberg LLP) as a reason for the Court to allow its putative class action to proceed and stay a competing proceeding brought by a different firm;143 whereas the competing Ontario firm argued that Milberg’s involvement was a negative factor. The Court ultimately concluded that the proposed role of the US firm to provide investigative services, document management and strategic advice was a ‘neutral factor’ in determining carriage,144 although it ultimately awarded carriage to the firm aligned with Milberg.145 However, Justice Perell warned that an Ontario firm could be disqualified from seeking carriage of an Ontario proceeding ‘if the Ontario firm entered into an arrangement where an American law firm, or any foreign law firm for that matter, assumed de jure or de facto the role of the lawyer of record for the representative plaintiff, unless the foreign law firm obtained permission to practice law in Ontario with a right of audience before the court.’146 It could also be disqualified ‘if the foreign firm had a proprietary interest in the claims of the representative plaintiff and the class’147 that would ‘take the foreign firm’s involvement into the territory of champerty and maintenance and impermissible fee splitting.’148

IV. Conclusion The availability of class actions in Canada has significantly increased the number of private antitrust claims made. However, it is questionable whether 142 See, eg, Vitapharm Canada Ltd v Hoffman-Laroche Ltd, [2005] OJ No 1117 (SCJ). But see Sharma v Timminco, [2009] OJ No 4511 (SCJ) [Sharma]. 143 Sharma, ibid. 144 Ibid at para 18. 145 Subsequently, a lawyer from Milberg joined the Ontario bar so that he could practice Ontario law with the Canadian firm that won the carriage motion. See Kevin Van Paassen, ‘Top US class-lawyer coming to Canada’ in The Globe & Mail, 10 May 2011, http://www.theglobeandmail.com/report-onbusiness/industry-news/the-law-page/top-us-class-action-lawyer-coming-to-canada/article2017397/ (date accessed: 18 August 2011). 146 Sharma, cited above note 142, at para 75. 147 Ibid. 148 Ibid at para 78. The Court found that some of Milberg’s services might be chargeable as disbursements but others would be in the manner of fees for education that could not be charged to class members: ibid at para 79.

Collective Redress for Cartel Damages in Canada  233   this litigation has a significant role in deterring antitrust violations, given the investigative powers and punitive measures already available to the Commissioner of Competition. During the period when indirect purchaser claims predominated, it was also questionable whether class actions provided meaningful compensation, as distributions on behalf of such purchasers were often made on a cy-pres basis, given that it was not feasible to identify the proper individual claimants. The adoption of the Illinois Brick/Hanover Shoe approach in Canada may now result in significant awards to direct purchasers, as well as more manageable proceedings with fewer parties sharing more common issues. The recent history of class actions in Canada also demonstrates the complexity of collective redress that extends across borders (either within a federation or internationally). As the topics discussed above demonstrate, virtually every step in a class action can raise jurisdictional and fairness issues with a cross-border component. Variations on these multi-jurisdictional concerns are likely to arise in the EU as well.

234  Brian A Facey and David Rosner APPENDIX A – SECTION 36 (1) Any person who has suffered loss or damage as a result of (a) conduct that is contrary to any provision of Part VI, or (b) the failure of any person to comply with an order of the Tribunal or another court under this Act, may, in any court of competent jurisdiction, sue for and recover from the person who engaged in the conduct or failed to comply with the order an amount equal to the loss or damage proved to have been suffered by him, together with any additional amount that the court may allow not exceeding the full cost to him of any investigation in connection with the matter and of proceedings under this section. (2) In any action under subsection (1) against a person, the record of proceedings in any court in which that person was convicted of an offence under Part VI or convicted of or punished for failure to comply with an order of the Tribunal or another court under this Act is, in the absence of any evidence to the contrary, proof that the person against whom the action is brought engaged in conduct that was contrary to a provision of Part VI or failed to comply with an order of the Tribunal or another court under this Act, as the case may be, and any evidence given in those proceedings as to the effect of those acts or omissions on the person bringing the action is evidence thereof in the action. (3) For the purposes of any action under subsection (1), the Federal Court is a court of competent jurisdiction. (4) No action may be brought under subsection (1), (a) in the case of an action based on conduct that is contrary to any provision of Part VI, after two years from (i) a day on which the conduct was engaged in, or (ii) the day on which any criminal proceedings relating thereto were finally disposed of, whichever is the later; and (b) in the case of an action based on the failure of any person to comply with an order of the Tribunal or another court, after two years from (i) a day on which the order of the Tribunal or court was contravened, or (ii) the day on which any criminal proceedings relating thereto were finally disposed of, whichever is the later.

Mario Siragusa *

Options for Collective Redress in the EU

1. Introduction: Effective private enforcement in the EU: the way so far In recent years, one of the main concerns of the European Commission has been to bolster the development of actions for damages caused by antitrust infringements within European national legal systems. Debates and initiatives in this connection are clearly animated by the objective of granting better and more efficient enforcement of competition law, while allowing customers to recover losses suffered as a consequence of antitrust infringements. In particular, the Commission indicates that the absence of an effective legal framework for antitrust damages actions: (i) deprives victims of a considerable amount of compensation; and (ii) hampers full enforcement of the antitrust rules, with a negative impact on vigorous competition in an open internal market. A number of initiatives have been undertaken to identify the obstacles to a more efficient system of damages claims in the EU and to propose different options to overcome such obstacles. Let me just recall the 2005 Green Paper on Damages Actions for Breach of the EC antitrust rules, followed by the 2008 White Paper and, in March 2009, by a draft proposal for a Council Directive on rules governing actions for damages for infringements of Articles 81 and 82 of the Treaty (the ‘draft Directive’ – see Annex III to this volume), as well as by the publication of a comprehensive study on the methods of damages calculation, expressly aimed at establishing a guidance tool for national courts in assessing antitrust damages (Oxera, ‘Quantifying antitrust damages’, December 2009). This effort has continued, though in a more nuanced way, with the Barroso II Commission. Yet another public consultation was concluded at the end of April 2011.1 This is likely to be the first step of the new approach that Commissioner Almunia has taken with respect to collective redress initiatives since he began his work at DG Competition. In short, the objective of this new path is to present collective redress as a general means of supporting private enforcement of EU law in general – not only in the field of competition law – and to draw on a deeper involvement of the European Parliament. * Cleary, Steen, Gottlieb and Hamilton, Rome. 1 On 2 February 2011, the Commission launched the public consultation Towards a Coherent European Approach to Collective Redress. Stakeholders’ replies to the consultation are available at: http://ec.europa.eu/competition/consultations/2011_collective_ redress/index_en.html.

236  Mario Siragusa

2. The proposed options for collective redress In this general framework, a number of options for collective redress have been discussed. Their rationale is closely linked to the observed circumstance that victims rarely bring actions individually when they have suffered lowvalue damages and that, therefore, in cases where the damages caused by the infringement are fragmented and diffuse, almost no compensation is granted to the victims. This tendency can lead to the absurd result that the more widespread the damage is, and the greater number of victims involved, the fewer chances there are to see any action for damages filed, with the consequence that the infringer may retain a net illegitimate profit from the violation even after the imposition of a fine by a competition authority. This result is basically due to the disadvantageous risks/benefit ratio of such an action for damages for a single complainant, the difficulties in identifying and demonstrating the violation, the damages suffered and the causal link, and the costs of civil litigation, which may well be higher than the compensation itself. To allow these victims to receive appropriate and fair compensation, it is therefore crucial to establish a collective redress mechanism in order to improve access to justice and efficiency of civil litigation. This need has been clearly expressed by the Commission, which has proposed in the White Paper, and set out in the draft Directive of 2009, two different procedural instruments: (i) opt-in collective actions (group actions), and (ii) representative actions: (i) Group actions combine in one single procedure the claims of all victims of an infringement who have expressed their intention to file an action. This type of action has the advantage of making litigation more attractive for low-damage victims by improving the risk/benefit ratio and allowing the claimants to share evidence in the same proceeding. This kind of collective action is closer to the legal traditions of the EU Member States than is the opt-out class action familiar from U.S. practice.2 (ii) Representative actions, by contrast, allow a ‘qualified entity’, representing interests of a category or of a defined group of subjects, to bring an action for damages on behalf of the individuals or companies represented, who are not, in this case, parties participating formally in the proceeding. The interests represented must be sufficiently precise so as to clearly identify the 2 With respect to opt-in collective actions, Article 5 of the draft Directive provided as follows ‘1. Member States shall ensure that two or more injured parties who suffered harm caused by the same infringement of Article 81 or 82 of the Treaty can jointly bring a group action for damages. // 2. National courts shall treat a group action as one single action. However, the defendant shall not be prohibited from raising any defence that might be relevant with regard only to one or several of the claimants. // 3. Member States shall ensure that the court, having heard the parties, can allow other injured parties who suffered harm caused by the same infringement to join a group action after the court has been seized of such action, provided that this does not impair the sound administration of justice.’

Options for Collective Redress in the EU  237   subjects represented and allegedly harmed. The draft Directive of 2009 and the White Paper of 2008 suggested the establishment of two different kinds of qualified entities: (1) those representing legitimate and defined interests and having specific requirements set by law, which are defined in advance by each Member State; and (2) those which are certified on an ad hoc basis by each Member State in relation to a particular antitrust infringement. The representative action model proposed in the draft Directive provides that any injured party belonging to the group of subjects allegedly harmed should be entitled to opt out of the action brought by the representative entity.3 The two mechanisms are considered to be complementary. Opt-in mechanisms are closer to European legal culture but they are also more complicated compared to opt-out mechanisms because they require the identification of the claimants and proof of the damage suffered by each plaintiff. Opt-out actions, by granting the advantage of a wider representation of the victims, are more efficient and offer greater deterrence and corrective-justice effects; but they are also more likely to lead to excesses. A two-track approach was developed in the draft Directive prepared by the Barroso I Commission shortly before the end of its mandate. The draft Directive embraced both types of actions, with a view to enlarging the range of instruments at the victim’s disposal, thus improving the chances that victims would secure compensation. At the same time, the twofold solution helps to maintain a balance between the need to innovate and the need to preserve the discretion of Member States in transposing the Directive, while at the same time compelling them to gradually introduce mechanisms such as opt-out actions, which are currently allowed in only a minority of Member States. Fierce objections by stakeholders arose against certain aspects of the draft Directive (including in particular the rebuttable presumption in favour of indirect purchasers/final consumers),4 which in any event failed to address some major 3 With respect to representative actions, Article 6 of the draft Directive provided: ‘1. Member States shall ensure that qualified entities within the meaning of Article 7 can bring a representative action for damages on behalf of injured parties who suffered harm caused by the same infringement of Article 81 or 82 of the Treaty. // 2. The qualified entity shall define the group of injured parties on behalf of which it brings the representative action. The definition of the group shall be sufficiently precise so as to allow the court to conduct the proceedings on the action in a manner consistent with the rights of all parties and in particular with the rights of defence. To this end, the qualified entity shall define the injured parties that belong to the group by describing the infringement on which the claims of the injured parties are based and the harm it has caused to the injured parties. The qualified entity shall not be required to individually identify the injured parties that belong to the group. Member States shall ensure that the court seised of the action ascertains whether the definition of the group meets the requirements laid down in this Paragraph. If these requirements are not met, the court shall have the power to make the appropriate rulings, in particular to dismiss the action, after having heard the parties. // 3. […]. // 4. Member States shall ensure that any injured party belonging to the group can exercise its right not to be represented by the qualified entity in the representative action. Any decision by the court on the merits of the case shall be binding on all injured parties represented by the qualified entity at the time when such decision is adopted. […]’. 4 The passing-on defence was one of the hotly debated issues. Article 11 of the draft Directive provided: ‘Member States shall ensure that the defendant in an action for damages can invoke as

238  Mario Siragusa practical points related to collective redress, such as methods for allocating and calculating damages. Moreover, the draft Directive was criticized for unduly interfering with domestic procedural systems, and for its potential inconsistency with other EU initiatives in the field of collective redress, such as those in the area of consumer law and environmental law.5 This was the background to Commissioner Almunia’s new stance with respect to collective redress. As we will see, the latest proposals as of this writing aim at presenting a general and coherent framework in which private enforcement of antitrust law will be debated and defined in the future.

3. The collective redress consultation: Main issues and the next steps The Commission’s proposed directive in the field of competition enforcement has been withdrawn, and a new consultation on collective redress has been launched. The scope of the new consultation started earlier this year (ie, 2011) is by far wider than the one discussed in the 2008 White Paper. Indeed, following stakeholders’ worries about possible ‘inconsistencies between the different Commission initiatives on collective redress’ (Reding, Almunia and Dalli (2010), Renforcer la cohérence de l’approche en matière de recours collectif: prochaines étapes, § 11), the new consultation encompasses private enforcement of ‘any EU legislation creating substantive rights’ (Commission (2011), Towards a Coherent European Approach to Collective Redress, § 7). At the same time, the scope of the intervention seems to be more focused, given its main thrust on mechanisms and procedures to grant effective access to justice, and it seeks to identify ‘common legal principles on collective redress’ (Ibid, § a defense against a claim for damages the fact that the claimant passed on the whole or part of the overcharge imposed upon him. The burden of proving that the overcharge was passed on shall rest with the defendant. // 2. Member States shall ensure that, without prejudice to claims for loss of profit, the claimant in an action for damages is compensated for the harm resulting from passing on of the whole overcharge where he has shown that: (a) the defendant has infringed Article 81 or 82 of the Treaty; (b) the infringement resulted in an overcharge for the customer of the defendant; and (c) the claimant purchased the goods or services that were the subject of the infringement, or purchased goods or services derived from or containing the goods or services that were the subject of the infringement. This provision is without prejudice to the defendant’s right to show that the overcharge was not passed on to the claimant.’. According to the majority of NCAs and lawyers’ associations, these presumptions would be hard to rebut and would systematically lead to multiple liability for defendants (see, eg, the specific criticism reported in Luigi Prosperetti, Il danno antitrust, Bologna, 2009, p. 201). 5 In this regard, the Parliament’s Resolution of March 2009 is significant, insofar as it stressed the need to involve the Parliament via the co-decision procedure in any legislative initiative concerning collective redress. The Commission, however, chose to ignore the Parliament’s wishes. It appears that, in light of its contentious nature, Barroso may have intervened personally to withdraw the directive.

Options for Collective Redress in the EU  239   12). Furthermore, the subsidiarity principle requires that ‘any action at EU level should address the specific cross-border dimension of collective redress’, thereby limiting the extent of interference in national procedures (Ibid, § 14). Consequently, it appears that collective redress mechanisms are to be proposed and analyzed on the basis of the private enforcement of EU law in general. This implies that effective mechanisms for indirect purchasers and final consumers will have to be adopted in order to avoid any possible unjust enrichment of the infringers, which would ultimately result in under-compensation and weak deterrence. The new initiative on collective redress also underlines the need to involve all ‘EU citizens and businesses’ (Ibid, § 7) and not only final consumers. Collective redress mechanisms therefore clearly represent a means to balance and counteract the procedural strength of defendants in antitrust claims, thereby guaranteeing a fair and effective process, since, as the Commission claims, ‘collective redress is a matter of fundamental rights’ (Ibid, § 16).6 The stakeholders’ replies to the public consultation have shown almost unanimous support for the Commission’s initiative. The replies portray collective redress mechanisms as an essential tool to guarantee more effective enforcement of individuals’ rights, in particular where the monetary value of their harm would not otherwise justify bringing an action before national courts. However, many respondents have also pointed out that several Members States have already enacted national class action laws, and that superimposing an EU model of class action would fail to respect the principle of subsidiarity. Accordingly, many have suggested that the EU’s intervention should be non-binding. The observations received by the Commission also showed substantial agreement regarding the complementary – but, at the same time, independent – role that collective redress systems should have with respect to public enforcement. As to the necessary independence of collective actions, many respondents observed that, if collective actions were to be permitted only as follow-on actions after the intervention of public enforcement authorities, this would amount to a limitation of the right to access to justice pursuant to the constitutional law of a majority of Member States and pursuant to Article 6 of the European Convention on Human Rights. In any event, many stakeholders also recommended that various forms of coordination between private and public enforcement should be implemented. In this regard, some suggested that, when setting the level of fines, public authorities should consider the amount of damages possibly already paid pursuant to civil proceedings. Others pointed out that, where public authorities find an infringement, a new limitation period to bring damages actions should start, and that the findings of public authorities should be recognized as significant evidence before civil courts. 6 The Charter of Fundamental Rights of the EU establishes the right to an effective remedy for everyone whose rights and freedoms are violated (Article 47(1)). See also Article 19(1) TEU, which incorporates the principle of effective judicial protection.

240  Mario Siragusa As to the content of a possible EU class action model: on the one hand, trade associations showed particular concern about the risk of abusive litigation; on the other hand, consumer organizations stressed that consumers should be guaranteed the widest possible participation in class actions. Accordingly, in order to avoid abusive litigation, it has been recommended (mainly by trade associations) that the EU model of class action should: (i) (ii)

be based on an opt-in mechanism; have a certification phase, to filter out cases where the action is manifestly unfounded; (iii) have a strict compensatory approach (no treble damages); (iv) award damages directly to victims and not to their representative organizations; (v) be based on the ‘loser-pays’ principle; (vi) give publicity to the class action only after the court has certified it, in order to ensure that inadmissible class actions do not have negative effects on the defendant’s reputation; (vii) avoid plaintiffs’ financing mechanisms; and (viii) be brought only by publicly funded bodies, as otherwise abusive financial incentives, possibly to seek extortionate settlements, might prevail over the intention to compensate victims. On the other hand, consumer organizations expressed their preference for an optout model of class action. Moreover, some consumer organizations considered that, in order to avoid the risk of abusive litigation, it would be sufficient to limit the standing to sue to qualified and reliable organizations, whose reputation in their view represents a sufficiently reliable guarantee. As to the financial support of class actions, consumer organizations recommended mitigating the loser-pays principle and creating a public fund, which could be financed through the fines levied by the public enforcement authorities as well as by mandatory contributions of business companies. With respect to the scope of application of the EU collective redress model, many respondents pointed out that it should be limited to competition law and consumer law infringements. In this regard, it was observed that a sector-specific approach would be preferable, also in light of the specific need to coordinate private and public antitrust enforcement, which has already been deeply scrutinized by the Commission in the preparation of the draft Directive. The latter suggestion may be compared with the program illustrated by Commissioner Almunia, according to which the next step will be the ‘definition of a general legal framework to collective redress’ and, subsequently, the use of this framework ‘to launch specific legislative initiatives in the different policy domains’.7 7 Joaquín Almunia ‘Common standards for group claims across the EU’, speech, 15 October 2010, Valladolid.

Options for Collective Redress in the EU  241  

4. Italian achievements in the field of collective redress 4.1. Legislative developments On 1 January 2010, Article 140-bis of the Italian Consumer Code made it possible for consumers to bring class actions in Italy against certain breaches of contract or torts that occurred after 15 August 2009. In particular, under the Italian Consumer Code, class actions may be brought against business enterprises by any consumer or user – either on his own, or through associations acting on his behalf or committees to which he belongs – seeking damages or declaratory relief for a violation of rights which is ‘identical’ to those of other consumers or users and which arise from certain actionable breaches of contract or torts, including, inter alia, antitrust infringements and unfair commercial practices. Since Article 3(a) of the Italian Consumer Code defines ‘consumers and users’ as individuals who act for purposes falling outside their trade, business or profession, the rules on class actions do not apply to claims brought by (or on behalf of) parties who are not individuals or who are acting within the scope of their trade, business or profession, including their employment contract. The Italian consumer class action procedure is based on the opt-in system and contemplates two stages. First, following an initial hearing, the court decides on the admissibility of the action, which, for this purpose, has to satisfy the following requirements: (i) the action must not be manifestly unfounded; (ii) there must be no conflict of interest between class members; (iii) the rights claimed by the class members must appear to be identical; and (iv) the first claimant must seem able to protect the interests of the class in an adequate manner. At this stage, the court may suspend the proceedings if the facts on which the class action is based also form the object of an investigation of an independent enforcement authority, or of review proceedings pending before an administrative court. If the court deems the class action admissible, it will issue an order setting out: (i) rules for the notification of the proceedings to the other members of the class; (ii) the characterization of the rights that are at stake in the proceeding; (iii) the deadline for the exercise of other consumers’ or users’ right to opt in; and (iv) rules governing the ensuing investigatory phase. Prior to the deadline set by the court, consumers or users may opt in to the class action by filing a written declaration accompanied by evidence supporting their claim, without any lawyer assistance being required. Class members who opt in waive their right to initiate any other legal proceedings based on the right(s) that the class action seeks to protect. However, consumers or users who have opted in to the class action can initiate separate legal proceedings if:

242  Mario Siragusa (i) the claimant who filed the class action settles his claim (on an individual basis or jointly with other class members); or (ii) the class action is otherwise abandoned before the tribunal renders its final judgment. Moreover, once the class action is filed and the opt-in period has expired, no other class actions can be brought against the same defendant in connection with the same matter. Before the expiration of the opt-in period, any other class action brought against the same defendant in connection with the same matter is either consolidated with the first class action (if the actions are pending before the same tribunal), or cancelled from the register of pending cases (if the actions are pending before other tribunals, in which case all such class actions may be resumed before the tribunal adjudicating the initial class action within 60 days). Following the investigatory phase of the procedure, if the court issues a final ruling in favour of the plaintiffs, it may either award a fair estimate of damages to each of the individual consumers or users who have elected to opt in to the class action, or establish a criterion to quantify damages. Any judgment rendered in such a proceeding is binding upon the defendant, the plaintiff, and all other individuals who opted in to the action. On the other hand, consumers and users who did not opt in are free to bring their own individual actions against the business enterprise.

4.2. Italian case law Given the still recent introduction of the consumer class action in Italy, the case law that has developed so far only relates to the first phase of the procedure, where courts decide on the admissibility of the action and verify whether the requirements mentioned in section 4.1 above are satisfied. As of this writing, only one consumer class action has been allowed to go forward by a court in Italy. The majority of class actions filed have been dismissed. For example, on 27 May 2010, the Turin Tribunal declined to hear the merits of such a claim, finding that the consumer on behalf of whom the class action was brought did not have standing to sue the defendant bank, as the bank had never applied against him the allegedly unlawful contractual clause included in their agreement. Thereby, the court clarified that only consumers who have themselves suffered damages may initiate a class action; mere alleged illegality is not enough. On 30 March 2011, the Rome Tribunal dismissed two class actions brought against two banks, where a consumer claimed that he had incurred illegitimate costs arising from a contractual clause that should have been declared null and void. The court found that the consumer class action law was inapplicable, ratione temporis, because the agreements containing the contested clause had been concluded prior to 15 August 2009.

Options for Collective Redress in the EU  243   By an order of 11 April 2011, the Rome Tribunal dismissed as manifestly unfounded a class action brought by three consumers against a cigarette manufacturer. The plaintiffs had claimed that the defendant was responsible for failing to take all reasonable measures to ensure that cigarette smokers would not be affected by the negative effects of tobacco smoke. Moreover, the court found that the rights claimed by the class members were not identical, since the personal situations of the three claimants were all different, and could not be treated in a single class action. The only case so far in which a consumer class action has been admitted in Italy was brought before the Milan Tribunal, which, by an order of 20 December 2010, accepted to hear an action initiated by a consumer against a company selling flu tests, who claimed that the company had engaged in misleading advertising of its products. Following an appeal brought by the defendant, the Milan Court of Appeal decided, on 3 May 2011, to uphold the order of the Milan Tribunal. In light of the few cases described above, it seems that the two-stage procedure provided by the Italian consumer class action law provides an effective mechanism to prevent abusive litigation. Nonetheless, courts have so far always compensated the costs of the dismissed actions, in consideration of the novelty and complexity of the issues raised by the Italian class action law, thereby obviating the risk that consumers and their associations might be discouraged from bringing further class actions in the near future. It is noteworthy that none of the class actions brought thus far in Italy appear to have been related to antitrust infringements.

5. Insights on specific issues One of the most common scenarios of antitrust claims for damages are follow-on actions, in which claimants can rely on a decision already issued by a competition authority for proving the infringement. This is either because the positive findings of an antitrust authority are legally binding on courts (as happens in Germany with respect to the decisions of the competition authorities of the Member States, pursuant to Section 33(4) of the Act against Restraints of Competition; and as happens in all Member States with respect to the EU Commission’s decisions) or simply because judges may consider conclusions already drawn by an antitrust authority to be accurate and persuasive. Stand-alone actions, by contrast, leave the burden of proof of the infringement completely on the claimant. However, competition authorities’ decisions generally do not address all factual and legal elements that claimants are required to prove in the context of followon actions for antitrust damages. One of the main reasons for this is that antitrust violations – and especially cartels – are usually seen as per se infringements and competition authorities do not need to investigate their effects on the market to establish a competition law violation.

244  Mario Siragusa This means that in both stand-alone and follow-on actions, plaintiffs will have to: (i) prove the individual damage and its causal link with the anticompetitive conduct (section 5.1); and (ii) support the judge in the quantification of the damage claimed (section 5.2). In this context, the issue of access to evidence may play a crucial role (section 5.3).

5.1 Burden of proving individual damages and their causal link with the impugned conduct Proving damages and the existence of a causal link with the allegedly anticompetitive conduct can be extremely complex. This is clearly acknowledged in the Staff Working Paper accompanying the Commission’s White Paper on antitrust damages actions, where it is observed that, even in follow-on suits, ‘having to demonstrate in detail the causation and quantification of [...] damages remains a particular difficulty’.8 In Italy, this issue was also addressed by the Italian Supreme Court in SAI v. Nigriello,9 a cartel infringement case where a policyholder sued his insurance company to obtain a refund of part of an insurance premium he had paid. According to the claimant, the premium had been increased as a result of an upstream cartel in which the defendant had participated, as previously ascertained by the Italian Competition Authority. With a view to alleviating the burden of proof of possible plaintiffs in the context of follow-on actions for cartel damages, the Supreme Court held that both the individual harm suffered by claimants and the causal link with the infringement can be presumed if it is proved that: (i) a cartel did in fact exist, which can be demonstrated on the basis of an infringement decision already issued by a competition authority; and (ii) the claimants purchased cartelized goods or services: to this end, it is sufficient to provide copy of contracts entered into with the undertakings concerned or other relevant documents (such as invoices). According to the Supreme Court, it is then up to the defendant to rebut this presumption by showing that claimants did not suffer any real damage. To this end, the defendant can employ economic analysis to show that the cartel did not have any material effect on the market and that, in the absence of the infringement, prices and other purchase conditions would have been essentially the same. Indeed, economic research confirms that some cartels can be unsuccessful and cause no anticompetitive effects.10 In this respect, the following analysis may be relevant: 8 Staff working paper accompanying the EU Commission’s White Paper on antirust damage actions, http://eur-lex.europIa.eu/LexUriServ/LexUriServ.do?uri=SEC:2008:0404:FIN:EN:PDF, para 89. 9 Italian Supreme Court, 2 February 2007. 10 John M. Connor, Price-Fixing Overcharges: Legal and Economic Evidence, selected paper for the annual meeting of the American Agricultural Economics Association, Providence, RI, 2005.

Options for Collective Redress in the EU  245   – Price trend comparison. If price trends before and during the cartel period are similar, it can reasonably be assumed that the violation did not impact on competition. Along the same line, when there are no substantial price trend differences between the geographic market where the cartel was in place and areas where there was no infringement, the collusion was probably ineffective. – Price dispersion. If there is not any significant variation in prices after the end of the cartel, this can be seen as evidence that the cartel was not limiting competition. – Monitoring. When cartelists cannot monitor the ‘correct’ implementation of the collusive arrangements, the incentive to cheat and deviate from the agreed conduct is stronger. – Number of undertakings and products involved. Economic theory shows that it is extremely difficult to run a cartel when the parties or the products concerned are too numerous. – Competitive pressure by other players. Parties to the cartel may have been unable to implement the cartel due to the competition constraints imposed by players that did not participate in the violation. In this respect, it is also important to determine whether customers can easily switch suppliers. – Customer buyer power. It can be difficult for undertakings to implement a cartel when customers enjoy strong countervailing buyer power.

5.1.1 The passing-on defence Another possible economic argument that defendants can put forward in order to resist claims for damages is the passing-on defence (‘POD’). The POD stems from a simple observation of market dynamics: in multi-tier distribution chains, purchasers that have to pay supracompetitive prices resulting from an antitrust infringement may be able to pass overcharges downstream to their own customers. These ‘indirect purchasers’ can be producers deploying the cartelized component in their manufacturing process, or distributors, or end consumers. The phenomenon arises typically in a price-fixing setting, which represents the reference point of the debate, even if it can also be discussed in the case of an abuse of dominant position and, to a lesser extent, in the field of vertical restraints (where the producer’s interest is often aligned with that of the consumer, in the sense that both typically aim to compress the distributor’s margin). Connor’s study examines the effects of 674 cases of hard core cartels. He concludes that effects of cartels on prices vary widely; the cartel overcharges in the sample range from -10% (that is, a decrease in price of 10%) to +322%, with the median overcharge of 25%. Significantly, however, 6% of all cartels in the sample were ‘unsuccessful.’ See also Stefan Fölster and Sam Peltzman, The Social Cost of Regulation and Lack of Competition in Sweden: A Summary in The Welfare State in Transition: Reforming the Swedish Model, University of Chicago Press, 1997, pages 315–352; Michael F. Sproul, ‘Antitrust and Prices’, 101 Journal of Political Economy 742 (1993).

246  Mario Siragusa Therefore, in this context the question arises as to whether the passing-on of the anticompetitive charges has any relevance within the context of adjudicating antitrust claims and, if so, how the burden of proving the passing-on of the anticompetitive overcharge is to be allocated. Unfortunately, there is no best solution to these problems. The approaches proposed in legal scholarship and followed in national litigation systems tend to be different, ranging between two extreme positions, each inspired by a particular enforcement philosophy. On the one hand, we find the so-called ‘zero option’, where both the POD and the standing of indirect purchasers are recognized. This approach is strictly compensatory in nature, insofar as it remedies the potentially multiple liability of the defendants and ensures that direct purchasers do not obtain compensation greater than the harm actually incurred. This approach is common to most EU Member States. On the other hand, we find the opposite approach, which denies both the POD and indirect standing. This scenario relies heavily on direct purchasers as the leading force in antitrust private enforcement, given their closer ties with the infringers and their presumably superior knowledge of the dynamics of the relevant market. At the same time, denying the POD strongly incentivizes direct customers to claim damages, and it may lead to substantial over-deterrence and excess of litigation (in this regard, the literature refers to direct purchasers as the ‘most efficient enforcers’ or ‘better detector’ (CEPS (2007), at § 470). As for the Italian system, we note that the POD is not recognized as such by Italian legislation or case law. Therefore, the issue must be considered in accordance with general civil liability principles, which provide that a claimant may only seek compensation for damages it actually suffered, provided that it did not contribute to causing them. If we look at the few antitrust cases focusing on these issues, we find that Italian courts have indeed applied general civil law principles, concluding that claimants have no standing with respect to damages they have passed on to their customers. This reasoning appears to have been endorsed in a number of Italian cases. For example, in Indaba v. Juventus,11 a travel agency entered into a contract with Juventus Football Club, undertaking to sell the tickets of the 1997 Champions League final match in Munich only in a bundle with travel packages that included services not normally needed by football supporters. The venture proved to be unsuccessful and the travel agency sued Juventus for antitrust damages. The court found that the agreement unduly restricted competition and that Juventus had abused its dominant position in the relevant market for the sale of those tickets by imposing excessive prices and illegally tying the sale of the tickets to the sale of the travel package. However, no damages were awarded to the plaintiff. According to the court, the plaintiff had actually signed the restrictive agreement with an intention to ‘pass on the damage’ to its customers. It was therefore not entitled to any damages, because it had intentionally contributed to causing the harm. 11

Turin Court of Appeal, 6 July 2000.

Options for Collective Redress in the EU  247   In that circumstance, it seems that Italian judges have applied the rule provided for by Article 1227 of the Italian Civil Code, which attaches relevance to the contribution of the injured party to his own damages. In light of Article 1227, an injured party has to behave with due diligence in order to mitigate the harmful effects of the unlawful conduct, and can only claim damages to the extent that they are not a consequence of his own misconduct. In a second case, Unimare v Geasar,12 the judge, while concluding that there was no violation of antitrust law, incidentally added that, in any event, the claimant would not have been entitled to recover any damages because it had passed on its additional costs to its customers. Indeed, Unimare, a former provider of handling services at the Olbia airport (Sardinia), claimed that Geasar, the current management body of Olbia Airport, had abused its dominant position by (i) excessively increasing its fees, without any justification, and (ii) stealing Unimare’s main client, the US Naval Service Order (‘NSO’), by presenting itself as the only entity qualified to supply groundhandling services at the airport. The court maintained that there was no abuse, but went on to state that, in any event, it would not be possible to claim damages caused by an abuse of dominance when the damages have been completely passed on to the claimant’s customers. In the case at stake, the additional fees imposed by Geasar had been passed on from Unimare to the NSO, as explicitly stated in the contract between the parties, which provided for the reimbursement of all the expenses incurred by Unimare. More specifically, the court found that: (i) the NSO had actually reimbursed any fees paid by Unimare, pursuant to a specific contractual clause; and (ii) the tariff increase had not caused the NSO to switch to another supplier because the new fees equally applied to all operators. Therefore, since Unimare had entirely passed on the tariff increase to its client, it could claim no damages. More generally, Italian Courts have often dealt with the issue of passing on in the context of tax/subsidy refund cases, because under Italian law the public administration must not refund illegally levied taxes if it is able to prove that the claimant has passed on the charge to its customers.13 In these cases, the courts have maintained that: (i) the refund to the claimant of an illegally levied tax, a charge for which the claimant has already passed on to its customers, does not necessarily imply unjust enrichment; (ii) the claimant may obtain damages for any harm caused by the undue or discriminatory application of the tax (eg, the shortage of goods imported from other countries); and (iii) the claimant may also obtain damages for any reduction of its sales caused by the passing-on (ie, by Cagliari Court of Appeal, 23 January 1999. According to the Supreme Court in a tax refund case, ‘in our legal system, and particularly in the tax sector, there is no general principle which would rule out the recipient’s duty to return undue payments when the payer has already obtained elsewhere the restitution of the unduly paid sum’ (judgment of 24 May 2005, No. 10939). By the same token, absent special circumstances, any indemnifications collected owing to an own insurance policy do not reduce the damages that the injured party is entitled to claim from the defendant (see, eg, Supreme Court, 12 May 2003, No. 7269, and 15 April 1993, No. 4475). 12 13

248  Mario Siragusa the increase of prices aimed at compensating the increase in costs caused by the illegally applied tax).14 The standing of indirect purchasers is generally recognized by Italian courts. As an example, in Indaba v. Juventus (cited above), the Turin Court of Appeal maintained that the actual victims of Juventus’ abuse were its football supporters (who had been forced to spend more than they otherwise would have done), not the travel agency. The court thus stated in an obiter that, because of the passingon, those indirect customers ‘would be the ones entitled to claim damages for the overcharges they did not want’.

5.1.2 Negative declaratory actions The POD and, more generally, the absence of a causal link can also be claimed by infringers in the context of negative declaratory actions for damages. The peculiarity of this type of action has to do with the fact that the claim is brought by the offender against the possible victims, rather than vice versa. It is the infringer who files a claim with the court and requests the court to confirm that his anticompetitive conduct (for which he has already been sanctioned by a competition authority) did not cause any individual harm. Through this form of action, the infringer has the possibility to: – anticipate possible suits for damages; and – secure the jurisdiction of a given national civil court pursuant to Articles 27 and 28 of Regulation 44/2001 (which provide that, where proceedings involving the same parties and cause of action (Article 27) or related actions (Article 28) are pending before judges of different Member States, the court other than the first seized must/may stay its proceeding and/or decline jurisdiction). These negative declaratory actions were regarded as legitimate by Advocate General Tesauro in the leading case, Tatry, where he stated that the ‘bringing of proceedings to obtain a negative finding, which is generally allowed under the various national procedural laws and is entirely legitimate in every respect, is an appropriate way of dealing with genuine needs on the part of the person who brings them’.15 However, negative declaratory actions in the antitrust field have been extremely rare. To my knowledge, in the EU there have been only two cases like this. The first case originated when, following the infringement decision adopted by the Commission in the organic peroxide case (COMP/E-2/37.857), one of the 14 See, for all instances, Supreme Court, 24 May 2005, No. 10939, cited previous footnote. With respect to a previous law, pursuant to which claimants could obtain a refund only if they proved that the illegally levied taxes had not been passed on to their customers, the Constitutional Court stated that ‘the inversion of the burden of proof’, which required from the claimant ‘the (negative) proof of the absence of passing on’ was so manifestly unreasonable that it violated the Italian Constitution (Constitutional Court, 9 July 2002, No. 332). 15 Case C-406/92, Tatry [1994] ECR I-5439, para 23.

Options for Collective Redress in the EU  249   parties to the contested cartel (Akzo Nobel) decided to anticipate the damages actions of two customers (BASF and Basell) and thus brought an action before a Dutch court ‘asking the [court] to rule that Akzo Nobel was not liable towards BASF or Basell or their respective group companies’.16 Very limited information is available on this proceeding. The action was probably withdrawn shortly after it was filed. The second case is one where I have been personally involved. In November 2006, the Commission adopted an infringement decision concerning an alleged cartel on the markets for Butadiene Rubber (BR) and Emulsion Styrene Butadiene Rubber (ESBR) (synthetic rubbers primarily used in tire production) involving six undertakings, namely Bayer, Dow, Eni, Shell, Kaucuk and Stomil. In July 2007, Eni brought a negative declaratory action before the Milan Tribunal against five large purchasers of BR and ESBR products,17 maintaining, inter alia, that the alleged cartel did not have any price effect on the relevant markets. Therefore, the misconduct could not have caused any harm to these companies. The Milan Tribunal retained jurisdiction but rejected the claim brought by Eni,18 stating that the request of the plaintiff was: – void for reasons of vagueness, since it would have been necessary ‘to clearly specify the behavior – hypothetically lawful or at least not harmful – attributable to the plaintiff, which, depending on the manner in which the plaintiff behaved, would exclude any potential compensation claim’; in fact, the plaintiff failed to present ‘facts or conduct from which such absence could be implied, that are not necessarily common to each of the defendants’ and to consider ‘each specific action causing damage to each individual customer’; and – taking account of Article 16 of Regulation 1/2003 (which provides that ‘[w] hen national courts rule on agreements, decisions or practices under Article 81 or Article 82 of the Treaty which are already the subject of a Commission decision, they cannot take decisions running counter to the decision adopted by the Commission’), the Milan Tribunal held that granting the claimant’s request would have resulted in a judgment running counter to the decision already adopted by the Commission. As regards this second point, however, it is to be noted that in its decision, the Commission had expressly stated that the actual impact of the alleged infringement could not be measured, and that the Commission therefore did not give it consideration, either for establishing the contested violation or for calculating the fine to be imposed on the undertakings concerned. In other words, the Commission’s decision in the BR/ESBR case did not provide for any assessment of the actual impact of the contested conduct on the market. Arguably, 16 See Akzo Nobel’s press release, http://www.akzonobel.com/news/pressreleases/2006/ akzo_ nobel_rejects_damages_claim.aspx. 17 Pirelli, Michelin, Continental, Goodyear, Bridgestone and Cooper Tyre & Rubber. 18 Milan Tribunal, 5 May 2009.

250  Mario Siragusa the Milan Tribunal erroneously interpreted the claim regarding the cartel’s effects as a request to find that no cartel had been formed. This specific matter, along with other issues that led to the rejection of Eni’s claim, is currently being contested before the Milan Court of Appeal.

5.2. Quantification of damages Another important issue where economics plays a leading role is the quantification of the harm effectively suffered by the victims. Where feasible, courts are required, with the assistance of the claimant and (possibly) economic experts, to quantify the damages to be awarded as precisely as possible. Different economic methods may be used for this purpose, including in particular the following: – The econometric method. This methodology implies the development of models simulating normal market conditions. This is achieved by plugging estimates of demand elasticity and other variables such as marginal costs, prices and quantity, into an economic model of how firms compete with one another and how they respond to their rivals’ competitive decisions. – The cost-based method. This methodology involves obtaining information on the average unit cost of production of products affected by the cartel and estimating a competitive price by adding to this cost a profit margin considered to be appropriate under competitive conditions. – The yardstick method. This approach involves the comparison of prices in the market where collusion is alleged to have occurred with a similar market where prices are unaffected by collusion. This could be either a comparison of identical product markets in other geographic areas, different product markets in the same geographic areas, or different product markets in different geographic areas. The benchmark market would ideally have competitive characteristics similar to the allegedly collusive market (ie, similar cost structures and demand characteristics, thus allowing differences in prices between the two markets to be attributed largely to the effects of the cartel as opposed to other market conditions). – The before-and-after method. This methodology involves a comparison of prices during the period of the alleged cartel with prices in the period before and/or after, on the assumption that the latter provides a reasonable approximation of price levels in the absence of the cartel. This latter method (‘before and after’) is probably the one most commonly used by Italian civil courts. In this regard it is useful to cite three examples of Italian cases where it has been applied.

Options for Collective Redress in the EU  251   The first is Viaggi del Ventaglio,19 where the plaintiff (a travel agency) claimed to have suffered damages as a result of a boycott by several upstream tour operators which collectively refused to supply it with travel packages from April to June 2001. In quantifying the damages suffered by the claimant, the court: – first considered the claimant’s monthly average turnover during the period before the infringement (January OR March); then – multiplied this figure by the monthly percentage increase of sales that the plaintiff could have reasonably expected during the period in which the boycott took place (from March 2001 onwards, this kind of business usually shows a material increase); and finally – multiplied the resulting amount by the number of months that the boycott lasted, concluding that this was a reasonable estimate of the turnover that the plaintiff would have earned through the sales of travel packages of tour operators involved in the infringement if it had not been victim of the boycott. A similar approach was followed in Inaz Paghe,20 where the court awarded damages based on a loss of profits arising from contracts terminated by clients of a software provider as a result of a collective boycott organized by national and local employment consultant associations. In order to quantify the harm suffered by the plaintiff, the judge: – first identified the number of contracts that had been supposedly terminated because of the boycott, comparing (i) the number of contracts terminated by the plaintiff’s clients in the two-year periods before and after the boycott to (ii) the number of contracts terminated during the two-year boycott; and then – multiplied the resulting number by the average profit that would have reasonably been earned from lost clients, concluding that this was a good estimate of damages effectively suffered by the claimant. The third case is International Broker,21 where a company acting as a broker of bitumen brought an action for damages against certain bitumen producers that had participated in an upstream cartel aimed at fixing prices and partitioning the market. The court awarded the claimant both actual loss and loss of profits. When calculating the latter, in particular, the judge: – first assumed that, in the absence of the cartel, the plaintiff would have generated the same turnover as in the year prior to the commencement of the infringement; hence, he calculated the delta between the turnover generated by the claimant both during and prior to the cartel; and then – applied to the resulting turnover figure a profit margin of 40%, which was deemed the average profit percentage for the activity of bitumen brokerage.

19 20 21

Milan Court of Appeal, 30 April 2003. Milan Court of Appeal, 10 December 2004. Rome Court of Appeal, 31 March 2008.

252  Mario Siragusa

5.3. Access to evidence Access to evidence is without doubt the central and at the same time the most complex issue in the context of the EU’s efforts to achieve effective private enforcement of competition law. The information asymmetry between the claimant and the defendant in antitrust damage claims is the primary obstacle to an efficient system of indemnification. Indeed, especially in stand-alone cases against cartel participants, access to evidence is crucial to proving the existence of the antitrust infringement, which may in this case be much more difficult to prove compared to damages and/or the causal link. On the other hand, developing specific instruments that may allow appropriate access to evidence is not only extremely complicated but also a very delicate issue. In fact, granting competing companies or consumers access to documents, files and reports of a company allegedly responsible for an antitrust infringement is a brutal intrusion into the business and management of that company, and there have to be adequate guarantees that its confidential information will not be divulged. In this area there is always the risk of burdensome solutions.

5.3.1. The Commission’s Proposal of March 2009 In this complex context, the draft Directive (included in this volume as Annex III) tried to adopt a balanced approach, changing the position previously expressed by the Commission in the White Paper. The aim as expressed in the draft Directive is to ensure a minimum threshold of access to evidence through disclosure orders in every Member State, while at the same time avoiding excessive intrusions into the national jurisdictions and leaving Member States free to adopt broader measures to ensure access to evidence. Pursuant to Article 8 of the draft Directive, national judges may order the disclosure of evidence where the party requesting a disclosure order has: – presented reasonably available facts and evidence showing plausible grounds to suspect that he/she has suffered harm caused by an antitrust infringement; – shown that evidence in the control of the other party or a third party is relevant to substantiate his/her claim; – specified either pieces of this evidence, or sufficiently precise and narrow categories of the evidence on the basis of reasonably available facts; and – shown that he/she is unable, applying reasonable efforts, to produce the requested evidence. Given the different possible readings of words such as ‘relevant’ or ‘reasonable effort’, a margin of discretion obviously exists in the assessment of the aforementioned requirements. The draft Directive also provides that the disclosure order should not be issued if it seems disproportionate, in light of: (i) the likelihood that the antitrust

Options for Collective Redress in the EU  253   infringement was actually committed; (ii) the value of the claim for damages; (iii) the scope and cost of disclosure, especially for any third parties concerned; and (iv) whether the evidence to be disclosed contains confidential information, especially that of any third parties. Another obstacle to the use of disclosure orders arises from leniency programs. Pursuant to Article 9 of the draft Directive, leniency applications (and settlement submissions) could not be the object of a disclosure order. In this regard, it is noteworthy that the Commission has changed his mind with respect to the White Paper, where it had also considered the possibility of limiting the civil liability of successful immunity applicants. With respect to the need to balance the right of access to evidence with the protection of leniency programs, the Court of Justice in its judgment of 14 June 2011 established that ‘European Union law on cartels […] must be interpreted as not precluding a person who has been adversely affected by an infringement of European Union competition law and is seeking to obtain damages from being granted access to documents relating to a leniency procedure involving the perpetrator of that infringement’.22 However, since ‘leniency programs are useful tools if efforts to uncover and bring to an end infringements of competition rules are to be effective and serve, therefore, the objective of effective application of Articles 101 TFEU and 102 TFEU ’,23 and ‘[t]he effectiveness of those programs could […] be compromised if documents relating to a leniency procedure were disclosed to persons wishing to bring an action for damages’,24 the Court also stated that ‘[i]t is […] for the courts and tribunals of the Member States, on the basis of their national law, to determine the conditions under which such access must be permitted or refused by weighing the interests protected by European Union law’.25 Indeed, the risk that a more effective private enforcement could compromise the attractiveness of EU and national leniency programs – and, ultimately, the public enforcement of competition law – was also stressed by many respondents in the Commission’s consultation on collective redress.

5.3.2. Access to evidence in Italy In the Italian legal system, access to evidence in antitrust damages actions is regulated by ordinary rules of civil procedure. In particular, pursuant to Article 210 of the Civil Procedure Code (‘CPC’), following a party’s request the court may order the other party or a third party to disclose a document or any other evidence that it considers necessary in order to decide the case. The disclosure order may be issued subject to the same requirements that apply to search orders pursuant to Article 118 CPC, that is, it may be issued only when: (i) the requested evidence is essential to ascertaining Case C-360/09, Pfleiderer [2011] ECR I-5161, para 32. Ibid, para 25. 24 Ibid, para 26. 25 Ibid, para 32. 22 23

254  Mario Siragusa the disputed facts; and (ii) the disclosure would not cause serious damage to that other party or to third parties. Moreover, pursuant to Article 94 of the CPC’s implementing provisions, the party’s request for a disclosure order must: (i) precisely identify the requested evidence; and (ii) when necessary, prove that the requested evidence is held by the other party or by a third party. Article 2711 of the Civil Code provides that the court may order the disclosure of all books of account, in order to enable the records referring to the disputed facts to be extracted. Finally, pursuant to Article 213 CPC, courts, even on their own motion, may request the public authority to provide information and documents that are deemed necessary in the civil proceeding. To date, courts have followed quite a restrictive approach in the enforcement of the aforementioned provisions, requiring that requests for disclosure orders refer to specifically identified – or at least identifiable – documents, whose content is known to be – or at least appears to be – relevant to deciding the dispute. Moreover, courts generally consider that disclosure orders can only be issued where the requesting party has no other possibility to obtain the evidence (eg, in follow-on cases, through an access to the file of the competition authority). Finally, even where a disclosure order is issued, if the addressee of the order refuses to comply with it, no sanction can be imposed against him/her. However, pursuant to Article 116 CPC, the judge may draw adverse inferences from the party’s refusal to provide the requested evidence. Nonetheless, in (at least) one follow-on case the court’s information request to the Italian Competition Authority has provided the claimant with useful evidence, as the Italian Competition Authority disclosed to the court the minutes of a cartel hearing of the defendants’ representatives as well as the documents seized in a dawn raid at the defendants’ premises.26 However, the Italian Competition Authority would arguably refuse to provide courts with information and documents it has received through leniency applications filed by investigated parties. Indeed, pursuant to the May 2010 amendments to the Italian leniency program, third parties, including those that have been admitted to intervene in the investigation procedure, are barred from access to written (or the transcripts of oral) leniency statements and the supporting documentation in the Authority’s case file. Moreover, the other parties to the investigation may have access to the leniency statements only after the date of notification of the statement of objections, provided that they undertake not to make copies of the leniency statements, and to use the information contained therein only for the purposes of judicial or administrative proceedings for the application of the competition rules at issue in the Authority’s investigation. The disclosure tool provided for by the Italian civil procedure rules is therefore much more limited than the one proposed in the draft Directive and certainly cannot provide an effective solution to the access to evidence problem. *     *     * 26

Rome Court of Appeal, 31 March 2008.

Options for Collective Redress in the EU  255   In light of this brief overview of some of the distinctive features of private antitrust litigation, it is evident that, in order to effectively improve access to justice against antitrust infringements, any legislative initiative on collective redress mechanisms should not only set out a system to aggregate individual actions, but should also address a number of other issues, which would otherwise continue to impede the development of an effective system of private antitrust enforcement.

Silvia Pietrini*

The Future of Collective Damages Actions in Europe

How long do we have to wait for the European Commission to legislate in the field of collective action? Since 2005, we have seen a play whose acts take years to play out. The first act began in 2005. At that time, the European Commission published a Green Paper to revive the debate on collective action at European level in the context of a reflection on actions for damages in the area of free competition. A new consultation followed in 2008 on the White Paper.1 In parallel with the initiatives of the Directorate General for Competition Policy (DG Competition), the Directorate General for Health & Consumers published in 2008 a Green Paper on collective redress for consumers, followed by a working paper in 2009.2 The hard work of the various departments of the European Commission during these years might suggest the adoption of a text in a reasonable time horizon, because in 2009 DG Competition prepared a draft version of a directive with the intention to make robust private enforcement a reality.3 Unfortunately, this draft text was never published and was finally withdrawn in 2010 when the new college of European Commissioners was installed. The 4th of February 2011 marked the beginning of the second act of the play. At that time, DG Competition, the Directorate General for Health and Consumers (DG SANCO) and the Directorate General for Justice, Freedom and Security (which had until then been excluded from the relevant debates) launched a new public consultation to ensure a coherent approach to collective redress4 in the context of the construction of a European area of ​​freedom, security and justice. This document has shown, however, the delay of concrete actions. In fact, the Commission did not take into account the results of the many discussions held and limited itself to presenting a list of questions formulated in rather abstract and vague terms. The European Parliament on 4 February 2012 adopted a resolution in favour of EU legislation that would lay down common rules in this area,5 and its Committee on Economic and Monetary Affairs commissioned a study on * Maître de conférences, Université Lille 2, Droit et Santé, Centre Réné Demogue-CRDP. 1 See http://ec.europa.eu/competition/antitrust/actionsdamages/index.html. 2 See http://ec.europa.eu/consumers/redress_cons/collective_redress_en.htm. 3 See Mel Marquis, ‘Cartel Settlements and Commitment Decisions’, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2008: Antitrust Settlements under EC Competition Law, Hart Publishing, 2010, pages xxix et seq, at lxviii-lxx; and Mario Siragusa, ‘Options for Collective Redress in the EU’, this volume. The 2009 draft is included in this volume as Annex III. 4 Towards a coherent European approach to collective redress, SEC(2011) 173, 4.2.2011. 5 European Parliament resolution of 2 February 2012 on ‘Towards a Coherent European Approach to Collective Redress’ (2011/2089(INI).

258  Silvia Pietrini Collective Redress in Antitrust, published in June 2012.6 But as of early 2013, the Commission still had not come forward with any proposals. We have now entered the third act of the play. At the beginning of the year, Director General Alexander Italianer stated that ‘[i]n collective redress, the Commission continues to work on a follow-up to the Consultation on a coherent European approach to collective redress that took place in 2011’.7 According to Vice-President Joaquín Almunia, speaking at the ‘New Frontiers of Antitrust 2013’ conference held in Paris on 22 February 2013,8 the Commission is planning to propose a legislative initiative which would harmonize some of the rules governing antitrust damages actions and protect corporate statements submitted by leniency applicants from disclosure in private litigation. However, it seems that collective actions would not be included among the new provisions. Finally, on 4 March 2013, during the meeting of the European Parliament’s Committee on the Internal Market and Consumer Protection, when questioned on the issue of the introduction of collective redress, Viviane Reding, Vice-president and Commissioner responsible for justice, fundamental rights and citizenship, stated that ‘the Commission did not consider collective redress as an end in itself’ but that ‘a proposal ensuring appropriate guarantees and preventing abusive litigation would be tabled in time before the end of the term’, ie, no later than 2015.9 However, there is no specific schedule and the subject is not mentioned in the 2013 Work Programme. What future, then, for collective redress in Europe? It’s hard to say. The different Directorates General of the Commission have not yet given a clear indication of having reached common views on how to proceed. Moreover, it is not easy to anticipate what impact any initiative of the European legislator might have. This depends not only on the extent of the intervention (I) but also on the type of legislative instrument used (II). Unfortunately, many indicia suggest that the EU’s attempt to establish common rules could be very limited, and that only a minimum of harmonization will be sought.

I. Towards procedural harmonization a minima ? According to DG Competition, EU legislation is needed in this area. Certainly, the impulse of the Commission for the development of collective redress in the Member States has been very important, considering that a dozen countries have 6 Collective Redress in Antitrust, IP/A/ECON/ST/2011-19 PE 475.120, June 2012, http://www. europarl.europa.eu/committees/fr/studiesdownload.html?languageDocument=EN&file=74351. The authors of the report were Paolo Buccirossi, Michele Carpagnano, Lorenzo Ciari, Massimo Tognoni and Cristiana Vitale. 7 Alexander Italianer interview, in 12(2) Competition Law Insight, 19 February 2013, page 3. 8 The proceedings from this conference are due to be published. 9 A partial summary record of the meeting may be found at www.register.consilium.europa.eu.

The Future of Collective Damages Action in Europe  259   adopted reforms or are in the process of reforming the national legal arsenal. However, this informal approach is insufficient because it does not address the significant asymmetry in Europe that operates at the expense of victims of anticompetitive behaviour.10 For the moment, DG Competition is thinking about how to define guiding principles applicable to all private actions in the antitrust domain (A). Nothing is said, however, about specific aspects of collective action, mainly because the Commission is still awaiting the results of the impact assessment in accordance with the request made by the European Parliament in its resolution of 2 February 2012 (‘Towards a coherent European approach to collective redress’). A new impetus to intervene could come from United Kingdom, which already represents an attractive jurisdiction in which to bring private claims,11 and which (with a new bill on private antitrust enforcement, published in January 2013) could become one of the main forums of jurisdiction in Europe for collective action.12 Nevertheless, harmonization on competition issues alone would be disappointing since competition litigation represents a mass dispute and it is necessary to consider specific features relevant to collective action (B).

A. Proposals for a lege lata: toward the definition of guiding principles in the field of private enforcement According to recent statements by its representatives, DG Competition is presently defining three guiding principles that would apply to all private antitrust actions, be they individual or collective.

1. Actions that benefit all victims At present, it is mainly large businesses that seek compensation via litigation when they suffer antitrust-related harm. Final consumers do not act at all; only a few consumer associations use procedural tools made available to them by national legislatures. And only a small proportion of infringement decisions adopted by the Commission are followed by actions for damages in national courts. In practice, litigation is concentrated in a small number of Member States. On 5 July 2012, the UK’s Competition Appeal Tribunal awarded both compensatory and (for the first time) exemplary damages to the victim of an abuse of a dominant position. See Patrick Harrison, ‘The UK Competition Appeal Tribunal imposes exemplary damages in a predatory pricing case in the bus services market (2 Travel / Cardiff Bus)’, 5 July 2012, e-Competitions, No 48795, www. concurrences.com. 12 In January 2013, the UK Government published its response to the Department for Business Innovation and Skills (BIS) consultation on options for reform of the private actions regime in the competition sphere. See Private Actions in Competition Law: A consultation on options for reform – government response, http://www.biicl.org/files/6310_10_12_2012_urn_13_501_front_and_back_ covers_22-jan-13_private_actions_in_competition_law__a_consultation_on_options_for_reform__ government_~_response_pdf.pdf. 10 11

260  Silvia Pietrini Relying in particular on the famous judgments Courage13 and Manfredi,14 DG Competition intends to enshrine the right of all victims (consumers but also businesses) of anticompetitive practices to ensure effective damages actions. The UK reform project goes in this direction without further specifying the concept of business.15 It is interesting to note that the ‘Collective Redress in Antitrust’ study that emerged in June 2012 at the request of the European Parliament is more cautious about this concept, stating that ‘only very small enterprises (ie, micro businesses whose situation and needs are comparable to that of consumers) are allowed to participate in collective redress mechanisms’.16 DG Competition also appears favourable to the recognition of indirect purchasers’ right to bring private damages actions. If confirmed, this recognition could pose the difficult question of the establishment of the conditions of liability and especially the assessment of damages, which is very complex.17 On this subject, it appears the Commission will probably issue (during the second quarter of 2013) a Communication that will provide economic insight into the harm caused by anticompetitive practices and the methods commonly used to quantify such harm.

2. Collective redress subject to the principle of primacy of public enforcement According to DG Competition, public investigations and sanctions must remain at the heart of European competition law enforcement. Collective redress is regarded only as a complement to public action. Given this fundamental background assumption, three guiding principles seem to be on the horizon. First, with a view to ensuring the integrity of public enforcement, the Commission would first promote follow-on actions (which already represent the majority of cases, whereas stand-alone antitrust suits remain rarer still) by establishing a rule whereby the decisions of the national competition authorities of the Member States would be binding on national courts. This proposal is not new, as it was already in the draft Directive of 2009 on damages actions for breach of European antitrust law, and it seems to reflect the position of the European Parliament in its resolution of 2 February 2012. Such a uniform approach may indeed be needed to give all EU consumers and businesses the same level of legal certainty, and to mitigate forum shopping. It remains to be seen whether the Commission will limit recognition to the decision of the national authority (as requested by the European Parliament) or whether it will take on board the position Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297. Joined Cases C-295 to C-298/04, Manfredi v Lloyd Adriatico Assicurazioni [2006] ECR I-6619. 15 See Private Actions in Competition Law: A consultation on options for reform – government response, cited above note 12, page 26. 16 Collective Redress in Antitrust, cited above note 6, page 78. 17 See Commission, Draft Guidance Paper Quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty, http://ec.europa.eu/competition/consultations/2011_ actions_damages/draft_guidance_paper_en.pdf. 13 14

The Future of Collective Damages Action in Europe  261   of the European Economic and Social Committee. The latter, in an opinion of 25 March 2009, asserted that under the principle of equivalence of procedural rules, a provision similar to that provided for by Article 16 of Regulation 1/2003 should be applied to all decisions made by national competition authorities which establish that Article 101 or Article 102 TFEU has been infringed.18 The argument is that this solution should be imposed on the principle of mutual recognition of national competition authorities. Such an approach would be consistent with the Joint Statement by the Council and the Commission on the functioning of the Network of Competition Authorities, according to which the ‘Member States accept that their enforcement systems differ but nonetheless mutually recognize the standards of each other’s system as a basis for cooperation’.19 Second, in order to promote follow-on actions, DG Competition intends to maintain its former proposal to require a minimum of harmonization of national statutes of limitations. According to this approach, for purposes of both individual and collective actions, a new period of limitations would start on the day an infringement decision is adopted. Third, it is considered imperative that private enforcement actions should not jeopardize the effectiveness of leniency programs. DG Competition has for years insisted on the need to protect leniency as an essential means of detecting and punishing cartels. But in its Pfleiderer judgment,20 the European Court of Justice upset this principle, stressing that the need to protect leniency programs does not automatically override victims’ right of access to pertinent documents, and that the two opposing interests should rather be balanced on a case-by-case basis by national judges. In that regard, as pointed out by Advocate General Jääskinen in the Donau Chemie case,21 ‘[t]he right of access to a court is not […] absolute. It can be subject to limitations, provided that they do not undermine the very core of the right of access, pursue a legitimate aim, and where there is a relationship of proportionality between the means employed and the legitimate aim sought to be achieved’.22 The solution adopted by the ECJ in Pfleiderer has been a significant evolution of the place that must be accorded to private enforcement, but it also exposed its operational limits. Indeed, a system based on case-by-case balancing is scarcely satisfactory, as national approaches differ considerably,23 and ex ante uncertainty 18 See Opinion of the European Economic and Social Committee on the White paper on damages actions for breach of the EC antitrust rules, COM(2008) 165 final, (2009/C 228/06), 2009 OJ C228/40, point 4.3.4. 19 http://ec.europa.eu/competition/ecn/joint_statement_en.pdf, point 8. 20 Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161. 21 Opinion of Advocate General Jääskinen, delivered 7 February 2013, in Case C-536/11, Bundeswettbewerbsbehörde v Donau Chemie AG & alii, judgment pending as of this writing. 22 Ibid, paragraph 54. 23 See, eg, Local Court Bonn, decision of 18 January 2012, WuW/E DE-R 3499 – Pfleiderer II (described by Florian Bien, ‘Germany – Access to documents: The civil court of Bonn upholds the Bundeskartellamt’s refusal to grant access to the files voluntarily submitted by a leniency applicant (Pfleiderer)’, Concurrences 2-2012, No 45915, at pp 174-176); English High Court, National Grid Electricity Transmissions Plc v. ABB Limited & Others [2012] EWHC 869 (Ch) (described by Nikos

262  Silvia Pietrini may potentially have serious consequences for leniency programs, and hence for the exposure of secret cartels. It is therefore unsurprising that, following the adoption of the Pfleiderer judgment, many voices have called for European legislation to provide more reliable guidance on how the conflict between access to evidence and incentives to self-report can be resolved.24 In June 2012, Commissioner Almunia reaffirmed that he would propose legislation designed for that purpose.

3. Actions aiming to full compensation for the damage suffered According to representatives of DG Competition, any private action relevant to antitrust would be subject to the principle of full compensation for the damage suffered. Punitive damages would be excluded. This marks an evolution in relation to the draft Directive of 2009, which established the principle of full compensation but did not present it as an absolute principle. Since then, the Commission has more firmly embraced this principle.25 Similarly, in the resolution on collective redress of 2 February 2012, the European Parliament was clearly hostile to punitive damages. Moreover, in Europe even those countries in which the courts may in principle grant exemplary damages are currently considering the adoption of the principle of full compensation. Most notably, in UK the Government’s view is that treble damages have no place within any UK collective redress regime.26

B. Proposals for a lege ferenda: the need to define specific guidelines for collective redress Pending the publication of the impact assessment of a European initiative in the matter, the Commission is undoubtedly paying close attention to the bill of the British government published in January 2013. This bill has the merit of affirming the primary role of collective actions as part of a broader overhaul of private antitrust enforcement. If collective redress is to contribute to the construction of a harmonized European area of justice, some of the UK proposals should be taken into account as Europe proceeds with its own reforms.

Dimopoulos, Mark Israel and Malcolm Walton, ‘United Kingdom: Private Antitrust Litigation’, The European Antitrust Review 2013, http://www.globalcompetitionreview.com). 24 See, eg, Private Actions in Competition Law: A consultation on options for reform – government response, January 2013, cited above note 12, page 58. 25 Alexander Italianer, ‘Public and Private Enforcement of Competition Law’, speech of 17 February 2012, Brussels, http://ec.europa.eu/competition/speeches/text/sp2012_02_en.pdf. 26 Private Actions in Competition Law: A consultation on options for reform, cited above note 12, page 40.

The Future of Collective Damages Action in Europe  263  

1. Launching a collective action A threshold question in relation to the launching of collective actions for damages is that of which parties are eligible to bring such claims. In its draft Directive of 2009, the Commission envisaged that various representative litigants might initiate such actions: trade or consumer associations and bodies established on an ad hoc basis as well as victims themselves. Following the instalment of the new college of Commissioners in 2010, the situation has changed. In particular, Commissioner Almunia has favoured a monopoly that would be given to public bodies and to eligible not-for-profit organizations (whose eligibility could be withdrawn in case of abuse). In its Resolution of 2 February 2012, the European Parliament confirmed this approach by proposing to define European standards to clearly identify the entities qualified to bring a collective action. However, in the ‘Collective Redress in Antitrust’ study, it was noted that beyond consumers’ associations ‘there seems to be no reason to restrict the ability of other subjects to bring a collective action to claim compensation for diffuse damage suffered due to anticompetitive conduct’.27 A new impulse could also come from the UK bill. According to the UK Government, ‘only those who have a genuine interest in the case, such as genuinely representative bodies (such as trade associations or consumer associations) or those who have themselves suffered loss should be allowed to bring cases’.28 This implies the exclusion of law firms, third party funders and special purpose vehicles because there could be a risk of abuse. There is also the question of the constitution of the group. In the 2009 draft Directive, the Commission made a​​ distinction depending on whether it was a representative action or group action, assuming an opt-out mechanism in the former case and an opt-in mechanism in the latter. The issue is being debated not only in the Member States but also within the European institutions. In its Resolution of 2 February 2012, the Parliament was very hostile to opt-out actions. By contrast, the European Economic and Social Committee proposed the creation of a judicial remedy with both opt-in and opt-out mechanisms.29 At the national level, some countries such as Denmark have adopted hybrid solutions that mix characteristics of the two models. This hybrid approach is also being considered by the UK Government. The new bill purports to introduce a limited opt-out collective redress regime for competition law, with cases to be heard only in the Competition Appeal Tribunal (CAT). According to this proposal, the CAT would be required to certify whether a collective action brought under the new regime is suitable for collective action and whether it should proceed under an opt-in or an opt-out basis.30 The certification of the court would be a strong process of judicial Collective Redress in Antitrust, cited above note 6, page 78. Private Actions in Competition Law: A consultation on options for reform, cited above note 12, page 34. 29 See, eg, Opinion of the European Economic and Social Committee on the White paper on damages actions for breach of the EC antitrust rules, cited above note 18. 30 Private Actions in Competition Law: A consultation on options for reform, cited above note 12, page 31. 27 28

264  Silvia Pietrini certification, including a preliminary merits test, an assessment of the adequacy of the representative and a requirement that a collective action would be the best way of bringing the case.31 An opt-in basis would then be used in cases brought by a small number of businesses. To avoid encouraging forum shopping behaviour, the ‘opt-out’ aspect of a claim would only apply to UK-domiciled claimants, though non-UK claimants would be able to opt in to a claim if desired.32 By contrast, a French bill filed on 28 January 2013 by a group of senators would establish a collective redress system based only on an opt-in approach.33 Certainly, the opt-in model limits the risk of frivolous actions and better preserves individual liberty. But it may also discourage meritorious actions when the value of the harm suffered by each potential claimant is very low and prevents defendants from extinguishing their liability with a single settlement agreement.34 It might be useful to also consider hybrid solutions whereby courts are granted discretion as to whether the opt-out model is necessary to guarantee that a significant proportion of injured parties are compensated for harm suffered.35 This solution, however, requires a reflection on the thorny issue of publicizing the action. Any means chosen should guarantee adequate publicity so that the opt-out mechanism might be transplanted in countries such as France.36

2. Funding mechanisms for collective actions Particular attention should be devoted to funding mechanisms because collective legal actions are costly and raise coordination problems due to dispersed harm. However, studies show that the solutions adopted across Member States differ widely.37 The UK bill proposes to prohibit contingency fees in order to avoid the development of a ‘litigation culture’. Furthermore, any unclaimed sums would have to be paid to the Access to Justice Foundation, although defendants would be free to settle on other bases, including on a cy près or reversion-to-the-defendant basis, subject to approval by the judge.38 The Commission should consider how to create effective public financing, as the US approach ‘has been traditionally viewed with scepticism in Europe for both ethical and efficiency reasons. In terms of ethics, the idea of making a business out of a suffered injury is often rejected; in efficiency terms, the American system is thought to conduce to excessive litigation’.39 For instance, the Quebecois Class Ibid, page 40. Ibid. 33 See http://www.senat.fr/leg/ppl12-297.pdf. 34 Collective Redress in Antitrust, cited above note 6, page 78. 35 Ibid, page 89. 36 See Silvia Pietrini, L’action collective en droit des pratiques anticoncurrentielles. Perspectives nationale, européenne et internationale, Bruylant, 2012, pages 357 et seq. 37 See, eg, Christopher Hodges, Stefan Vogenauer, Magdalena Tulibacka, ‘Costs and Funding of Civil Litigation’, Oxford Legal Studies Research Paper 55-2009, papers.ssrn.com/abstraxt=1511714. 38 Private Actions in Competition Law: A consultation on options for reform, cited above note 12, page 42. 39 Collective Redress in Antitrust, cited above note 6, page 73. 31 32

The Future of Collective Damages Action in Europe  265   Action Assistance Fund (Fonds d’aide aux recours collectifs), which supports all stages of the procedure, turned out to be effective without creating such a ‘litigation culture’.40

3. On the possibility of using alternative dispute resolution (ADR) In the context of the collective redress debate, the European Parliament has repeatedly stressed the need for a common reflection on the role of alternative dispute resolution. Most recently, in connection with the draft Directive on alternative dispute resolution for consumer disputes of 12 March 2013, the Parliament stated that ‘[t]he existence of an effective system for collective claims and easy recourse to ADR should be complementary and they should not be mutually exclusive procedures’.41 Some national legislative bills are also considering the use of ADR in the context of collective actions. For example, the French bill of 28 January 2013 foresees the possibility for the judge to propose, at any moment of the procedure, the use of mediation. The UK bill goes further, distinguishing between two scenarios. First, the Government intends to introduce the possibility of a settlement in the context of the collective redress procedure. Any opt-out settlement must be judicially approved, with the approval conditional on a consideration of the reasonableness of the fees being paid to legal representatives, and on giving the underlying claimants an opportunity to opt out of the settlement.42 Second, the Government has also decided to introduce an opt-out collective settlement regime for competition law, a system similar to the Dutch Mass Settlement Act (2005). This instrument allows businesses to quickly and easily settle cases on a voluntary basis.43 As with the collective actions regime, the ‘opt-out’ nature of a settlement would only apply to UK-domiciled claimants, while claimants outside the UK could opt in if they so choose. The judge would have to certify, via a mechanism similar to the one described above, that the case is suitable for such a settlement, and that the settlement is ‘fair, just and reasonable’. Having approved the settlement, the judge would then be able to issue any directions he thinks appropriate regarding the settlement mechanics and procedure. The Commission should include this issue in its discussion of antitrust class actions. Indeed, alternative mechanisms to resolve disputes outside the judicial process provide a quicker and less costly route than court litigation. They allow for the compensation of injured parties in a timelier manner, while at the same 40 See Pietrini, L’action collective en droit des pratiques anticoncurrentielles, cited above note 36, pages 288 et seq. 41 Position of the European Parliament adopted at first reading on 12 March 2013 with a view to the adoption of Directive 2013/.../EU of the European Parliament and of the Council on alternative dispute resolution for consumer disputes and amending Regulation (EC) No 2006/2004 and Directive 2009/22/EC (Directive on consumer ADR), point 27. 42 Private Actions in Competition Law: A consultation on options for reform, cited above note 12, page 43. 43 Ibid, page 50.

266  Silvia Pietrini time placing less stress on the judicial system. Adequate procedural rules may be put in place to encourage both claimants and defendants to settle before reaching the trial stage, and to facilitate such settlements.44

II. Towards harmonization, but how? During the ‘New frontiers of Antitrust 2013’ conference of 22 February 2013, the Director of the DG Competition’s ‘Policy and Strategy’ Section – Carles Esteva Mosso – said that the fact that there are 27 Member States with different systems does not necessarily preclude the effective application of principles developed at European level. However, if the action of the Commission were limited to the adoption of a soft law instrument, such as a simple communication, this action in practice would be ineffective for the evolution of private enforcement in Europe. Certain European capitals, such as London and Amsterdam, could become particularly attractive forums of jurisdiction, which could reinforce a disturbing trend toward multi-tiered justice in Europe. It must not be forgotten that the ‘Brussels I’ Regulation allows forum shopping, and the amendments that will take effect when Regulation 1215/2012 enters into force in 201545 do nothing to change this. The Commission should therefore opt for a binding instrument. A formally harmonized regime for collective redress would contribute to the realization of a European area of justice and to the proper functioning of the internal market. But what would be the most suitable instrument to accomplish this harmonization? Studies show that directives are not sufficient to ensure the effectiveness of EU law (A). For several reasons, therefore, the EU legislator should adopt a regulation as the appropriate instrument to govern collective legal actions (B).

A. The problems with harmonization by directives Directives have the advantage of allowing for an adjustment of rules to local preferences and situations. Directives are not designed to be directly applicable; they are intended to take effect only once they have been transposed into national law by each Member State. Indeed, according to the jurisprudence of the Court of Justice, ‘the transposition of a directive into domestic law does not necessarily require that its provisions be incorporated formally and verbatim in express, specific legislation; a general Collective Redress in Antitrust, cited above note 6, pages 76 et seq. Regulation (EU) 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, 2012 OJ L351/1. 44 45

The Future of Collective Damages Action in Europe  267   legal context may, depending on the content of the directive, be adequate for the purpose provided that it does indeed guarantee the full application of the directive in a sufficiently clear and precise manner so that, where the directive is intended to create rights for individuals, the persons concerned can ascertain the full extent of their rights and, where appropriate, rely on them before the national courts’.46 In practice, as noted by one author, ‘[t]he impact of the harmonisation programme to date therefore has not been the creation of a single, consistent and coherent body of […] law common to all the EU Member States; instead, there are now 27 national rules on doorstep selling, distance selling and so on’.47 Therefore, the risk of the adoption of non-harmonized solutions across Member States is one of the main obstacles to the efficient and effective functioning of the single market.48

B. The need for full harmonization in light of the principle of the effectiveness of European Union law In his report on ‘A New Strategy for the Single Market’, Mario Monti stated that a ‘Regulation brings the advantages of clarity, predictability and effectiveness. It establishes a level playing field for citizens and business and carries a greater potential for private enforcement.’49 Indeed, a Regulation is directly applicable and binds the competent authorities in the Member States; it does not require lengthy transposition and it ensures clarity and coherence of standards and obligations across the Community. Therefore, from a European perspective, a regulation should be preferred to a directive. Of course, it is true that ‘choosing a Regulation would create its own challenges: it would adopt terminology and concepts distinct from any national law, and Member States could not utilise more appropriate national legal terminology as they can when transposing a directive’.50 However, the use of a regulation may be needed to have a coherent European framework, to protect victims of anticompetitive practices and to give adequate expression to the constitutional principle of the effectiveness of EU law. Regulations are the instruments often used in the field of judicial cooperation. A regulation could be adopted on the basis of Article 81 TFEU in order to ensure effective access to justice. Some authors consider that Article 103 TFEU would be the most appropriate Treaty legal base for European legislation on collective Case C-59/89, Commission v Germany [1991] ECR I-2607, paragraph 18. Christian Twigg-Flesner, ‘Good-Bye Harmonisation by Directives, Hello Cross-Border Only Regulation? – A way forward for EU Consumer Contract Law’, http://ec.europa.eu/justice/news/ consulting_public/0052/contributions/309_en.pdf, page 5. 48 See European Parliament resolution of 14 September 2011 on better legislation, subsidiarity and proportionality and smart regulation (2011/2029(INI), http://www.europarl.europa.eu/sides/getDoc. do?type=TA&reference=P7-TA-2011-0381&language=EN. 49 Mario Monti, A New Strategy for the Single Market at the Service of Europe’s Economy and Society, 9 May 2010, http://ec.europa.eu/bepa/pdf/monti_report_final_10_05_2010_en.pdf, page 41. 50 Twigg-Flesner ‘Good-Bye Harmonisation by Directives’, cited above note 47, page 8. 46 47

268  Silvia Pietrini redress in antitrust.51 In any case, a regulation would also ensure the application of Article 47 of the Charter of Fundamental Rights of the European Union, under which access to the courts to assert the rights and freedoms guaranteed by the law of the Union must be guaranteed in all Member States. Several studies have shown that a European instrument on collective action would respect the principles of subsidiarity and proportionality because the absence of such an instrument could jeopardize the application of EU law, the protection victims’ entitlement to compensation, and the right of access to justice.

III. Conclusion The Commission has proceeded very cautiously towards a legislative initiative. After announcing, at the end of 2011, that a proposal was on the way, the Commission undertook to present a Communication by the end of 2012 and a legislative proposal on actions for damages for infringement of the antitrust rules. The new Commissioner for Health and Consumer Policy – Tonio Borg – stated at his confirmation hearing before the European Parliament that a major priority of his mandate would be to implement effective systems of redress, including collective redress. After a great deal of consultations and work conducted in the field of consumers and competition over a number of years, the detailed rules in respect of effective redress have been sufficiently discussed. The Commissioner responsible for Competition also seems to support a system of collective redress. By contrast, the Commissioner responsible for Justice, Fundamental Rights and Citizenship has never concealed a cool attitude toward the aggregation of legal claims.52 If a new legislative initiative is essential for Member States that have not yet adopted new measures to deal with collective redress, it is also essential to rein in the growing diversity of national procedures. Indeed, only formal action at the European level will ensure equality of consumers and businesses in an integrated area consistent with the principles of the European Union. Moreover, robust private enforcement ‘is a way to bypass Member States who are unwilling to fully implement EU law. […] Finally, it is likely to increase the legitimacy of EU law, as it relies on national courts whose legitimacy is beyond doubt’53. The UK bill proposes a number of quite radical changes to the existing regime of private actions in competition law in the UK. Our hope is that, at EU level, the European Commission will do the same! 51 Paolo Buccirossi and Michele Carpagnano, ‘Is it Time for the European Union to Legislate in the Field of Collective Redress in Antitrust (and how)?’, 4(1) Journal of European Competition Law and Practice 3 (2013). 52 See Viviane Reding, ‘Europe, The Law and The German Legal Profession: A Comment From Luxembourg’, 18 September 2012, SPEECH/12/614. 53 Simon Vande Walle, Private Antitrust Litigation in The European Union and Japan, Maklu, 2013, page 160.

John Ratliff *

Integrating Public and Private Enforcement of Competition Law: Implications for Courts and Agencies

The object of this chapter is to address the issue of how public authorities’ enforcement of the EU competition rules relates to private actions for compensation arising from infringements of those rules,1 including some ‘takeaway’ conclusions from our workshop. This question is topical because public enforcement has become more intense, driven mainly by public immunity/leniency programmes, while at the same time private enforcement has taken off in comparison to a few years ago. Private claims have come a very long way since 1966, when the European Commission published its first study on the subject2 and since 2001, when similar issues were last debated in these proceedings.3 Looking at recent reporting of damages cases in specialist reporting services like MLex and other sources, my understanding is that there are actions going on at the moment concerning some 20 EU cartel decisions and my impression is that there are at least as many again involving other EU competition law issues and national cases. That may be a little high, since some cases may have settled, but it could also be low, given the few countries on which information is available. There have also been some publicised settlements,4 to which others should be added which have happened without publicity. * Partner, Wilmer Cutler Pickering Hale and Dorr LLP (‘WilmerHale’), Brussels. The views expressed in this paper are personal and do not necessarily reflect those of WilmerHale. This paper is a revised version of a background paper prepared for the EU Law and Policy Workshop at the EUI, Fiesole, in June 2011. It was updated for publication in January 2013. With thanks to Katarzyna Bojarojć and Katrin Guéna for their research assistance, and to Jacques Bourgeois and Frédéric Louis for their comments on earlier drafts. 1 See generally ICN Cartels Working Group Report, ‘Interaction of Public and Private Enforcement in Cartel Cases’ (May 2007) (‘ICN Report’), Part III; OECD Policy Roundtable, ‘Private Remedies’ (2007) (‘OECD Roundtable’), Chapter 4 (‘Collective Action and the Interface between Public and Private Enforcement’); Wouter Wils, ‘The Relationship between Public Antitrust Enforcement and Private Actions for Damages’, 32 World Competition 3 (2009); Wouter Wils, ‘Should Private Enforcement Be Encouraged in Europe?’, 26 World Competition 473 (2003). 2 La Réparation des conséquences dommageables d’une violation des Articles 85 et 86 du Traité Instituant La CEE, CEE Série Concurrence No 1. 3 Claus-Dieter Ehlermann and Isabela Atanasiu, eds, European Competition Law Annual 2001: Effective Private Enforcement of EC Antitrust Law, Hart Publishing, 2003. 4 For example, Parker ITR in the Marine Hose case. See Mlex, 6 March 2009. It appears that Parker ITR agreed to set up a fund to reimburse purchasers if they gave up their right to sue the company. This did not include direct purchasers of marine hose in the US, covered by a series of separate settlements. It is reported that Parker ITR paid a sum equal to 16% of its hose sales between 2002 and 2007 into an interest bearing escrow account. See www.hausfeldllp.com.

272  John Ratliff Some may argue that this is still a very small number overall, but it is a huge increase in comparison with ten years ago, when there appeared to be just the odd case from time to time. With this growth have come new pressures in the interplay of public enforcement and private compensation claims, which are the focus of this workshop. Scope of this chapter: I propose to focus on ‘follow-on’ claims for damages for cartel infringements, although clearly other claims arising from the breach of the competition rules are important, notably injunctions in cases alleging the breach of Article 102 of the Treaty on the Functioning of the European Union (TFEU).5 My understanding is that stand-alone private actions in cartel cases are rare. I am not proposing to discuss the wider legislative role of the European Commission as regards private claims, but rather to focus on features of administrative or court practice.6 However, I would note that the Commission has been considering specific legislation to promote private actions for damages in competition cases for some years now, and this was still on its legislative agenda for 2012. The issue is also bound up with questions as to whether there should be more general legislation to promote collective redress claims by consumers (for competition and other issues). I also do not propose to discuss the implications of settlements for private actions, since we debated that in our 2008 workshop.7 Plan: My main themes are three: (i) Public enforcement and private compensation proceedings are very different and, to some extent, complementary. (ii) The passage of time is very important and regrettably, even if there are valid reasons for it, these proceedings (both public and private) still take far too long, both from a plaintiff and defendant perspective. (iii) In recent years there have been important developments in the case law as regards the relationship between public and private proceedings, as private actions have become more frequent and there have been increasing efforts by private parties to obtain access to public enforcers’ files through applications to the Commission under the EU Transparency Regulation,8 and/or applications for disclosure of evidence to competition authorities before national courts in follow-on actions. 5 This is also an area which needs further development, notably in jurisdictions which only give injunctive relief if damages are not available. The possible award of damages many years after the event is no use to a plaintiff which has been driven out of business in the meantime and no substitute for effective injunctive relief, preventing dominant companies from denying the competitive process. See the forceful dissenting opinion of Wilberforce LJ in Garden Cottage v Milk Marketing Board, [1983] 3 Weekly Law Reports 143, 160. 6 See generally the section on ‘Actions for damages’ at http://ec.europa.eu/competition/antitrust/ actionsdamages/index.html. 7 See Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2008: Antitrust Settlements under EC Competition Law, Hart Publishing, 2010. 8 Regulation 1049/2001 of the European Parliament and the Council, 2001 OJ L145/43.

Public and Private Enforcement: Implications for Courts and Agencies 

273  

I propose to address these themes in turn and then to touch on some of the wider issues in this area: (i) ‘Confidentiality’ in the Commission decision. (ii) To what extent should the decision of a competition authority bind a court? (iii) Should immunity applicants also be protected in follow-on claims? (iv) Compensation as mitigation. Finally I set out some overall ‘takeaway’ conclusions.

1. Public and private proceedings are very different When considering the ‘integration’ of public and private enforcement, I think the most important point is to note that the two types of proceeding are very different. The two may be complementary, if a private action follows from a public action, but the two types of claim are directed at different objectives and function differently. In private claims, as follow-on actions, typically what happens is that a plaintiff, who is generally a customer of a cartel member, will bring a claim relying on the Commission’s decision, or the decision of a national competition authority (NCA), applying the EU rules. Sometimes a prejudiced competitor or another affected party (indirect purchaser) may sue, but so far this has been rare. The claim is generally directed to certain defendants who have supplied the plaintiff and brought in the most favourable jurisdiction for the claim in the plaintiff’s view. The jurisdictions developing such litigation fastest are England and Wales, The Netherlands and Germany (although clearly there are other cases in other Member States, notably in Scandinavia as regards asphalt). Typically the Commission decision will be pleaded as proof of an infringement (relying on EU case law such as Masterfoods9 and/or national statutory provisions) and any findings of fact in the decision will be relied on to show liability. The plaintiff aims to seek damages, so still has to prove causation and the quantum of damage. Typically the defence will argue that the cartel had no effect in practice, ie even if unlawful behaviour occurred, it was not applied to the plaintiff in practice; or, if the plaintiff is a direct purchaser, that any effect was just ‘passed on’ to the plaintiff’s own customers so that the plaintiff suffered no injury; and/or that any effect on indirect purchasers is too remote in terms of causation. It is important to appreciate how different an issue it is to claim damages in this way, in comparison to the Commission’s proceedings, where a cartel infringement is found as a ‘restriction by object’ (by nature, rather than effect). 9

Case C-344/98, Masterfoods v HB Ice Cream Ltd. [2000] ECR I-11369.

274  John Ratliff In such a case, typically the Commission will show that the defendant ‘undertaking’ or group, in particular certain subsidiaries, participated in an infringement through relevant conduct over a period of time. Usually this is done by proving that a number of meetings among competitors took place in which anticompetitive conduct was discussed, and that other contacts were made by phone, by e-mail or in person. From this the Commission will conclude that anticompetitive behaviour (eg price increases, or market or customer sharing) was agreed. Importantly, the Commission does not need to show actual effect, ie that a particular company was affected in a particular way. If there is an infringement ‘by object’, where an infringement is found ‘by its nature’, such as price-fixing amongst competitors, that is enough for the Commission’s purposes. It does not need to ask how much particular suppliers sold to particular customers, nor does it have to show any particular price increase from the cartel behaviour. Sometimes the Commission will make some statements in its decision on ‘actual impact’ just to reinforce its case, or to justify higher fines, eg under the Commission’s 1998 Fining Guidelines.10 Typically, the Commission will show what it calls ‘implementation’ of a cartel, ie that at least efforts were made after an unlawful meeting to try and carry out what was agreed. The Commission may also include some references to evidence of effect in its file, but usually these will not be detailed or systematic, nor will they necessarily show that a particular price target increase or other unlawful arrangement actually occurred in a particular case. If the Commission makes statements about effect, typically this will also lead to a counter from the defence and may lead to appeals to the European Courts, in part because the defence may argue this was wrongly relied on to increase the fine and in part because of concern about follow-on damages actions. Sometimes the Commission has itself noted that proving such effect is very difficult. For example, in Citric Acid, the Commission stated: There is no need to quantify in detail the extent to which prices differed from those which might have been applied in the absence of the arrangements. Indeed, this cannot always be measured in a reliable manner, since a number of external factors may simultaneously have affected the price development of the product, thereby making it extremely difficult to draw conclusions on the relative importance of all possible factors.11

More specifically, many issues come up in private claims, which are entirely different from those in the Commission proceedings, yet they may well have to be addressed, depending on the circumstances. For example: – Proper defendant: o Is the company sued the correct defendant? o Did the defendant supply the plaintiff, applying the cartel? 10 1998 OJ C9/3. Under those Guidelines, ‘actual impact on the market’ is a factor relevant to the assessment of the gravity of the infringement. 11 Citric Acid, 2002 OJ L29/18, recital 211. However, the Commission otherwise noted that the arrangements had been ‘carefully implemented’ (see recitals 210–228).

Public and Private Enforcement: Implications for Courts and Agencies 

275  

o Should the defendant be ‘joining’ others, (ie pulling them into the case as a co-defendant) given that the claim is for joint and several liability? – Proper jurisdiction: o Is the defendant in the relevant jurisdiction, so that a claim can be brought? o Can the defendant be an ‘anchor defendant’, allowing the whole cartel to be sued in the same jurisdiction?12 o Are there any contractual ousters in favour of another jurisdiction or arbitration which are relevant?13 o What happens if some customers sue in one country and other customers sue in another in the same case?14 o What happens if claims are brought in different countries, but the plaintiffs have different interests (ie some as direct purchasers, others as indirect purchasers, others as prejudiced competitors)? – Limitation: o Is the plaintiff’s claim brought in time or barred by the relevant (national) limitation rules? – Did the plaintiff in fact suffer damage from the conduct found to be a ‘restriction by object’? o Was the cartel, in fact, not applied in terms of a price overcharge or market or customer sharing arrangement vis-à-vis the plaintiff, in full or in part? o If implemented, how much was sold to the plaintiff and what was the extent of the damage? o Even if there was implementation of the cartel, did the plaintiff in fact suffer damage, or did it pass on the alleged overcharge to its own customers, in all or part, with no or limited loss and no or limited reduction in sales volume?15 o If so, how does one integrate the two types of claim (direct and indirect purchaser). Or do you just take them one by one on the merits, depending on claims and evidence? – What about other circumstances which may have affected market behaviour over the period of the cartel? For example: o General market issues, such as a recession affecting sales and making price increases difficult. 12 See, in English law, Provimi [2003] EWHC 961 (Comm); Cooper Tire, [2009] EWHC 2609 (Comm.), on appeal: [2010] EWCA Civ 864; Council Regulation 44/2001, 2001 OJ L12/1, Articles 2(1), 5(3) and 6(1). 13 See Council Regulation 44/2001, cited previous footnote, Article 23(1). 14 For example, what is now known in England as the ‘Italian torpedo’ can arise. In this scenario, a defendant seeks to pre-empt a plaintiff’s choice of jurisdiction by commencing proceedings for a negative declaration (ie, a declaration that certain conduct has no effect or causes no damage) in another jurisdiction. This happened in Cooper Tire, cited above note 12, but the argument failed on the facts. 15 This is a topical issue in national proceedings, partly owing to its complexity and partly because diverging judgments are emerging. See, eg, judgments of the French Cour de Cassation, Cass Com, 15 June 2010, Ajinomoto Eurodyne v Ceva santé and of the Karlsruhe Court of Appeals, 11 June 2010, 6 U 118/05 (Kart).

276  John Ratliff o Or more specific issues, such as the countervailing purchasing power of the plaintiff, making it illusory to think that a price increase, even if agreed, could be implemented. o Or extraneous events, such as a factory fire, or another bid at the other end of the world, which could explain why a company did not bid for a particular contract at a given time in Europe? – Joint and several liability and contribution: o Even if cartelist A did not supply the plaintiff at an overcharge, can A be liable, nevertheless, for damage which his co-cartelists B, C and D may have caused to the plaintiff? o If so, how can that claim be handled, without compromising confidentiality relevant to current supply arrangements? o If so and say company D is not involved in the private claim, can A, B and C still sue it for its ‘contribution’ on the joint liability, even though this is long after the limitation period has expired on the plaintiff’s original claim (as appears possible in some national laws)?16 – How are all these things to be proved (whether on the plaintiff or defence side) and to what standard. Notably, what proof of the claimed damage will the judge require? Will strict proof be required, or will some degree of equitable assessment be enough? One may add that such complexities may be multiplied in some cases by enforcing authorities and claims around the globe, each with its own specifics and timetable. I have set out such a long list just to emphasise that private litigation is another world in comparison to public proceedings before the Commission and the specialised appeal system to the European Courts. These considerations are very far from those in Commission proceedings, which typically start with an undertaking (ie a group of companies) giving information to the Commission in return for immunity from fines, Commission dawn raids, further leniency applications, requests for information to the companies, a Statement of Objections (SO) and a Hearing, leading to a decision. Recently, there may also be a settlement negotiation. Typically the decision will impose huge fines on most of those involved, followed often by one or two rounds of appeal to the European Courts. All of which is focussed on the finding of an infringement by ‘restriction by object’, ie the facts showing the conduct which was unlawful, but not what actual effect there was on customers. So the key point is that public and private proceedings are very different, addressing different issues. They are complementary, if and insofar as the public decision and proceedings may be relevant to the private follow-on proceeding. Both insofar as the public decision may bind the national court as to whether there was an infringement or not and insofar as the public proceedings may disclose evidence which the plaintiff will seek to use. They are different, because many of the issues raised are different. 16 See the UK Civil Liability (Contribution) Act 1978, Sections 1 and 2. This is also a topical issue because it may discourage some early settlements (unless all the defendants are involved).

Public and Private Enforcement: Implications for Courts and Agencies 

277  

2. The passage of time is very important If a follow-on claim is brought, a case which has already gone through two complex public enforcement phases (ie the Commission’s proceedings followed by appeals to the General Court and possibly the Court of Justice) enters the private action phase, which is no less complex. This may also be in two parts, one dealing with the plaintiff’s claim (and related appeals), another so-called ‘contribution’, if all defendants are not brought into the private claim through joinder, so that a defendant which has had to pay compensation on the basis of joint and several liability sues other cartelists for their ‘contribution’ to that payment. Follow-on actions are, by their nature, usually much later and sequential on a definitive ruling as to liability. Commission proceedings may often take five to seven years and the European Courts’ review may be as long17 (five years at the GC is not unusual; further appeals may be a further three years at the ECJ). So one can easily have a situation where the claim for possible compensation starts to be litigated some 13–15 years after the public proceedings started. The events concerned will have occurred even earlier and, with a continuing infringement in a long cartel, that may mean that evidence is sought, going back as much as 10 to 20 years before the Commission’s decision. Clearly, this is very important because: – The relevant personnel may no longer be with the defendants, or even working or alive! – Records may also not be available, as computer systems may have changed, perhaps several times. The same applies if the staff in the relevant department have moved. As a result, files may have been deleted, or records may not be as comprehensive, as apparently unneeded files are no longer retained. – The defendant company or business may also have been sold, or have been split up and sold, sometimes more than once, a phenomenon which happens regularly during Commission proceedings and may well occur after also. This may also have consequences for liability and file retention. – Some files will have been retained, because they are considered relevant to the proceedings concerned, but it is only recently, with the new wave of follow-on claims, that companies have started to realise the importance of keeping the old records related to actual sales, as opposed to what was required for the Commission proceedings. (Unless they are involved in US claims, where counsel are quick to tell you to keep everything, because if not adverse inferences may be drawn by the courts.) – Last but not least, often the in-house or external counsel may have changed, so ensuring continuity, or understanding of the facts and issues may be difficult. 17 Clearly if a settlement with the Commission occurs this may be shorter, as in the recent cases generally there have been no appeals.

278  John Ratliff The result is that, in practice, running the follow-on private action case can be very difficult and impractical from both the plaintiff and defence sides and, in practice, these cases take ages: 20–25 years appears quite possible if matters do not first come to settlement. That may change to some extent if more Commission cartel settlements occur and as more of the procedural ‘satellite’ issues to follow-on litigation are resolved, but in general the structure of all this suggests that these cases will continue to take a very long time. Such a passage of time is perhaps understandable, given that multi-party ‘conspiracy’ cases are inherently complex, but in my view it is far too long. At present, the impression one has in practice is that private claims settle not on their merits, dealing with the actual issues, but just on pure litigation tactics with settlements based on nuisance value, rather than any real assessment of actual harm. One may say that happens often in litigation, but it is not a very satisfactory position in terms of dealing with the real issues.

3. Recent developments as regards access to the Commission’s file This leads us to the issue of access to the Commission’s file, a topic which is very controversial and has resulted in a number of very important European Court judgments in recent years.18 There are two situations in practice, applications to the Commission and applications before the courts. The issues raised are, however, similar.

(a) Applications to the Commission Generally, a plaintiff will seek to obtain as much information from the Commission as possible, as soon as possible, to benefit from the Commission’s wide investigative powers and any immunity or leniency statements given to the Commission. Such efforts may be made by a plaintiff seeking to join the Commission’s procedure as a complainant or, more specifically, through applications under the EU Transparency Regulation.19 Complainants (and to some extent third parties) have procedural rights in the Commission’s proceedings and may obtain some information through participation therein, ie a non-confidential version of the Statement of Objections or through attendance at the Oral Hearing, if any.20 18 See generally Van Bael & Bellis, Competition Law of the European Community, 5th Edition, Kluwer Law International, 2010, pp 1041–1059. See also Gaëtane Goddin, ‘Recent Judgments Regarding Transparency and Access to Documents in the Field of Competition Law: Where Does the Court of Justice of the EU Strike the Balance?’, 2 Journal of European Competition Law & Practice 10 (2011). 19 2001 OJ L145/43 (based on Article 255 EC, now Article 15 TEU).

Public and Private Enforcement: Implications for Courts and Agencies 

279  

Otherwise, as regards EU Transparency Regulation claims, the Commission emphasises that release of its file may undermine its process and that its task and priority is to use its resources as a public enforcer, not as a discovery tool for private follow-on claims. The Commission has therefore rejected claims for all or parts of its file under the EU Transparency Regulation, arguing that:

20

– As a matter of principle, it should not have to reveal immunity/leniency statements, or settlement statements, because such disclosure may stop them coming in to the authorities, removing the vital cooperation which allows cartels to be found and proved in the first place. – Practically, it should not have to wade through confidentiality claims again under EU Transparency Regulation standards, having already done so during its proceedings (i) under its own rules on access to file and confidentiality among defendants; and (ii) for the subsequent publication of its own decision, in order to assist a follow-on claim by certain third parties,21 instead of using its resources to pursue the next cartel case. However, in December 2011 (ie after our workshop in July 2011) the GC found that the Commission should disclose its case-file list in the CDC Hydrogene Peroxide case.22 The Commission argued on the basis of principle that disclosure would undermine its leniency programme, because those cooperating would be more likely to be sued for damages, with the result that their commercial interests would be undermined. However, the GC rejected this as a ground for denying disclosure.23 Then, in May 2012, the GC ruled that the Commission should have carried out a ‘concrete and individual’ assessment of its file in the Gas Insulated Switchgear cartel case, in response to an application by EnBW, a customer, for most of its file under the Transparency Regulation.24 The GC also held that the Commission could not deny access based on general grounds of principle, such as that this would undermine its investigation (which was in this case over by the time of the Transparency Regulation application), or that this would deter leniency applicants in other cases.25 Nor could the Commission just broadly exclude categories of documents in its file. The GC emphasises that there is no special exception under the Transparency Regulation for competition proceedings and requires the Commission to take this detailed review process even for its internal documents. Importantly, it appears that such disclosure could even apply to leniency materials, ie immunity leniency 20 See Article 6 of Commission Regulation 773/2004, 2004 OJ L123/18; and Van Bael & Bellis, cited above note 18, pp 1041–1042. 21 See, eg, Case T-2/03, VKI [2005] ECR II-1121. 22 Case T-437/08, CDC Hydrogene Peroxide, [2011] ECR II-8251. 23 Ibid, paras 46–49. 24 Case T-344/08, EnBW, judgment of the GC of 22 May 2012, not yet reported. On appeal: Case C-365/12 P, 2012 OJ C287/29. 25 Ibid, paras 113–129.

280  John Ratliff statements and documents quoting them. The EnBW judgment also emphasised that confidentiality is to be assessed as of the date of the application for disclosure, which may be some time after the Commission’s decision, so what was confidential during the Commission’s proceedings may not be later, when access to the file is sought. Predictably the Commission is appealing,26 in particular because in the Odile Jacobs and Agrofert cases,27 a rather different approach was taken. The ECJ ruled there that the Commission could refuse access to documents in merger proceedings and annulled the GC judgment to the opposite effect. The ECJ balanced the interests of the Merger Control Regulation process with the Transparency Regulation and held that third parties should not have access, because that would inhibit the Commission’s review process.

(b) Applications for information before the courts Normally the Commission cannot stop a court from ordering the defence to disclose relevant material depending on the relevant procedural rules. The Commission also appears not to object to a national court asking for information from its file, seeing this as part of its duty of cooperation with national courts,28 provided (i) that the receiving Court regulates the confidentiality of any disclosure properly;29 and (ii) that the Commission objects to disclosure of the immunity/leniency statement or settlement submissions in order to protect its enforcement system.30 However, this is still a general concern to defendants, because they fear that what will happen is that the whole file, including confidential business secrets, will be released to the plaintiffs without proper control. It is argued that, before a file is transferred, the Commission has to check confidentiality.31 Another approach is for a plaintiff to make a more focussed disclosure request through the national court, asking for specific information from the defendant, allowing both the national court and the defence to check the relevance and confidentiality concerned. However, in practice, in some jurisdictions such specific discovery powers are not widely used or effective (although they may be provided for).

The Commission’s appeal is at 2012 OJ C287/29. Cases C-404/10 P, Odile Jacobs and C-477/10, Agrofert, judgments of the ECJ of 28 June 2012, not yet reported. See also ECJ, Press Release 92/12 of 28 June 2012. 28 Article 15 of Regulation 1/2003 and Article 4(3) TEU, formerly Article 10 EC. 29 See Case T-353/94, Postbank [1996] ECR II-921, paras 64–73. 30 See the Commission’s approach in the recent National Grid case. 31 See Postbank, cited above note 29, at paras 90–96; Case 53/85, Akzo [1986] ECR 1965. 26 27

Public and Private Enforcement: Implications for Courts and Agencies 

281  

(i) Pfleiderer As our group met in June 2011, fate would have it that the ECJ had given judgment in the Pfleiderer case32 just days before. It may be recalled that the ECJ ruled on the question of access to leniency submissions in the context of a reference from a German court in a follow-on damages claim for infringement of the EU competition rules. The issue was access to the Bundeskartellamt’s file in the German decor paper cartel case. Pfleiderer was a customer of the companies concerned. Predictably, as a result there was extensive discussion in our workshop as to the ramifications of the judgment. It may be useful therefore to summarise briefly what the case was about and then to consider how the related issue of access to immunity/leniency materials is developing. Background to Pfleiderer In January 2008, the Bundeskartellamt (BKA), acting pursuant to Article 81 EC (now Article 101 TFEU) imposed fines totalling 62 million euros on the three largest European producers of decor paper (which is special paper for the surface treatment of engineered wood) and on five individuals personally, for price-fixing agreements and agreements on capacity closure. Those decisions were based, inter alia, on information and documents which the BKA had received in the context of its German leniency programme. The companies concerned did not appeal and the decisions became final. Pfleiderer then submitted an application to the BKA seeking access to the file (which in German law here meant the right of a lawyer to inspect the file). Pfleiderer stated that it had purchased goods worth more than 60 million euros from the companies concerned. The BKA replied by sending the three decisions imposing fines on the companies concerned, from which identifying information had been removed and a list of the evidence recorded as having been obtained during the search. Pfleiderer then sent a second letter to the BKA expressly requesting access to all the material in the file, including the documents relating to the leniency applications. The BKA partly rejected that application and restricted access to the file to a version from which confidential business information and internal documents had been removed. Pfleiderer then brought an action before the Amtsgericht Bonn challenging that decision. In February 2009, that Court ordered the BKA to grant Pfleiderer access to the file, in accordance with the relevant national provisions. The BKA objected and the Court then stayed enforcement and made the reference to the ECJ. The ECJ’s judgment in Pfleiderer The Amtsgericht asked whether the provisions of EU competition law, in particular Articles 11 and 12 of Regulation 1/2003 and the second paragraph of 32

Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161.

282  John Ratliff Article 10 EC (now Article 4(3) of the Treaty on the European Union, or TEU), in conjunction with Article 3(1)(g) EC were to be interpreted as meaning that, for the purpose of bringing civil law claims, parties adversely affected by a cartel may not be given access to leniency applications or to information voluntarily submitted by leniency applicants which a NCA has received, pursuant to a national leniency programme, within the framework of proceedings for the imposition of fines which are intended to enforce Article 81 EC (now Article 101 TFEU).33 In short, the ECJ said ‘no’. The EU provisions concerned did not preclude such access. It was for national courts to assess whether such disclosure was necessary and proportionate to protect a claimant’s right to damages, given the public interest in enforcement based on such leniency statements. More specifically, the Court noted that in the absence of a binding EU Regulation on the issue, it was for the Member States to establish and apply national rules on the access to leniency documents.34 The ECJ emphasised, however, that Member States must ensure that the rules which they establish or apply do not render the implementation of EU law impossible or excessively difficult. Specifically, in the area of competition law, they must ensure that the rules which they establish or apply do not jeopardise the effective application of EU competition law.35 The Court acknowledged that leniency programmes are useful tools to uncover and bring to an end competition law infringements and that their effectiveness could be compromised if leniency documents were disclosed to private damages claimants.36 Further, the Court noted that the view could reasonably be taken that the possibility of such disclosure would deter a person involved in a competition infringement from seeking leniency, particularly if that information might then be exchanged with other competition authorities.37 However, the Court also noted that an individual has the right to claim damages for loss caused to him by anticompetitive conduct (referring to Courage v Crehan38 and Manfredi39). According to the Court, the existence of such a right also strengthens the working of the EU competition rules and has a deterrent effect on potential infringers, so such actions are also important to the maintenance of effective competition in the EU.40 The Court therefore noted that, in considering access to leniency documents, the applicable national rules must not be less favourable than those governing 33 See ibid, para 18. Article 11 of Regulation 1/2003 provides for cooperation between the Commission and the competition authorities of the Member States, including the transmission of the most important documents in its file. Article 12 of the Regulation provides that such authorities have the power to provide one another with any matter of fact of law, including confidential information. Competition authorities may include courts. 34 Ibid, paras 20–23. 35 Ibid, para 24. 36 Ibid, paras 25–26. 37 Ibid, para 27. 38 Cases C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297. 39 Joined Cases C-295/04 to C-298/04, Manfredi v Lloyd Adriatico Assicurazioni [2006] ECR I-6619. 40 Pfleiderer, cited above note 32, paras 28–29.

Public and Private Enforcement: Implications for Courts and Agencies 

283  

similar domestic claims; and that they cannot operate in such a way as to make it ‘practically impossible or excessively difficult’ to obtain such compensation. Finally, the Court held that it is necessary to weigh the interests in favour of disclosure of the information against those in favour of protection of such information. This had to be done by national courts on a case-by-case basis, according to national law and taking into account all the circumstances.41 Discussion and subsequent events This was an immediately controversial ruling. Partly because the advocates for disclosure and those for non-disclosure were both disappointed, each hoping for a clearer rule in their favour. Partly because some argued that leaving national judges to decide whether to disclose leniency materials ‘on a case-by-case basis’ was problematic. Notably, it was argued that uncertainty alone would deter leniency applications, because sometimes a judge might allow access to all or part of a leniency application. The judgment also appears to have caught many by surprise, insofar as Advocate General Mazák had taken a different line, proposing a distinction between two types of documents, those prepared specifically as part of leniency and those which were pre-existing documents given to the competition authorities. AG Mazák would have denied access to the former, but allowed release of the latter.42 This was a view also favoured by many, as a reasonable compromise between the public and private interests concerned. Moreover, AG Mazák had argued in his Opinion that, while there is no de jure hierarchy between public and private enforcement, in his view public enforcement is of far greater importance than private actions in ensuring compliance. He stressed that it is the public cartel decisions which, in fact, allow private parties to bring damages claims in follow-on actions (there being few standalone actions and many more cartel cases since immunity/leniency incentives were introduced). In short, he suggested that making leniency statements accessible might be counter-productive to damages claims overall and the balance should be in favour of supporting the public procedure. It may be useful to quote what he said: In my view, the disclosure to civil litigants of the contents of voluntary self-incriminating statements made by leniency applicants, in the course of a leniency procedure and for the purpose of that procedure, in which the applicants effectively admit and describe to a competition authority their participation in an infringement of Article 101 TFEU, could substantially reduce the attractiveness and thus the effectiveness of a national competition authority’s leniency programme. This in turn could undermine the effective enforcement by the national competition authority of Article 101 TFEU and ultimately private litigants’ possibility of obtaining an effective remedy. Thus, while the denial of such access may create obstacles to or hinder to some extent an allegedly injured party’s fundamental right to an effective remedy, I consider that the interference with that right is justified by the legitimate aim 41 42

Ibid, paras 30–32. AG Mazák’s Opinion of 16 December 2010, paras 40–47.

284  John Ratliff of ensuring the effective enforcement of Article 101 TFEU by national competition authorities and indeed private interests in detecting and punishing cartels’.43

Again, many agree with this. The ECJ’s approach appears to be not that this appraisal of the importance of leniency is wrong, but that there may be cases where an absolute protection of leniency materials cannot be guaranteed (ie because, in a particular case, there is no other evidence available to bring the private claim to damages) and therefore the ‘rule’ as to whether leniency material can be disclosed has to drafted in a more measured way, so as to ensure that a private claimant’s right to damages is not jeopardised. It is also important to recall the context in which the ECJ had to rule – notably, where there is no binding EU Regulation on this issue and therefore national procedural rules apply. The Court was not asked ‘Should the leniency materials be disclosed’? Rather, it was asked ‘Does EU law preclude such disclosure under national rules?’ Hence the more qualified response than some people expected. Another key point in the judgment, which came out in our discussion at the workshop, is that the private right to damages set out in Courage v Crehan,44 even if it is a ‘fundamental right’, may be limited by the public interest in protecting leniency materials in the appropriate case (as indicated by the Court). It may be argued that the Court had sympathy for the protection of the leniency materials and qualified that only to the extent that it considered that there might be a case for access, in the event that it was ‘excessively difficult or manifestly impossible’ for the plaintiff to bring his action otherwise. Again, this was also well explained by AG Mazák. The Commission is now planning to propose legislation designed to protect leniency materials.45 It will be interesting to see how the Commission will take this balancing of interests into account. In other words, whether the Commission will suggest that there should be a general, blanket protection for leniency materials (because the uncertainty of case-by-case assessments is too harmful to leniency, as noted above); or whether the EU will simply provide that access to leniency documents should be only in the last resort after a national court has exhausted all other means, including specific discovery, in order to ensure that a plaintiff has a right to damages. In other words, echoing the high standard set by the Court with the ‘excessively difficult or manifestly impossible’ test, but structuring more what that means. Normally, one would expect the Commission to favour the former. The Heads of the NCAs have also made it clear that they do not want their leniency programmes undermined, with a joint resolution stating: Ibid, para 44. Cited above note 38. 45 See, eg, Alexander Italianer, ‘Public and private enforcement of competition law’, speech, 17 February 2012, Brussels, http://ec.europa.eu/competition/speeches/text/sp2012_02_en.pdf, pp 4–6; Joaquín Almunia, ‘Antitrust enforcement: Challenges old and new’, speech 12/428 of 8 June 2012, St. Gallen, http://europa.eu/rapid/press-release_SPEECH-12-428_en.htm. 43 44

Public and Private Enforcement: Implications for Courts and Agencies 

285  

‘[A]s far as possible under the applicable laws in their respective jurisdictions and without unduly restricting the right to civil damages, CAs [ie the NCAs] take the joint position that leniency materials should be protected against disclosure to the extent necessary to ensure the effectiveness of leniency programmes.’ 46

However, it may be noted that this formulation also leaves open how far the protection of leniency materials may go.

(ii) Applications of Pfleiderer before the national courts Since the ECJ’s judgment, and as of this writing, national courts have applied Pfleiderer twice. In both cases, the result was that the leniency materials were not disclosed, but the two courts’ approaches were somewhat different. Following the ECJ’s judgment, the Amtsgericht Bonn in the second Pfleiderer case applied the principles explained by the ECJ and chose to reject access to the leniency documents in the files of the BKA.47 The court held that such disclosure could undermine the effectiveness of the BKA’s leniency programme. Cartel members could be deterred from making leniency applications with selfincriminating information if private plaintiffs are later entitled to receive access to these documents. This could impede the effective detection and prosecution of cartels. In weighing the relevant interests in effective public and private competition law enforcement protected by EU law, the court ruled that the refusal to provide access to the leniency applications would not make it ‘practically impossible or excessively difficult’ for Pfleiderer to obtain compensation for the damages suffered. Pfleiderer would receive access to the non-confidential versions of all other documents in the BKA’s file, including non-confidential versions of documents seized during dawn raids (which took place before the leniency applications were filed). The Court also stressed that, under German law, German courts dealing with follow-on damage claims are bound by findings made in final decisions by cartel authorities. Accordingly, Pfleiderer would not need the leniency documents to show that a competition law infringement occurred, as it could rely on the BKA’s fining decisions. The issue of disclosure of leniency materials also came up in proceedings brought by National Grid in the UK, related to claims involving the Gas Insulated Switchgear cartel. In fact, National Grid raised the impact of the Pfleiderer judgment on disclosure in its case in London the very next day after the ECJ’s judgment.48 Interestingly, Mr. Justice Roth decided to invite the Commission to 46 Resolution of the Meeting of Heads of the European Competition Authorities of 23 May 2012, ‘Protection of leniency material in the context of civil damages actions’, http://ec.europa.eu/ competition/ecn/leniency_material_protection_en.pdf. 47 Case No. 51 Gs53/09, decision of 18 January 2012. See MLex, 30 January 2012. 48 MLex, 15 June 2011.

286  John Ratliff make observations on how he should proceed, which the Commission duly did in some detail.49 Mr. Justice Roth then gave an important judgment in which he examined, in a case-specific and detailed way, whether the documents requested should be disclosed and to what extent they should be redacted, taking into account the national procedural rules. He then allowed parts of the Commission’s confidential decision and certain passages from requests for information to be disclosed to the plaintiffs, while protecting material based on leniency. More specifically, he ruled that the ECJ’s Pfleiderer judgment applied to the Commission’s own leniency programme (as well as national leniency programmes)50 and conducted a balancing exercise, weighing the interest in disclosure against the need to protect an effective leniency programme.51 In assessing the proportionality of the disclosure request he considered in detail (i) whether the information concerned was available from other sources; and (ii) the relevance of the leniency materials to the issues in this case.52 Interestingly, Mr. Justice Roth considered that there were not other means available for National Grid to derive the information sought, ‘at least not without excessive difficulty’. He also accepted that much of the information sought might be relevant to National Grid’s case, but found that was not clear for all the documents sought. He went on to say: The mere fact that it is common ground that according to the relatively low threshold for standard disclosure these documents would otherwise fall to be disclosed is not determinative when there is a powerful countervailing factor to be weighed against disclosure. It is necessary to ascertain whether the particular documents or parts of the documents really are of such potential relevance that specific disclosure should be ordered. Just as a blanket objection to disclosure on the grounds that this might prejudice the Commission’s leniency programme cannot, without more, be accepted, so also would it be wrong to permit disclosure of the entirety of the leniency materials sought without closer examination53

He then inspected the documents himself and ruled that the confidential version of the decision could be revealed to those in the ‘confidentiality ring’ in the case, although he ruled that certain passages should be redacted. As for the other documents sought, which appear to have included responses to Commission requests for information, he ruled that limited passages should be disclosed, for the most part providing explanations of the pre-existing documents supplied. Other leniency materials were not disclosed, on the basis that they were not of such relevance to the proceedings, or because the interest in non-disclosure outweighed the interest in disclosure.54 49 National Grid Electricity Transmissions Plc v ABB Limited & Others [2012] EWHC 869 (Ch). Both the European Commission’s observations and the judgment are available at http://ec.europa.eu/ competition/court/antitrust_amicus_curiae.html. 50 Ibid, para 26. 51 Ibid, paras 30–60. 52 Ibid, para 39. 53 Ibid, para 52. 54 Ibid, para 59.

Public and Private Enforcement: Implications for Courts and Agencies 

287  

(iii) Comment At our workshop in 2011, I had already suggested that the overall enforcement priority should be to bring companies to the Commission’s door to tell their story, since immunity/leniency has been a key feature of modern anti-cartel enforcement. The older cases, without the benefit of such corporate statements, appear to have been much harder to bring and there were clearly less of them. Notably, since the Commission introduced the first Leniency Notice in 1996, the number of cartel cases decided by it has substantially increased: from 20 cases decided between 1990 and 1999 to 63 cases between 2000 and 2009.55 These figures speak for themselves in suggesting that, at least for now, immunity/leniency programmes should be protected. All the more so as the number of cartels being revealed suggests that continued enforcement in this area has to be a high priority. I therefore favour protection of the corporate immunity/leniency statement and any related material (ie answers to questions provided as part of the immunity/ leniency applicant’s duty to cooperate with the Commission) to the maximum extent possible so that, as far as possible, applicants are not placed in a worse position than other members of the cartel through their cooperation.56 Fear of fines and in some systems criminal sanctions is huge, especially given the scale of fines in recent years57 so, even if there is an increased risk of follow-on claims, most companies may still be expected to go to the competition authorities to seek immunity/leniency.58 However, as noted in Pfleiderer, it may be that some will be deterred from approaching the Commission.59 It may also be reasonably argued that immunity/ leniency applicants should not be singled out as the easy target to focus on for private claims.60 It will be interesting to see how legislation on these issues will fare and whether the Commission will wait for the EnBW judgment at the ECJ before putting forward its proposals. In the meantime, we must expect further European Court cases on these ‘access to file’ issues. At the time of writing, there are at least two more cases coming (apart from the ECJ’s judgment in EnBW), namely, Case C-536/11, Donau Chemie and Others61 and Case T-380/08, The Netherlands v European Commission (Dutch Bitumen).62 Another aspect of National Grid is also being considered by the Court of Justice. Mr. Justice Roth also sought copies of Alstom and Areva’s Responses to the SO in the Gas Insulated Switchgear cartel case, which the Commission The statistics can be found at http://ec.europa.eu/competition/cartels/statistics/statistics.pdf. In the same vein, see the OECD Roundtable, cited above note 1, UK Report, at pp 349–350. 57 See http://ec.europa.eu/competition/cartels/statistics/statistics.pdf. 58 See the discussion in the ICN Report, cited above note 1, at pp 43–44. 59 See Case C-360/09, cited above note 32, paras 26–27. 60 See the ICN Report, cited above note 1, pp 43–46. 61 2012 OJ C13/5. See now AG Jaaskinen’s Opinion of 7 February 2013. 62 2008 OJ C285/51. 55 56

288  John Ratliff decided to provide, but those parties appealed and that decision is currently, in part, suspended.63 As we discussed in our workshop, it would be good to see courts in the meantime using their specific discovery powers more. In other words, it would be good to see them following the Amtsgericht Bonn and Mr Justices Roth’s line: not ordering disclosure of the leniency materials; but, in a practical way, ordering disclosure of existing documents and other specific materials relevant to the plaintiff’s claim.

4. Confidentiality A related, parallel issue to this debate about access to file is the amount of information in the Commission’s actual decision, which it has to publish.64 As private actions have grown, this has also become a much more contested issue. Some of the older decisions are quite detailed, but still with ‘redactions’ for confidentiality, meaning business secrets. More recent texts in cartel cases are often shorter and may be heavily ‘redacted’ as to the outline of the facts, if the relevant text is heavily based on immunity/leniency statements, because then, as already explained, in principle the Commission will not disclose passages based on immunity/leniency statements. The amount of information in a decision is reduced even more in settlements, because then the Commission only publishes a short summarising decision.65 In other words, debate over the content of the decision mirrors that over access to file. There are mixed views on this, depending on whether you are on the plaintiff or defence side. Plaintiffs would like as much information as they can obtain, while the defence argues that, provided the essential reasoning of a decision is set out, both the confidential elements and those based on leniency should be deleted. We should be clear that when people speak of a need for ‘confidentiality’ here there are usually two separate issues: First, ensuring that business secrets are not disclosed (ie Company X’s prices and commercial terms for supply to a customer) and second, protecting a company’s disclosures in its immunity/ leniency applications, as discussed above. There is consensus that the first type of confidentiality should be protected, where the information is still a business secret. The controversy is about the second type of claim and more specifically about whether that disclosure is contrary to the ‘commercial interests’ of the company in question (as provided for under the EU Transparency Regulation). The current position is that the GC has stated in EnBW and CDC Hydrogene Peroxide that it does not consider that protecting cartelists from damage claims 63 See Case T-164/12 R, Alstom v Commission, Order of 29 November 2012, not yet reported. See also MLex, 5 December 2012. 64 See Article 30 of Regulation 1/2003. 65 See generally the section on settlements on the Commission Competition page on its website, http://ec.europa.eu/competition/cartels/legislation/cartels_settlements/settlements_en.html.

Public and Private Enforcement: Implications for Courts and Agencies 

289  

is a commercial interest deserving of protection,66 although the Commission is appealing. It would appear that position also has to be assessed against the position of the ECJ in Pfleiderer that leniency materials are worthy of protection (because their disclosure may deter others from coming in to the competition authorities in the future), provided that denying their disclosure does not make the plaintiff’s task in bringing a damages claim ‘practically impossible or excessively difficult’.67 Again, it will be interesting to see what happens in EnBW and in the other coming judgments on related issues.

5. To what extent should a competition authority’s decision bind a court? We should now turn to consider the following sources, each of which bears on the public/private enforcement relationship: the Masterfoods judgment of the Court of Justice;68 Article 16 of Regulation 1/2003; and the Enron case in the United Kingdom.69 Masterfoods makes it clear that national courts cannot adopt judgments ‘running counter’ to a Commission decision.70 Article 16 carries that over into Regulation 1/2003. Enron is an English Court of Appeal judgment. The Court of Appeal found that, even if Enron had been found by the Office of Rail Regulation (ORR) to have been discriminated against, contrary to what is now Article 102 TFEU and its English equivalent, Enron was limited in the findings of fact on which it could rely to those which were ‘a clearly identifiable finding of fact to a given effect’ in the ORR’s decision, not just ‘passages from which a finding of fact might arguably be inferred’. Further, the Court of Appeal held that courts could not ignore direct relevant evidence before them on an issue which may not have been the focus of the ORR’s decision.71 That meant, in practice, that Enron still had to show causation, ie that even if it had been discriminated against, it had not lost a contract in a bidding process for other reasons. This is an interesting judgment, and it was initially controversial. Did it not undermine the Masterfoods principle? However, common sense would suggest 66 CDC Hydrogene Peroxide, cited above note 22, at paras 47–50; EnBW cited above note 24, at paras 147–148. 67 See Pfleiderer, cited above note 32, at paras 25–27 and 30. 68 Cited above note 9. 69 Enron Coal Services Ltd v English Welsh & Scottish Railway Ltd, [2011] EWCA Civ 2. 70 See also Sections 47A and 58 and 58A of the UK Competition Act 1998 as amended; Section 33(4) of the German Act Against Restraints of Competition; and Iberian UK Ltd v BPB Industries, [1996] 2 CMLR 601, where Mr Justice Laddie held that a decision of the Commission finding an Article 102 TFEU infringement (then Article 86 EC) could be relied on by third parties as proof of the infringement concerned. 71 See Lloyd LJ at para 56.

290  John Ratliff otherwise. The Court of Appeal was not saying that a competition authority decision should be ignored. It was saying that the decision had to be clear in its findings and relevant to the issues before the national court. The Court of Appeal was also making a point similar to that explained above in Section 1: There might be a finding of breach of Article 102 TFEU through discrimination and yet no effect in fact. Another question raised by the judgment is whether this sort of ruling should mean that authorities like the Commission should do more, checking and formulating their findings of fact for the private parties’ follow-on claims, or even proving effect? I do not think so, for the reasons already discussed above. We come back to the principle that the competition authority’s task and priority is to ensure public enforcement of the competition rules, as best it can. Typically the competition authority already aims to say whatever it reasonably can as regards the infringement. However, it will usually not do more, if not necessary for its purposes, because its own decision has to withstand scrutiny on appeal and it will not want that decision put in jeopardy by going further than it has to. Clearly a plaintiff does not like this and finds it frustrating if there are not the sort of findings in an Commission decision which would make its case easier, especially if otherwise it may be difficult to find the proof required. However, there is also a cost-benefit analysis in public enforcement. The work required to prove specific effects in cartel cases for particular defendants would be considerable and, as noted by the Commission itself in Citric Acid above, sometimes very difficult, if not impossible. I do not think that public enforcers should be involved in doing that ‘in the general interest’, rather than tracking the next cartel.

6. Should immunity applicants also be protected in follow-on claims? Some have suggested that the successful immunity applicant should also obtain some immunity from follow-on damages, provided that it cooperates with the plaintiffs. This appears to be an amended import from recent US law, under which the successful immunity applicant is only exposed to single damages, rather than treble damages, if he cooperates.72 The idea is to increase further the incentives of companies to seek immunity, while assisting the subsequent private enforcement. A European version of this is to be found in the Hungarian Competition Act, which provides for the successful immunity applicant to pay damages only if the follow-on cartel plaintiffs have not obtained full compensation from the other jointly and severally liable defendants.73 72 See the US Antitrust Criminal Penalty and Reform Act 2004, Pub L No.108–237, §215, 118 Stat. 665, 668 (2004), summarized in the OECD Roundtable, pp 359–360. 73 Section 88 D of the Hungarian Competition Act, 1996 as amended. See Csongor István Nagy,

Public and Private Enforcement: Implications for Courts and Agencies 

291  

Both variants are complex. Both were introduced by legislation. Some question the constitutionality of shielding a ‘tortfeasor’ from damage to his/its victims.74 Others argue that, but for the immunity applicant’s initiative, there would be no detected tort to sue for in the first place. There is also a material difference between the US and Hungarian models. In the US version, the special trebling of damages is only ‘de-trebled’; ‘single damage’ liability remains. By contrast, in the Hungarian version, the successful immunity applicant only pays if his/its fellow cartelists do not. The liability is still there, but it may not be called on. There are also some delicate issues and tensions in the cartelists’ positions. One would think the defendant should be obliged to cooperate, in the sense of providing information, but should not have to settle, since it may still not concede causation and quantum. In the Hungarian model one may also think that the ‘other cartelists’ may have less incentive to settle, because that way the immunity applicant may have to pay, in all or part.

7. Compensation as mitigation Some suggest that instead of fines, instead of paying all of a fine, defendants could offer compensation to victims.75 The general idea is that, in practice, everything paid by the defence adds to deterrence, so why not pay part of the fine into a fund held for the cartel victims, either giving all those identified a flat payment or, with suitable publicity, giving all those potentially affected time to make their claim and, if it is not claimed within say two or three years, pay the balance to the Commission? It is argued that such a system would be more likely to result in compensation going to victims and more promptly. It is also suggested that this would not prevent follow-on litigation, if the amount offered did not appear adequate, but it might do so in fact, if realistic figures were put forward. All this would be without prejudice to the Commission’s general position that fines are about punishment and deterrence, while private claims are about ‘The new Hungarian rules on damages caused by horizontal cartels: presumed price increase and limited protection for whistleblowers – an analytical introduction’, 32 European Competition Law Review 63 (2011). It appears this was introduced because of the Hungarian rule establishing a presumption in cartel cases that they resulted in a 10% price increase (Section 88 C), so as not to deter immunity applicants. Interestingly it appears that in German law an immunity applicant may also not be the subject of the Bundeskartellamt decision, which it is argued may assist its position in any follow-on claims. See the OECD Roundtable, cited above note 1, at p 307. 74 See the Comments of the Association of European Competition Law Judges on the Commission’s White Paper on Damages Actions, 2008, at para 18. 75 See Jacques Bourgeois and Stéphanie Strievi, ‘EU Competition Remedies in Consumer cases: Thinking Out of the Shopping Bag’, 33 World Competition 241 (2010); Wils, ‘The Relationship’, cited above note 1, at pp 24–26. This paper does not address Article 9 settlements, where in principle fines are not in issue and there is no admission of liability, but in practice the willingness to settle may be influenced by both the risk of the Commission switching to an Article 7 decision with a finding of liability and then the risk of follow-on damages claims.

292  John Ratliff compensation and corrective justice. In essence, it is said that it would be better to apply at least some of the fines to prompt consumer relief now, rather than paying them into the EU budget and leaving the parties to fight the long battle through the current landscape for private claims. Such payment is also part of the deterrence for defendants, simply because they have to pay it. These are interesting ideas. Mainly because, as noted above, private claims take so long, and because, although there are far more private claims than before, many issues are still being debated and/or uncertain, so that the prospects of success are still unclear. This is especially so for indirect purchasers with more complex causation and quantum issues, or in cases of many small claims. For these types of claim, one would think that this route might have particular attractions. It may also be an interesting idea for defendants, insofar as it might allow them to clean up the cartel damage more quickly, rather than facing extended and expensive follow-on litigation. Especially if it meant that the sum of fines, damages and legal fees may also be less because there might be some discount for paying early, before any causation and damage has actually had to be shown. Defendants may also like the idea of avoiding the accumulation of interest and legal fees, which can be a very significant issue, especially if, as now, claims take a long time to be litigated.76 On the other hand, one may doubt that defendants would like plaintiffs to have two ‘bites of the cherry’. In other words, if you are going to be sued anyway, why concede anything and pay into a fund? It may be that you would prefer just to wait and deal with it all later in any litigation. Clearly, defendants will also expect that what is paid into a fund should be given adequate credit in the overall fine, not some small, token reduction. Ideas like this have been tried and accepted by the Commission sometimes, so apparently a new legal base is not required. The main examples are Pre-insulated pipe77 (a cartel case) and Nintendo78 (a parallel import case). Sometimes this has been part of an apparent strategy to wind up a case without admitting liability, sometimes with liability.79 It may also be interesting to note that, at paragraph 18 of the European Competition Authorities’ Principles of Convergence on Sanctions, it is suggested that ‘[t]he applicable fine may be reduced if the offender takes active steps to mitigate the adverse consequences of the infringement, in particular by providing voluntarily, timely and adequate compensation to those who have suffered damage as a result of it’.80 76 See Clifford Jones, ‘A New Dawn for Private Competition Law Remedies in Europe? Reflections from the US’, in Ehlermann and Atanasiu, eds., Effective Private Enforcement, cited above note 3, pp 95 et seq, at pp 103–105. 77 1999 OJ L24/1, para 172. 78 2003 OJ L255/33, paras 440–441. 79 See also Rover, Press Release IP/93/993 of 16 November 1993, where a fund was set up in a resale price maintenance case, leading to an informal settlement with the Commission. See also Commission, XXIIIrd Report on competition policy (1994), para 228. 80 However (and as Wils observes in ‘The Relationship’, cited above note 1, at p 20, footnote 100), paragraph 18 of the Principles also states that, ‘[w]here compensation is taken into account as a

Public and Private Enforcement: Implications for Courts and Agencies 

293  

There remains the issue that the Commission may not want to be obliged to commit resources in ‘trustee monitoring’, or assessing the adequacy of compensation offered, again because that would be burdensome, too difficult and out of line with the focus of its own investigation, as discussed above. But perhaps a flat rate system could be found, whereby the Commission would allow say 20% off its fine, if a compensation scheme was set up by the defence before the Commission decision or within three months after it, which would meet certain defined access, quantum and payment criteria. Finally, the idea that a company would have to pay the current levels of fine and huge amounts of damages appears to go too far. While I can see the current argument that, while damages claims are uncertain, fines should be high, I think fines should be reduced if compensation is paid. Punishment and deterrence is partly accounted for in such a case by the compensation paid. Part of the interest of the fund idea is that it seeks to address that uncertainty of recovery in a prompt and practical way. That should result in some reduction in the public fine.

Conclusions The main ‘takeaways’ I have from this review and discussion are the following: – The interplay of public enforcement and private compensation claims could not be more topical at this stage in the evolution of EU competition law. – The two types of action are different, but complementary. – Given the difficulties of detecting cartels without public immunity/leniency programmes, they must be protected to the maximum extent possible. – It is clearly important that the right to damages be respected, but one should not forget that, at present, it is public enforcement which is permitting private claims in the first place. – We are seeing these issues come up in many different ways before the European and national courts now all the time and, as a result, the law is being clarified fast. – It will be interesting to see if this rapidly evolving area of law also leads to EU legislation, protecting the immunity/leniency process generally and facilitating a common approach to the related disclosure issues across the whole EU. – However, these proceedings all take too long, which has implications for the process and the outcomes, and merits further consideration. One idea to be considered further is whether some type of suitably incentivised damages settlement system could be devised to speed things up. mitigating circumstance, this reduction should not in any case be such as to undermine the deterrent effect of the fine.’

Ian S Forrester and Mark D Powell*

Market Forces and Private Enforcement: A Start but some Way Still to Go

The successes of the European competition regime have been remarkable. Competition disciplines have been accepted by Member States (reluctantly and unevenly, but overwhelmingly), national judges and business leaders. In contrast to the United States, where public enforcement was, in the 1980s (as it is today), a small fraction of the legal effort deployed to pursue compliance with the antitrust rules, the efforts of the Commission were relatively more important. It was the Commission which drafted the block exemptions, which set its enforcement priorities, which chose which law to make via its rare flagship decisions, which imposed penalties, and which spoke up for the principle of competition inside the deliberations of the European institutions. Since the case of BRT v SABAM,1 it was known that Articles 85 and 86 of the Treaty of Rome had direct effect and could be invoked in national courts. However, there was an unreality which muddled the process. The prohibition of Article 85(1) was deemed to have a broad, extensive scope. It caught (according to the Commission’s conservative choice of enforcement priorities in the early 1960s) both basically good agreements which had restrictive features but achieved and pursued a pro-competitive goal, and basically bad agreements which could not be redeemed. The category of former agreements was vast. Tens of thousands of agreements signed each year containing exclusivity provisions were in theory caught by Article 85(1), rendered null by Article 85(2) and even eligible for fines. If they did not fit the narrow confines of the available block exemptions, the theoretically indispensable remedy against unenforceability and fines was the filing of a notification seeking an exemption. In theory, exemptions would be granted on their merits. In reality, exemptions were almost never granted in specific cases. Specific exemption decisions were reserved for interesting and important cases where the Commission wished to make new law (or pursue soft targets, like trade fairs, which could be good for the statistics).2 The Commission was the only entity which could grant * Respectively: Queen’s Counsel at the Scots Bar and Honorary Professor, University of Glasgow; and Solicitor, England & Wales; both of White & Case, Brussels. Warm thanks are expressed to Joanna Cohen, Katarzyna Czapracka, Strati Sakellariou, Morris Schonberg, Calum Warren and the other colleagues who have cheerfully shared their knowledge. The opinions expressed are wholly personal. 1 Case 127/73, BRT v SABAM [1974] ECR 313. 2 Among the trade fair and bourse cases were decisions on office equipment (VIFKA, 1986 OJ L291/46); freight futures (Baltic Freight Futures Exchange Limited (BIFFEX), 1987 OJ L222/24);

296  Ian S Forrester and Mark D Powell exemptions. Courts could not do so. As a result, a lot of legal and procedural creativity was deployed: national courts preferred to find that Article 85(1) was inapplicable, to avoid ceding jurisdiction to the Commission. This history is still relevant today. There is a link between procedural and substantive law. National courts will seek to escape the constitutional consequences of European competition law if the consequences of strictly applying it seem unpalatable. If courts feel that the procedures strictly applicable can lead to unjust results, the substantive law will be adapted. It is not a coincidence that in the United States, while dire procedural entanglements await those credibly accused in the civil courts of antitrust infringements, the courts are conservative, indeed reluctant, in recognising new categories of antitrust infringement. Intellectual creativity by plaintiffs and enforcers is better received in Brussels (that said, national judges are also ready to reject novel theories of harm). Because of the primary role of the Commission, great (maybe excessive) weight was in the past attached to any indication of its opinion on the merits, and the Commission was correspondingly exceedingly cautious in expressing any view to assist the courts in applying EU law in national litigation. It developed the ‘comfort letter’, which was less than an exemption but more than a silence in response to a notification. During this period, ‘private enforcement’ before national courts was unquestionably hindered by the intellectual and procedural conundrum at the centre. There were a few national cases, but their prosecution was dominated – to an undesirable and even perverse degree – by questions along the lines of ‘What would the European Commission say about this case?’. Since there were almost no positive decisions affirmatively recording the acceptability of some arrangements, the law developed on the basis of condemnations. There were curiosities such as Camera Care3 (a libel action arising out of an accusation of competition infringements), and Garden Cottage Foods,4 concerning ‘breach of statutory duty’ as a means of characterising competition law infringements in national law. While it was not seriously doubted that damages should be available in principle,5 it was a theoretical, academic, possibility rather than a practically relevant opportunity. There arose cases of tension between the determinations of national judges and European proceedings. The handling of national litigations unfolds according to quite well-established procedures, according to a predictable timetable. dentistry equipment (International Dentalschau, 1987 OJ L293/48 and British Dental Trade Association, 1988 OJ L233/15); and machine tools (CECIO/EMO, 1989 OJ L37/11). Case 792/79 R, Camera Care Ltd. v Commission [1980] ECR 119. Garden Cottage Foods v Milk Marketing Board, [1983] 2 All ER 770 (H.L.). John Temple Lang, then of the Commission Legal Service, was an assiduous advocate of the proposition that national damages for breaches of European competition law were competent and desirable as well. See Temple Lang, ‘EEC Competition Actions in Member States’ Courts – Claims for Damages, Declarations and Injunctions for Breach of Community Antitrust Law’, 7 Fordham International Law Journal 389 (1983). 3 4 5

Market Forces and Private Enforcement  297   Commission proceedings were far from predictable. Decisions were rare events, sometimes as few as ten per year, reserved for flagship cases where the Commission wanted to advance the law. Indeed, this unpredictability made judges reluctant to cede responsibility to a public authority which was notably slow. Although the issues at stake were not as serious as today, and although there was more acceptance of the Commission’s role as defender of the faith, it was evident that a Commission decision was more of a prosecutorial than a judicial document. And since in many cases there was likely not to be a decision, an excessive amount of energy was devoted to arguing propositions such as ‘If the Commission were to decide, it would find in my client’s favour’. And when they did emerge, decisions would often lack the judicious review of both sides of the argument which a court would normally deploy. The textual nature of many infringements was ill-adapted to resolving genuine debates about truly difficult competition law problems, since there was a gap between the infringement and the remedy. In Kerpens,6 the buyer sought to evade paying a debt on the ground that an export ban had been imposed. Some claims were made on the back of Commission decisions:7 when the judgment by the ECJ was pronounced in Magill,8 there were literal cries of relief from Magill’s representatives, who knew they would be indemnified for the snuffing-out of their television guide.

The problems of Mr Crehan There was one celebrated court case where the ‘victim’ believed he and others like him had an excellent case both in law and fact. But it illustrates many of the problems of private enforcement. Courage v Crehan9 presented a number of fundamental questions. The Commission had for years favoured the opening up of national beer markets, by imposing limitations on the extent to which brewers could compel that public house tenants must sell only the landlord’s beers. The battles on ‘tied houses’ went back to Brasserie de Haecht,10 and a separate case on the impact of a network of similar deals (Delimitis).11 So the policy and the 6 Case 319/82, Société de Vente de Ciments et Bétons de l’Est SA v Kerpen & Kerpen GmbH und co. KG [1983] ECR 4173. 7 Commission Decision 89/205/EEC of 21 December 1988 in Case IV/31851, Magill TV Guide/ ITP, BBC and RTE, 1989 OJ L78/43. 8 Joined Cases C-241/91P and C-242/91 P, Radio Telefis Eireann (RTE) and Independent Television Publications Ltd (ITP) v Commission [1995] ECR I-743. 9 Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297. On this judgment, see Assimakis Komninos, ‘New Prospects for Private Enforcement of EC Competition Law: Courage v Crehan and the Community Right in Damages’, 39 Common Market Law Review 447 (2002). 10 Case 23/67, Brasserie de Haecht v Wilkin [1967] ECR 127; Case 48/72, Brasserie de Haecht II [1973] ECR 77. 11 Case C-234/89, Delimitis v Henninger Bräu [1991] ECR I-935.

298  Ian S Forrester and Mark D Powell pro-competitive goal of the Commission’s interpretation of the law were clear. But what were the practical consequences for a public house tenant, Mr Crehan, who claimed that the contractual terms imposed on him were anticompetitive, and ineligible for an exemption, such that he deserved compensation?12 Mr Justice Park has described the experience of hearing the parties in the case of Crehan v Courage. The plaintiff was an honest man who had made a bad investment of lottery winnings in an English public house. The pub withered and he lost his money. Was that loss caused by the fact that there were discrepancies between the requirements of the relevant block exemption on exclusive purchasing agreements13 and the contracts of the Courage Brewery? There was evidence from beer drinkers, accountants, window cleaners and others as well as the plaintiff. However, the claim of Mr Crehan turned out, properly analysed, not to be wellfounded in the eyes of the judge. Contrary to his claims, the beer market was not foreclosed in the sense required by the Delimitis doctrine. Yes, there were contractual burdens on Mr Crehan, and yes, he was honestly trying to make his pub succeed. But, essentially his failure was not attributable to a breach of the competition rules but to a badly conceived business model which meant his beers were too costly. The Commission, by contrast, after several years of anxious reflection, had previously conveyed that the particular beer supply agreements were caught by Article 85(1), but might perhaps be eligible for an exemption; but then the Commission had second thoughts about foreclosure, and finally issued a cautious comfort letter which did not settle the situation of Mr Crehan. Now, do we regard the case as proving that the law is defective? Mr Crehan was an honest man who had entered a contract with a much larger business partner, and could have been entitled to hope that the contract would fit the strictest requirements of European competition law. He had taken a big chunk of savings and lottery winnings, invested them, and lost them. He had gone through the winnowing process of convincing his chosen lawyers, dealing with budget problems, coping with the stress of the trial, facing the miseries of defeat and, even worse, protracted litigation14 about all the rich and varied subtleties the case presented constitutionally. The protraction was indeed prodigious: some ten years with trips to Luxembourg and the House of Lords. Surely he deserved something? Does the story show that he should have gone for a settlement, encouraged by the Commission, and extracted something from the mess? Or, do we regard the story as perhaps sad in human terms but sound in terms of legal analysis? We suspect that if today Mr Crehan were to come to Brussels and persuade the Commission to investigate the complaint, the outcome might be different: the Inntrepreneur Pub Company (CPC) et al. v Crehan, [2006] UKHL 38. At the time, Commission Regulation (EEC) No. 1984/83 on the application of Article 85(3) of the Treaty to categories of exclusive purchasing agreements. 14 Courage v Crehan, cited above note 9; Bernard Crehan v Inntrepreneur Pub Company Limited and Brewman Group Limited [2003] EWHC 1510 (Ch); Bernard Crehan v Inntrepreneur Pub Company CPC [2004] EWCA 637; Inntrepreneur Pub Company (CPC) et al. v Crehan [2006] UKHL 38; Crehan v Inntrepreneur Pub Company Ltd and another [2007] EWHC 90081. 12 13

Market Forces and Private Enforcement  299   Commission would likely have sought to persuade Courage to come to a financial settlement as the price of a global peace with the Commission, while warning Mr Crehan that he should not be too optimistic about his own case. Be that as it may, courts apply, and ought to apply, the same rigour to claims about road accidents, medical negligence, construction disputes and competition problems. The 2006 gathering in this series heard the story of the man who complained to DG IV about a cartel which put him out of business, but who received a letter from DG IV Director Colin Overbury to the effect that resources did not permit the enquiries which would be necessary to pursue his allegations. Some years later, the cartel was condemned by the Commission, and some time after that the ‘victim’ started the process of seeking compensation.15 So far as we can know, he was not successful. In the meantime, competition law has been decentralised from Brussels, national enforcement is a reality, the Commission’s primacy is even more assured, and over a hundred countries have a competition law, often modelled on Europe’s. It is not surprising that the Commission aspires to see a world where ‘private enforcement’ functions as vigorously as it does in the United States. At the same time, the Commission wishes to avoid American supposed excesses: the lawyer not as renderer of a service for a fee, but as a privateer16 incentivised by the chance of treble damages; class actions where the class is overly broad; discovery pursued for the purposes not of achieving enlightenment but inflicting pain and inconvenience;17 and other perceived excesses which are described in the literature and commented on by other papers submitted to this gathering.

Shift in attitudes Now, what can be observed about the European legal marketplace to offer guidance on the integration of public and private enforcement of competition law in Europe? It used to be that large corporations were unlikely to litigate against an entire economic sector, and it was equally unlikely that the European law firms which advise such corporations would lightly turn to being ‘plaintiffs’ lawyers’. In the United States, by contrast, there is a large and prosperous plaintiff bar, one section of which routinely pursues competition claims, often on the back of public enforcement actions. It cannot have surprised anyone that these firms would set 15 See Claus-Dieter Ehlermann and Isabela Atanasiu, eds, European Competition Law Annual 2006: Enforcement of Prohibition of Cartels, Hart Publishing, 2007, at pages 38 and 162–163. 16 A warship which was privately owned but publicly licensed and sought plunder for its owner from enemy shipping. 17 We note in passing that discovery’s pains might not flow in only one direction. It certainly forces the defendants to reveal possibly embarrassing details of past conduct; but it can also permit defendants to confirm that plaintiffs passed on the ‘cartellised’ price to their customers.

300  Ian S Forrester and Mark D Powell up shop in London. England offers discovery, the convenience of the English language, familiar though not identical court procedures, and an expert local bar. Conditional fees are permitted for the moment, such that in the event of success, solicitors can effectively double their hourly rate. It is not sure that this rule will remain, and double rates are not as tempting as a percentage of treble damages, but it is a significant incentive. While English courts award costs against the losing party, and such costs can be immense, the new class of lawyer entrepreneurs has created an attractive prophylactic mechanism: insurance. If the would-be plaintiff loses, an insurance policy, paid for by the law firm, will indemnify the loser. As a result of these various phenomena, even large and sober enterprises are liable to conclude that bringing litigation seeking competition law damages may be not merely conceivable but even required as a matter of duty to shareholders. Other routes to litigious opportunity also exist: there exists a group of companies named CDC (Cartel Damages Claims) which is in the business of pursuing competition claims in a number of fora, including Germany, Finland and the Netherlands. The German courts have recently found that its activities are not contrary to public policy.18 There is indeed an emerging ‘claim investment marketplace’ for those who wish to place money in funding promising litigation where the plaintiffs may lack the resources or the will to carry the case forward. There are specialist brokers, lawyers and other market-makers. Companies with names like Juridica Capital are listed on ‘alternative investments’ bourses. They have survived challenges based on champerty19 and barratry,20 old themes of public policy, and seem to be rewarding investors with attractive returns. The cost of insurance against losing and being ordered to bear ruinous costs is proving less expensive than one would have expected, as most cases settle. And the business model favours moderate settlements rather than stunning victories.21 Insolvency Management Fund Limited (IMF) started out in Australia by funding litigation by bankruptcy trustees who enjoyed statutory authority to sell off estate assets in order to fund measures to recoup assets which had been improperly alienated by the bankrupt estate. In exchange for 30% of the total settlement, the company offers to plaintiffs security against being on the losing side as well as covering their own lawyers’ fees. Comparable initiatives emerged in the UK, Germany, Switzerland and the Netherlands, so that litigation finance can be found on either the plaintiff or the defence side. In the former case, if there is no recovery, the financier loses. The terms will reflect the probability of settling or winning: the financier will be ready to invest where the cost is lower than the realistic percentage chance of winning the realistic value of the claim. In the case BGH, judgment of 7 April 2009, Case No KZR 42/08. ‘The illegal proceeding, whereby a party not naturally concerned in a suit engages to help the plaintiff or defendant to prosecute it, on condition that, if it be brought to a successful issue, he is to receive a share of the property in dispute.’ 20 ‘The offence of habitually exciting quarrels, or moving or maintaining law-suits; vexatious persistence in, or incitement to, litigation.’ 21 Further details of these developments can be found on the websites of specialist law firms such as Linzer & Associates, PC. 18 19

Market Forces and Private Enforcement  301   of a defence, the financier will purchase a share of the claim being defended in exchange for a share of the damages minimised by good lawyering. For example, a skilled motion for summary judgment may be a lucrative investment for a litigation financier. From the client’s point of view, such an arrangement can help contain the legal fees while getting good legal services: the financier will pay the legal fees on a sliding scale depending on the amount of the damages awarded (or minimised, depending on one’s viewpoint). The financier will prefer cases with few defendants as opposed to many defendants, cases which are straightforward as opposed to factually complex, and plaintiff clients who want a settlement rather than a public triumph. Thus there are numerous viable alternatives to the classical route of seeking a lawyer, launching a claim and litigating matters to a conclusion four years later. These changes, and new services and new business models, do not at all depend on legislation or other initiatives by antitrust enforcement agencies like the Commission.

There is a competitive marketplace for resolving disputes There is competition as between different countries to resolve competition law disputes, and between alternative procedural routes within a single country. Thus in England and Wales litigants currently have a choice between the Competition Appeal Tribunal (CAT) and the High Court. ‘Follow-on’ claims, brought in reliance on a previous infringement decision by the Commission or the national competition authority, may be brought before the CAT, while both ‘follow-on’ and ‘stand-alone’ claims may be brought before the non-specialist High Court. The CAT has narrowly construed the limits of its jurisdiction in relation to ‘follow-on’ claims, with the result that the High Court is emerging as the preferred forum for all competition law claims in England and Wales. For the moment, the High Court has a narrow but perceptible lead over the CAT. Most national court systems have handled competition law claims. The English courts appear to be the present market leader, but they have competitors: the Netherlands and Germany have efficient court systems which are much cheaper than those of England. On the other hand, England is keeping up its lead by addressing briskly a number of fundamental problems. Our colleagues at White & Case in Brussels prepared a chart of litigations pending before the UK courts: we felt somewhat diffident about producing such a list, as it was sure to be wrong or incomplete. However, it seems to have been a useful initiative, more right than wrong, and others present favoured the idea of recording how many actions are currently pending and who is involved. So the chart entitled ‘Follow-On Damages Claims in the UK’ (see pp 310–13) contains a non-exhaustive list of competition claims brought before the English courts (either the High Court or the

302  Ian S Forrester and Mark D Powell CAT), together with the law firms involved in the various proceedings (additions and corrections to the chart are welcome and are being supplied). A number of conclusions can be drawn from this table. Firstly, we may note the preponderance of follow-on damage actions being brought in the UK. The English courts have become a convenient forum in which to bring such actions, as their lawyers have become comfortable with such proceedings. Claims have expanded territorially: the recent case of Ryanair v ExxonMobil22 is based on a competition infringement decision not of the Commission, but the Italian Competition Authority. This is indeed a novel development. Secondly, it is clear that settlement should be, and is, a realistic goal in the minds of claimants when embarking on such litigation, and one which distinguishes the European pattern from the casino atmosphere which may prevail in the US, notably Texas, where jury verdicts have been remote from any economic reality (and where immense verdicts benefit the local economy). Finally, it is interesting to observe that in a number of cases, manufacturers and customers have lined up against each other. Thus we have seen paraffin wax producers defend litigation brought by candle makers, and tyre producers claiming against producers of styrene butadene rubber.23 Indeed, each side is represented by substantial law firms. Hausfeld LLP is frequently on the plaintiff side, but UK-based firms are there also. There is no longer any inhibition on making competition claims in national courts in Europe. Public companies no longer refrain from making competition claims against other companies.

The passing-on defence What should the courts do where the victim of the wrong has unintentionally taken effective steps to palliate the wrong? What about the pizza shop which charges its customers a price which reflected the price of the cartelised tomatoes? What if the victim is a regulated industry like a public utility which charges for its services the fee prescribed by the national regulator? What if the regulator took account of the value of the regulated assets (assets whose price had been affected by a cartel) as part of its justified cost base? The end customers might have a claim against the cartelist, but that is a different, much more remote, matter. In the German courts, it is currently disputed whether the passing-on defence should be permitted as a matter of public policy, on the grounds that its validity could impede the effective enforcement of the competition rules. A decision on this matter from the Bundesgerichtshof is awaited. In one recent case, a German 22 Ryanair v ExxonMobil Aviation International Ltd., Esso Italian s.r.l., Esso Italian s.p.a., claim lodged as 2010–1259. 23 Sintesi e Ricerca S.p.A. and Others v Royal Dutch Shell plc (paraffin wax cartel) HC-09-CO2672; Cooper Tire v Bayer plc and others (rubber cartel) [2009] EWHC 2609 (Comm).

Market Forces and Private Enforcement  303   court introduced a new approach to the passing-on defence, whereby direct and indirect purchasers were designated joint creditors. The effect, roughly speaking, would be that the direct purchaser can claim the entirety of the loss once from the cartelist and is then obliged to compensate the indirect purchaser. The decision has been appealed and is pending before the Federal Supreme Court.24 One day we might have guidance from the ECJ concerning the consistency with EU law of the passing-on defence, but not for the moment.

Class actions By way of exception to the business-to-business claims brought by companies described above, consumer actions have not been particularly successful. Claims on behalf of consumers about cartelised football shirts25 were, broadly speaking, a financial failure. The Commission has resorted to a novel way of drumming up follow-on claims. Charmingly, there is a page26 on the website of the Directorate General for Competition which depicts a virtual room full of cartelised products (for example, paraffin wax candles, an LCD television set, a bowl of bananas). The idea is to facilitate the initiation of private damages actions by informing consumers of products purchased which vest them with a claim. One can browse round a wellfilled room, full of litigious opportunity. There is no doubt that proving breach, causal link, and loss must for many private citizens and small businesses be an immense task. Attempts have been discussed at EU level to facilitate such claims by a common framework for collective redress across the EU.27 The Commission has launched a consultation. Should competition law claims be facilitated above other species of claim? This presents an obvious problem of principle: why should competition claimants have right to a privileged status? It is not obvious why competition claims are entitled to precedence (rules of evidence, burden of proof, causality, duty to mitigate damage, and limitation periods) over claims arising out of other misfortunes about which other claimants may wish to litigate. Those who have suffered negligent medical treatment, whose family and surroundings have been damaged by environmental degradation, or who have been lured into imprudent financial investments should in principle be eligible for the prompt examination and disposition of their claims. 24 Higher Regional Court of Berlin, judgment of 1 October 2009, 2 U 10/3 Kart. Ready-Mix Concrete II. 25 The Consumers Association v JJB Sports PLC, Case number 1078/7/9/07, Competition Appeals Tribunal, 5 March 2007, settled. 26 See http://ec.europa.eu/competition/consumers/how/index_en.html (accessed on 15 June 2011). 27 See http://ec.europa.eu/consumers/redress_cons/collective_redress_en.htm (accessed on 15 June 2011). The consultation closed on 30 April 2011. Over 300 organisations submitted replies to the consultation.

304  Ian S Forrester and Mark D Powell Should those whose claim is based on the competition rules be eligible for swifter justice, more tolerant rules on proof and evidence, different prescriptive periods or easier access to counsel’s services? In its resolution of 26 March 2009 on the Commission’s White Paper on damages actions for breach of the EC antitrust rules,28 the European Parliament noted that: measures at Community level must not lead to arbitrary or unnecessary fragmentation of procedural national laws and that, therefore, careful consideration should be given to whether, and if so to what extent, a horizontal or integrated approach should be chosen to facilitate out-of-court settlements and the prosecution of actions for damages.

Substantive harmonisation of certain elements of the competition laws of the Member States already exists in theory and mostly in reality.29 To the extent that a national claim follows on from a Commission decision, procedural consistency is desirable. Perhaps the need for consistency is less compelling where there has been no Commission activity: the claim would then be comparable to one for compensation from a Member State for denying free movement rights. No doubt there will be Ph.D theses written on the desirability of procedural harmonisation as to private enforcement of competition law. The 2006 judgment of the ECJ in Manfredi30 makes clear that any individual who has been harmed by an anticompetitive practice enjoys an EU law-based right to damages, as long as there is (i) ‘harm’, (ii) a competition law violation, and (iii) a ‘causal relationship’ between the harm and the violation. Developments in collective redress are already taking root in various Member States, driven by a perceived need to help consumers more than any sentiment of cross-border activism.31 Any EU initiative may be overtaken by the effect of market forces at national level. As some cartelised products are consumer-orientated, facilitation of collective actions might lead to a rise in private damages claims: however, in actual practice the sums paid to individual consumers have been very modest, far less meaningful than the pain suffered by the defendant. To counter this phenomenon, the Commission might consider giving a significant reduction in fines to infringing companies which would undertake to compensate victims fairly, subject to an arbitration remedy in case of dispute about the amount. It may be more useful, and perhaps even more appropriate, for the Commission to inform consumers of condemnations, rather than to consider enacting legislation which purports to harmonise the bringing of private actions across the Member States and which runs the risk of inadvertent and undesired consequences. We 28 2008/2154(INI), http://www.europarl.europa.eu/sides/getDoc.do?type=TA&reference=P6-TA2009-0187&language=EN. 29 For example, in 2006 the European Competition Network (ECN) adopted a model leniency programme. See http://ec.europa.eu/competition/ecn/model_leniency_en.pdf. 30 Joined Cases C-295/04 to C-298/04, Vincenzo Manfredi and Others v Lloyd Adriatico Assicurazioni SpA and Others [2006] ECR I-6619. 31 For instance, in Poland, opt-in collective actions are available under the Act on Pursuing Claims in Group Proceedings of December 2009, which came into force as of July 19, 2010, while in the Netherlands, the Dutch Class Action (Financial Settlement) Act 2005 enables binding opt-out settlements of mass disputes.

Market Forces and Private Enforcement  305   have seen that national courts, such as those in England, are fashioning good mechanisms for dealing with private damages actions. The judicial marketplace may come up with better solutions, such as a healthy tradition of favouring settlements, to resolve such actions. A more felicitous role for the Commission may be to continue to encourage the taking of private damages actions and to improve its own procedures with a view to enhancing the credibility of its decisions in the eyes of national judges and practitioners.

The problems of primacy and reliability There is no doubt that determinations by the Commission as to the existence of a competition law infringement are constitutionally entitled to important weight in the context of a national litigation. This flows from the judgments of the ECJ in Masterfoods32 and Delimitis.33 However, what degree of deference is required, and to what portions of the decision? This is a matter of very great constitutional and practical significance. The Commission is an administrative body whose procedures have been adapted regularly but not vastly since their establishment in 1962. Fundamentally, the Commission is not a trier of fact. And there lies the problem. How should a national judge whose job is to be trier of fact treat the recital in a Commission decision to the effect that representatives of companies A, B, C, D and Z attended a meeting on 13 January 1998, another in 2002, and a further one in 2007? Suppose that company Z says its employee was not present and that dishonest rivals were telling falsehoods? But the Commission decision nevertheless states that Z participated. Can the national judge feel free to decide for himself if company Z should be one of those found jointly and severally liable for supplying cement to Dutch customers at prices which reflected illicit price-fixing agreements and practices? Or is the judge bound by the recital in the decision? The question has been addressed in an Opinion by Advocate General Walter van Gerven,34 which distinguishes between the findings of fact that underlie the final decision and those that do not.35 Departing from the latter would not likely give rise to the risk of conflicting decisions, while departing from the former would be tantamount to challenging the decision. This issue has recently been considered in an English case concerning the follow-on effect of decisions taken by national regulators.36 While enunciating a similar distinction, based on the violence that might be done to the finding of Case C-344/98, Masterfoods Ltd. v HB Ice Cream Ltd. [2000] ECR I-11369. Case C-234/89, Delimitis v Henninger Bräu [1991] ECR I-935. 34 Case C-128/92, Banks v British Coal [1994] ECR 1-1209. 35 Paragraph 61 of the Opinion. 36 Enron Coal Services ltd v English Welsh & Scottish Railway ltd [2011] EWCA Civ 2 (concerning a decision by the ORR, considered as the OFT in the railway sector). 32 33

306  Ian S Forrester and Mark D Powell infringement, the judges also drew a distinction between the nature of the inquiry before a regulator, and in private damages actions.37 Facts about causation and damages will not necessarily be subjected to the greatest of scrutiny by the regulator if they are not directly relevant to establishing the infringement. An undertaking may not have been concerned to contest such findings of fact for the purposes of the regulatory proceedings. But those findings may come back to haunt it during private damages actions. Similarly, the parties claiming damages will not have been party to the proceedings before the regulator, and will not have had the opportunity to be fully heard. The court noted that there is potential for injustice: findings made by the regulator on incomplete evidence exacerbated by the impossibility of attacking them later.38 To what extent should the national judge distinguish shades of grey in terms of culpability? It is common enough for geographically remote industry members (Greece, Portugal, Ireland, for example) to be peripherally or occasionally involved in a cartel led by a small number of large companies. Commission decisions rarely make fine rankings of culpability (indeed, this is a frequent source of debate on appeal). It is not difficult to imagine that a judge will be asked to calibrate the damages to take account of level of culpability, frequency of participation in the meetings, market share and geographic remoteness from the area principally affected by the cartel. Such subtle nuances have not yet been addressed but they should be. Joint and several liability as between all the guilty parties will be too crude and potentially too unjust. The recitals of a decision usually will be inadequate to resolve these questions, so it will be for national judges to determine. A related issue concerns the degree of uncertainty felt in reaching the conclusions. Judgments in the European Courts are thus far unanimous. The Advocate General’s opinion in the ECJ can serve the function of a dissent, by revealing the validity of alternative viewpoints. In a recent article,39 Judge Azizi considers dissents as a means of reinforcing the judgment by confirming that certain points were carefully considered, and rejected. Although he hesitates to recommend that dissents be published, he concludes by noting that ‘the on-going debate on a possible disclosure of dissenting opinions should be seen by EU courts as a very serious motive to pay due attention to the standard of reasoning in their judgments’. It seems clear that national judges would be aided by the availability of dissents in competition authorities, so as to assist in identifying and weighing critical issues.40 Ibid, paragraph 150. Ibid. 39 Josef Azizi, ‘Unveiling the EU Courts’ Internal Decision-Making Process: A Case for Dissenting Opinions?’, 12 ERA Forum 49 (2011). 40 For examples of a recent increase in rigour of the appellate function in Luxembourg, see Case T-348/08, Aragonesas Industrias y Energia, SAU v Commission [2011] ECR II-7583; Case T-349/08, Uralita SA v Commission [2011] ECR II-373*; Case T-11/06, Romana Tabacchi srl v Commission [2011] ECR II-6681; Case T-44/07, Kaučuk a.s. v Commission [2011] ECR II-4601. 37 38

Market Forces and Private Enforcement  307  

How binding is binding? Commission findings are not binding on a national court considering an issue arising between different parties in respect of a different subject matter. Where the legal and factual context of the case before the Commission is not identical to that before the national court, the Commission decision, though it is not binding on the national court, may be regarded as highly persuasive. Can it be more? Lord Hoffmann said, in Inntrepreneur Pub Company v Crehan [2006]: For my part, I do not find deference in this context a very helpful expression. It is commonly (if not altogether happily) used in administrative law when a court decides that the decision-making power on a particular question properly belongs to someone else and that the court should not substitute its own view. But the decision-making power on whether article 81(1) applies plainly belonged to the English court, exercising concurrent jurisdiction, and I find it difficult to see how the exercise of this power can be combined with ‘deference’ to the decision of someone else. The correct position is that, when there is no question of a conflict of decisions in the sense which I have discussed, the decision of the Commission is simply evidence properly admissible before the English court which, given the expertise of the Commission, may well be regarded by that court as highly persuasive. As a matter of law, however, it is only part of the evidence which the court will take into account. If, upon an assessment of all the evidence, the judge comes to the conclusion that the view of the Commission was wrong, I do not see how, consistently with his judicial oath, he can say that as a matter of deference he proposes nevertheless to follow the Commission. Only a rule of law, in the nature of an issue estoppel which obliges him to do so, could produce such a result and the Court of Appeal accepted that there was no such rule.41

The procedural imperfections of the Commission’s regime have been extensively debated. For as long as those problems remain, there will be reluctance on the part of judges and parliaments to tolerate that Commission findings must be deferred to in the manner contemplated by Lord Hoffmann. We suggest that in the present procedural circumstances anything in the administrative decision may be departed from by a national judge, in light of the evidence properly appraised.

Due process in Brussels and its relevance to private enforcement The deficiencies in the Commission’s investigatory/prosecutorial/adjudicatory role in Article 101 and 102 proceedings, and the traditional deferential habits of appellate judges in Luxembourg, are relevant to our discussion. If Commission findings of fact are binding, parties will vigorously contest any factual finding that may have a bearing on later liability for private damages actions, even if the 41

Inntrepreneur [2006] UKHW 38, paragraph 69.  

308  Ian S Forrester and Mark D Powell establishment of such facts are not crucial for the infringement. Cartels do not require proof of damage for a ‘conviction’. What about related civil liability? There must be a link between the quality of the ‘federal’ procedures in Brussels and the credibility of ‘national’ follow-on actions which arise out of these procedures. Lots of us have written on these topics and there is no need to rehearse the familiar concerns and the familiar responses of the institution. However, there is a genuine problem. It will be increasingly difficult to encourage the adoption of new measures to enhance ‘following-on’ in light of the lack of a more rigorously convincing method of fact finding; the adverse consequences of being found to have participated in a cartel or other infringement; and the sharp differences between administrative decisions and judicial determinations. To quote Kovacic, Hollman and Grant: In recent years, the global competition community has gained a deeper appreciation of what engineers have understood for ages: brilliant theory without skilful implementation is a bad match. Great ideas from economics, law or other disciplines require equally great implementing institutions to move a system of competition policy forward.42

This observation is made in a constructive and not in a carping spirit. It would be desirable to effect some reforms for the benefit of talented officials, perplexed national judges, potential claimants seeking compensation and potential infringers who deserve that their conduct be assessed by using high quality procedures.

Conclusions Competition law in Europe has come a very long way, and the process is getting broader, deeper and stronger. Looking through the past 16 years of these gatherings in Fiesole confirms the nature and extent of these powerful evolutions. Probably there will be more private enforcement claims, and more rapid claims, than the follow-on claims brought after the adoption of the Product Liability Directive. Large industrial companies are now ready to make competition claims, which they would not have done in the past. New economic models are being proposed by renderers of legal services – success fees, legal aid, insurance against the costs of losing, financing of litigation, contingent fees. It is unclear where these changes will lead, but they are irreversible, and they are a concomitant manifestation of the role of market forces. The individual victim of a cartel in 2012 has a much better chance of an effective remedy than in 2006 or 2000. Now that the big constitutional principles are settled, institutional resistance is not significant. We are in a phase of considering many detailed questions of court procedure and damages law. Courts and litigators and teachers are debating 42 William Kovacic, Hugh Hollman and Patricia Grant, ‘How Does Your Competition Agency Measure Up?’, 7 European Competition Journal 25, 25 (2011) (footnote omitted).

Market Forces and Private Enforcement  309   them: some decisions will be good, some less good, and later ones will learn accordingly. It is not clear that new legislation is the best way of smoothing the path of a litigious victim to compensation for breaches of the competition rules. It is not clear that alleged victims of breaches of competition law deserve better treatment than the alleged victims of other misfortunes. Legal systems across Europe offer competing alternatives as fora to determine questions of liability. We are in a learning period for the moment. It is likely that judicial market forces, a powerful means to deliver innovation and solutions to practical problems, will carry us further than legislation.

(Abuse of Dominance)

18/06/2010

04/07/2010

Volvo v Pilkington Car Glass Group

Albion Water Limited v Dwr Cymru Cyfyngedig

01/09/2010

Copper Tubes & Fittings

15/12/2010

Carbon Graphite

Grafton v IMI

late 2010

Jet-fuel

Ryanair v Exxonmobile Aviation International Ltd., Esso Italian s.r.l., Esso Italian s.p.a. Deutsche Bahn AG and Others v Morgan Crucible PLC and others

21/02/2011

Date (claim issued)

Car Glass

Cartel

Honda v Asahi, Saint-Gobain and Pilkington

Case Name

Director General of Water Services Decision, 26 May 2004

EU Decision, 28 November 2007

EU Decisions, 3 September 2004

EU Decision, 3 December 2003

Italian Competition Authority Decision, 2006

EU Decision, 28 November 2007

Infringement Decision (Date)

Claimant Lawyers

Ongoing

Ongoing

Ongoing

Ongoing (NB related to 09/02/2007 claim against members of the same cartel - see below)

Ongoing

Palmers Solicitors

Hausfeld

Hausfeld

Hausfeld

Nabarro

Ongoing (NB related Hausfeld to 04/07/2010 claim see below)

Status

Follow-on Damages Claims in the UK

UK Court

Wilmer Cutler

Herbert Smith

Allen & Overy

Herbert Smith, Freshfields, Ross & Co

CAT

High Court

High Court

CAT

Herbert Smith, High Court Clifford Chance, Linklaters, Cleary Gottlieb, Freshfields. Hogan Lovells High Court

Defendant Lawyers (nonexhaustive)

310  Ian S Forrester and Mark D Powell

21/12/2009

23/11/2009

Moy Park & Methionine Others v (Animal Evonik Degussa & Feeds) Others

LCD / CRT

Candle Wax

Marine Hose

Nokia v Toshiba & Others

Sintesi e Ricerca, Gies Kerzen, Richard Wenzel v Shell, Exxon & Others

Waha Oil v Dunlop Oil

09/07/2009

29/07/2009

EU Decision, 2 July 2002

01/05/2010

Marshall Foods Methionine (Animal Group & Ors. v Evonik Degussa Feeds) & Ors

Ongoing

Ongoing

Ongoing

Ongoing

EU Decision, 28 January 2009 Ongoing (NB and OFT Decision, 11 June related settlement 2008 with another marine hose cartel member, Parker ITR)

EU Decision, 1 October 2008

EU Decision, 8 December 2010

EU Decision, 2 July 2002

EU Decision - not yet issued

04/05/2010

Tom Tom v LCD / CRT Samsung & Others

Ongoing

EU Decision, 24 January 2007 Ongoing

EDF v Areva, Gas Insulated 10/05/2010 Alstom, Mitubishi, Switchgear Siemens and Others

Hausfeld

Hausfeld

Bird & Bird

Hausfeld

Maclay, Murray & Spens

Origin

Dundas & Wilson

Freshfields and others

High Court

White & Case, Allen High Court & Overy, Covington & Burling, and others High Court White & Case, Baker & McKenzie, Freshfields, Lovells, Norton Rose, DLA Piper, Cleary

CAT

CAT

White & Case, Ashurst

White & Case, Ashurst

High Court

High Court

Covington & Burling

?

Market Forces and Private Enforcement  311  

School Fees

(Abuse of Dominance)

Gas Insulated 01/11/2008 Switchgear

N J and D M Wilson v Lancing College Limited

Freightliner Limited v English Welsh & Scottish Railway Limited

National Grid v Alstom, ABB & Areva

13/03/2008

British poultry firms v BASF

Vitamins

18/09/2008

Claims Funding Air Cargo International v BA & Others

07/11/2008

03/12/2008

18/12/2008

Animal Feed Phosphates

Trouw Ltd and others v Britphos Ltd and others

Date (claim issued)

Cartel

Case Name

EU Decision 21 November 2001

EU Decision, 9 November 2010

Claimant Lawyers

Damages claim heard by CAT and dismissed. Leave to appeal CAT decision granted.

Settled

On Appeal (NB related Devenish Nutrition Ltd case (2008) where claimants were not awarded damages, High Court and CA)

Ongoing

Taylor Vinters

Hausfeld

High Court

High Court

CAT

CAT

High Court

UK Court

Cleary Gottlieb, CAT Freshfields, Maclays Murray & Spens, Mayer Brown

Slaughter & May & others

Freshfields, Lovells, Shearman Sterling, CC

Freshfields, Orrick

SJ Berwin

Berwin Leighton Paisner

?

White & Case, Herbert Smith, Bird & Bird, Linklaters, Baker & McKenzie

Defendant Lawyers (nonexhaustive)

?

Ongoing (High Court SJ Berwin stayed proceedings)

Status

EU Decision, 24 January 2007 Ongoing

Office of Rail Regulation Decicion, 19 December 2006

OFT Decision, 21 December 2006

EU Decision, 20 July 2010

Infringement Decision (Date)

312  Ian S Forrester and Mark D Powell

Rubber

Football Shirts

Carbon Graphite

Vitamins

Beer Ties

The Consumers Association v JJB Sports Plc

Emerson Electric Vo and Others v Morgan Crucible Company plc

Healthcare at Home v Genzyme Limited

Inntrepreneur v Crehan

ME Burgess and (Abuse of Others v Austin & Dominance) Sons and Harwood Park

Cooper Tire & Others v Bayer and Others

01/10/1997

05/04/2006

09/02/2007

05/03/2007

03/08/2007

22/02/2008

EU Decision, 28 February 1991

OFT Decision, March 2003

EU Decision, 3 December 2003

OFT Decision, 26 November 2007

OFT Decision, 29 June 2004

EU Decision, 29 November 2006

Crowell & Moring

Clyde & Co.

Damages not awarded (HL overturned CA decision)

Maitland Walker

Sprecher Grier Halberstamr

Manches

Herbert Smith, Freshfields, Ross & Co

DLA Piper

White & Case, Herbert Smith, Freshfields, Linklaters, Baker & McKenzie Young & Lee ?

SJ Berwin

Order made for claim Ashurst to be withdrawn

Ongoing

Settled

Settled

Ongoing (Shell settled)

High Court

CAT

CAT

CAT

CAT

High Court

Market Forces and Private Enforcement  313  

Bruno Lasserre*

Integrating Public and Private Enforcement of Competition Law: Implications for Courts and Agencies

Private antitrust enforcement has fostered a large debate across Europe for more than ten years, since the landmark Courage judgment of the European Court of Justice.1 Public debate on the subject was revived by the publication of the Green2 and White3 papers on antitrust private enforcement, and, again by the public consultation on collective redress,4 which also addressed other sectors. In my view, no genuinely new facts or events have emerged to nurture further debate about the desirability of European standards for collective redress in the antitrust field, or about the ways and means to achieve, within the specific European antitrust institutional design, an optimal interplay between private and public enforcement. However, there is plenty of novelty in terms of the general political and legal background, which makes the case for building a European model more convincing. First, the level of substantive convergence has dramatically increased since the entry into force of Council Regulation 1/2003 in 2004. Admittedly, there remain some areas of substantive law where national procedural autonomy still prevails. But is it still so clear-cut if one looks at the key issue of fines? In this regard it is significant that the European Competition Authorities have adopted common principles on fines,5 and furthermore the Court of Justice have ruled that ‘the effectiveness of the penalties imposed by the national or Community competition authorities is a condition for the coherent application of EU competition rules’.6 I believe that virtually no area of antitrust law can remain exempt from this process of convergence. In this respect, the European experience stands in sharp contrast * President, Autorité de la concurrence, Paris. The written contributions of the Autorité to the Green Paper of the European Commission and to the 2011 public consultation may be downloaded from http:// www.autoritedelaconcurrence.fr/doc/classactions.pdf (French) and http://www.autoritedelaconcurrence. fr/user/standard.php?id_rub=389&id_article=1609 (French and English versions). 1 Case C-453/99 Courage Ltd v Bernard Crehan [2001] ECR I-6297. 2 Commission, Green Paper on damages actions for breach of the EC antitrust rules, SEC(2005)1732, 19 December 2005, COM/2005/0672 final. 3 Commission, White paper on damages actions for breach of the EC antitrust rules, SEC(2008) 404, SEC(2008) 405, SEC(2008) 406, 2 April 2008, COM/2008/0165 final. 4 Commission, Staff working document, public consultation: Towards a Coherent European Approach to Collective Redress, 4 February 2011, SEC(2011)173 final. 5 ECA Working Group on Sanctions – Pecuniary sanctions imposed on undertakings for infringements of antitrust law – Principles for convergence, http://www.bundeskartellamt.de/ wDeutsch/download/pdf/ECA/ECA_principles_for_convergence.pdf. 6 Case C-429/07, Inspecteur van de Belastingdienst v X BV [2009] ECR I-4833, especially paragraph 37.

316  Bruno Lasserre with other fields of European law such as EU contract and consumer law, where only minimal harmonization has been achieved so far and where the solution for the future may rather lie in the institution of a 28th regime at the European level that would add a new layer on the millefeuille of national laws in those fields. In this context, EU antitrust law has become a privileged and fitting area for the development of European standards for collective redress. Second, class action legislation throughout Europe has spread far beyond expectations. The three ‘happy few’ in 2004 when the Green Paper was drafted, ie, the UK, Portugal, and Sweden, have been joined since then by many others. While the scope of these regimes certainly vary, today (in 2011), 13 Member States7 now offer either joint litigation or mediation mechanisms for complaints based on the same infringement, or a test case system whereby the liability of an undertaking established in the lead case is presumed to be established in other cases based on the same infringement. Interestingly, although some Member States in 2004–2005 invoked the need to preserve their civil law procedural model and strongly opposed any EU legislative intervention in the antitrust area on this ground, both the civil law and common law traditions are represented among the group of 13 Member States. Third, forum shopping and ‘law shopping’ have revealed the ‘winners’ in the competition between legal marketplaces: London and Amsterdam. With the array of options provided by EU jurisdictional rules,8 it is easy for an undertaking with 7 Bulgaria (consumer law; Article 188 of the Law on Consumer Protection and Chapter 33 of the Code of Civil Procedure); Denmark (contractual and extra-contractual liability; Sections 254 (a) to (k) of the Administration of Justice Act); Finland (consumer law; the ‘Ryhmakannelaki’ law No 444/2007 of 13 April 2007), Germany (Capital Investors’ Model Proceedings Act, or ‘KapMuG’ law of 16 August 2005; in addition, and since 2008, individual claims for redress may be brought in their own names by specialized companies that have acquired litigation rights in exchange for a partial entitlement to damages if the claim is successful; this was the case in the award by the Court of Appeal of Dusseldorf in the framework of the cement cartel); Hungary (action in the public interest under the competition law which may subsequently provide grounds for individual claims for redress; Article 92 of the competition law as applicable from 1 November 2005); Italy (competition law and consumer law; Article 140 bis of the Consumer Code); the Netherlands (collective mediations in contractual and extra-contractual matters, submitted to validation by the Court of Appeal of Amsterdam; Articles 7.907–910 of the Civil Code and Articles 1013 to 1018 of the Code of Civil Procedure); Poland (contractual and extra-contractual liability, except for bodily injuries; the Act of 17 December 2009 on Collective Redress Litigation); Portugal, (consumer, public health and environment law; Law No 83/1995 on class actions); Romania, (consumer law; Government Ordinance No 21/1992); Sweden (contractual and extra-contractual liability; Law No 2002/599 on class actions; Spain (consumer law; Sections 206.1.6, 221, 223 paragraphs 3 and 519 of the Code of Civil Procedure and Article 11 of Law No 15/2007 on competition); United Kingdom (law of general scope applicable in both contractual and extra-contractual issues; Article 19, III of Civil Procedural Rules). 8 Pursuant to Council Regulation 44/2001 of 22 December 2000 relating to jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (‘Brussels I’), actions for extra-contractual liability seeking compensation for antitrust damages may be initiated either in the Member State in which the defendant has its habitual residence (Article 2), or before the courts of the State where the damaging event occurred (Article 5(3)), or, if there are multiple defendants, before the court where one of them resides, ‘provided the claims are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings’ (Article 6(1)). Furthermore, Article 28 allows the competent court to join related cases.

Enforcement of Competition Law: Implications for Courts and Agencies 

317  

sufficient legal and financial resources, having suffered from an anticompetitive practice, to gather the factual elements establishing that this practice may have had an effect in the UK or in the Netherlands, where the substantive and procedural framework is favourable to its claim. This is less so, however, for individual and dispersed consumers as long as no common EU principles are transposed into the national legislations of the 27 Member States; those who live in Member States where collective action mechanisms are available can be compensated for their harm, while those in other Member States are essentially deprived of their right to damages. Hence, the question is whether the situation is sustainable in the long term and how it will be resolved. The most promising scenario in terms of justice and legal security, both for defendants and claimants, would be the establishment of common EU principles. Fourth, the questions tested by the Commission have evolved. In 2005, the Commissioner asked stakeholders whether the EU should introduce class actions at EU level in the antitrust area. In 2010–2011, the question mutated into: ‘In the light of the experience of Member States and third countries, how will we build our own European model while providing for adequate safeguards against unmeritorious and abusive litigation?’ Last but not least, if one studies the results of the latest public consultation, it can be seen that while no consensus has emerged on the former question (EU-level class action rules), there is consensus on the latter, even between consumer associations and business. Almost all stakeholders agree that essential characteristics of US-style class actions should be avoided, such as doorstepor media-based promotion combined with contingency fees, disproportionate discovery conducive to fishing expeditions, and punitive damages. Further, almost all agree that adequate private international law rules are necessary to accommodate the specific nature of collective actions, and they agree that the development of mediation is a necessary complement to any EU intervention – even if opinions vary as regards its interplay with legal avenues for appeal and the role of the judiciary. And they insist that the attractiveness of leniency should be preserved: consumer associations recognize that many cartels would otherwise not be discovered at all, while business associations are keen to preserve the possibility to use this attractive tool. From the perspective of a national competition authority, three main issues deserve specific developments in this new context: the reasons why the Commission should concentrate on follow-on actions; the role of NCAs in the new private enforcement system; and the need for objective and dispassionate reflection on the dispute over possible over-enforcement as the framework for private claims is strengthened.

318  Bruno Lasserre

1. Why should follow-on actions be the cornerstone of an EU model? The deeper reflection taking place since 2005 has invited elected representatives, in the light of the enforcers’ and other stakeholders’ views, to design European principles that would take into account the specific interplay between public and private enforcement in the EU. I believe that a reasoned comparison between the mirror situations across the Atlantic should lead them to conclude that follow-on actions should be the cornerstone of our EU model. If one tries to sum up and sketch – in a somewhat crude manner – the main features of the interplay between public and private enforcement system in the US, two main elements stand out. On the one hand, private enforcement is the standard antitrust enforcement tool,9 and it is viewed as playing both a compensatory and a punitive role.10 As a consequence, the law does not require the plaintiff to take the risk of paying the defendant’s legal fees if the suit fails. By comparison, public enforcement remains exceptional and focuses on the biggest issues, and antitrust authorities seek to maintain coherence in the enforcement of antitrust rules by submitting amicus curiae briefs to courts where they feel it necessary to do so. On the other hand, the US legislator and the judiciary have sought to compensate for the downsides of the handling of complex legal and economic issues via trial by lay juries. For example, the Supreme Court and federal appellate courts have raised the standard of proof in monopolization and retail price maintenance cases, and they have disallowed the passing-on defence. For its part, the US Congress has mitigated the effects of punitive damages by ‘de-trebling’ the possible damages awarded to claimants where the anticompetitive practice was uncovered through a leniency application. (With regard to leniency, see also section 3 below.) By contrast, in Europe, enforcement by the Commission and by NCAs is and will remain the main pillar of antitrust enforcement. Private enforcement serves an exclusively compensatory role;11 it entails economic risks for the claimant, in the form of ‘loser pays’ rules; and civil judges (including highly specialized tribunals in some EU Member States), rather than juries, are competent to hear and adjudicate private enforcement cases. Follow-on actions are therefore perfectly tailored to EU-style private antitrust enforcement and should be the privileged way to trigger claims for damages. Follow-on actions also offer advantages for all stakeholders. For defendants, they ensure stronger procedural guarantees, as they are offered three opportunities 9 It is usually estimated that between 90 and 95 percent of antitrust cases in the federal courts are privately initiated. See Clifford Jones, Private Enforcement of Anti-trust law in the EU, UK and USA, Oxford University Press, 1999. 10 This has been acknowledged by the Supreme Court. See Atlantic Richfield Co v USA Petroleum, 495 US 328, 360 (1990); Illinois Brick Co v Illinois, 431 US 720, 746 (1977) and ibid at 748, 749 (Brennan, J, dissenting); Fortner Enterprises. v US Steel Corp, 394 US 495, 502 (1969). 11 See further details below about so-called punitive damages in the UK; to my knowledge, in Ireland and Cyprus, no punitive damages have been awarded so far in the antitrust area.

Enforcement of Competition Law: Implications for Courts and Agencies 

319  

to be heard by at least one and possibly two specialized entities which enjoy sufficient legal and economic expertise: first, before the competition authority and then possibly on appeal, if the decision of the competition authority is challenged; and eventually during the private enforcement trial before the civil judge. Moreover, the decisions of competition authorities include factual findings which have already been subject to an adversarial exchange of arguments, and provide useful information for the subsequent debate before the civil court. Hence, the risk that a ruling might be made on the basis of different files if claims are lodged before several courts is reduced, as is that of subsequent contradictory decisions on the issue of fault, especially if the status of NCA decisions is strengthened (see section 2). For consumer associations, follow-on actions are attractive too: complaints before an NCA incur much lower expenses than stand-alone actions, and follow-on claims are built on sounder foundations, since the issue of whether a fault was committed will be a matter already decided upon by a public authority. EU legislation in this area would also be made simpler if the European principles of collective redress are centred on follow-on actions, as fewer rules would be needed to frame discovery, to avoid the dissemination of business secrets,12 and to address the issue of the chilling cost of access to justice for end consumers.

2. How may NCAs further contribute to the construction of this new system? Another critical issue for the construction of an EU model is the concrete role of NCAs. As already touched upon by the various consultations of the Commission, the strengthening of NCA decisions would provide significant encouragement for follow-on actions. By contrast, stand-alone actions may undermine the efficiency of antitrust enforcement if a concurrent administrative investigation by an NCA is already underway and if it has led to on-site inspections,13 in which case legal certainty for defendants and claimants is likewise undermined. To foster follow-on actions, EU legislation could provide that the operative part and the facts of the case described in the recitals of the final14 NCA decision are sufficient in and of themselves to substantiate the existence of wrongdoing that gives rise to a right to redress. Yet they would not be ‘binding’ in the common meaning of the term, as the civil judge would still be responsible for adjudicating a number of pivotal issues, in particular by deciding whether or not to certify a group of claimants, by checking whether assigned firms which were not addressees 12 In a follow-on case, NCAs will already have made extensive use of their powers of investigation and established that a fault has been committed; the frequency and intensity of the need for disclosure will necessarily be lower. 13 See paragraphs 86–91 of the opinion of 21 September 2006 of the Conseil de la concurrence. 14 NCA decisions do not enjoy the same presumption of validity as decisions of the European Commission.

320  Bruno Lasserre of the NCA decision can also be held liable and whether group structures have evolved, by assessing the causal link between the infringement and the injury, and by setting the level of damages. The legislation or case law in fourteen EU Member States15 already provide that NCA decisions provide sufficient proof of fault as described above. Such legislation would also eliminate risks of reverse discrimination depending on whether an antitrust case is handled by the Commission or an NCA. While victims of anticompetitive practices established by a decision of the Commission can point out, pursuant to the Masterfoods ruling16 and Article 16(1) of Regulation 1/2003, that the competent civil court is not permitted to take a decision running counter to that of the Commission on the issue of fault, they do not in all Member States enjoy an equivalent mechanism as regards a final NCA decision. A debate has also arisen as to whether NCAs should assess the level of damages in individual claims or not. The experience of the French Autorité may be of interest in this respect, as it is the sole NCA which is bound by its domestic legislation to take into account, when setting the level of administrative fines, besides the seriousness of the infringement and the individual situation of the infringer, the criterion of ‘damage to the economy’. However, this unique concept remains clearly distinct from that of the individual injury borne by a claimant, and distinct even from the sum of individual injuries to all the buyers (including final consumers).17 Thus, non-cartel members in the same economic sector may benefit from a cartel overcharge, which can be incorporated into the damage to the economy. The French Temporary work case18 illustrates another category of harm done to market players in general that may be encompassed within the criterion of damage to the economy. In that case, the Autorité found that the world’s three leading temporary work agencies, representing 70% of the turnover in the French market, had discussed the price charged to their biggest clients, including the rate of the social security contributions relief that 15 In Denmark, Italy, Poland and Romania, case law provides for a reversal of the burden of proof. In 10 other Member States, pursuant to the law (Germany, Bulgaria, Greece, Hungary, Czech Republic, United Kingdom, Slovakia and Slovenia) or case law (Belgium, Finland), NCA decisions which have determined that there has been an infringement of competition rules and which have become final are conclusive. In the UK, the Competition Act provides that the English courts are bound by the decisions of the competition authorities in relation to findings of fact or infringement, but not all statements in such a decision necessarily constitute a finding of fact. See Enron Coal Services Ltd v English Welsh and Scottish Railway Ltd, [2011] EWCA Civ 2. 16 Case C-344/98, Masterfoods Ltd and HB Ice Cream Ltd [2000] ECR I-11369, paragraphs 49 et seq. 17 See paragraph 27 of the Notice of 16 May 2011 on the Method Relating to the Setting of Financial Penalties of the Autorité, which states that ‘the harm done to the economy is not limited to the illegal profits that competition law infringers may have expected to earn thanks to the anticompetitive practices at stake, since it encompasses all the disturbances that these practices are liable to cause to the competitive operation of the activities, sectors or markets directly or indirectly concerned, as well as to the economy in general. It incorporates not only the transfer and loss of welfare that the infringement is liable to cause to intermediary or end-consumers and to the citizens as a whole, but also, notably, its adverse impact on the incentives of other market players, for instance in terms of innovation. It is therefore not limited to a precisely measurable loss.’ 18 Decision 09-D-05 of 2 February 2009 relative to practices in the Temporary Work sector, http:// www.autoritedelaconcurrence.fr/user/avisdec.php?numero=09-D-05.

Enforcement of Competition Law: Implications for Courts and Agencies 

321  

they passed on to them. The damage to the economy assessed by the Autorité therefore included the ‘hijacking’ of the Government tax breaks and the loss of chance of more jobs or of higher wages if the ‘social wedge’ had been addressed. The estimated value of the damage amounted to 20 million euros. In light of this difference between the concept of aggregate economic harm, on the one hand, and the individual injury at stake in civil damages claims on the other, it seems reasonable not to require NCAs to provide a quantitative assessment of individual harm in their decision. Even the French NCA has no duty to quantify aggregate harm.19 To impose such a requirement would lead NCAs away from their core mission and encroach upon the duty of civil courts, which have made significant efforts to refine their methodology for assessing damages.20 Nonetheless, in antitrust cases where their investigations help to uncover reliable, relevant and key data, it is useful for NCAs to issue detailed decisions specifying overcharges and lost sales, after conducting an inter partes debate on counterfactuals or simulation models. This avoids duplication of investigation measures in follow-on claims, and it simplifies the administrative preparation of tort actions. The Autorité has set out scenarios of economic harm on various occasions, including for example the Historic monuments bidrigging case,21 where the estimated 24.5% overcharge may help local buying authorities claim adequate damages. Another example is the Multilateral interchange fees on cheques case,22 where an overall overcharge of 220 million euros was detected. Other competition authorities that measure the effects of their actions on consumer welfare, such as the UK Office of Fair Trading or the Dutch NMa, likewise may occasionally include such information in their decisions. Such practices also have the advantage of reducing the subsequent need to disclose to potential claimants documents accessory to leniency applications or oral statements, thus mitigating or avoiding the possible adverse ripple effects on the attractiveness of leniency. This leads me to a third area where NCAs may usefully contribute to building an adequate private enforcement system, that of the protection of leniency materials. In its Pfleiderer judgment,23 the Court of Justice acknowledged that a balance was needed between the interests of private claimants and the public interest in the 19 Court of Appeal of Paris, 26 January 2010, Adecco France SAS, 23 March 2010, SEM G.Eg, and 28 January 2009, Epsé Joué Club. 20 The ‘Draft Guidance Paper − Quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty’ published by DG Competition on 17 June 2011 (http://ec.europa.eu/ competition/consultations/2011_actions_damages/draft_guidance_paper_en.pdf) contains numerous examples of the experience of selected courts in this respect (see the tables at pp 66–67). 21 Decision 11-D-02 of 26 January 2011 relative to practices implemented in the monuments restoration sector, http://www.autoritedelaconcurrence.fr/user/standard.php?id_rub=389&id_article =1548. 22 Decision 10-D-28 of 20 September 2010 relative to fees and conditions applied by banks and other financial institutions for the handling of cheques, http://www.autoritedelaconcurrence.fr/user/ standard.php?id_rub=368&id_article=1472. 23 Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161.

322  Bruno Lasserre effective enforcement of Articles 101 and 102 TFEU, in respect of which leniency programmes play a key role.24 The Court held that it ‘is for the courts and tribunals of the Member States, on the basis of their national law, to determine the conditions under which […] access [to documents relating to a leniency procedure] must be permitted or refused by weighing the interests protected by European Union law’. However, this does not preclude NCAs from urging their respective governments to adopt legislative rules that strike a proper balance between transparency and the effectiveness of leniency programmes under a common approach defined within the ECN; nor does it preclude the European legislator from defining common rules in the future. The principle enshrined in the EU Model Leniency Programme in relation to oral statements might provide a useful basis for reflection in this regard. According to the Model Programme, informants should not be worse off than other parties, whether at national or EU level, and equivalence of protection should be recognized throughout the Union.25 The European legislator could also help draw a line with respect to demands of disclosure arising from third country applicants, whether private claimants or public authorities, taking into account all relevant existing international instruments.

3. Interplay between administrative proceedings and private enforcement: is there an issue of over-enforcement? The third ‘hot topic’, as regards private enforcement, is not new but it has gained all the more salience and intensity as the probability of the enactment of EU legislation has increased. The right terms of the debate ought to be clearly laid out before any position is taken on this issue. Of paramount importance is that, due to the specific interplay between public and private enforcement in Europe, punishment and deterrence on the one hand, and compensation of harm on the other hand, are served by two distinct instruments and two distinct public authorities: NCAs in the first case,26 civil 24 See paragraphs 25–27: ‘25. […] leniency programmes are useful tools if efforts to uncover and bring to an end infringements of competition rules are to be effective and serve, therefore, the objective of effective application of Articles 101 TFEU and 102 TFEU. 26. The effectiveness of those programmes could, however, be compromised if documents relating to a leniency procedure were disclosed to persons wishing to bring an action for damages, even if the national competition authorities were to grant to the applicant for leniency [an] exemption, in whole or in part, from the fine which they could have imposed. 27. The view can reasonably be taken that a person involved in an infringement of competition law, faced with the possibility of such disclosure, would be deterred from taking the opportunity offered by such leniency programmes, particularly when, pursuant to Articles 11 and 12 of Regulation No 1/2003, the Commission and the national competition authorities might exchange information which that person has voluntarily provided.’ 25 ECN Model Leniency Programme, paragraphs 28–30 and 47–49 http://ec.europa.eu/competition/ ecn/model_leniency_en.pdf. 26 Although the criminal judge is mainly competent in a minority of Member States (Denmark, Estonia, Ireland); it has a complementary role in the United Kingdom.

Enforcement of Competition Law: Implications for Courts and Agencies 

323  

judges in the second. By contrast, where both aims are mixed and served with the same instrument, as with private enforcement in the United States, it is conceivable that adjustments will be made to address possible issues of over-enforcement – or, to be more specific, of unjust enrichment – which actually does not derive so much from private claims per se as from the recourse to punitive damages. This is the rationale behind the de-trebling of damages in US legislation.27 In the EU, as already stressed, no Member State embraces deterrence as an objective of private enforcement. Where the issue has arisen, courts in Europe have preserved the mutual autonomy of public and private enforcement. For instance, the UK High Court has affirmed that claimants seeking damages from cartelists in a private action based on a finding of infringement of Article 101 by the Commission are entitled to compensatory damages but not to exemplary damages, restitutionary damages, or to an accounting of profits to punish and deter wrongdoers.28 Another important feature to consider, when trying to gauge in an objective manner the possible risk of over-enforcement, is that the right to compensation for a tort is a fundamental principle of law. It would therefore be illegitimate – and probably contrary to ECHR principles – to grant full or partial ‘civil immunity’ to a firm just because it has already been fined in antitrust administrative proceedings. The fact that until recently infringing firms have enjoyed quite a large de facto civil immunity in Europe should be no pretext for preserving such immunity in the future. The issue is different, however, if the fundamental right to compensation is to be balanced against the need to maintain the integrity of leniency programmes, which serve the general interest objective of the effective application of Article 101 by uncovering secret cartels.29 An array of proposals for preserving the attractiveness of those programmes have already been put forward, especially concerning the lead applicant to be granted administrative immunity.30 Such proposals need to be given careful attention and should be further tested by stakeholders, provided they merely alter the allocation of compensation between infringers and treat claimants on the basis of objective criteria. Claimants should not be overall worse off as a result of such proposed rules, and due attention should be paid to solvency issues.31 The two above-mentioned aspects involve, in my view, a matter of principle. They are widely sufficient in themselves to obviate the need for measures in Europe to reduce administrative fines. Fines are fixed by reference to legislative principles in the Member States that incorporate common criteria, including in particular the seriousness of the infringement. These principles are absolutely independent of private enforcement. Equally, in light of the foregoing discussion any initiative to 27 Antitrust Criminal Penalty Enhancement and Reform Act (Antitrust Reform Act) of 2004, Pub L No 108–237, 118 Stat. 665 (2004). 28 High Court, Devenish Nutrition Ltd & Ors v Sanofi-Aventis SA (France) & Others [2007] EWHC 2394 (Ch), [2008] EWCA Civ 1086. 29 See footnote 24. 30 Green Paper, options 29 and 30; White Paper, paragraphs 304–305. 31 See, eg, Jaynie Randall, ‘Does De-Trebling Sacrifice Recoverability of Antitrust Awards?’, 23 Yale Journal on Regulation 101 (2006).

324  Bruno Lasserre reduce compensatory damages on the basis of any sanctions already imposed by public authorities would be inappropriate. Beyond those principles, there would be altogether insuperable practical obstacles to measures aimed at articulating administrative fines and civil compensation to reduce the level of one or the other. When NCAs decide a case, their purpose is to establish whether an infringement has occurred and, where relevant, to set the level of fines accordingly; it is therefore not possible at that stage to incorporate any compensation paid to private parties into the calculation of an immediately payable fine. Commitments or injunctions would not be a viable way either to allow for timely and adequate compensation. Recitals 11 and 3 and Article 5 of Regulation 1/2003 only assign to these instruments the objective of putting an end to competition concerns or anticompetitive practices; compensation is a concept quite alien to that objective. Furthermore, if NCAs ventured in that direction, national courts would surely find that they have failed to respect both the principle of proportionality principle and separation of duties between judges and competition authorities.

Conclusion The 2011 public consultation conducted by the European Commission has not produced evidence of decisive arguments against collective redress in the antitrust field, but it has invited the European and national legislators to refine further and secure appropriate solutions to avoid abusive litigation. If an EU legislative initiative is proposed by the College of Commissioners, the debate would revolve, in competition law terms, around: the technical modalities of enshrining safeguards against unmeritorious litigation; the quest for adequate and perhaps innovative solutions to preserve the attractiveness of leniency programmes; and the precise role of NCAs in the new private enforcement system. This chapter has provided some insight from the perspective of an NCA. The final determinants of the political decision may, however, lie elsewhere. A few highly political issues begun to emerge, including in particular the financial incentives that may or may not be granted to consumer associations, the balance between mediation and judicial redress, and a way to find possible compromise solutions between opt-in and opt-out structures that would guarantee an individual right of access to justice. The ‘buy-in’ of the judiciary will be a test of these developments. This is particularly true as regards the latter two topics. Mediation is only attractive where the judges exert in-depth control on the fairness and publicity of the voluntary settlement and/or where private enforcement is genuinely effective.32 As for the 32 The compared developments following the decisions of the French competition authority relating to the mobile telephony cartels, which involved end-consumers whose access to justice is difficult in

Enforcement of Competition Law: Implications for Courts and Agencies 

325  

thorny choice between opt-in and opt-out, intermediate options might draw their inspiration from the proposals of the British Civil Justice Council33 or from the Danish example,34 which leave some degree of discretion to the judge on a caseby-case basis depending on the size of the class, its representativeness, the level of financial stake, and the status of the claimant. (For example, in Denmark, opt-out actions are reserved for claims brought by the Ombudsman.) More generally, judges will obviously play a key role. They will be essential decision makers not only when assessing damages35 but also, if the experience of foreign jurisdictions is any guide, when interpreting general principles of tort law in the specific context of follow-on claims, or, even further, when led to recognize that the directors of firms that have been the victim of an anticompetitive practice owe a fiduciary duty to their shareholders to at least consider whether a claim should be filed.

practice, as legal costs would greatly exceed the likely damages, and the temporary work cartel, where the victims were one of the biggest French firms, is telling. See also Willem van Boom, ‘Collective settlement of mass claims in the Netherlands’, in Matthias Casper, André Janssen, Petra Pohlmann and Reiner Schulze, eds, Auf dem Weg zu einer europäischen Sammelklage?, Sellier, 2009, pp 171–192. 33 Improving Access to Justice through Collective Actions: Developing a More Efficient and Effective Procedure for Collective Actions, Draft Court Rules For Collective Proceedings (2010). 34 Danish Parliament (Folketinget), 22 February 2007, 2006–07 – L 41 (som vedtaget), Forslag til lov om ændring af retsplejeloven og forskellige andre love. (Gruppesøgsmål m.v.) 35 In this regard, judges may draw inspiration from the methods already used in antitrust cases (see Draft Guidance Paper, cited above note 20), but they may also be inspired by jurisprudence in the field of intellectual property.

Horst Butz*

Integrating Public and Private Enforcement in Europe: Issues for Courts

Introduction According to the European Court of Justice, any citizen or business who suffers harm as a result of a breach of the EU antitrust rules (i.e., Articles 101 and 102 TFEU) should be able to claim reparation from the party who caused the damage. However, despite this requirement under European law to establish an effective framework enabling victims to exercise their right to compensation, victims of EU antitrust infringements have so far only rarely obtained reparation for harm they have suffered. The success of antitrust damage actions and full compensation of victims rest not only upon the existence of an effective legal framework for compensation, but also on a high effectiveness of the antitrust actions for damages in the courts of Europe. In this context, the following ten issues seem to be very important.

1. Reduction of court fees To promote effectives antitrust damages actions it is important to set court fees in an appropriate manner so that they do not become a disproportionate disincentive to antitrust damage claims. Under German law, the judge has, in antitrust damages actions under certain circumstances, the possibility to reduce the value of the matter in dispute. Since that value determines the court fees, the fees and hence the risk of bringing a claim are correspondingly reduced. In order to benefit from this possibility, the claimant must be facing financial hardship (§ 89a GWB). Moreover, the Commission suggests, in its 2008 White Paper, that national courts should be given the possibility of issuing cost orders derogating, in certain justified cases, from the normal cost rules, preferably upfront in the proceedings. Such cost orders would guarantee that the claimant, even if unsuccessful, would not have to bear all costs incurred by the other party. The latter suggestion may be *

VRLG a.D. Landgericht Düsseldorf; Heinrich Heine University, Düsseldorf

328  Horst Butz helpful, but there is not much hope that it will be adopted under German law. It is quite opposite to our whole traditional system of court fees.

2. Standing to sue in class actions With respect to collective redress, the Commission considers that there is a clear need for mechanisms allowing aggregation of the individual claims of victims of antitrust infringements. Individual consumers, but also small businesses, especially those who have suffered ‘scattered’ and relatively low-value damages, are often deterred from bringing an individual action for damages by the costs, delays, uncertainties, risks and burden involved. The Commission therefore suggests a combination of two complementary mechanisms of collective redress to overcome those issues in the field of antitrust. The mechanisms are: – representative actions, which are brought by qualified entities, such as consumer associations, state bodies or trade associations, on behalf of identified or, in rather restricted cases, identifiable victims; and – opt-in collective actions, in which victims expressly decide to combine their individual claims for harm they have suffered in a single action. I am sceptical that these suggestions will be successful in practice. It is unclear whether this approach could overcome the obstacles to private antitrust enforcement. For example, it is far from clear whether such qualified entities can assume the administrative burden of bundling a multitude of claims and of collecting and analyzing the purchase data required for the calculation of damages. The running of representative actions with thousands or even hundreds of thousands of claimants over several years requires a high level of know-how, management, manpower, technical and IT infrastructure as well as the resources necessary to cope with the complexity of antitrust damages actions. They will also have to bear great financial risk, and the benefits are questionable. Opt-in collective actions are in line with European legal cultures and traditions. However, opt-in actions are no effective solution for the special requirements in antitrust damages actions. Each claimant would be party to the legal proceedings and would thus have to substantiate and adduce evidence of the individual damage caused by the cartel. The claimants would also have to bear the costs and financial risks of the antitrust damage action. Small and medium-sized corporate victims, in particular – for whom the opt-in collective action is notably designed for – may often be unwilling to bear the considerable costs and risks associated with antitrust litigation. Possibly, there will be other ways of securing collective redress without the introduction of new procedural tools, and using means that differ from those employed in the U.S. class action system. One example is the model adopted by

Enforcement in Europe: Issues for Courts 

329  

Cartel Damage Claims SA (CDC) in Brussels. The CDC approach is based on the idea that antitrust damage claims are valuable and marketable assets. CDC evaluates the cartel-related damage on a market-wide basis and enforces the claims in its own name and on its own account. In 2005, CDC brought an action for damages against six cement producers before the Regional Court (Landgericht) of Düsseldorf. It had purchased the cartel-related claims of 28 damaged companies, mostly medium-sized companies, which are active in the concrete production and manufacturing sector. The antitrust damages claimed now amount to approximately 176 million euros. In 2007, by interlocutory judgment the court confirmed the admissibility of the action, and thereby the admissibility of this model of bundling claims. The judgment was contested by one of the defendants before the Higher Regional Court of Düsseldorf and the German Federal Court of Justice (Bundesgerichtshof). The Federal Court of Justice dismissed the appeal and confirmed that CDC’s damages action was admissible. Obviously, as litigation of this kind becomes increasingly encouraged, it is also important to avoid the risk of abusive litigation.

3. Forum shopping In Germany we have a limited number of specialized courts dealing with competition law cases (damage claims, claims for injunctive or declaratory relief, contract defence claims, etc.). In the Regional Court of Düsseldorf (Landgericht) there are more than 100 competition law cases per year, with a few damages actions. The Düsseldorf court is probably the one with the most competition law cases in Germany. Of course, in competition law cases – as in intellectual property cases – forum shopping is common. With regard to international cartels that operate in Europe, there will be forum shopping between the courts of different Member States. This issue will become particularly prominent with the increase in litigation involving specialized claimants, such as CDC and others. The nature and extent of forum shopping will depend on the complexities and requirements of damage claims, and the different costs and decisions within Member States.

4. Agency decisions and national court proceedings In the last years the decisions of the Bundeskartellamt in hard core cartel cases have increased. From 2002 to 2005, the Bundeskartellamt started 11 proceedings and decided 9 cases, whereas from 2006 to 2009, 20 proceedings were started and 14 were decided.

330  Horst Butz Competition cases are particularly fact-intensive. Much of the key evidence necessary for proving a case for antitrust damages is often concealed and, being held by the defendant or by third parties, it is usually not known in sufficient detail to the claimant. On the other hand, nearly all antitrust damages cases are followon actions: they are triggered by a decision of the Bundeskartellamt. This gives the claimant the possibility to have more information before filing a claim. Between the Bundeskartellamt and the courts there is a very efficient and supportive contact. If necessary the Bundeskartellamt provides the courts with information and access to records. It also sends representatives to court hearings in certain cases of broader interest.

5. Access to documents Under German law the court has the power to order the production of documents. This includes the production of documents in the possession of the Bundeskartellamt and of other competition agencies. The court can also ask the Commission for comments and pertinent information. Before starting a follow-on action, the victim also has the possibility, via his legal representative, to access the records of the Bundeskartellamt. However, parties must specify, with more or less precision, the individual documents that they wish to be disclosed. Often this is not easy. But there are other limits as well. Discovery can be refused on grounds of legal privilege, or for reasons linked to the protection of business secrets. In a particular case this will be a discretionary decision of the judge. In this context, the court may order that a certified public accountant inspect the relevant documents and that he divulge not the details but only the results to the requesting party.

6. Burden and standard of proof Generally, the burden of proof is on the claimant. Under German law the judge has to be (fully) convinced that a certain fact is true; a mere ‘likelihood’ is not sufficient. This standard of proof also applies in competition law cases. However, sometimes this burden of proof is lightened or reversed. This is the case, for example, when there is prima facie evidence, or when there are facts considered to be of public knowledge. Other factors that may attenuate the claimant’s burden of proof include the existence of a prior court judgment, decisions of the Commission or the Bundeskartellamt, or decisions of the other NCAs of the EU Member States finding a breach of Article 101 or Article 102 TFEU. If these decisions are final, victims

Enforcement in Europe: Issues for Courts 

331  

of anticompetitive conduct can rely on them as binding proof in civil proceedings for damage pursuant to § 33 Abs. 4 GWB and Article 16 of Regulation 1/2003.

7. Expert evidence Of course, expert evidence is admissible in competition law cases. Experts can be appointed by either the court or the parties. Notwithstanding the ‘more economic approach’ in EU competition policy, court experts are still nearly never appointed in German competition law cases. However, an increasing number of expert economic opinions are delivered by parties to the courts, although expert evidence generated in this way is generally given less evidential value.

8. Calculation of damages The calculation of damages is in practice often the most expensive part of an action for damages in competition law cases. The Court of Justice has emphasized that victims of antitrust infringements must, as a minimum, receive full compensation for the real value of the loss suffered. The entitlement to full compensation therefore extends not only to the actual loss due to an anticompetitive price increase, but also to the loss of profit as a result of any reduction in sales, and it also encompasses a right to recover accrued interest. This calculation is the same under German law, as provided for by § 33 Abs. 3 GWB. Once the scope of damages is clear, the quantum of these damages must be calculated. This calculation, implying a comparison with the economic situation of the victim in the hypothetical scenario of a competitive market, is often a very cumbersome exercise. It can become excessively difficult or even practically impossible, if the idea that the exact amount of the harm suffered must always be precisely calculated is strictly applied. To assist national courts and parties involved in antitrust damages actions, the Commission announced in its White Paper the intention to draw up pragmatic and non-binding guidance for the quantification of harm in antitrust damages actions. Such guidance could include economic insights into harm caused by anticompetitive practices and information on methods commonly used to quantify such harm. As part of the expertise sought in the preparation of this guidance, the Commission arranged for a group of external economic consultants and lawyers to produce a study on the quantification of harm suffered by victims of antitrust infringements (December 2009). Following the publication of this study, the Directorate-General for Competition organized a workshop with external economists to discuss a range of issues concerning quantification of antitrust harm in actions for damages.

332  Horst Butz When calculating damages, German courts, including the German Federal Court of Justice (Bundesgerichtshof), are guided by the ‘Vergleichsmarktkonzept’ (BGH Beschluss vom 19.6.2007 – KRB 12/07, Rdn. 19 ff.). This means that they take into consideration other markets which can usefully be compared to the market concerned. However, the court can also choose another method, for example the beforeand-after method. With this method, the overcharge is calculated by comparing prices set during the cartel period and the prices set in the period before and/or after the price-fixing occurred. This before-and-after method has also been applied by German courts, including for example the Higher Regional Court Berlin in the ready-mixed concrete case in 2009. As a third option, the court can resort, under § 287 ZPO, to its best estimate of the harm that has been suffered on the basis of all the facts. A further issue is the passing-on defence. This can be relevant if the direct customer of the infringer fully or partially passed on the illegal overcharge to his own customer (i.e., to an ‘indirect purchaser’). Under § 33 GWB, the damage is not excluded when the direct customer has sold the goods. In these cases the court compares offsetting losses with advantages gained from the infringement, and the defendant has the burden of proof in establishing his defence.

9. Damages actions and leniency applicants It is important, for both public and private enforcement, to ensure that leniency programmes are and remain attractive. Adequate protection against disclosure in private damages actions must be ensured for corporate statements submitted by a leniency applicant in order to avoid placing the applicant in a less favourable situation than that of its co-infringers. Otherwise, the threat of disclosure of the confession submitted by a leniency applicant could have a negative influence on the quality of his submissions, or even dissuade an infringer from applying for leniency altogether. Therefore, the Commission suggests in the White Paper that such protection should apply: – to all corporate statements submitted by all applicants for leniency in relation to a breach of Article 101 TFEU, – regardless of whether the application for leniency is accepted, rejected or leads to no decision by the competition authority. This protection would apply where disclosure is ordered by a court, whether the order is made before or after adoption of a decision by the competition authority. The reasons for these suggestions by the Commission, and the suggestion itself, are convincing. However, I am not sure that such a general legal regulation would be feasible under German law because it would contradict general principles of German procedural law, especially the principles that govern the burden of proof.

Enforcement in Europe: Issues for Courts 

333  

The Commission also puts forward for further consideration the possibility of limiting the civil liability of the immunity recipient to claims made by his direct and indirect contractual partners, and not by others. That would help to make the scope of damages to be paid by immunity recipients more predictable and more limited, without unduly sheltering them from civil liability for their participation in an infringement. These suggestions would certainly be helpful to encourage leniency applicants, and as noted above this would support the success of public and private enforcement. Nevertheless, limiting civil liability and joint liability in damages actions in competition cases would lead to many problems and would undermine fundamental principles of the German law of damages. It might furthermore be contrary to German constitutional law, and specifically Article 14 of the Grundgesetz, which protects the right to property. For these reasons, the German ministry for trade and technology and the Bundeskartellamt have already expressed objections against suggestions made along these lines in the Commission’s Green Paper of 2005.

10. Problems in court practice In addition to the issues mentioned above, there is a great general problem in the practice of the German courts, especially the courts of first instance. There is a significant imbalance between judges, on the one hand, and the litigating parties on the other. This imbalance involves the court’s manpower as well as its equipment and infrastructure. It may be that this problem also exists in other civil cases, but my experience is that, in significant antitrust cases, the problem nearly always emerges. I will give you a concrete example in a few words. In the cement cartel case I mentioned already, the parties are the following. First of all, there is the claimant, CDC, a company which is specialized in filing damages claims. This company, which is based in Brussels, has a rather large staff that includes specialized lawyers, external economists, IT experts and so on. CDC also has expertise and experience resulting from many antitrust cases lodged in several European states. The six defendants are the largest German producers of cement. These are big companies with the best specialized lawyers, economists, and experts in their market. Then there are 28 summoned third parties and each of those parties’ lawyers and staff. These parties have presented to the court files of some 10,000 of pages and 100,000 documents, bills and data (paper and electronic). On the other side there is a single professional judge who has to deliver a judgment in this case as well as judgments in many other cases in a given year. This imbalance of manpower and resources is of course a problem of the administration of justice, but it is also an issue for the courts and for the effectiveness of antitrust damages actions. So the point is that we need courts

334  Horst Butz with a competent staff, available economists, and adequate equipment and infrastructure.

Conclusion As the Commission explained in its White Paper in 2008, the absence of an effective legal framework for antitrust damages actions hampers the full enforcement of the antitrust rules and thus has a negative impact on vigorous competition in an open internal market. The current ineffectiveness of antitrust damages actions is best addressed by a combination of measures at both European and national level. The problem is that the Commission’s suggestions, which could lead to greater effectiveness in the realm of private enforcement, are often incompatible with fundamental principles of national procedural law or damages law. It will be the task of the next years to find legal and practical solutions to this delicate and legally intricate problem.

Philip Lowe

Conclusions

1.

2. 3.

4.

5. 6.

7.

Competition law derives from the prescription of the legislator, confirmed and extended by the courts, that certain forms of business conduct are, or can be, harmful to other businesses or individuals and are, or can be, harmful to the public interest. (‘Hard core’ infringements are those where the establishment of the fact of the infringing conduct is sufficient to be considered harmful, whereas for other types of conduct, the establishment of likely harm is dependent on sufficient evidence of the damage they cause or are likely to cause to consumers (meaning either businesses or individuals both), either directly or indirectly by damaging the competitive process and weakening the effective functioning of markets.) It is in the public interest that there be effective deterrence and sanctioning of anticompetitive conduct through both public and private enforcement. There is a natural interdependence in the effectiveness of public and private enforcement. Commercial practices will be influenced not only by public enforcement decisions (fines, etc.) but also by the assessment of their likely exposure to private action. While the main focus of public enforcement is deterrence and sanctioning, the focus of private action is, on the one hand, taking action against conduct which is harmful to the specific plaintiffs concerned but which may not have been sanctioned by public enforcement, and on the other hand, obtaining redress for the damage caused by the anticompetitive action concerned. Private enforcement action can relate to anticompetitive agreements, in particular cartels, but also to abusive unilateral conduct. Where that is the case, the nature of private actions may be different. Private action in respect of antitrust frequently overlaps with action to stop abusive conduct and obtain redress under contractual, intellectual property and commercial law. Relief and redress may also be foreseen in EU and national regulation for specific sectors, such as network industries. The effective integration of public and private enforcement of competition law also needs to take these possibilities into account. Because public agencies must inevitably focus limited resources on priority enforcement areas, there is an important role for private action in pursuing infringements which are not addressed by public enforcement. Private action can therefore complement the work of antitrust authorities and strengthen overall competition law enforcement. Its role should not therefore be limited to follow-on actions. It performs an essential function, for example, in

336  Philip Lowe

8.

9.

10.

11.

12.

13.

bringing abusive conduct to an end, or forestalling it, through injunctions. This may be relevant, for example, for the many claims of exploitative abuse at national level. Consideration could also be given to proposals for redress in the context of an overall settlement of a dispute. Public authorities need to monitor more closely the number and scope of private actions, including those which end in settlements or are induced by arbitration. The role for mediation also needs to be reviewed to establish the precise role and responsibility of mediators, and the compatibility of their action with competition law. Despite significant progress, not least as a result of the debates initiated at the European level, there remains room for further development of the framework for private action in EU jurisdictions. The Lisbon Treaty’s provisions on citizens’ fundamental right to redress strengthen the legal basis for legislative proposals which the Commission may put forward at the EU level, following up on the Green and White Papers of 2005 and 2008. The extent of private antitrust litigation in the US is sometimes exaggerated. So too is the extent of unmeritorious actions. However, there are indeed a number of undesirable features of the US system, including treble damages, which affect the risk profile of cases and encourage defendants to settle, in some cases with little regard for the merits of the case. The EU has an opportunity to avoid some of the excesses of the US system while at the same time achieving a better balance between public enforcement and private actions than is presently the case in most Member States. But the EU’s review of the possibilities for strengthening the scope for private action needs to pay as much attention to the experience and legislative framework now established in a number of Member States (for example in Germany, the Netherlands, Spain and the United Kingdom) as it does to the situation in the US. It remains clear, however, that in a number of Member States no effective framework for private action exists. The effective integration of public and private enforcement depends on a judicious and proportionate combination of rules at national and European level, on for example discovery and access to documents, disclosure of leniency applications and subsequent enforcement decisions (SOs and final decisions) and facilitation of settlements. The Pfleiderer judgment of the European Court of Justice, decided on 14 June 2011, appears to leave open the question of whether and in what conditions documents can be disclosed to allegedly injured parties under European Union law. In order to facilitate private actions, it is in the interests of plaintiffs, defendants and judges that public agencies provide as much guidance as possible, for example on the nature of unilateral conduct under Article 102 TFEU, and whether such conduct can be regarded as per se anticompetitive or whether it should be examined on a ‘rule of reason’ basis. In this respect, due attention has to be paid as much to the rights of defendants and to the general public interest as to facilitating private rights of action for plaintiffs.

Conclusions 

14.

15.

16.

17.

18.

19. 20.

337  

However, any guidance provided by the Commission cannot detract from the ultimate discretion of the courts to interpret the law and to examine specific cases in the context of the particular procedural system in which they arise. Some economic studies suggest that the characteristics of certain leniency schemes could, on certain assumptions, encourage cartel activity. In those circumstances, public authorities could be better off shifting resources away from prosecuting cases triggered by leniency to more effective detection of cartels. However, in general, the introduction of leniency programmes has significantly increased detection compared to other enforcement tools. Public agencies should generally be in a position to assist courts by indicating in enforcement decisions the likely scope of the overall consumer harm of the abusive conduct which is condemned. However, the establishment of the quantum of harm and of the rights of individual parties is a matter for the courts. The resources of courts to examine complex economic and commercial issues are necessarily limited. Judges are also generally bound to restrict themselves to issues submitted to them by the parties. It is incumbent on the Commission to afford as much evidence as may be necessary to support the establishment of an infringement or the rejection of a complaint, in particular where it appears likely that there will be follow-on private action. The Commission has also published guidance on certain issues to assist courts, for example as to how damages can be quantified. The conditions under which courts – beyond the evidence and pleading furnished by the parties – may obtain necessary clarification of issues of fact or economic analysis, for example in the form of amicus curiae briefs, could be examined further by the legislator. There has been substantial progress in facilitating business-to-business private action in the US and Europe. However, there is some concern about the potential abuse by defendants of recourse to arbitration, particularly where collective redress might be excluded by means of carefully crafted arbitration clauses. With regard to collective redress, there also appears to be a legal and cultural divide, as US colleagues traditionally tend to favour opt-out solutions while most EU colleagues traditionally prefer opt-in. Some hybrid solutions could be considered, partly to take account of the potentially different interests and capacities of final consumers and perhaps SMEs on the one hand and large, sophisticated enterprises on the other. While currently emphasis is placed more on collective action for redress, in practice collective action has been more visibly successful in bringing abuses to an end and encouraging new behaviours or forms of agreement on the market. Several colleagues emphasise the value of sanctions against the individuals who were personally responsible for infringements, notwithstanding the benefits of corporate fines. The development of national frameworks for encouragement of private action gives the EU the opportunity to learn from best practice and to move

338  Philip Lowe towards a minimum level of harmonisation which does not frustrate national initiatives or attempt to impose a ‘once size fits all’ solution that is insensitive to a variety of different institutional structures and procedures, in particular in smaller jurisdictions. However, markets in Europe are increasingly integrated and anticompetitive conduct by companies operating throughout Europe is usually replicated in many jurisdictions. There is a strong case for putting in place a minimum level of harmonisation both in the interests of providing predictability for companies and offering consumers the opportunity for collective action on a stronger Europe-wide basis. 21. Issues of redress are by no means related only to antitrust violations, and specific provisions in the antitrust field need to be carefully designed to avoid creating unintended consequences in other fields. 22. Finally, the US, Canadian and EU experiences of private action prompt colleagues to emphasise the importance of a careful interplay of mechanisms for public and private enforcement and the need for continuous review, experimentation and improvement in the search for the most effective enforcement system based on an optimal use of the resources at the disposal of public administrations, of the judicial system, and of businesses and consumers.

Veljko Milutinović*

The ‘Right to Damages’ in a ‘System of Parallel Competences’: A Fresh Look at BRT v SABAM and its Subsequent Interpretation

Introduction A decade elapsed between the two European Competition Law Annual Workshops at the EUI which have dealt specifically with the private enforcement of EU antitrust law.1 The first Workshop in 2001 took place at a uniquely interesting and uncertain time: the 1999 White Paper on Modernisation was still stirring controversy;2 what ultimately became Regulation 1/2003 was being drafted by the European Commission; and the seminal Courage case was still pending before the European Court of Justice.3 In the intervening period between the 2001 and 2011 Workshops, the private enforcement debate has, of course, moved forward. Following Regulation 1/2003, which set out the basics of decentralized (and thus, also, private) enforcement of EU antitrust rules, the Commission published its Green Paper in 2005 and its White Paper in 2008.4 In the same general period, the Courage judgment was reaffirmed and built upon by the ECJ in Manfredi5 in 2006. * At the time of writing, Jean Monnet Fellow, Robert Schuman Centre for Advanced Studies, EUI, Florence; then and currently, Assistant Professor at the Graduate School of Business Studies, Megatrend University, Belgrade. 1 See Claus-Dieter Ehlermann and Isabela Atanasiu, eds, European Competition Annual 2001: Effective Private Enforcement of EC Antitrust Law, Hart Publishing, 2003. 2 White Paper on Modernisation of the Rules Implementing Articles 85 and 86 of the EC Treaty, 1999 OJ C132/1. See also the discussion in Claus-Dieter Ehlermann and Isabela Atanasiu, eds, European Competition Annual 2000: The Modernisation of EC Antitrust Policy, Hart Publishing, 2001; and see generally Veljko Milutinović, The Right to Damages under EU Competition Law: from Courage v Crehan to the White Paper and Beyond, Kluwer, 2010, chapter 2 (with references to further literature at footnote 13). 3 See Proposal for a Council Regulation on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty and amending Regulations (EEC) No. 1017/68, (EEC) No. 2988/74, (EEC) No. 4056/86 and (EEC) No. 3975/87 (‘Regulation implementing Articles 81 and 82 of the Treaty’), COM (2000) 582, and Council Regulation (EC) No. 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, 2003 OJ L1/1; Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297. The Opinion of Advocate General Mischo in the Courage case was already known, and his recommendation as far as the specific outcome in the case was consistent with the judgment later handed down by the Court; what changed was the Court’s rationale, as will be seen below. 4 Green Paper on Damages Claims for Breach of the EC Antitrust Rules, COM (2005) 672 final; White Paper on Damages Claims for Breach of the EC Antitrust Rules, COM (2008) 165 final. 5 Joined Cases C-295 to C-298/04, Manfredi v Lloyd Adriatico Assicurazioni [2006] ECR I-6619.

342  Veljko Milutinović Since Manfredi was decided, and up to the time of writing (ie, 2013), while a lively academic and political debate has continued, legislative action by the Council has remained elusive.6 In addition to the shifts forward, occasioned mainly by the Green Paper and the White Paper, the period after the first European Competition Law Annual on private enforcement saw the debate wandering away somewhat from its origins and, ultimately, from its proper place in the overall narrative of the evolution of EU law. Two (by now seemingly well-established) notions developed by the ECJ are of special concern in this respect. While laudable in principle, through their subsequent interpretation and application the twin notions of a ‘system of parallel competences’ (as a rule of interaction between public and private enforcement) and the ‘right to damages’ established in Courage have become a source of confusion and difficulty. This is so in particular as regards the ‘binding effect’ of public enforcement decisions in civil cases, the relationship between leniency programmes and ‘follow-on’ claims for damages, and prudential limitations on such claims. The aim of this paper is to re-examine those twin notions in their historical context, in an attempt to clarify and, where necessary, limit their scope. It is hoped that this exercise will help to untangle some of the recurring conundrums of private enforcement which are of relevance to its future development.

The ‘foundational myth’: BRT The twin notions of a ‘right to damages’ and a ‘system of parallel competences’ are almost four decades in the making. Their joint ‘foundational myth’7 can be traced back to the ECJ’s (retrospectively) seminal judgment in BRT.8 In that case, the Court was asked whether the duty set out in Article 9 of the thenapplicable Regulation 17/62,9 which deprived NCAs of jurisdiction whenever the Commission opened proceedings to investigate a potential Article 101 or Article 102 infringement, also deprived national courts of their jurisdiction. The answer was in the negative, and the grounds were as follows: 6 Commission Work Programme 2012, Annex I, point 7, http://ec.europa.eu/atwork/pdf/cwp2012_ annex_en.pdf (last viewed 1 March 2013). However, as will be seen in the postscript to this chapter, after the conclusion of most contributions to this volume the Commission did produce, in June 2013, a significant legislative proposal. In this respect, the body of the chapter will not take into account the Commission’s latest proposal, in order not to bestow ‘retrospective wisdom’. Rather, it will examine the situation prior to June 2013. In doing so, it will make evident the need for legislative intervention, which now seems to be on its way. 7 The term ‘foundational myth’ is borrowed from sociology, in particular the work of Antoine Vauchez on the origins of direct effect and primacy of EU law. See Antoine Vauchez, ‘‘Integrationthrough-law’: Contribution to a Socio-history of EU Political Commonsense’, EUI Working Paper (RSCAS) 2008/10, at p 7. 8 Case 127/73, BRT v SABAM [1974] ECR 51. 9 Council Regulation (EEC) 17/62, First Regulation implementing Arts 85 and 86 of the Treaty [1959–1962] OJ Eng. Spec. Ed. 87.

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

343  

As the prohibitions of Articles [101](1) and [102] tend by their very nature to produce direct effects in relations between individuals, these Articles create direct rights in respect of the individuals concerned which the national courts must safeguard. To deny, by virtue of the aforementioned Article 9 [of Regulation 17/62], the national courts’ jurisdiction to afford this safeguard, would mean depriving individuals of rights which they hold under the Treaty itself.10

This dictum has had a persistent influence on the thinking of the Commission. When a new enforcement regulation (Regulation 1/2003) was proposed and enacted a little less than thirty years later, the relevant part of the rule remained unaltered: the Commission could remove the competence of national competition authorities (‘NCAs’), but not that of national courts.11 This view was reinforced in Delimitis,12 a judgment that can (in retrospect) be considered a decisive blow to the system of Article 101(3) mandatory notification/prior exemption of restrictive agreements administered by the Commission.13 In that case, the Court found that, not only were national courts not obliged to await a decision of the Commission if they were certain that there was no restriction of competition under Article 101(1) (and thus, by definition, no infringement),14 they could even (de facto, if not de jure), decide the issue of exemption under Article 101(3) in those cases where it seemed certain that the Commission would not grant an exemption.15 Subsequently, in Masterfoods,16 the Court was called upon to decide whether national courts had a duty to stay proceedings before them while a case based on the same alleged infringement which was the subject of a Commission decision was being heard by what is now the General Court (‘GC’) in an action for annulment. The Advocate General argued that, if the validity of the national court’s ruling depended on the validity of the Commission’s decision,17 the national court should stay its proceedings and await the outcome of the action for annulment.18 The action for annulment before the GC should not be circumvented via a preliminary reference to the ECJ, for two main reasons. The first was that ‘cases where exactly the same legal issue is raised in two parallel legal procedures, completely independent of each other, by the same parties, are clearly due to an abnormality of a procedural system’ and should therefore be avoided.19 A second, related point, BRT, cited above note 8, paras 16–17. See Recital 17 of Regulation 1/2003, cited above note 3. 12 Case C-234/89, Delimitis v Henninger Brau [1991] ECR I-935. 13 The final blow came, of course, when Article 101(3) was granted direct effect by virtue of Article 1 of Regulation 1/2003. 14 See also BRT, above note 8, para 22. The wording in BRT is narrower, as it allows national courts to proceed only when either (a) there is no appreciable effect on competition (de minimis), or (b) there is no appreciable effect on inter-state trade. In contrast, the Court in Delimitis (above note 12, para 50) did not limit the grounds on which the inapplicability of Article 101(1) could be found. It merely referred to situations where ‘the conditions for the application of Article [101(1)] are clearly not satisfied’. 15 Delimitis, above note 12, para 50. 16 Case C-344/98, Masterfoods v HB Ice Cream Ltd. [2000] ECR I-11369. 17 Which it inevitably does, if one assumes that Commission decisions have a ‘binding effect’, as the ECJ did in that case. See the discussion below (‘From Parallel Competences to ‘Binding Effect’’). 18 Opinion of Advocate General Cosmas in Masterfoods, cited above note 16, at para 108. 19 Ibid, paras 45–48. 10 11

344  Veljko Milutinović was that the two types of proceedings are different and the preliminary reference procedure does not entail the same detailed review of the Commission’s decision. Paradoxically (and subject to ultimate resolution through an appeal to the ECJ), in the absence of such a duty to stay proceedings, the ECJ and the GC could issue mutually conflicting rulings – while both courts remain within their rights.20 The Court refused to follow the Advocate General’s logic. Instead, it gave national courts a choice: they could either stay their proceedings while the action for annulment is pending or submit a preliminary reference to the Court.21 The Court followed the logic of ‘parallel competences’ as embraced in BRT and Delimitis, preferring not to interfere with national courts; in doing so, it opened the door to parallel decision-making by itself and the GC. The Court then established what subsequently came to be known as the ‘binding effect’ of Commission decisions as an exception to ‘parallel competences’ while refusing to find another exception that would force national courts to await the GC’s decision. The implications of this choice will be discussed further below. The preceding trilogy of cases represents the main judicial architecture of the system of ‘parallel competences’. According to this notion, the Commission (and the NCAs) can and does (do) pursue infringements in parallel with national courts, so that public and private enforcement can be competent for the same case simultaneously; the ‘abnormality of the procedural system’ that ‘parallel competences’ create is not taken into account. Public and private enforcement are, in principle, ‘equal’ in rank. While BRT, Delimitis and Masterfoods provided the ‘groundwork’ for equality as non-interference, it was clear in the mind of the Commission, as late as 2000, that there was a division of labour between competition authorities, which primarily served to protect the public interest, and national courts, which primarily served to protect private interests; private enforcement was to be mainly a ‘complement’ to public enforcement.22 The notional equality of importance between the two systems was completed through the ‘Courage turn’ and its rights-based discourse, discussed below.

Ibid, paras 50–53. See paras 57 and 60 of the Court’s judgment. 22 Notice on cooperation between national courts and the Commission in applying Articles 85 and 86 of the EEC Treaty, 1993 OJ C39/05, point 4. While this division may be (partly) attributed to the existence of the Commission’s ‘exemption monopoly’ prior to Regulation 1/2003, the same tone was maintained in the explanatory notes to the Commission’s proposal for that Regulation (Proposal for a Council Regulation, cited above note 3), where the Commission stated at p 6: ‘Unlike national authorities or the Commission, which act in the public interest, the function of national courts is to protect the rights of individuals. They can grant damages and order the performance or nonperformance of contracts. They are the necessary complement to action by public authorities’; and at p 9: ‘As the proposal aims at increasing the level of private enforcement before national courts, an initial increase in Article [267 TFEU] references can be expected. A significant increase, however, is unlikely, as it is expected that most litigation before national courts will concern areas where the law has been clearly established’ (emphasis added). 20 21

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

345  

The ‘Courage turn’ As is well known to those who study private antitrust enforcement, the question before the ECJ in Courage was relatively limited in scope; it was essentially the following: is the application of the English doctrine of in pari delicto, insofar as it would prevent a party to a restrictive agreement from recovering the loss he has suffered thereunder, contrary to EU law?23 It is evident from the opinion of Advocate General Mischo, who reached the same result as the Court but on somewhat different grounds, that the question could have been answered purely as a matter of the effectiveness of EU law, without recourse to any ‘inherent’ right to damages arising under EU law. The Advocate General essentially reasoned from the following premise: assuming that English law envisages a right to damages for antitrust infringements (which it does), is an exception to this right, which would exclude any party to a restrictive agreement from claiming such damages, contrary to EU law?24 The Advocate General considered that it was. Although it reached the same operative outcome, the Court proceeded from a different direction: it constructed a link to the past, and in particular to Van Gend en Loos and BRT, basing its arguments on the ‘legal order’ ‘created’ by the Treaty, which ‘give[s] rise to rights which become part of [individuals’] legal assets’.25 In the Court’s view, the rights conferred by the legal order as conceived in Van Gend en Loos were such that ‘any individual’ is entitled to ‘claim damages for loss caused to him by a contract or by conduct liable to restrict or distort competition.’26 Another significant turn, which preceded and enabled Courage, was the state liability ‘turn’, which occurred in Francovich and Brasserie du Pêcheur/ Factortame.27 There, the ECJ found – with no explicit basis in the Treaty – that Member States could be liable in damages for failing to implement a directive or enacting legislation that runs counter to directly effective Treaty provisions. The legal requirements for incurring liability were transposed, grosso modo, from the ECJ’s case law on the liability of the EU Institutions (which, unlike Member State liability, was envisaged in the Treaty).28 As Advocate General Van Gerven foreshadowed seven years prior to the Courage judgment, in his Opinion in Banks,29 the ECJ made a deductiveteleological use of state responsibility, extending it to the horizontal plane and Courage, cited above note 3, para 16. Ibid. The Advocate General’s Opinion should be read in its entirety, but paras 42–44 are especially pertinent. 25 Ibid, para 19 of the judgment. 26 Ibid, para 26. 27 Joined Cases C-6 and C-9/90, Francovich and Bonifaci v Italy [1991] ECR I-5357; Joined Cases C-46 and C-48/93, Brasserie du Pêcheur v Germany and R v Sec. of State for Transport ex parte Factortame [1996] ECR I-1029. 28 Francovich, ibid, paras 41–42. For discussion of this process of transposition and its potential setbacks, see Milutinović, above note 2, section 3.3.2. The non-contractual liability of the EU is set out in Article 340(2) TFEU. 29 Case C-128/92, Banks v British Coal [1994] ECR I-1209. 23 24

346  Veljko Milutinović finding that a ‘Community right to damages’ exists between individuals under Article 101.30 Effectively, in Courage, the Francovich/Brasserie du Pêcheur rules against recalcitrant Member States were extended to undertakings that fail to comply with Treaty rules addressed to them. This transposition is largely unobjectionable as such (ie, provided that one accepts the basic premise of Member State liability somehow being ‘inherent in the system of the Treaty’).31 Indeed, one might say that the transposition reflects the purported symmetry between state- and privately-induced partitioning of the internal market, which the Court established decades earlier in Consten and Grundig.32 Prior to Courage, it was easy enough to confine the notion of ‘parallel competences’ to its functional origins: arguably, the Court in BRT and Delimitis sought to lighten the Commission’s burden using the same reasoning as in Van Gend en Loos,33 whereby ‘private attorneys general’ supplement the enforcement of EU law.34 This supplementary enforcement was arguably at the root of the development and centralization of the concept of direct effect. Quite simply and rationally, it was clear that centralized enforcement proceedings pursued by the Commission against recalcitrant Member States would never suffice to make EU law effective, as the Commission could not detect and prosecute all (or even most) infringements of EU law, considering both institutional capacity constraints and the cumbersome nature of the infringement procedure. BRT represents, in a sense, a horizontalization of the Van Gend en Loos principle, just as Courage represents a horizontalization of the Francovich/ Brasserie du Pêcheur principle. Functionally, this horizontalization probably seemed necessary in a time when national competition authorities (NCAs) were not obliged to apply EU law and some Member States did not even have an NCA.35 As late as the time of the 1999 White Paper, private enforcement was still viewed by the Commission in the broadly functional light of ‘decentralization’: national courts and NCAs were to relieve the Commission of a big part of its enforcement burden,36 as the Commission simply could not and would not pursue all competition law infringements in an ever-larger Union.37 National courts and See Opinion of the Advocate General, paras 38–45. Milutinović, above note 2, section 4.3.2. 32 Cases 56 and 58/64, Consten and Grundig v Commission [1966] ECR (Eng. Spec. Ed.) 299, at 341: ‘The Treaty, whose Preamble and content aim at abolishing the barriers between States, and which in several provisions gives evidence of a stern attitude with regard to their reappearance, could not allow undertakings to reconstruct such barriers. Article [101(1)] is designed to pursue this aim.’ 33 Case 26/62, Van Gend & Loos v Netherlands Inland Revenue Administration [1963] Eng. Spec. Ed. 1. See especially the dictum at p 13: ‘The vigilance of individuals concerned to protect their rights amounts to an effective supervision in addition to the supervision entrusted by Articles [258] and [259] to the diligence of the Commission and of the Member States.’ 34 See Paul Craig, ‘Once upon a Time in the West: Direct Effect and the Federalization of EEC Law’, 12 Oxford Journal of Legal Studies 453, 455 (1992); Milutinović, above note 2, section 3.2.1. 35 The obligation to apply EU law in certain circumstances was introduced by Article 3(1) of Regulation 1/2003. Among the nine Member States that belonged to the EEC in 1974, Italy did not at that time have a competition authority; the AGCM was ultimately established in 1990. 36 See 1999 White Paper, cited above note 2, especially points 46 and 91–100. 37 Ibid; Case T-24/90, Automec Srl v Commission (II) [1992] ECR II-2223, paras 75-77, where the GC upheld the Commission’s discretion as to whether to investigate a complaint. 30 31

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

347  

NCAs would fill the enforcement gap and allow the Commission to focus on legislation, policy-making and the prosecution of the most serious infringements. The significance of the Courage turn is that it surpasses the functional, burdensharing explanation, places private enforcement in the forefront and ‘explains’ (with hindsight) which ‘direct rights’ the Court ‘had in mind’ in BRT – one such right being the right to damages. Henceforth, any force that stood in the way of ‘parallel enforcement’ would stand not only in the way of the competence of national courts, which could be legislated away (as it was in Regulation 17/62 and, arguably, in the two successive Merger Regulations38) or reshaped (whenever the ECJ found it inadequate for the effective application of EU law – as in Simmenthal39) but also in the way of individual rights. Arguably (and this is the essence of the turn), individual rights granted by the Treaty could not be legislated away or judicially removed, at least not without significant difficulty and definitely not without a concept for resolving conflicts between rights, a concept which is almost entirely absent from EU law. If individual rights were derived ‘directly’ from the Treaty, national courts would obtain a role that is seemingly independent of the competition policy conducted by the Commission.

Escaping the retrodiction of ‘individual rights’ When the Court decided BRT, it did not make clear which rights exactly it had in mind, and how it was exactly that they emanated ‘directly’ from Articles 101 and 102. While the Court relied explicitly on BRT to construct the ‘right to damages’ in Courage,40 to say that the Court in BRT had damages (or any other remedy) ‘in mind’ would be a ‘teleological retrodiction’: an intellectual construct that interprets the past based on ‘logical outcomes’ of major ECJ judgments.41 Although the ECJ formally follows the Continental method and does not follow a common law doctrine of stare decisis,42 the operation of precedent is 38 Regulation 17/62, cited above note 9, Article 9(1); Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings, 2004 OJ L24/1, Article 22, paras (1) and (2). 39 Case 106/77, Amministrazione delle finanze dello Stato v Simmenthal [1978] ECR 629, paras 20– 22, effectively empowering all national courts to disapply national law on the basis of incompatibility with EU law, regardless of the internal allocation of judicial competence in a given Member State. 40 Courage, note 3 above, para 23. 41 See Vauchez, note 7 above, p 4. The present author has once contributed to perpetuating the retrodiction of BRT. See Veljko Milutinović, ‘Private Enforcement: Upcoming Issues’, in Giuliano Amato and Claus-Dieter Ehlermann, eds., EC Competition Law: A Critical Assessment, Hart Publishing, 2007, p 725, at p 726, stating, without qualification, that BRT provided a self-contained EU ‘legal basis’ for antitrust damages claims. 42 Or, as Anthony Arnull has observed, the Court is not formally ‘bound’ by its previous decisions but it does, generally, observe them and it tends to use a process of ‘distinguishing’ past case law when ruling against it; see Anthony Arnull, The European Union and its Court of Justice, OUP, 1999, Chapter 15.

348  Veljko Milutinović amply evidenced by both the frequent and seemingly binding (or at least highly persuasive) references to past judgments and the doctrine of acte clair, whereby Article 267 TFEU preliminary references may be rejected, inter alia on the ground that the point of law being raised has already been decided by the Court in a previous case.43 The way in which the Court reconciles acte clair with the absence of stare decisis is by saying that it does not pronounce ‘new’ law: it merely interprets the law as it always was,44 thus opening the way for teleological retrodictions.45 The dangers entailed in this approach can easily be demonstrated in the example of BRT and its subsequent interpretation. In 1974, it was not at all obvious that Articles 101 and 102 had direct effect, for several reasons. Until that time, Van Gend en Loos was the ‘law of the land’ as regards direct effect. The case law on direct effect has mutated significantly over the past fifty years,46 but Van Gend en Loos inferred direct effect for those Treaty provisions that contain a ‘clear and unconditional prohibition, which is not a positive but a negative obligation’. Such an obligation, furthermore, ‘is not qualified by any reservation on the part of states which would make its implementation conditional upon a positive legislative measure enacted under national law.’ Articles 101 and 102 TFEU were and remain negative obligations (conduct X ‘shall be prohibited’). Likewise, their application was not subject to any measure of national law (save in the period prior to the enactment of Regulation 17/62).47 However, to say that the prohibitions in Articles 101 and 102 were clear requires a long stretch of interpretation. In the period between the enactment of the Treaty of Rome and BRT, the Court had to clarify, word by word, virtually the entire text of Article 101. Among other things, the Court had to answer very basic questions regarding, inter alia: what is meant by ‘agreement’;48 what is meant by the term ‘undertaking’;49 what ‘affects trade between the Member States’;50 whether and how Article 101 applied to ‘intra-brand’ restrictions of competition;51 and indeed, Case 283/81, CILFIT v Ministry of Health [1982] ECR 3415, para 21. Case C-453/00, Kühne & Heinz v Produkschap voor Pluimvee en Eiren [2004] ECR I-837, para 21, with references to earlier judgments to this effect. 45 Interestingly, in CILFIT, cited note 43 above, the Court seems to have left room for a historically conscious interpretation: at para 20, it found that, when interpreting EU law, regard must be had, inter alia, to ‘its state of evolution at the date on which the provision in question is to be applied’. 46 Van Gerven, in Banks, note 29 above, refers at para 27 to ‘minor differences’ in the wording of direct effect; he then points, however, to quite significant evolutions of substance, such as the introduction of direct effect in the context of directives. 47 Article 104 TFEU gave Member States competence to apply Articles 101 and 102 in accordance with national law, ‘until’ the entry into force of the instruments referred to in Article 103. Strangely, this provision has survived since the Treaty of Rome (1957) and remains in place today, despite the fact that not one but two relevant regulations were adopted by the Council in the meantime. 48 Grundig, cited above note 32. 49 Case 155/73, Italy v Sacchi [1974] ECR 409, paras 13–17. More precisely, the judgment in Sacchi was adopted three months after BRT. 50 Grundig, above note 32, at p 341; Case 5/69, Völk v Vervaecke [1969] ECR 295, paras 5 and 7; Case 8/72, Vereeniging van Cementhandelaren v Commission [1972] ECR 977, para 29 (‘effect on trade’ exists whenever an agreement affects the whole territory of one Member State). 51 Grundig, above note 32, pp 342–343. 43 44

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

349  

what a ‘restriction of competition’ is in the first place.52 All of these concepts were still undergoing significant evolution long after BRT was decided.53 With regard to Article 102, the very concept of dominance under Article 102 was, as yet, undefined: 1974 was four years before Hoffman La Roche and five years before United Brands, so the terms ‘dominance’ and ‘abuse’ had, at that time, virtually no content. In addition, in 1974 the Commission’s ‘exemption monopoly’ was very much alive and Article 101(3) was thus excluded, by definition, from being directly effective. Interestingly, almost two decades after BRT, in Banks, Advocate General Van Gerven sought to draw a parallel between Article 101(1) TFEU and Article 65(1) of the European Coal and Steal Community (‘ECSC’) Treaty; he used BRT by means of analogy in order to show that the ‘monopoly’ held by the Commission with regard to exemption does not exclude the prohibition from being directly effective.54 The ECJ flatly disagreed, ignored any analogy with BRT and found, instead, that since the exemption was not directly effective, the entire provision of Article 65 ECSC was likewise not directly effective.55 Last but not least, the BRT judgment itself seems uncertain about the direct effect of Article 101(1), as the Court only allowed the national courts to ignore Commission proceedings if the agreement before them was either de minimis or devoid of effect on inter-state trade. Indeed (and this should serve as another warning against teleological retrodictions), while BRT is regarded as the source judgment for direct effect of Articles 101(1) and 102, it was Delimitis and not BRT that allowed the national courts to carry out a substantive assessment under Article 101(1).56 How can the direct effect of Articles 101 and 102 be rationalized, then, when these provisions were not, judging from their text, ‘clear and unconditional’ within the meaning of Van Gend en Loos and when the Court itself was hesitating to allow national courts to apply Article 101(1) in full? The only rational explanation may lie along the lines of critical legal theory: the direct effect of Articles 101 and 102 had to be endorsed for a practical reason; subsequent retrodiction then solidified the narrative. In this respect, it should be recalled that 1974 was a year of ‘revolution’ for direct effect: during that year, the Court broadened this concept almost beyond recognition, when it significantly reduced the importance of unconditionality (Reyners) and of the ‘no national Ibid; Völk, above note 50. See, among many others, Case 107/82, AEG v Commission [1983] ECR 3151 (concept of ‘agreement’ and ‘effect on trade’); Case 26/76, Metro v Commission [1977] ECR 1875, paras 20–23 (‘intra-brand’ competition); Case C-73/95 P, Viho Europe v Commission [1996] ECR I-5457 (concepts of ‘agreement’ and ‘undertaking’). 54 Opinion of the Advocate General in Banks, cited above note 29, at para 30. 55 Ibid, para 17 of the judgment. Curiously, whatever today’s ‘teleological retrodictions’ may state, the Court did not explicitly address the impact of the removal of direct effect so far as Article 101(3) was concerned on the remainder of the Article. It merely listed Article 101(1) as being directly effective; this reading of BRT comes from Delimitis, cited above note 12, paras 44–45. 56 See the text and references above at note 14. 52 53

350  Veljko Milutinović measures needed’ criterion (Van Duyn).57 BRT can therefore be seen as part of a judicial activism trio in 1974 – rather than as a judgment based on settled legal principles. Turning to the vagueness of BRT’s reference to ‘direct rights’, there were arguably no direct rights (much less remedies) for individuals that could be directly inferred from Articles 101 and 102 in 1974. The only ‘right’ or ‘remedy’ that one might argue could be deduced from the text of the Treaty is the nullity of restrictive agreements under Article 101(2). The problem with this argument is that nullity is neither a right nor a remedy. Arguably, nullity could be construed as a (public law) sanction, with civil law effects.58 In this connection, one must bear in mind the fundamental distinction between voidance and nullity which is often ignored: the former normally operates upon invocation, while the latter operates ab initio and in the absence of invocation.59 When a provision of an agreement is null, it is not null because someone has declared it to be null – which is what happens in the case of voidance. Nullity does not exist because someone can claim it – which would make it a right. This is because, in principle, a civil law right entails the right to forego the right.60 Since no one can waive nullity,61 and since nullity operates, by definition, in the public interest, nullity cannot be a right. Nullity is also not a remedy, as it exists, regardless of harm, as a matter of public interest. This much seemed clear at least as late as 1986, when the ECJ last ruled directly on the issue and distinguished nullity (which operated as a matter of EU law) from the civil law consequences of nullity (which were a matter for national law).62

57 In Case 2/74, Reyners v Belgium [1974] ECR 631, paras 26–30, the Court found that, although what is now Article 49 TFEU (freedom of establishment) required the EEC to adopt implementing legislation, that did not prevent the core prohibition of discrimination from having direct effect in the meantime. In Case 41/74, Van Duyn v Home Office [1974] ECR 1337, paras 11–13, the Court found that provisions of directives could have direct effect – in that case because the Van Gend en Loos criteria were met and because the applicability of the relevant provision depended entirely on the conduct of individuals. 58 Milutinović, above note 2, section 7.2, without, however, expressing whether the sanction is ‘public’ or ‘private’. Arguably, the notion of ‘sanction’ is more appropriate to public law, which orders conduct and punishes non-compliance but does not typically balance private interests. The dichotomy between this sanction and its civil law consequences is consistently upheld in the case law of the ECJ. See Case 56/65, Société Technique Minière v Maschinenbau Ulm GmbH [1966] (‘STM’) ECR Eng. Spec. Ed. 235, at 250; Case 319/82, Société de Vente de Ciments et Bétons de l’Est v Kerpen & Kerpen [1983] ECR 4173, para 12. 59 See Milutinović, ibid, explaining the distinction and the problems arising from the failure to observe it. In the case of English law, the ‘nullity’ referred to in Article 101(2) may be closer to illegality than to voidance. 60 The principle is subject to the notable exception of the right to (one’s own) life, although this too is debatable in some jurisdictions and some contexts. 61 The ECJ often referred to nullity as ‘automatic voidance’, which indicates that what is at stake is indeed nullity. See, eg, STM, cited above note 58, especially at p 250. 62 Case 10/86, VAG France v Établissements Magne [1986] ECR 4071, paras 14–16, referring to STM and Kerpen & Kerpen (each of which is cited above note 58).

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

351  

From ‘parallel competences’ to ‘binding effect’ After the Court found, in BRT, that there is a system of parallel competences, subsequent judgments were drawn into a teleological-deductive path dependency. This was not inevitable, but it became so when the Court construed BRT as enunciating a principle – rather than a mere practical solution – in Delimitis. In the latter case, the Court painted itself into a corner by finding that national courts could essentially ignore the Commission in many or most cases.63 The problem became apparent immediately in Delimitis, even as the Court was composing its judgment: if national courts are not obliged to wait for the Commission, they could conceivably adopt judgments that conflict with Commission decisions. Thus, a conflict resolution rule was needed and the Court created one, in the same judgment and before endorsing parallel proceedings; the rule stated that national courts should ‘avoid’ taking decisions that might conflict with conceivable future decisions of the Commission.64 The rule was rudimentary and incomplete, as it did not state what would happen if a national court failed to avoid this scenario and a conflicting judgment were indeed adopted.65 The rule was also the first major crack in the notion of ‘parallel competences’. Although competences still ran ‘in parallel’ in principle, the scales were tipped in the Commission’s favour: it was the national courts that had to take heed of the Commission’s decisions and not the other way around.66 When the Court decided Masterfoods in 2000,67 the ‘unpainted corner’ shrank further. Rather than following its Advocate General and finding that national courts must await the outcome of the action for annulment before the GC,68 the Court extended Delimitis to scenarios where such actions are pending before the GC, thus indicating that the action for annulment was part of the public enforcement procedure to which national civil proceedings could run ‘in parallel’. As a result, a stronger conflict resolution rule was required: the Court made it clear that, if a national court adopted a decision that conflicts with a Commission decision, the latter would prevail.69 This is the so-called ‘Masterfoods rule’ (also known as the ‘binding effect’ of Commission decisions) which, combined with the rule in Delimitis,70 was soon afterwards enshrined in Article 16(1) of Regulation 1/2003. The ‘binding effect’ of competition authority decisions vis-à-vis national courts remains one of the thorniest issues in private enforcement and will be discussed separately below. Delimitis, cited above note 12, para 50. Ibid, para 47. 65 See Milutinović, above note 2, section 11.3.2. 66 Indeed, the Court relied on a notion of ‘legal certainty’ to justify this rule; it was unclear, however, why ‘legal certainty’ should be a one-way street, operating, as it were, only against the national courts but not against the Commission. See Milutinović, above note 2, section 11.2.3. 67 Masterfoods, cited above note 16. 68 See discussion above (‘The ‘Foundational Myth’: BRT’). 69 Masterfoods, above note 16, paras 48 and 52. 70 See above note 64. 63 64

352  Veljko Milutinović

The conceptual problem with parallel competences Looking at the ‘big picture’, the main conceptual problem with ‘parallel competences’ is that competences, are, by their very nature, not meant to run in parallel.71 In any proceedings where his rights and/or obligations are being determined, a subject of the law must be able to say, with certainty, which institution is competent to hear his case. This is evident from the countless national and supranational acts that set out the competences of courts and administrative authorities and the means to resolve the problem of overlapping (or parallel) competences. Indeed, many national constitutions and similar basic acts grant the power to resolve disputes over competences to the highest (or one of the highest, if there are several) national courts.72 Generally speaking, public international law, as a decentralized amalgamation of rules that does not fit the classical definitions of a ‘legal order’,73 tolerates parallel competences,74 not least due to the prevailing notion of state sovereignty. By contrast, the EU is supposed to be a supranational legal order of limited sovereignty, which tends, as its name suggests, towards creating something more akin to the classical model of a legal order.75 The traces of order are evident, in this context, inter alia in the so-called ‘Brussels Regulation’ (formerly the Brussels Convention),76 which exists specifically in order to minimize litispendence between Member States. Closer to the present discussion, the socalled European Competition Network operates under a system of rules that are specifically aimed at reducing competence overlaps not just between the Commission and the NCAs, but also among the NCAs themselves.77 Viewed 71 In this sense, see the Opinion of the Advocate General in Masterfoods, cited above note 16, paras 45–48. 72 These are typically supreme or constitutional courts. Thus, for example, in Italy, under Article 65 of the Basic Law on the Organisation of the Judiciary (Legge fondamentale sull’ordinamento giudiziario del 30 gennaio 1941 n. 12) the Supreme Court of Cassation has the competence to resolve competence conflicts among the various courts. In Austria, a similar competence rests with the Constitutional Court, under Article 138(1) of the Federal Constitutional Law (Bundes-Verfassungsgesetz). 73 Contrasting it with the Hartian concept of a legal order, Shaw explains that public international law lacks ‘a recognised body to legislate or create laws, a hierarchy of courts with compulsory jurisdiction to settle disputes over such laws and an accepted system of enforcing those laws’. Malcolm N. Shaw, International Law, 6th edition, Cambridge University Press, 2008, pp 2–3. 74 For an example of how this works (or rather does not work) in practice, see Veljko Milutinović, ‘El Acuerdo de Asociación entre Centroamérica y la Unión Europea: lecciones e indicadores futuros para la liberación del comercio europeo’, in Joaquín Roy, ed., Después de Santiago: Integración Regional y Relaciones Unión Europea-América Latina, Miami-Florida European Union Center, 2013, pp 89–104, at pp 98–101. 75 Or, as the ECJ famously held in Van Gend en Loos, cited above note 33, ‘this Treaty is more than an agreement which merely creates mutual obligations between the contracting states[...]’, and further: ‘the Community constitutes a new legal order of international law.’ 76 Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, 2001 OJ L12/1, especially Art. 27. On 10 January 2015, this Regulation will be replaced by Regulation 1215/2012, of the same name; the relevant provision will be Article 29. 77 Articles 11 and 13 of Regulation 1/2003, cited above note 3; Commission Notice on cooperation within the Network of Competition Authorities, 2004 OJ C101/03.

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

353  

in this context, ‘parallel competences’ seem to represent an anachronism and a legal imperfection suitable for public international law but out of place in the EU’s supranational legal order.

The ‘Binding Effect’ of ‘Non-Infringement’ Decisions The Commission and some national courts have referred to existing Commission decisions as being ‘binding’ on the courts of the Member States.78 In civil law terms, ‘binding effect’ means that, in a case before it,79 a civil court must treat the existence of illegal conduct and the identity of the perpetrators as a given (and not as a rebuttable presumption) resulting from the Commission’s decision. This notion was confirmed explicitly by the ECJ in its recent Elevator Cartel (Damages) judgment,80 where the Court held that, in a ‘follow-on’ damages claim, ‘the national court is required to accept that a prohibited agreement or practice exists,’81 while ‘the existence of loss and of a direct causal link between the loss and the agreement or practice in question remains, by contrast, a matter to be assessed by the national court’.82 Although it is now clear that the ‘binding effect’ of Commission decisions exists, the rule in Article 16(1) of Regulation 1/2003 leaves many questions unanswered regarding the exact material and personal scope of the ‘binding effect’ concept. Limitations of space and relevance preclude a thorough discussion of all of them.83 However, it is necessary to address one question of significant practical relevance, namely: whether the ‘binding effect’ also applies to decisions whereby the Commission finds, under Article 10 of the Regulation, that certain parties did not infringe the competition rules through certain conduct. These decisions are established in the Regulation under the heading ‘findings of inapplicability’. For 78 See, eg, point 2.3. of the 2008 White Paper, cited above note 4, and points 8 and 13 of the 2004 Notice on Co-Operation, cited previous footnote. In Inntrepreneur Pub Company (CPC) and others v Crehan [2006] UKHL 38, the House of Lords consistently referred to this effect as ‘binding’ (although it declined to find such an effect in that case, due to the lack of identity of the alleged infringers – between the Commission decision being invoked and the case before the court). 79 At least one author has suggested that, as the Commission has only ‘claimed’ binding effect for NCA decisions in damages actions in its 2008 White Paper (at point 2.3.), such a binding effect would not extend to nullity or injunctions. See Kosmas Boskovits, ‘Modernization and the Role of National Courts’, in Ioannis Lianos and Ioannis Kokkoris, eds., The Reform of EC Competition Law: New Challenges, Kluwer, 2010, pp 95–117, at p 111. That view must be rejected. First, nullity arises from the existence of an infringement, which is deemed proven under Article 16(1) of Regulation 1/2003. Second, the rule in Article 16(1) is itself based on Masterfoods, where ‘binding effect’ was to be applied, inter alia, to an injunction issued by a national court in favour of the infringers and running counter to the Commission decision. 80 Case C-199/11, European Union v Otis et al., judgment of the ECJ of 6 November 2012, not yet reported. 81 Ibid, para 65 (emphasis added). 82 Ibid. 83 For such a discussion, see Milutinović, above note 2, chapter 11.

354  Veljko Milutinović the sake of clarity, the discussion that follows uses an arguably more apt term: ‘non-infringement decisions’. While the Commission and the ECJ have yet to address the issue directly, the Council, the GC and academic commentators (present author included) have seemed to cast doubt on the ‘binding effect’ of non-infringement decisions. Against such a ‘binding effect’ it has been questioned whether, since Articles 101 and 102 have direct effect, it must remain open to national courts to apply them – even when the Commission has established that they do not apply.84 It is now submitted that neither Masterfoods nor Article 16(1) can be used to support this view, as no useful distinction was made, either legislatively or judicially, with regard to the ‘positive’ or ‘negative’ nature of a Commission decision. It is not disputed today that Article 16(1) has displaced the direct effect of Articles 101 and 102 with regard to infringement decisions: parties found guilty of an infringement cannot rely on direct effect in order to ‘overrule’ a Commission decision before a national court. In the absence of a distinguishing feature, therefore, there is no reason to conclude that Article 16(1) does not displace direct effect with regard to non-infringement decisions. Like infringement decisions, these decisions establish a certain legal situation and that legal situation binds the national courts. A shadow of a doubt was cast by the GC in 2005: in First Data Corp,85 the court found, in an interim order, that Masterfoods does not apply to what were known, under Regulation 17/62, as ‘negative clearance decisions’, for three reasons. First, negative clearance decisions merely state that ‘for the Commission, on the basis of the facts in its possession, there is no need to intervene. Negative clearance does not therefore constitute a definitive assessment’.86 Second, negative clearance decisions also do not constitute: [...] the adoption of a position which falls within the exclusive competence of the Commission. As Article [101(1) TFEU] is directly applicable, as the Court of Justice has held on various occasions, it follows that individuals may rely on it before national courts and derive from it rights.

And third: [...] as national courts may also have other information on the particular circumstances of the case, they are naturally bound to reach their own opinion, on the basis of the information in their possession, on the applicability of Article [101(1)] to certain agreements.87

The first of the three reasons may have been valid under the then-applicable law, as the GC was applying Article 2 of Regulation 17/62, which provided for Ibid, section 11.4.5. Case T-28/02, First Data Corp et al. v Commission [2005] ECR II-4119. This judgment was cited (also with doubts) by Assimakis P Komninos, ‘Effect of Commission Decisions on Private Antitrust Litigation: Setting the Story Straight’, 44 Common Market Law Review 1387, 1394 footnote 34 (2007). 86 First Data Corp, cited previous footnote, para 50. 87 Ibid. (emphasis added). 84 85

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

355  

negative clearance decisions properly so-called. However, although one may be tempted to use the same term for the sake of ostensible historical continuity, Article 10 decisions under Regulation 1/2003 are substantially different. They do not provide that there are ‘no grounds for action’ by the Commission ‘based on the facts in its possession’, as Article 2 of Regulation 17/62 did. Instead, according to the wording of the current Regulation, the Commission finds under Article 10 that Article 101 or Article 102 is ‘not applicable’ to a certain factual situation. It is submitted that the change of wording is substantial and not accidental.88 Some confusion may have been brought about by Recital 14 of the Regulation, which states that Article 10 decisions are of a ‘declaratory nature’, thus ostensibly hinting at a sort of ‘softer’, ‘non-binding’ intent on the part of the Council. Nevertheless, as a matter of law, it must be recalled that, under the same Regulation, all infringement decisions are also, in principle, declaratory in nature, as constitutive individual exemptions under Article 101(3) no longer exist.89 Thus, with regard to Article 10 non-infringement decisions, this first argument no longer applies.90 It could be applied, arguably, if the Commission decided, under Article 10, that there are ‘no grounds for action’ on its part. However, it is submitted that such a finding (as opposed to a finding that Article 101 or Article 102 is not applicable) would run counter to the nature and purpose of Article 10 decisions. 88 This conclusion is supported by other provisions of Regulation 1/2003: in Article 5, the ‘no grounds for action’ formulation is retained for NCAs, who may not adopt non-infringement decisions but may find that there are no such grounds for action; in addition, Article 9(1) retains the ‘no grounds for action’ principle for commitment decisions adopted by the Commission. 89 Instead, according to Article 1(2) of Regulation 1/2003, ‘Agreements, decisions and concerted practices caught by Article [101(1)] of the Treaty which satisfy the conditions of Article [101(3)] of the Treaty shall not be prohibited, no prior decision to that effect being required.’ Interestingly, while the Commission’s Proposal for what became Regulation 1/2003 (see above note 3) also proclaimed Article 10 decisions to be ‘declaratory’ in nature, the Commission made it clear (at page 19) that their purpose was to provide greater legal certainty in the new, decentralized system. In addition, according to the Commission, ‘Article 16 of the proposed Regulation creates a general obligation for national competition authorities and national courts to make every effort to avoid decisions conflicting with decisions adopted by the Commission. A finding of inapplicability by the Commission pursuant to Article 10 can therefore make an important contribution to the uniform application of Community competition law.’ Be all of that as it may, the declaratory/constitutive dichotomy is useful only to the extent that it clarifies that, post-Regulation 1/2003, there is no exemption ‘monopoly’ for Article 101(3), whereas a monopoly on block exemptions under the same Treaty provision (which are ‘constitutive’) remains. Theoretically, one could claim that, due to the direct effect of Articles 101 and 102 and the consequently ‘declaratory’ nature of all decisions adopted in the application thereof, ‘binding effect’ cannot exist, by definition. Yet this is precisely the type of conundrum to which the subsequent interpretation of BRT leads, making any type of ‘primacy’ of public over private enforcement logically impossible, under a notion of directly effective ‘rights’, which competition authorities cannot override. That conundrum has, however, been solved by the ECJ in Delimitis and Masterfoods and endorsed by the EU legislator in Regulation 1/2003, through a conflict-resolution rule that leads to ‘binding effect’. 90 In the Commission Staff Working Paper accompanying the White Paper on Damages actions for breach of the EC antitrust rules, COM(2008) 165 final, points 152–153, the Commission specifically excluded, from its proposal in the 2008 White Paper, the establishment of binding effect as regards ‘negative’ NCA decisions, ie, decisions or parts of a decision stating that there seems to be no infringement. The basis for doing so was Article 5 of Regulation 1/2003, which does not list non-infringement decisions. From this one could deduce that, since non-infringement decisions are envisaged for the Commission, these too should be binding.

356  Veljko Milutinović As for the GC’s second and third arguments, they too ought to be rejected – both at the time they were pronounced and today – as they seem to expand the notion of ‘parallel competences’ constructed in BRT while ignoring the conflict resolution rule laid down in Masterfoods.91 The latter judgment can only be reconciled with the First Data Corp judgment precisely because of this conflict resolution rule: both the Commission and the national courts can apply Article 101(1) and Article 102 but, if there is a conflict, the application of these provisions as adopted by the Commission shall prevail. This determination is not dependent on whether the Commission has an ‘exclusive competence’ (as it once did, under Regulation 17/62, for the application Article 101(3)), and the binding effect of Commission decisions does not distinguish according to whether such a decision is ‘constitutive’ or ‘declaratory’. The only argument that may, perhaps, raise legitimate concerns with regard to the binding effect of non-infringement decisions is the judicial protection/due process argument. Unlike infringement decisions, non-infringement decisions are unlikely to be challenged by those to whom they are addressed. Instead, the interested party would normally be a third party, which, unlike the addressee, must show that he is ‘directly and individually concerned’ in order to have standing under Article 263 TFEU.92 The parallel, twin avenues for challenging Commission decisions left open by the ECJ in Masterfoods make sense in this scenario, as they can be used to plug a potential gap in the judicial protection of third parties. This ‘plug’ seems superfluous, however, as another one is already built into the system under TWD:93 as it may prove difficult for many or most third parties to prove ‘direct and individual concern’ in challenging a non-infringement decision, they can mount an indirect challenge under Article 267 TFEU, without the need for any Masterfoods parallelism.94 It is not entirely clear, however, whether Masterfoods has displaced TWD specifically with respect to competition law, as the Court in Masterfoods did not address the issue.95 It is submitted, however, that parallelism between annulment proceedings and preliminary references has no place beyond the context envisaged in TWD. Indeed, as Advocate General Cosmas has argued – to no avail – the Article 263 action for annulment is the proper procedure for challenging Commission decisions;96 the constraints and special features – ratione Similarly, see Komninos, cited above note 85. For problems with regard to applying the so-called Plaumann test in competition cases, see Milutinović, cited above note 41, at pp 732–735. 93 Case C-188/92, TWD Textilwerke v Germany [1994] ECR I-833, para 26. 94 In TWD the ECJ held, essentially, that the ‘indirect challenge’ of EU acts through Article 267 references is inadmissible if the affected parties were informed of their right to file an action for annulment under Article 263 and if their standing in such an action would have been certain; therefore, it may be admissible if either (a) the parties were not thus informed, or (b) their standing was uncertain. See also Case C-239/99, Nachi Europe v Hauptzollamt Krefeld [2001] ECR I-1197, para 29, where TWD was cited with approval; the latter judgment was adopted three months after Masterfoods. 95 The relevance of TWD was raised by the Italian Government, however, as may be seen at para 36 of the judgment. One may presume that the Court did not address the issue because, in that case, an action for annulment was filed by the addressees in a timely manner, and TWD would therefore not apply in any event. 96 See above notes 19 and 20. 91 92

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

357  

temporis,97 ratione personae98 and ratione materiae99 of this procedure should not be circumvented via Article 267 preliminary references.

Apparent problems emanating from the concept of ‘binding effect’ While the ‘binding effect’ of Commission decisions is now settled in the case law and in legislation, the extension of binding effect to the decisions of NCAs has remained controversial since the 2008 White Paper was published.100 Viewed under the hypothesis that there is a ‘system of parallel competences’, the proposition of introducing a ‘binding effect’ of NCA decision was criticized, inter alia, on the ground that it turns national courts into ‘mere assessors of damages’,101 and that it violates the principle of independence of the judiciary.102 Elsewhere I have set out and addressed most of the recurring objections to endowing NCA decisions with ‘binding effect’, and that discussion will not be replicated here.103 In brief, it was shown that the principle of independence of the judiciary is not relevant in this context, as this is not the type of situation that the theory (or practice) of the separation of powers was intended to address. It may now be added that imposing limitations on the sphere of competence of national courts (ie, limiting their role to the assessment of damages) would hold only in ‘follow-on’ cases: insofar as national courts hear new cases (ie, ‘stand-alone’ cases not previously investigated by a competition authority), the competence of national courts would remain unfettered. This solution is commensurate with national courts’ greater contribution to enforcement in ‘stand-alone’ cases: they do not ‘merely’ contribute to greater rectification of known infringements; they also 97 Under Article 263, the affected parties have a two-month time limit during which they must bring their action; under Article 267, there is no time limit, as the challenge to the Commission’s decision is refashioned as a ‘question regarding the interpretation of EU law’ which is, in principle, timeless. 98 Article 263 imposes significant constraints with regard to standing. See generally Paul Craig and Gráinne De Búrca, EU Law: Text, Cases and Materials, 4th edition, OUP, 2008, pp 509–528; and Milutinović, n. 92 above. By contrast, in Article 267 proceedings the case is initiated by a national court and not by the interested party; therefore, only limitations on admissibility, and not on the standing of individuals, apply. 99 In Article 267 proceedings, the ECJ interprets EU law; in an Article 263 case, the GC (and, on appeal, the ECJ), review a Commission decision and thus apply EU law. In addition, the Article 263 procedure allows for much closer scrutiny of a Commission decision. See, eg, Advocate General Cosmas, cited above note 19. 100 See the comments submitted in the public consultation on the 2008 White Paper, http://ec.europa. eu/competition/antitrust/actionsdamages/white_paper_comments.html, especially those of the CCBE, Confcommercio, Confindustria, SEV, Verband der Chemischen Industrie and others. 101 Komninos, cited above note 85, at 1404 and 1428. 102 See, among others, the responses to the 2008 White Paper by the Council of Bars and Law Societies of Europe: http://ec.europa.eu/competition/antitrust/actionsdamages/white_paper_ comments/ccbe_en.pdf, at p 4. For a broader discussion, see Milutinović, cited above note 2, chapter 12. 103 Milutinović, ibid.

358  Veljko Milutinović contribute to the detection and punishment of unknown ones, which may promote separate and additional deterrence against unlawful conduct.104 It is also consistent with an understanding of the relationship between public and private enforcement as an integrated, complementary system – and not two systems that merely run in parallel to each other. More to the point, it is consistent with the view that, ultimately, what constitutes effective private enforcement is the effectiveness of Articles 101 and 102 and not the effectiveness of the ‘right to damages’ as such: the need for the effectiveness of the latter is derived from and conditional on the need for effectiveness of the former. An argument to the contrary can only be sustained using a teleologically retrodictive interpretation of BRT.

Real problems related to ‘binding effect’ A recurring and seemingly legitimate criticism of attributing binding effect to administrative decisions (of the Commission and of NCAs) is its compliance, or rather non-compliance, with standards of fundamental rights protection.105 The argument can be summarized thus: if administrative decisions are binding on civil courts in a damages claim, then the civil defendants (ie, the infringers) are denied the right to a fair trial, as the issues of infringement and the defendants’ culpability therefor are not examined (independently) by a civil court. Instead, ‘civil rights and obligations’, within the meaning of Article 6 of the European Convention on Human Rights’ (ECHR) are determined by an administrative body, which is not an ‘impartial tribunal’.106 The problem with this line of criticism is that it is not specific to private enforcement. If anything, it points to a defect in the public enforcement systems to which it refers. The issue of fundamental rights compliance in the context of antitrust enforcement is an old one; it became prominent in the period between the early 1980s and the middle 1990s but was subsequently laid to rest, for better or worse, as regards proceedings before the Commission.107 Initially, the right to 104 CEPS Brussels, Erasmus University Rotterdam and LUISS Rome, Making antitrust damages more effective in the EU: welfare impact and potential scenarios, Brussels, 2008 (‘Impact Study’), sections 2.1 and 2.2. 105 See the discussion in Milutinović, above note 2, sections 12.2 and 12.3; the possibility of a problem is acknowledged by the Commission in its Staff Working Paper (above note 90), points 155 and 162. 106 Council of Europe, Convention for the Protection of Human Rights and Fundamental Freedoms of 1950. 107 For pre-Regulation 1/2003 perspectives, see, eg, Richard Brent, ‘The Binding of Leviathan? The Changing Role of the European Commission in Competition Cases’, 44 International and Comparative Law Quarterly 255 (1995); Frank Montag, ‘The Case for a Radical Reform of the Infringement Procedure’, 8 European Competition Law Review 28 (1996); Veljko Milutinović, The Compatibility of the Fact-Finding Powers of the Commission in Competition Law Investigations with the European Convention on Human Rights, LL.M. thesis, University of Bristol, 2001. For a postRegulation 1/2003 perspective, see Milutinović, above note 2, sections 12.2 and 12.3.

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

359  

a fair trial arose as a ‘general principle of law’ applied by the ECJ, on the basis of the applicability of this right in all Member States, which are also parties to the ECHR.108 After some initial dithering by the ECJ,109 the GC found that public enforcement by the Commission does give rise to a right to a fair trial and that this right is sufficiently safeguarded by the action for annulment before the GC (subject, ultimately, to an appeal before the ECJ).110 Whatever the case may be, it is no longer controversial that a right to a fair trial exists in this context,111 regardless of whether or not there is a civil claim before a national court. However, the issue of fundamental rights compliance by NCAs lingers on, not least due to the great diversity of public enforcement models adopted by the Member States.112 In several Member States, one finds the ‘judicial model’, whereby the NCA acts as a sort of ‘prosecutor’ before a court of law.113 But the majority of Member States use the ‘monist’ model, which entrusts both investigations and decisions to a single authority (with variations as to the exact internal allocation of tasks).114 Among the ‘monist’ jurisdictions, the level of scrutiny in judicial review varies significantly, from an ‘administrative review’ to a full-blown ‘merits’ appeal on both facts and law.115 In at least one Member State (Italy), opposition to the binding effect of NCA decisions has been linked explicitly to a ‘light touch’ judicial review conducted in that jurisdiction.116 The diversity of enforcement and judicial review procedures among the various public enforcement systems may give rise to legitimate concerns, especially as regards the cross-border binding effect of NCA decisions, which is what the Commission had in mind in 2008.117 In a cross-border scenario, the civil courts of one Member State (State of recognition) would be bound by decisions adopted in another Member State (State of origin), which may or may not have applied the same human rights standards as those applicable in the State of recognition. Here one finds a novel dimension: while binding effect within the same Member State could be justified by stating that it is up to the Member State to ensure that public 108 See, eg, Joined Cases 209/78, etc. Van Landewyck v Commission (‘Fedetab’) [1980] ECR 3125; Joined Cases 100–103/80, Musique Diffusion Française and Others v Commission [1983] ECR I-11005. 109 Ibid. 110 Case T-348/94, Enso Española v Commission [1998] ECR II-1875, paras 57–63. 111 The position was also confirmed explicitly by the ECJ in Elevator Cartel (Damages), cited above note 80, paras 45–77, with Article 6(1) of the Convention now being effectively ‘replaced’ by Article 47 of the EU Charter of Fundamental Rights (‘CFR’), 2000 OJ C364/01 (see Article 6 TEU and the Declaration Annexed to the Final Act of the Intergovernmental Conference which Adopted the Treaty of Lisbon, of 13 December 2007). 112 For a comprehensive overview, see ECN Working Group Cooperation Issues and Due Process Decision-Making Powers Report, 31 October 2012, http://ec.europa.eu/competition/ecn/decision_ making_powers_report_en.pdf (last viewed 1 April 2013). 113 Ibid, point 2.3. 114 Ibid, point 2.1. 115 Ibid, point 3.1.7, with an overview of the various systems. 116 Comments by Aldo Frignani at the fourth Convegno biennale di Trento sull’applicazione delle regole di concorrenza: ‘La cultura della concorrenza in Italia e nell’Unione Europea’, University of Trento, 18–19 April 2013. 117 Point 2.3 of the White Paper, cited above note 4.

360  Veljko Milutinović enforcement in its jurisdiction is human rights-compliant, one cannot reasonably expect Member States to effectively supervise each other. In response to this concern, the Commission has suggested that the recognition of the binding effect of foreign NCA decisions could be subject to an exception, roughly along the lines of Article 34(1) of the Brussels Regulation.118 In retrospect,119 such a rule may be suboptimal, as it does not seem to cover cases of insufficient judicial review below the (arguably, very high) threshold of being ‘manifestly contrary to public policy’. In addition, limitations on recognition based on human rights compliance entail the risks of both underuse (failing to detect a human rights problem in the State of origin) and overuse (refusing to recognize a foreign decision for a different reason and then justifying the refusal on shaky or non-existent human rights grounds). In closing, it is submitted that the issue of human rights compliance should not be focused on (nor, worse still, relegated to) private enforcement. What is not yet widely recognized in the field of competition policy, but ought to be, is that – in the post-Lisbon Treaty era – the Charter of Fundamental Rights (CFR) constitutes positive EU law.120 Rather than laying the blame on ‘binding effect,’ therefore, an examination of the robustness of judicial review across the various Member States may be preferred. As the ‘Guardian of the Treaties’, the Commission is (no longer) bound ‘merely’ to ensure the effectiveness of the Treaty provisions on competition; it is likewise bound to ensure the effectiveness of Article 47(2) of the CFR, which guarantees the right to a fair trial.

Public and Private Enforcement Collide: Pfleiderer Beyond ‘binding effect’, there are serious open issues regarding the relationship between public and private enforcement in the context of the collection and use of evidence. In Pfleiderer,121 the ECJ was asked whether EU law prevented an NCA from disclosing leniency documents to ‘follow-on’ private plaintiffs. In reply, the Court said, essentially, ‘no’.122 It then added however that, when determining whether to order an NCA to disclose such documents, a national court must ‘determine the conditions under which such access must be permitted or refused by weighing the interests protected by European Union law’.123 The ‘interests’ to be ‘weighed’ are principally two: the interest in having a successful leniency programme, and the interest in protecting the ‘right to damages’ of the victims of a cartel.124 118 Staff Working Paper, cited above note 90, point 162. Under the new Regulation 1215/2015, the relevant provision is in Article 45. 119 I have supported this idea in the past. See Milutinović, above note 2, section 12.3. 120 CFR, cited above note 111. 121 Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161. 122 Ibid, para 32 and the operative part of the judgment. 123 Ibid. 124 Ibid, paras 25–27 (leniency), paras 28–29. (damages) and para 30 (synthesis of the two).

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

361  

No guidance was provided as to exactly how a national court ought to perform this balancing exercise. Pfleiderer itself was a benign example and a positive outcome was still likely for the victims: the national rule in that case made disclosure the default position, non-disclosure being the exception.125 What remains unclear is what happens in a less optimistic scenario, where national law adopts non-disclosure as the default position, and the judge is called upon to justify an exception, on the grounds of a Courage-based ‘right to damages’. Is it reasonable to suppose that a national judge would override national law and what is (presumably) the public interest for the sake of facilitating a private damages claim? Arguably – no. Another inadequacy of the Pfleiderer judgment is that it only answers the question of whether EU law prohibits NCAs from disclosing leniency documents. At the EU level, the European Commission has maintained a hard line of nondisclosure of leniency (or other) documents to ‘follow-on’ damages claimants,126 although that position is now being eroded somewhat, as will be seen below.127 The paradox of stressing, persistently, the existence and importance of the ‘right to damages’ while,128 at the same time, denying access to evidence for private plaintiffs (with equal or greater persistency) was not lost on the Commission. This is why the Commission made a point, in its recent claim against the ‘Elevator Cartel’, of claiming damages before a national court solely on the basis of the public version of its infringement decision against the cartel,129 which is available to everyone. The ECJ found that, in view of the fact that the public version was used, there was no violation of ‘equality of arms’ between the Commission (as damages claimant) and the infringers (as damages defendants).130 Combined with the ECJ’s finding that Commission decisions are binding,131 the narrative can be Ibid, paras 7 and 14. In Article 7 of the Commission Notice on the rules for access to the Commission file in cases pursuant to Articles 81 and 82 of the EC Treaty, Articles 53, 54 and 57 of the EEA Agreement and Council Regulation (EC) No 139/2004, 2005 OJ C325/07, the Commission defines the personal scope of the right to access to the file thus: ‘[...] both in cases under Articles [101] and [102] [TFEU] and in cases under the Merger Regulation, access is granted, upon request, to the persons, undertakings or associations of undertakings, as the case may be, to which the Commission addresses its objections (hereinafter, ‘the parties’)’. Exceptionally, points 30–34 of the Notice provide that access may be granted to complainants in antitrust proceedings and involved parties in merger proceedings – however, neither of the indicated exceptions is applicable to follow-on damages claimants. As regards cartel settlement proceedings, the Commission refuses to disclose settlement submissions without the consent of the relevant applicants. See Commission Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel cases, 2008 OJ C167/01, point 39. 127 Indeed, the Commission spearheaded a joint position of non-disclosure throughout the European Competition Network (ECN), in the immediate aftermath of Pfleiderer, even though its own legal position was unaffected by that judgment. See Resolution of the Meeting of Heads of the European Competition Authorities of 23 May 2012: Protection of leniency material in the context of civil damages actions, http://ec.europa.eu/competition/ecn/leniency_material_protection_en.pdf (last viewed 15 March 2013). 128 Throughout the 2005 Green Paper and the 2008 White Paper. 129 Elevator Cartel (Damages), cited above note 80, at para 73. 130 Ibid. 131 Ibid, para 65. 125 126

362  Veljko Milutinović summarized as follows: disclosure of leniency documents is uncalled for, because the infringement is established and the national court is bound as regards the existence of the infringement and the identity of the infringers. This narrative supports the Commission’s argument in favour of ‘binding effect’ insofar as it enhances the utility of public enforcement decisions.

Towards a more nuanced approach: CDC In CDC,132 the Commission’s position of non-disclosure of documents suffered a significant ‘dent’ in its ‘armour’. There, the GC found that the Commission can be compelled to disclose so-called ‘statements of contents’ (lists of documents collected during a Commission investigation – including leniency documents) to a follow-on private plaintiff.133 That in itself narrows the list of non-disclosable documents, the essence of the Court’s reasoning being that a statement of contents merely lists the contents of the Commission’s file but does not disclose any of the documents contained therein.134 This ‘core’ part of the judgment is less interesting than what the GC had to say more generally regarding the relationship between leniency and actions for damages: a) the interest of cartel members not to be exposed to damages claims is not a legal interest worthy of protection;135 b) by contrast, a victim’s right to damages is worthy of protection;136 c) competition policy does not differ from other EU policies with regard to disclosure – Regulation 1049/2001, the relevant instrument,137 makes no exception for competition policy;138 d) in order to create an exception to the general rule of liberal disclosure, it is not sufficient to make an a priori argument to the effect that disclosure of documents in one case could undermine the effectiveness of the leniency programme in general.139

Case T-437/08, CDC Hydrogene Peroxide v Commission [2011] ECR II-8251. Ibid, paras 79 and 81. 134 The reason why the statement of contents is useful for follow-on plaintiffs is because many or most Member States do not have a common law system of discovery; rather, they require the plaintiff to identify, with a substantial degree of precision, the documents held by the defendant that are useful to its case. 135 CDC, cited above note 132, paras 47 and 49, read in conjunction with paras 69–71. 136 Ibid, para 49. 137 Regulation (EC) No 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, 2001 OJ L145/43. 138 CDC, cited above note 132, para 72. 139 Ibid, para 54 read in conjunction with paras 69–71. 132 133

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

363  

What CDC and Pfleiderer have in common is that both cases can be seen as ‘crying out’ for legislation. Public and private enforcement are not really ‘parallel’, as private plaintiffs often seek evidence from public authorities, whereas the reverse is far less common. On the one hand, the Commission’s approach of denying access to documents a priori can hardly be seen as treating private enforcement as ‘equal’ to its public counterpart. On the other hand, the ‘right to damages’ prevents the leniency programme from ‘prevailing.’ It is therefore obvious that reasoning in terms of ‘parallel competences’ and ‘balancing acts’ has hit a ‘wall’, made of positive law. If one is to have a functioning, coherent system of antitrust enforcement in the EU, legislation is necessary. One must agree with the GC’s findings to the effect that Regulation 1049/2001 does not make the distinction between competition and other policies, because it quite simply does not. As a matter of positive law (and the conspicuous absence of teleological interpretation is duly noted), one must also agree that the interest of leniency applicants to be shielded from damages claims is not a recognized legal interest. That, however, could be changed via EU legislation. In addition, the GC’s judgment in CDC indicates that ‘balancing acts’ of the type envisaged in Pfleiderer are not really suitable for national courts. While it is arguable that the Commission has failed, in CDC, to demonstrate a causal link between disclosure of the documents sought in that particular case and the endangerment of its leniency programme in general, it is also arguable, in the Commission’s defence, that it is impossible to say, a priori, at which point the disclosure of leniency documents would reach a ‘tipping point’, where would-be leniency applicants would cease to volunteer information.140 This choice too could be made by the EU legislator – albeit arbitrarily, in the sense that there might not be definitive ‘scientific proof’.

The Disabling Effects of the ‘Right to Damages’ For the most part, the ‘right to damages’ established in the Courage judgment was hitherto viewed as an enabling principle. Not only did plaintiffs have an EU law right to damages, that right had a certain EU law content. Thus, the Court in Manfredi reaffirmed that ‘any individual’ could claim damages and proceeded to lay down some further procedural and substantive rules. The Commission picked 140 To the knowledge of the present author, no empirical study has been carried out regarding the impact of prospective damages liability on would-be leniency applicants in the EU. This is logical, since the data is only as abundant (or rather, as scant) as actual damages cases in the EU. However, some interesting data has surfaced and been analysed in the U.S. See Impact Study, cited above note 104, p 507. More recently, see United States Government Accountability Office, Criminal Cartel Enforcement: Stakeholder Views on Impact of 2004 Antitrust Reform Are Mixed, but Support Whistleblower Protection (Washington, D.C., GAO, 2011), http://www.gao.gov/products/GAO-11619 (last viewed 1 March 2013).

364  Veljko Milutinović up on this theme in the 2008 White Paper, considered the rules from Courage and Manfredi to be established, and sought to build on them by promoting EU-level legislation. In contrast, little or no attention seems to be given to the potentially disabling effects of the ‘right to damages.’ In Manfredi, the Court interpreted the right to damages as a right to compensation, while rejecting an EU law right to punitive damages, the latter being left to the discretion of the Member States.141 In retrospect, it may be wondered why such evident disharmony between the Member States should be allowed. The belief that damages should be purely compensatory can be explained by a belief that the right to damages is autonomous and inherent in the system of the Treaty, ie, that there is a self-justifying notion that exists under EU law, according to which all loss should be compensated. However, this view ignores the utilitarian origin of direct effect and, consequently, of the ‘right to damages,’ which was established not for some unfathomable or selfjustifying reason but for the sake of the greater effectiveness of the prohibitions in Articles 101 and 102. The Court has found, rightly, that there is no right to punitive damages under existing EU law; however, it has not precluded the introduction of such a right via EU legislation. What happened post-Manfredi, however, is that the Commission did not propose an EU rule imposing punitive damages in its 2008 White Paper. Even though it was found, in the Impact Study conducted on behalf of the Commission and for the purposes of the very same White Paper, that, in certain circumstances (hard core cartels), punitive damages would be justified as a matter of deterrence (and thus as a matter of the effectiveness of Article 101),142 and although the Commission discussed double damages in the 2005 Green Paper,143 the idea disappeared from the Commission’s agenda by the time of the 2008 White Paper. The relegation of punitive damages to national law is dissonant with the harmonization of substantive law applicable to restrictive agreements. Indeed, Article 3(2) of Regulation 1/2003 prohibits stricter national rules governing restrictive agreements whenever there is an effect on trade between the Member States (which is, as regards major infringements, more or less always the case);144 the EU legislator was therefore careful, in Regulation 1/2003, to avoid overdeterrence of agreements under national law. What Manfredi does (as subsequently applied by the Commission) is to specifically allow overdeterrence under national law, through discretion over punitive damages, which Member States can apply even in economically suboptimal scenarios, ie, even in scenarios of antitrust infringements other than hardcore cartels. Manfredi, cited above note 5, paras 92 and 99. Impact Study, cited above note 104, tables 29–31, pp 192–195, pp 419–421, arguing essentially that punitive damages may be optimal for hard core cartel cases but not for other types of cases. 143 Green Paper, cited above note 4, Option 16. 144 See Veljko Milutinović, Enforcement of Articles 81 and 82 EC before National Courts PostCourage: Enhancing a Community Policy or Shifting a Community Law Paradigm?, Ph.D thesis, European University Institute, 2008, sections 1.4.2 to 1.4.4. 141 142

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

365  

Another disabling effect of the ‘right to damages’ comes from the Commission’s reading of the term ‘any individual who has suffered a loss’ – the identity of the antitrust damages plaintiff as proclaimed in Courage and Manfredi. In the 2008 White Paper, the Commission interpreted that term as implying that there can be no indirect purchaser-type limitation on standing for damages plaintiffs;145 it was felt that the rule established by the U.S. Supreme Court in Illinois Brick (effectively, though not formally, excluding the standing of indirect purchasers) could not exist in Europe,146 due to the ECJ’s reference to recovery by ‘any individual’.147 As was argued extensively elsewhere,148 the presence of the words ‘any person’ (analogous to ‘any individual’) in Section 4 of the Clayton Act149 did not prevent the U.S. Supreme Court from laying down the rule in Illinois Brick. Rather than focusing on the identity of ‘any person’, the Supreme Court focused on the notion of ‘injury’ (loss) and found that indirect purchasers are not ‘injured in their business or property’ (the second part of the rule in Section 4) for purposes of U.S. (federal) antitrust law. A similar process was repeated in Brunswick,150 where the Supreme Court found that plaintiffs who suffered damage not as a result of the anticompetitive but as a result of the pro-competitive aspects of a competition infringement did not suffer ‘antitrust injury’; they were thus likewise excluded.151 The rejection of Illinois Brick in Europe is to be welcomed, whatever the underlying motivation or justification, as the rule is unjust and its apparent doctrinal soundness has been rejected by some of the most eminent American authorities on the subject.152 Nevertheless, even if one accepts, for the sake of argument, that the rejection of this rule is inherent in the case law of the ECJ, what is at issue here is the way in which the ECJ came to lay down the relevant case law and the way in which it is likely to interpret it in future cases. Rather than reasoning inductively and establishing, case-by-case, which plaintiffs are ‘meritorious’ and why,153 the Court started out by laying down a general rule, which, through the process of deduction which is characteristic of the Court, seems to point towards the inclusion of all persons harmed, in one way or another, by an infringement of White Paper, cited above note 4, point 2.1, first paragraph. Illinois Brick Co. v Illinois, 431 U.S. 720 (1977). 147 Many years before the 2008 White Paper (indeed, before Courage), at least one author claimed that Illinois Brick-type limitations on standing might be contrary to the ‘constitutional’ nature of Articles 101 and 102 TFEU. See Clifford A Jones, Private Enforcement of Antitrust Law in the EU, UK and USA, OUP, 1999, p 187. 148 Milutinović, above note 2, section 10.2 (findings with regard to the EU), section 9.3 (situation in the US). 149 15 U.S.C. § 15. 150 Brunswick Corp v Pueblo Bowl-O-Mat, Inc., 429 U.S. 477 (1977). 151 Ibid, at 486–489. 152 See Antitrust Modernisation Commission, Report and Recommendations (Washington, D.C. 2007), pp 265–284, http://govinfo.library.unt.edu/amc/report_recommendation/amc_final_report.pdf (last viewed 1 April 2013). For a thorough discussion of the problems surrounding Illinois Brick, see Milutinović, above note 2, section 9.3. 153 For criticism of the treatment of the in pari delicto rule by the ECJ in Courage, see Giorgio Monti, ‘Anticompetitive Agreements: the Innocent Party’s Right to Damages’, 27 European Law Review 282 (2002). 145 146

366  Veljko Milutinović the competition rules. Beyond indirect purchasers, a time may come when difficult decisions will have to be made regarding, eg, derivative claims and claims by shareholders and/or ‘stakeholders.’154 Last but not least, whether or not one agrees with the assessment that EU law precludes a Brunswick-style concept of ‘antitrust injury’ in Europe,155 there may be something to be said for this concept, as the range of possible damages plaintiffs says a lot about the underlying purpose and goals of antitrust law.

Conclusion Following their initial success as ‘groundbreakers’ for private enforcement, the twin notions of a ‘system of parallel competences’ and the ‘right to damages’ are in danger of becoming an ‘ouroboros’.156 Like the symbolic animal, while representing, in general, a force of good, they are, upon closer inspection and as the Greek word suggests, a serpent apt to eat its own tail. The limitations posed by the twin notions are complex and, in many respects, uncertain; the one implication that is simple and certain is that they should not be seen as inherent or self-justifying. Deductive path dependencies should yield to practical, inductive solutions: EU antitrust law should be allowed to develop in a manner that best suits the effectiveness of its prohibitions – which must have primacy over rights derived from that effectiveness (eg, the ‘right to damages’). Indeed, the further one drifts from the goal of effectiveness of the Article 101 and Article 102 prohibitions, the further one gets from the source of legitimacy of the rules adopted (or enunciated) in this field. As CDC and Pfleiderer have shown, it is the EU legislator that is uniquely suitable to make difficult choices, which at times may have to be arbitrary: hiding behind (un)’settled’ case law only delays and complicates the search for a functional solution that would integrate public and private enforcement into a single, coherent system. Likewise, the socially useful ‘binding effect’ of NCA decisions should not be precluded by intellectual constructs that seek to make public and private enforcement ‘equal’ at all costs. For better or worse and for the time being, the system of competition enforcement in the EU is still overwhelmingly public; this fact must be taken into account when considering the integration of competences that were, perhaps prematurely, found to be ‘parallel.’ This is not to say that EU antitrust law should not be made more ‘private’ in the future; it is to say that current realities must be taken sufficiently into account. See Milutinović, above note 2, section 10.3. Ibid, section 10.4. 156 The use of the term in a legal context is borrowed from Guilherme Vasconcelos Vilaça, Law as Ouroboros, Ph.D thesis, European University Institute, 2012. 154 155

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

367  

As for the ‘right to damages’ itself, it is submitted that it represents a useful and beneficial construct, provided its origins are not obscured by teleological retrodictions. Such retrodictions are apt to create path dependencies that complicate further the very problems they are meant to resolve – whether by creating a ‘come one, come all’ rule of standing for damages claimants or by confronting the Commission’s policy of protecting leniency documents with seemingly immutable ‘individual rights’. All other things being equal, introducing damages claims into the enforcement equation increases the potential punishment for unlawful conduct and incentivizes the prosecution of a greater number of infringements, thus enhancing deterrence. It is submitted that the time is now ripe to engage in a constructive dialogue that better reflects and furthers the underlying objective of the effectiveness of EU law. In closing, therefore, it is to be hoped that retrodictions and path dependencies will not unduly impede the EU legislator and judiciary in their efforts to integrate public and private enforcement coherently (albeit, if need be, somewhat ‘unequally’) in the near future.

Postscript: the Commission’s legislative proposal of 2013 After the main text of this chapter was written, the Commission published a damages actions ‘package’, which includes a proposal for a new Directive on damages actions,157 a Communication and practical guide on the quantification of harm in such cases.158 Including the relevant impact assessment,159 the package is more than two hundred pages long and covers two broad themes previously discussed and analyzed by the Commission in the 2008 White Paper and the 2011 Draft Guidance on Quantification.160 The size, scope and very recent date of the ‘damages package’ preclude a deeper analysis in this volume; numerous future workshops and conferences will no doubt be dedicated to its analysis. At this point, it is possible and necessary to outline a few of the key solutions as regards the legislative proposal stricto sensu, ie, the proposed Directive. 157 Proposal for a Directive of the European Parliament and of the Council 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, COM(2013) 404. 158 Communication from the Commission on quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty on the Functioning of the European Union, 2013 OJ C167/19; Commission Staff Working Document – Practical Guide on Quantifying Harm in Actions for damages based on breaches of Article 101 or 102 of the Treaty on the Functioning of the European Union, SWD (2013) 205. 159 SWD(2013) 204 final. 160 White Paper, cited above note 4; Draft Guidance Paper: Quantifying Harm in Actions for Damages Based on Breaches of Article 101 or 102 of the Treaty on the Functioning of the European Union (Brussels, 2011, http://ec.europa.eu/competition/consultations/2011_actions_damages/draft_ guidance_paper_en.pdf (last viewed 20 June 2013).

368  Veljko Milutinović Broadly speaking, the proposed Directive derives from and builds on the policy positions announced and/or discussed in the 2005 Green Paper and the 2008 White Paper. Thus, the Commission has proposed, inter alia, that there be, under EU law: – special, enhanced rules for disclosure in antitrust cases;161 – special protection for some categories of public enforcement documents;162 – ‘binding effect’ of existing NCA decisions in civil proceedings;163 – special rules on limitation periods;164 – a reduction in damages exposure for leniency recipients;165 – a ‘passing-on’ defence – where such can be proven and subject to an exception;166 – a (limited) presumption of ‘passing-on’ in favour of indirect purchaser claimants;167 – rules for co-ordinating separate claims arising from different levels of the supply chain;168 – a rule clarifying that lost profits can be obtained by the suppliers of the infringer, where applicable;169 – a rule creating a presumption that cartels cause harm;170 and – a rule enabling national courts to estimate the amount of damage.171 In addition, there are two aspects of the proposed Directive that were not directly anticipated earlier: – a special provision on settlements; and – the applicability of the proposed rules to cases decided under national competition law.172 Starting this discussion by reference to issues which were analyzed above in the main part of the chapter, and which feature prominently in the private enforcement 161 Articles 5 and 8 of the Proposal for a Directive, cited above note 162; see point 2.1, Question A of the Green Paper and point 2.2 of the White Paper, both cited above note 4. 162 Ibid, Articles 6 and 7; point 2.1, Question B and option 28 of the Green Paper; points 2.2 and 2.9 of the White Paper. 163 Ibid, Article 9; point 2.1, Question C of the Green Paper; point 2.3 of the White Paper. 164 Ibid, Articles 10 and 17; point 2.9, Question M of the Green Paper; point 2.7 of the White Paper. The proposal goes a bit further than what was discussed previously, as it specifically calls for a suspension of limitation periods while consensual dispute resolution proceedings are ongoing. 165 Ibid, Article 11; options 29 and 30 of the Green Paper; point 2.9 of the White Paper. 166 Ibid, Article 12; option 21 of the Green Paper, point 2.6 of the White Paper. 167 Ibid, Article 13; option 21 of the Green Paper, point 2.6 of the White Paper. 168 Ibid, Article 15; option 21 of the Green Paper; point 2.6 of the White Paper. 169 Ibid, Article 14. This option is vaguely related to the question posed in the Green Paper (point 2.3, question E) as to how damages should be defined, and to the promise made in the White Paper (point 2.5) to draw up guidelines on damages. In other words, what was not anticipated was a norm of positive law. 170 Ibid, Article 16(1). Again (see previous note), this solution is vaguely related to the category of facilitating the calculation of damages but it was not anticipated as a positive law norm. 171 Ibid, Article 16(2). The same considerations mentioned in footnotes 169 and 170 also apply here. 172 Ibid, Article 1(1).

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

369  

debate, the ‘binding effect’ of NCA decisions is set to become EU law. The proposed norm replicates one ‘half’ of Article 16(1) (the rule in Masterfoods) but not the other (the rule in Delimitis), as it refers only to existing and not ‘envisaged’ NCA decisions. This omission seems reasonable, as it is one thing to ask national courts to be on the lookout for Commission decisions and quite another to expect them to make inquiries with 27 (or now with Croatia, 28) NCAs. Despite concerns expressed by the Commission in 2008, no exception to the rule of recognition (human rights-based or otherwise) is provided.173 As was argued in this chapter, however, discrepancies in human rights protection between the Member States should be addressed separately and not as a matter of private enforcement. Perhaps the most positive aspect of the proposed Directive, which also represents a welcome (and sharp) policy U-turn, is the treatment of documents held by competition authorities – and leniency documents in particular. The Commission has decided to soften significantly its no-disclosure policy – with no compulsion from the ECJ or the GC (albeit with some encouragement, through the indeterminacy of Pfleiderer and the sweeping demolition of the Commission’s underlying premises in CDC). This softening essentially entails that, prima facie, all documents held by the Commission and the NCAs are subject to disclosure, except leniency corporate statements and settlement submissions,174 with a temporary ‘freeze’ on disclosure (pending the decision of the Commission/NCA) for other documents that were created specifically for the purpose of public enforcement proceedings (either by the parties or by the Commission/NCA).175 The U-turn on disclosure of leniency documents did not come about in total disharmony with the ECJ. Five days before the Commission published the proposed Directive, the ECJ issued its judgment in Donau Chemie,176 where it made clear that the balancing required by Pfleiderer is not a blank cheque for Member States to deny access for private plaintiffs. Indeed, in Donau Chemie the Court found that a national rule that requires the consent of all parties to proceedings before an NCA in order for documents gathered by the NCA to be made available to follow-on claimants runs counter to the principle of effectiveness of EU law.177 Thus, unlike the situation in Pfleiderer, the Court was willing to lay down a concrete rule on at least one specific scenario of (non) disclosure. The decision to lay down such a rule was justified on the ground that the national rule in question precluded the balancing act envisaged in Pfleiderer by effectively depriving the national court of competence to assess the pros and cons of disclosure.178 The Commission’s ‘change of heart’ on leniency documents should be applauded, as it aims at relieving national courts of much of the indeterminate See discussion above (‘Real Problems Regarding Binding Effect’). Articles 5 and 6 of the Proposal for a Directive, cited above note 157. Ibid, Article 6(2). 176 Case C-536/11, Bundeswettbewerbsbehörde v Donau Chemie et al., judgment of the ECJ of 6 June 2013, not yet reported. 177 Ibid, operative part of the judgment and para 49. 178 Ibid, paras 37 and 43. 173 174 175

370  Veljko Milutinović ‘balancing acts’, while making difficult but reasonable legislative choices as to which documents should remain excluded from disclosure. Quite reasonably, leniency corporate statements and settlement submissions are excluded from disclosure,179 as the disclosure of such documents would turn leniency applicants into ‘sitting ducks’ for damages claims, with no corresponding value added by the private claimant. In terms of corresponding added value, the Commission would make this very wide disclosure subject to a trade-off. At last, concrete measures are proposed to limit the damages liability of leniency recipients, using a combination of ‘dejoining’ of liability and ‘rebates’.180 By shifting more of the damages liability to cartel members that do not benefit from leniency, the proposed Directive enhances the leniency programme while at the same time ensuring, in principle, full compensation for the victims. This was a much-needed ‘squaring of the circle’, for which, like the relationship between leniency and damages in general, the legislator is uniquely placed. The Commission did not stop at addressing evidentiary issues that arise in relation to public enforcement’s investigative ‘output’. It proceeded to address the issue of civil disclosure inter partes.181 As anticipated in the White Paper,182 the Commission eschewed the much loathed and feared ‘US-style discovery’. It also avoided proposing a useless, minimal solution that cements the status quo, and opted instead for a middle ground. The solution in the proposed Directive would impose disclosure of evidence whenever a party ‘has presented reasonably available facts and evidence showing plausible grounds for suspecting that he [...] has suffered harm caused by the defendant’s infringement’.183 The proposed Directive would also impose safeguards, such as relevance and reasonable specificity of the evidence being sought, as well as proportionality of the request.184 The proposed solution should not adversely affect existing national rules on confidentiality, legal professional privilege, or similar rules of evidence.185 In essence, as regards disclosure inter partes, the Commission has defined a negative command in a positive way. When one adds up all of its ‘wishes’, one realizes that the same result can be reached by subtraction: the Commission wishes to eliminate civil procedure ‘impediments’ that require claimants to identify precisely each piece of evidence being sought.186 This makes sense in the Article 6(1) of the Proposal for a Directive, cited above note 157. Ibid, Article 11, paras (2) to (4). 181 Ibid, Articles 5 and 8. 182 White Paper, point 2.2; in the Green Paper, various options were presented (Options 1–3) which did include, inter alia, some form of discovery proceedings, but this was dropped by the time the White Paper was published. 183 Article 5(1) of the Proposal for a Directive, n. 157 above. 184 Ibid, Article 5(2) and (3). 185 Ibid, Article 5(4) and (5). 186 See Commission Staff Working Paper: Annex to the Green Paper Damages actions for breach of the EC antitrust rules, COM(2005) 672 final, points 34–35; Staff Working Paper, cited above note 90, points 87 and 90. 179 180

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

371  

context of infringements of Article 101, whose nature is, in the famous words of the ECJ, ‘frequently covert’.187 What does not make any immediate sense is why the entire provision on inter partes disclosure applies only to disclosure requests by claimants. There appear to be two problems with this solution, one evident and another perhaps less so. First and evidently (as was seen in Courage itself),188 the victim of a competition infringement will often not be the claimant in related civil proceedings. The solution in the proposed Directive seems biased in favour of cartel damages claims (most commonly, ‘sword’ cases, where the victim ‘attacks’ the infringer), without due regard for numerous ‘shield’ cases (restrictive distribution agreements, or the anticompetitive use of IP rights, etc.).189 The second objection is less evident: even ‘sword’ cases based on a demonstrable infringement may be, exceptionally, of questionable merit. A scenario that comes to mind is that of a firm that abuses its dominant position and then files a damages claim against parties that have concluded a restrictive agreement for the sole or predominant purpose of countering the abuse of dominance.190 Obviously, the tu quoque argument is primarily moral and not legal, so that both the abuse and the restrictive agreement should be investigated. It would be more economical, however, if the two claims were examined together by the same court. In this connection, it would have been useful to clarify, in the proposed Directive, that the term ‘claimant’ in Article 5(1) also refers to parties launching a counterclaim. It is at least arguable that national implementing legislation or judicial interpretation may eventually ‘iron out’ this ‘wrinkle’ – subject, of course, to a possible Article 267 reference to the ECJ. With regard to indirect purchasers – another prominent issue in the private enforcement debate – Recital 30 to the proposed Directive is quite interesting. It states as follows: While indirect purchasers are entitled to claim compensation, national rules of causality (including rules on foreseeability and remoteness), applied in accordance with principles of Union law, may entail that certain persons (for instance at a level of the supply chain which is remote from the infringement) are legally unable to claim compensation in a given case.191

Thus, the Commission seems to assume today, as it did in the 2008 White Paper,192 that the Courage and Manfredi ‘any individual’ language grants broad standing – including to indirect purchasers, although these cases are no longer cited to this effect. Yet, unlike the ‘codification’ of the principles of ‘equivalence’ and ‘effectiveness’ in Article 3 of the proposed Directive, the Commission fails to codify Courage, cited above note 3, para 27; Pfleiderer, cited above note 121, para 29. Courage, above note 3. 189 For IP-related ‘shield cases’, see, eg, in England: Chiron Corp v Murex Diagnostic [1994] 1 CMLR 410 (Court of Appeal); Sportswear SpA and Four Marketing Ltd v Stonestyle Ltd [2006] EWCA Civ 380 (Court of Appeal). 190 Some possible scenarios, illustrated by actual EU cases from the past (decided by the Commission) are discussed in Milutinović, above note 2, section 10.4.1. 191 Proposal for a Directive, cited above note 157 (emphasis added). 192 White Paper, point 2.1, first paragraph. 187 188

372  Veljko Milutinović such a rule. Instead, it seeks to alleviate the burden of proof of indirect purchasers and regulate the relationship with the passing-on defence – which is specifically allowed,193 unless the victims to whom the anticompetitive overcharge was passed on are legally precluded from claiming recovery.194 This is where, somewhat unexpectedly, room for disharmony between national laws reopens: while assuming (but not prescribing) that indirect purchasers can bring claims and while alleviating their burden of proof,195 the Commission then allows the possibility of indirect purchaser claims being excluded on the ground of ‘remoteness,’ subject to unnamed ‘principles of Union law.’ It is submitted that this solution is likely to result in an Article 267 reference to the ECJ, whereby the Court will be called upon to make an Illinois Brick-type choice regarding the compatibility of a national remoteness rule with EU law. The problem is therefore postponed but not avoided.196 The exception to the passing-on defence in cases where there is no indirect purchaser claim seems to be a positive development, as infringers are not allowed to escape liability solely for a lack of viable claimants. Having said that, this exception creates the possibility of unjust enrichment of direct purchasers who have passed their losses on to their customers. The proposed Directive addresses the unjust enrichment problem only in the context of two separate civil claims being launched, one by direct and another by indirect purchasers.197 If the only claim is that of direct purchasers, the possibility of unjust enrichment remains, and it may well be that Member States will act to stop this possibility from materializing. If they were to do so, based on previous ECJ judgments they would not, in principle, be running counter to EU law: the Court has consistently allowed unjust enrichment exceptions under national law that limit EU law rights to compensation/restitution, not least in Courage itself.198 Turning to limitation periods, ie, the maximum time periods under national law in which to bring an action, the proposed Directive follows the anticipated direction of the White Paper and Green Paper but goes much further than the ECJ’s judgment in Manfredi (which provided the impetus for dealing with this aspect of damages claims in the first place).199 In that judgment, the Court left it (formally) open to national courts to decide whether national limitation periods Proposal for a Directive, above note 157, Article 12(1). Ibid, Article 12(2). 195 Ibid, Article 13 paras (1) and (2). 196 See discussion above (‘The Disabling Effects of the ‘Right to Damages’’). In this respect, it will be recalled that Illinois Brick is not, strictly speaking, a rule that bars indirect purchaser standing; rather, it may be viewed as a sort of remoteness criterion. Standing is regulated quite generously by Article III of the U.S. Constitution, which ‘requires merely that the plaintiff present a genuine ‘case or controversy’, meaning that he is not requiring the court to issue an advisory opinion and that he has suffered ‘particularized damage’ or otherwise has a sufficient stake in the outcome of the case’. Milutinović, above note 2, p 303 (relying on Jones, cited above note 147, pp 159–160). 197 Proposal for a Directive, cited above note 157, Article 15(1). 198 Case 199/82, Amministrazione delle Finanze dello Stato v SpA San Giorgio ECR [1983] 3595, para 13 (an early case where national rules against unjust enrichment were allowed in the context of passing-on (of unlawfully levied taxes); Courage, cited above note 3, para 30, citing other judgments with approval. 199 See Manfredi, cited above note 5, paras 73–82. 193 194

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

373  

that are objective, short and not suspended pending NCA proceedings infringe the principle of effectiveness of EU law; meanwhile, in the grounds given in the judgment, the Court indicated that they probably do infringe that principle. In the proposed Directive, the Commission sought to tackle all three problematic aspects of limitation periods identified in Manfredi. First, the limitation period would now be subjective (and thus based on knowledge of the infringement by the victim), and the subjective element would be clarified to the fullest extent: in order for the limitation period to run, not only would the victim have to be aware of the infringement, he must also be aware of damage done to him, as well as the identity of the infringer who caused the harm. Second, the Commission addressed the shortness of limitation periods by providing a five year minimum. Third, the Commission proposed a suspensory effect for NCA decisions and judicial review proceedings, with a minimum one-year limitation period from the moment an NCA decision becomes final. Last but not least, the Commission’s proposal to introduce a presumption of harm for cartel cases is to be welcomed. As normally happens with legislative choices, this one is at least partly arbitrary but it is less so than most: if there is an overall consensus on anything in competition policy worldwide, it is that cartels tend to produce harm.200 A ‘reinventing of the wheel’ in national legal systems in this context (using civil law concepts of harm developed centuries, perhaps, before antitrust law emerged) would be a massive waste of resources and a serious setback for the effectiveness of EU antitrust law. Preponderance leaves room for exceptions, of course, and this is why, under the proposed Directive, the presumption of harm can be overturned. One major issue that arose in both the Green Paper and the White Paper is now conspicuous by its total absence: collective redress. As the Commission explains in the Explanatory Memorandum to the proposed Directive: In 2011, the Commission held a public consultation on a coherent European approach to collective redress. In the wake of stakeholders’ responses and the position of the European Parliament, the Commission opted for a horizontal approach on this matter rather than the inclusion of competition-specific provisions on collective redress in the current proposal. Adopting a horizontal approach allows for common rules on collective redress for all policy fields in which scattered harm frequently occurs and in which it is difficult for consumers and SMEs to obtain damages.

As I have argued before,201 while the ‘specificity’ of many of the solutions proposed in the 2008 White Paper may have been debatable (ie, how ‘specific’ the problems of antitrust damages claims are vis-à-vis the enforcement of EU law before national courts in general), the lack of specificity for collective redress was quite clear. The issues that militate in favour of collective redress (scattered, lowvalue claims and procedural inefficiencies) are prevalent in many or most forms 200 This conventional wisdom is, like all conventional wisdom, subject to criticism and, occasionally, to exceptions. See, eg, Iwan Bos and Erik Pot, ‘On the Possibility of Welfare-Enhancing Hard Core Cartels,’ 107 Journal of Economics 199 (2012). 201 Milutinović, above note 2, section 13.4.1.

374  Veljko Milutinović of ‘mass tort’ litigation (product safety, environmental, etc.)202 Thus, the decision to finally exclude collective redress from the antitrust damages effort comes as no surprise. It is to be hoped, however, that the efforts aimed at producing a quality horizontal document will bear fruit in the near future.203 Beyond specificity, a ‘constitutional’ query that may be raised with regard to the proposed Directive is why the Commission has opted to use a twin legal basis for its adoption. Rather than relying solely on Article 103 TFEU (giving ‘effect to the principles set out in Articles 101 and 102’), the Commission combined this legal basis with Article 114 TFEU (approximation of rules which ‘have as their object the establishment and functioning of the internal market’). After the exclusion of collective redress, the proposal became specific enough to be justified on the ground of effectiveness of Articles 101 and 102. The Commission then introduced an unexpected solution, however, which is the application of the Directive to cases decided under national antitrust law, insofar as Articles 101 and 102 also apply.204 It is this additional feature that, in the view of the Commission, necessitated the use of Article 114, as the application of common rules on damages claims to cases decided under national law could not be justified under Article 103. It is not entirely clear why the Commission has opted to extend the application of the rules to national law claims. According to the Commission’s Explanatory Memorandum: The current divergence of national rules governing damages actions for infringements of the EU competition rules, including the interaction of such actions with the public enforcement of those rules, has created a markedly uneven playing field in the internal market. [...] Because of this marked diversity of national legislations, the rules applicable in some Member States are considered by claimants to be much more suitable for bringing an antitrust damages action in those Member States rather than in others. [...] Similarly, these marked differences mean that undertakings established and operating in different Member States are exposed to significantly different levels of risk of being held liable for infringements of competition law.205

There would appear to be at least two main problems with these arguments. 202 The Commission may have harboured doubts in this regard even in 2008, since, in parallel with DG COMP’s efforts with regard to antitrust damages claims, DG SANCO was examining possibilities of a ‘horizontal’ solution that would address low-value, scattered claims by consumers. See White Paper, point 2.1, last paragraph; and points 62–64 of the Staff Working Paper, cited above note 90. 203 The latest stage, at the time of writing, is the Commission recommendation on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law, C(2013) 3539/3, http://ec.europa.eu/justice/ civil/files/c_2013_3539_en.pdf (last viewed 21 June 2013). 204 Article 1(1) read in conjunction with Article 4 second indent of the Proposal for a Directive, cited above note 157. 205 Proposal for a Directive, at p 9.

A Fresh Look at BRT v SABAM and its Subsequent Interpretation 

375  

First, discrepancies as regards claims under Articles 101 and 102 would largely disappear without having to extend the force of the new rules to national claims – after all, the Commission is tasked with protecting EU and not national competition rules. In addition to the actual harmonization of laws envisaged in the proposed Directive, the principle of effectiveness, combined with the duty of national courts to apply EU law in parallel to national law,206 could be deployed to override many possible discrepancies under national law. Such would be the case, for example, if a national law rule banned the pass-on defence in national proceedings. The principle of effectiveness of EU law, and in this case the effectiveness of the pass-on defence, would preclude the national judge from issuing a (nonsensical) judgment to the effect that the defence has succeeded in the ‘EU law claim’ but failed in the ‘national law claim’, when both claims are based on the same facts and the outcome would be, in effect, that the defence has failed for the ‘EU law claim’ as well (as the defendant would have to pay the same damages as he would have to pay if the defence did not exist). Likewise, in the event of disclosure of evidence, it is hard to see how a national court could order the disclosure of a document whose disclosure is prohibited in an ‘EU law claim’ for the purposes of the ‘national law claim’. Since the ‘national law claim’ and the ‘EU law claim’ run in parallel before the same court, disclosure in one is tantamount to disclosure in the other. Second, insofar as room for divergence does subsist once the rules are harmonized for ‘EU law claims’, the argument against divergence is still not entirely convincing. It seems almost as if divergence is seen as being per se an impediment to European integration. Diversity of national rules on procedure and remedies subsists in many or most areas of law; if anything, private enforcement of competition law is a minor problem in this respect, as it is a comparatively rare occurrence that does not affect most businesses on a daily (or even yearly) basis.207 ‘Competence creep’ therefore seems likely, because tort rules contain many divergences among the Member States whose continued existence may eventually come to be seen as perpetuating an ‘uneven playing field’ and ‘fragmenting’ the internal market. Perhaps (and this is a tentative conjecture), at least part of the real motivation behind using Article 114 may lie in the perceived need to include the European Parliament in the process. It is not inconceivable that the Parliament has brought its greater democratic legitimacy to bear in the Commission’s favour, in exchange for being involved in pro-consumer, high PR value legislation, including the present initiative. In light of the amendments made pursuant to the Lisbon Treaty, it is entirely clear that the EU has exclusive competence to adopt the rules necessary for the effectiveness of EU competition law. These rules are adopted under Article 206 Ibid, Article 3 (codification of ‘equivalence’ and ‘effectiveness’); and Regulation 1/2003, Article 3(1) (concurrent application of EU and national law). 207 How many actions for damages are ‘enough’ and how many have actually been filed in the EU are heavily debated issues. See Milutinović, above note 2, section 13.3.2. Nevertheless, it seems safe to say that, certainly outside of Germany, private enforcement in general and damages claims in particular are not a mass phenomenon.

376  Veljko Milutinović 103 TFEU, where the Council acts alone (upon a proposal from the Commission). Thus, the only way to enable the Parliament to participate is to extend the scope of the proposed Directive some ways beyond EU competition law. The only way to do that is through the ‘approximation of laws’ which are not, formally speaking, necessary for the implementation of Articles 101 and 102. One last argument arises at this point, which supports the approximation of laws in ‘national law claims’. Although the principle of effectiveness of EU law, combined with the fact that an ‘EU law claim’ and a ‘national law claim’ are heard as a single claim, would eventually lead to approximation, what the Directive does is to accelerate and cement this process, by doing away with the need for Article 267 preliminary references (and the need for trusting the Member States to make such references where necessary). In particular as regards leniency documents, once they are released to a damages claimant, the ‘cat is out of the bag’, so to speak. Thus, although national judicial practices that lead to documents being disclosed in a manner that undermines EU law would arguably be ‘ironed out’ through the preliminary reference procedure, it seems reasonable for the Commission to want to pre-empt such scenarios, which may have irreparable consequences. To conclude, the Commission’s proposal for a Directive definitely signifies a tangible improvement over the status quo, and it ought to be welcomed. A certain amount of criticism is to be expected, of course, on issues such as ‘competence creep’ or the incomplete regulation of indirect purchaser claims. The criticism should not detract from the fact that the ‘interface’ between public and private enforcement is still a work in progress, and that the Commission’s latest initiative demonstrates a strong will to tackle this complex issue head-on. Certainly, the law is evolving and will continue to evolve through case law (in particular, that of the ECJ). However, as was seen in the main text of this chapter, spontaneous evolution often leads to anomalies and internal tensions which must occasionally be resolved by legislative intervention. While a certain amount of judicial ‘balancing’ at the micro level is always to be expected (and welcomed) in any legal system, at the macro level it is the legislator who is best placed to resolve conflicts between policy goals – in this case, the conflict between effective enforcement of EU antitrust rules on the one hand and the seemingly immutable ‘individual rights’ derived from those rules, on the other.

Annex I Proposal for a Directive of the European Parliament and of the Council 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 (2013)

EUROPEAN COMMISSION

Strasbourg, 11.6.2013 COM(2013) 404 final 2013/0185 (COD)

Proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL 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

(Text with EEA relevance) {SWD(2013) 203 final} {SWD(2013) 204 final}

EN

EN

380  EU Proposal for infringements of the competition law provisions

EXPLANATORY MEMORANDUM 1.

CONTEXT OF THE PROPOSAL

1.1.

General context

Regulation No 1/20031 gives effect to the EU rules prohibiting anticompetitive agreements (including cartels) and abuses of dominant positions (‘the EU competition rules’), which are laid down in Articles 101 and 102 of the Treaty on the Functioning of the European Union (‘the Treaty’), by setting out the conditions under which the Commission, the national competition authorities (‘NCAs’) and national courts apply these provisions in individual cases. Regulation No 1/2003 gives the Commission and the NCAs powers to apply Articles 101 and 102 of the Treaty2. The Commission can impose fines on undertakings that have infringed these provisions3. The powers of the NCAs are set out in Article 5 of Regulation No 1/2003. The application of the EU competition rules by the Commission and NCAs is commonly referred to as the public enforcement of EU competition law. In addition to public enforcement, the direct effect of Articles 101 and 102 of the Treaty means that these provisions create rights and obligations for individuals, which can be enforced by the national courts of the Member States4. This is referred to as the private enforcement of the EU competition rules. Damages claims for breaches of Articles 101 or 102 of the Treaty constitute an important area of private enforcement of EU competition law. It follows from the direct effect of the prohibitions laid down in Articles 101 and 102 of the Treaty that any individual can claim compensation for the harm suffered, where there is a causal relationship between that harm and an infringement of the EU competition rules5. Injured parties must be able to seek compensation not only for the actual loss suffered (damnum emergens) but also for the gain of which they have been deprived (loss of profit or lucrum cessans) plus interest6. Compensation for harm caused by infringements of EU competition rules cannot be achieved through public enforcement. Awarding compensation is outside the field of competence of the Commission and the NCAs and within the domain of national courts and of civil law and procedure. Compliance with the EU competition rules is thus ensured through the strong public enforcement of these rules by the Commission and the NCAs, in combination with private enforcement by national courts. 1.2.

Grounds for and objectives of the proposal

The present proposal seeks to ensure the effective enforcement of the EU competition rules by 1

2 3 4 5

6

EN

Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, OJ L 1, 4.1.2003, p. 1. With effect from 1 December 2009, Articles 81 and 82 of the EC Treaty have become Articles 101 and 102 of the Treaty. Their substance has not been changed. Articles 4 and 5 of Regulation No 1/2003, respectively. Article 23 of Regulation No 1/2003. Article 6 of Regulation No 1/2003; see also Case 127/73, BRT v. SABAM, [1974] ECR 51, paragraph 16; Case C-282/95 P, Guérin Automobiles v Commission, [1997] ECR I-1503, paragraph 39. See Case C-453/99, Courage and Crehan, [2001] ECR I-6297; Joined Cases C-295/04 to C-298/04, Manfredi, [2006] ECR I-6619; Case C-360/09, Pfleiderer AG v Bundeskartellamt, [2011] ECR I-5161; and Case C-199/11 European Community v. Otis NV and others, [2012] ECR I-0000. Manfredi, see fn 5, paragraph 95.

2

EN

EU Proposal for infringements of the competition law provisions  

381  

(i)

optimising the interaction between the public and private enforcement of competition law; and

(ii)

ensuring that victims of infringements of the EU competition rules can obtain full compensation for the harm they suffered.

Optimising the interaction between the public and private enforcement of competition law The overall enforcement of the EU competition rules is best guaranteed through complementary public and private enforcement. However, the existing legal framework does not properly regulate the interaction between the two strands of EU competition law enforcement. An undertaking that considers cooperating with a competition authority under its leniency programme (whereby the undertaking confesses its participation in a cartel in return for immunity from or a reduction of the fine), cannot know at the time of its cooperation whether victims of the competition law infringement will have access to the information it has voluntarily supplied to the competition authority. In particular, in its 2011 Pfleiderer judgment, the European Court of Justice (hereinafter: ‘the Court’)7, held that, in the absence of EU law, it is for the national court to decide on the basis of national law and on a case-bycase basis whether to allow the disclosure of documents, including leniency documents. When taking such a decision, the national court should balance both the interest of protecting effective public enforcement of the EU competition rules and of ensuring that the right to full compensation can be effectively exercised. This could lead to discrepancies between and even within Member States regarding the disclosure of evidence from the files of competition authorities. Moreover, the resulting uncertainty as to the disclosability of leniency-related information is likely to influence an undertaking’s choice whether or not to cooperate with the competition authorities under their leniency programme. In the absence of legally binding action at the EU level, the effectiveness of the leniency programmes — which constitute a very important instrument in the public enforcement of the EU competition rules — could thus be seriously undermined by the risk of disclosure of certain documents in damages actions before national courts. The need to regulate the interaction of private and public enforcement was confirmed in the stakeholders’ responses to the public consultation on the 2008 White Paper on damages actions for breach of the EU antitrust rules (‘White Paper’)8 and the 2011 consultation on a coherent European approach to collective redress9. The May 2012 resolution of the Heads of the European Competition Authorities also stressed the importance of the protection of leniency material in the context of civil damages actions10. The European Parliament repeatedly emphasised that public enforcement in the competition field is essential, and called on the Commission to ensure that private enforcement does not compromise the effectiveness of either the leniency programmes or settlement procedures11. 7 8 9 10

11

EN

Case C-360/09, Pfleiderer AG v Bundeskartellamt, [2011] ECR I-5161. COM(2008) 165 final; see also Commission Staff Working Paper annexed to the White Paper, SEC(2008) 404. Public consultation ‘Towards a coherent European approach to collective redress’, see http://ec.europa.eu/competition/consultations/2011_collective_redress/index_en.html. Resolution of the Meeting of the Heads of the European Competition Authorities of 23 May 2012, Protection of leniency material in the context of civil damages actions, available at: http://ec.europa.eu/competition/ecn/leniency_material_protection_en.pdf. See the European Parliament resolutions of 2 February 2012 on ‘Towards a Coherent European Approach to Collective Redress’ (2011/2089(INI)): http://www.europarl.europa.eu/sides/getDoc.do?type=TA&language=EN&reference=P7-TA-2012-21

3

EN

382  EU Proposal for infringements of the competition law provisions

The first main objective of the present proposal is thus to optimise the interaction between public and private enforcement of the EU competition rules, ensuring that the Commission and the NCAs can maintain a policy of strong public enforcement, while victims of an infringement of competition law can obtain compensation for the harm suffered. Ensuring the effective exercise of the victims’ right to full compensation The second main objective is to ensure that victims of infringements of EU competition rules can effectively obtain compensation for the harm they have suffered. While the right to full compensation is guaranteed by the Treaty itself and is part of the acquis communautaire, the practical exercise of this right is often rendered difficult or almost impossible because of the applicable rules and procedures. Despite some recent signs of improvement in a few Member States, to date most victims of infringements of the EU competition rules in practice do not obtain compensation for the harm suffered. As long ago as 2005, the Commission identified, in its Green Paper on damages actions for breach of the EC antitrust rules12 (‘the Green Paper’), the main obstacles to a more effective system of antitrust damages actions. Today, those same obstacles continue to exist in a large majority of the Member States. They relate to (i)

obtaining the evidence needed to prove a case;

(ii)

the lack of effective collective redress mechanisms, especially for consumers and SMEs;

(iii)

the absence of clear rules on the passing-on defence;

(iv)

the absence of a clear probative value of NCA decisions;

(v)

the possibility to bring an action for damages after a competition authority has found an infringement; and

(vi)

how to quantify antitrust harm.

Besides these specific substantive obstacles to effective compensation, there is wide diversity as regards the national legal rules governing antitrust damages actions and that the diversity has actually grown over recent years. This diversity may cause legal uncertainty for all parties involved in actions for antitrust damages, which in turn leads to ineffective private enforcement of the competition rules, especially in cross-border cases. To remedy this situation, the second main objective of the present proposal is to ensure that throughout Europe, victims of infringements of the EU competition rules have access to effective mechanisms for obtaining full compensation for the harm they suffered. This will lead to a more level playing field for undertakings in the internal market. In addition, if the likelihood increases that infringers of Articles 101 or 102 of the Treaty have to bear the costs of their infringement, this will not only shift the costs away from the victims of the illegal behaviour, but will also be an incentive for better compliance with the EU competition rules. To achieve that objective, the Commission put forward concrete policy proposals in its 2008 White Paper. In the ensuing public consultation, civil society and institutional stakeholders

12

EN

and on the Annual Report on EU Competition Policy (2011/2094(INI)) http://www.europarl.europa.eu/sides/getDoc.do?type=TA&language=EN&reference=P7-TA-2012-31. COM(2005) 672final; see also Commission Staff Working Paper annexed to the Green Paper, SEC(2005) 1732.

4

EN

EU Proposal for infringements of the competition law provisions  

383  

such as the European Parliament13 and the European Economic and Social Committee14 largely welcomed these policy measures and called for specific EU legislation on antitrust damages actions15. 1.3.

Existing provisions in the area of the proposal



Council Regulation No 1/2003 on the implementation of the rules on competition laid down in Articles [101] and [102] of the Treaty Pursuant to Article 2, the burden of proving an infringement of Article 101(1) or of Article 102 of the Treaty shall rest on the party alleging the infringement. Should the defending party claim the benefit of Article 101(3) of the Treaty, it shall bear the burden of proving that the conditions of that paragraph are fulfilled. These rules apply both to public enforcement and to actions for compensation for the harm caused by an infringement of Article 101 or 102 of the Treaty.



Article 15(1) provides that, in proceedings for the application of Article 101 or 102 of the Treaty, national courts may ask the Commission to transmit to them information in its possession. The Commission Notice on the cooperation between the Commission and the courts of the EU Member States in the application of Articles 101 and 102 of the Treaty16 elaborates on the interpretation and practical application of this provision.



Article 16(1) provides that when national courts rule on agreements, decisions or practices under Article 101 or Article 102 of the Treaty which are already the subject of a Commission decision, they cannot take decisions running counter to the decision adopted by the Commission. National courts must also avoid giving decisions which would conflict with a decision contemplated by the Commission in proceedings it has initiated. To that effect, the national court may assess whether it is necessary to stay the proceedings pending before it.



Council Regulation No 44/2001 contains rules on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters17. Under the conditions set out in that Regulation, courts of the Member States have jurisdiction to hear antitrust damages actions, and judgments in such actions are recognised and enforced in other Member States.

13

European Parliament Resolution of 26 March 2009 on the White Paper on damages actions for breach of the EC antitrust rules (2008/2154(INI)). European Economic and Social Committee Opinion of 25 March 2009 on the White Paper on damages actions for breach of the EC antitrust rules, OJ C 228, 22.09.2009, p. 40. See European Parliament Resolution of 2 February 2012 on the Annual Report on EU Competition Policy (2011/2094(INI)). OJ C 101, 27.4.2004, p. 54. Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, OJ L 12, 16.1.2001, p. 1. This Regulation has been recently replaced by Regulation (EU) No 1215/2012 of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, OJ L 351, 20.12.2012, p. 1, which for the most part will enter into force on 10 January 2015.

14 15 16 17

EN



5

EN

384  EU Proposal for infringements of the competition law provisions



Council Regulation No 1206/2001 regulates the cooperation between the courts of different Member States in the taking of evidence in civil or commercial matters, thus including antitrust damages actions18.



Article 6(3) of Regulation No 864/2007 of the European Parliament and of the Council contains rules on the law applicable in antitrust damages actions19.



Regulation 861/2007 of the European Parliament and of the Council20 establishes a European procedure for small claims, intended to simplify and speed up litigation concerning small claims in cross-border cases, and to reduce costs.



Directive 2008/52/EC of the European Parliament and of the Council requires Member States to provide for a possibility to mediate in all civil and commercial matters, thus including antitrust damages actions21.



Article 15(4) of Commission Regulation No 773/200422 determines that documents obtained through access to the file of the Commission shall only be used for the purposes of judicial or administrative proceedings for the application of Articles 101 and 102 of the Treaty. The Commission Notice on access to the file23 provides for more detailed rules as regards access to the Commission file and the use of those documents.



The Commission Notice on immunity from fines and reduction of fines in cartel cases (the ‘Leniency Notice’)24 contains rules on the conditions under which undertakings can cooperate with the Commission in the framework of its leniency programme in order to obtain immunity from or a reduction of its fine in a cartel case. In paragraph 33 it determines that access to corporate statements is only granted to the addressees of a statement of objections, provided that — together with the legal counsels obtaining access on their behalf — they do not make any copy by mechanical or electronic means of any information in the corporate statement and that the information from the corporate statement is solely used for the purposes mentioned in the Leniency Notice. Other parties such as complainants are not granted access to corporate statements. This specific protection of a corporate statement is no longer justified when the applicant discloses its content to a third party. Furthermore, the Commission Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel cases (the ‘Settlement Notice’)25 sets out the framework for rewarding cooperation with the Commission in the conduct of

18

Council Regulation (EC) No 1206/2001 of 28 May 2001 on cooperation between the courts of the Member States in the taking of evidence in civil or commercial matters, OJ L 174, 27.6.2001, p. 1. Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II), OJ L 199, 31.7.2007, p. 40. Regulation (EC) No 861/2007 of the European Parliament and of the Council of 11 July 2007 establishing a European Small Claims Procedure, OJ L 199, 31.07.2007, p. 1. Directive 2008/52/EC of the European Parliament and of the Council of 21 May 2008 on certain aspects of mediation in civil and commercial matters, OJ L 136, 24.5.2008, p. 3. Commission Regulation (EC) No 773/2004 of 7 April 2004 relating to the conduct of proceedings by the Commission pursuant to Articles 81 and 82 of the EC Treaty, OJ L 123/18, 27.4.2004, p. 18. Commission Notice on the rules for access to the Commission file in cases pursuant to Articles 81 and 82 of the EC Treaty, Articles 53, 54 and 57 of the EEA Agreement and Council Regulation (EC) No 139/2004, OJ C 325/07, 22.12.2005, p. 7. Commission Notice on immunity from fines and reduction of fines in cartel cases, OJ C 298, 08.12.2006, p. 17. OJ 2008/C 167/1.

19 20 21 22 23

24 25

EN

6

EN

EU Proposal for infringements of the competition law provisions  

385  

proceedings commenced in view of the application of Article 101 of the Treaty in cartel cases ('settlement procedure'). Its paragraph 39 contains rules on the disclosure of settlement submissions to national courts. 2.

RESULTS OF CONSULTATIONS WITH THE INTERESTED PARTIES AND IMPACT ASSESSMENTS

2.1.

Consultation of interested parties

Both the 2005 Green Paper and the 2008 White Paper triggered a broad debate among stakeholders, and a large number of comments were submitted26. The public consultations showed broad support for the Commission’s general approach to enabling antitrust damages actions. Respondents welcomed the guiding principle of compensation and the consequential choice not to suggest measures such as US-style class actions, wide pre-trial discovery or multiple damages, which would pursue primarily an objective of deterrence. There was broad acknowledgement of the obstacles that prevent effective redress for victims of infringements of the competition rules. However, different opinions were voiced as to the substantive measures suggested with a view to remedying the problems. In 2011, the Commission held a public consultation on a coherent European approach to collective redress27. In the wake of stakeholders' responses and the position of the European Parliament28, the Commission opted for a horizontal approach on this matter rather than the inclusion of competition-specific provisions on collective redress in the current proposal. Adopting a horizontal approach allows for common rules on collective redress for all policy fields in which scattered harm frequently occurs and in which it is difficult for consumers and SMEs to obtain damages. As the first step of an horizontal approach on collective redress, the Commission adopted the Communication 'Towards a European Framework for Collective Redress'29 and a Commission Recommendation on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law.30 The Commission also held a public consultation in 2011 on a draft Guidance Paper on the quantification of antitrust harm31. This sets out insights into a range of methods used to quantify harm in damages actions and explains the strengths and weaknesses of these

26

27 28

29

30

31

EN

The written comments received by the Commission during the public consultation can be accessed here: http://ec.europa.eu/competition/antitrust/actionsdamages/green_paper_comments.html (Green Paper consultation) and here: http://ec.europa.eu/competition/antitrust/actionsdamages/white_paper_comments.html (White Paper consultation). See fn. 9 above. European Parliament resolution of 2 February 2012 on ‘Towards a Coherent European Approach to Collective Redress’ (2011/2089(INI)): http://www.europarl.europa.eu/sides/getDoc.do?type=TA&language=EN&reference=P7-TA-2012-21. Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, Towards a European Horizontal Framework for Collective Redress, COM(2013) 401 final. Commission Recommendation on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law, C(2013) 3539 final. Public consultation on a Draft Guidance Paper − Quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty, accessible at: http://ec.europa.eu/competition/consultations/2011_actions_damages/index_en.html.

7

EN

386  EU Proposal for infringements of the competition law provisions

methods. Institutional and other stakeholders generally welcomed the idea of issuing nonbinding guidance on quantifying harm caused by antitrust infringements32. 2.2.

Collection and use of external expertise

The Commission commissioned external studies for the preparation of the 2005 Green Paper33, for the 2008 White Paper34 and for the 2011 draft Guidance Paper on the quantification of antitrust harm35. 2.3.

Impact assessment

The proposed Directive was preceded by an Impact Assessment, which built largely on the findings of the Impact Assessment on the White Paper. In particular, measures that had been excluded in the White Paper because of their likely ineffectiveness or excessive costs were not reconsidered. The impact assessment report36 focused on four options for a follow-up initiative aimed at optimising the interaction between public and private enforcement of the EU competition rules and ensuring a more effective legal framework for damages actions for infringements of the EU competition rules across Europe. They ranged from no action at the EU level, through a soft-law approach, to two options for legally binding EU action. The preferred option — which is the basis of this proposal for a Directive — is considered to be the most cost-efficient way of achieving the set objectives. It takes due account both of the main comments received during the public consultations over the past eight years and of more recent legislative and judicial developments at EU and national level. 3.

LEGAL ELEMENTS OF THE PROPOSAL

3.1.

Legal basis of the proposal

The choice of a legal basis for a European measure must be based on objective factors which are amenable to judicial review. Those include the aim and content of the measure. The current proposal is based on both Articles 103 and 114 of the Treaty, because it pursues two equally important goals which are inextricably linked, namely (a) to give effect to the principles set out in Articles 101 and 102 of the Treaty and (b) to ensure a more level playing field for undertakings operating in the internal market, and to make it easier for citizens and businesses to make use of the rights they derive from the internal market. Regarding the first objective, the Court has clarified that the full effectiveness of the EU competition rules and, in particular, the practical effect of the prohibitions they contain would be put at risk if it were not open to any person to claim damages for loss caused to him/her by a contract or conduct liable to restrict or distort competition. It considered that damages actions strengthen the working of the EU competition rules and can thus make a significant 32 33 34

35 36

EN

The submissions received under the public consultation are available at: http://ec.europa.eu/competition/consultations/2011_actions_damages/index_en.html#contributions. ‘Study on the conditions of claims for damages in case of infringement of EC competition rules’, accessible at: http://ec.europa.eu/competition/antitrust/actionsdamages/study.html. ‘Making antitrust damages actions more effective in the EU: welfare impact and potential scenarios’, accessible at: http://ec.europa.eu/competition/antitrust/actionsdamages/files_white_paper/impact_study.pdf. ‘Quantifying antitrust damages — Towards non-binding guidance for courts’, accessible at: http://ec.europa.eu/competition/antitrust/actionsdamages/quantification_study.pdf. Commission Staff Working Document, Impact Assessment Report, Damages actions for breach of the EU antitrust rules, Strasbourg 11-6-2013, SWD(2013) 203 final.

8

EN

EU Proposal for infringements of the competition law provisions  

387  

contribution to maintaining effective competition in the EU37. In seeking to improve the conditions under which injured parties can claim damages and to optimise the interaction between the public and private enforcement of Articles 101 and 102 TFEU, the present proposal clearly gives effect to these provisions. This means that the proposed Directive must be based on Article 103 of the Treaty. However, that legal basis in itself does not suffice, because both the aim and the content of the proposed Directive transcend this legal basis. Indeed, the aim of the proposed Directive is wider than giving effect to Articles 101 and 102 TFEU. The current divergence of national rules governing damages actions for infringements of the EU competition rules, including the interaction of such actions with the public enforcement of those rules, has created a markedly uneven playing field in the internal market. These marked differences were already described in a 2004 comparative study38 and in the 2008 White Paper and its accompanying Impact Assessment. Since then, these differences have increased due to diverging legislative and judicial developments in only a limited number of Member States. One example of divergence is given by the different national rules applying to access to evidence. With the exception of a few Member States, the lack of adequate rules on the disclosure of documents in proceedings before a national court means that victims of a competition law infringement, who are seeking compensation for the harm suffered, have no effective access to evidence. Other examples concern national rules on passing-on (where existing differences have major implications for the ability of direct/indirect purchasers to effectively claim damages and, in turn, for the defendant’s chances of avoiding compensation for harm caused), the probative value of NCA decisions in subsequent damages actions, and national rules that are relevant to the quantification of antitrust harm (e.g. the existence of a presumption of harm). Because of this marked diversity of national legislations, the rules applicable in some Member States are considered by claimants to be much more suitable for bringing an antitrust damages action in those Member States rather than in others. These differences lead to inequalities and uncertainty concerning the conditions under which injured parties, both citizens and businesses, can exercise the right to compensation they derive from the Treaty, and affect the effectiveness of such right. Indeed, where the jurisdictional rules allow a claimant to bring its action in one of those ‘favourable’ Member States and where that claimant has the necessary resources and incentives to do so, it is thus far more likely to effectively exercise its EU right to compensation than when it cannot do so. As injured parties with smaller claims and/or fewer resources tend to choose the forum of their Member State of establishment to claim damages (one reason being that consumers and smaller businesses in particular cannot afford to choose a more favourable jurisdiction), the result of the discrepancies between national rules may be an uneven playing field as regards actions for damages and may affect competition on the markets in which these injured parties operate. Similarly, these marked differences mean that undertakings established and operating in different Member States are exposed to significantly different levels of risk of being held liable for infringements of competition law. This uneven enforcement of the EU right of compensation may result in a competitive advantage for undertakings that have breached Articles 101 or 102 of the Treaty, but which do not have their headquarters or are not active in one of the ‘favourable’ Member States. Conversely, uneven enforcement is a disincentive to the exercise of the rights of establishment and provision of goods or services in those Member 37 38

EN

See fn. 5 above. See fn. 33 above.

9

EN

388  EU Proposal for infringements of the competition law provisions

States where the right to compensation is more effectively enforced. The differences in the liability regimes may thus negatively affect competition and run a risk of appreciably distorting the proper functioning of the internal market. To ensure a more level playing field for undertakings operating in the internal market and to improve the conditions for injured parties to exercise the rights they derive from the internal market, it is therefore appropriate to increase legal certainty and to reduce the differences between the Member States as to the national rules governing actions for antitrust damages. The extent to which the approximation of national rules is pursued is not limited to damages actions for breaches of the EU competition rules, but also of national competition rules when they are applied in parallel. In particular, when an infringement that has an effect on trade between Member States also breaches national competition law, actions for damages based on it must comply with the same standards established for breaches of EU competition law. Approximating national substantive and procedural rules with the aim of pursuing undistorted competition in the internal market and enabling citizens and undertakings the full exercise of the rights and freedoms they derive therefrom is not merely ancillary to the objective of ensuring effective enforcement of the EU competition rules. This conclusion results not only from the aims, but also from the specific provisions of the proposed Directive. The content of the proposed Directive cannot be fully covered by Article 103 of the Treaty because it also modifies the applicable national rules concerning the right to claim damages for infringements of national competition law, even if that is only in respect to anticompetitive behaviour that has an effect on trade between Member States and to which EU competition law thus equally applies39. It follows that the scope of the proposed Directive, arising not only from the aims but also from the contents of the instrument, goes beyond giving effect to Articles 101 and 102 of the Treaty and means that the proposed Directive also has to be based on Article 114 TFEU. These interdependent, though distinct, aims of the proposed Directive cannot be pursued separately, through the adoption of two different instruments. For instance, it is not feasible to split the proposed Directive into a first instrument, based on Article 103 TFEU, that approximates national rules in damages actions for breaches of Articles 101 and 102 TFEU, and a second one, based on Article 114 TFEU, that requires Member States to apply the same substantive and procedural rules to damages actions for breaches of national competition law. This choice cannot be made for substantive and procedural reasons. From a substantive point of view, the indissociable link between the two independent objectives underpins the concrete measures that pursue them. For instance, the exceptions to disclosure and limitations of liability give full effect to Articles 101 and 102, even in claims based on breaches of national competition law, when this has been applied in parallel to the Treaty provisions. Moreover, because of the need for legal certainty and a level playing field in the internal market, the same rules must apply to breaches of EU competition rules and of national competition law (where these are applied in parallel to the EU rules). From a procedural point of view, and to avoid impairing the institutional balance within the EU legislature, the only way of achieving uniform rules for the two situations is to adopt a single legal instrument in the same procedure. For these reasons, the contents of the initiative are not split among separate instruments but are addressed jointly in the proposed Directive, which should thus be based on both Articles 103 and 114 of the Treaty. 39

EN

See further under section 4.1 below.

10

EN

EU Proposal for infringements of the competition law provisions  

3.2.

389  

Subsidiarity principle (Article 5(3) of the Treaty on European Union)

The proposed Directive is in line with the subsidiarity principle since its objectives cannot be sufficiently achieved by the Member States, and there is a clear need for, and value in, EU action. A legally binding act at EU level will be better capable of ensuring that full effect is given to Articles 101 and 102 of the Treaty through common standards allowing for effective damages actions across the EU, and that a more level playing field is established in the internal market. More specifically, the proposed Directive can be deemed to comply with the principle of subsidiarity for the following reasons:

EN



There is a significant risk that effective public enforcement by the Commission and NCAs would be jeopardised in the absence of EU-wide regulation of the interaction between public and private enforcement, and in particular of a common European rule on information from the file of a competition authority being available for the purposes of a damages action. The point can be illustrated most clearly with regard to information that undertakings have voluntarily given to competition authorities under their leniency programme. The unpredictability that follows from the fact that each national court has to decide on an ad hoc basis and according to the applicable national rules whether or not to grant access to this leniency-related information cannot be adequately addressed by — potentially diverging — national legislation. Indeed, since the Commission and the NCAs can exchange information within the ECN, potential leniency applicants are likely to take into account the national legislation which offers the lowest level of protection (for fear that their case may eventually be decided by that NCA). The perceived level of protection of leniencyrelated information will thus be determined by whichever national legislation offers the lowest level of protection, to the detriment of the applicable rules in other Member States. It is therefore necessary to establish a standard common to all Member States for the interaction between public and private enforcement. This can only be done at the EU level.



Experience shows that, in the absence of EU legislation, most Member States do not provide, on their own initiative, for an effective framework for compensation for victims of infringements of Articles 101 and 102 of the Treaty, as repeatedly required by the Court. Since the publication of the Commission’s Green and White Papers, only a small number of Member States have enacted legislation aimed at enabling antitrust damages actions, and even this is usually limited to specific issues and does not cover the whole range of measures envisaged by the current proposal. Despite the few steps taken by some Member States, there is thus still a lack of effective compensation for victims of infringements of the EU antitrust rules. Only further incentives at European level can create a legal framework that gives effective redress and guarantees the right of effective judicial protection as laid down in Article 47 of the Charter of Fundamental Rights of the European Union.



There is currently a marked inequality between Member States in the level of judicial protection of individual rights guaranteed by the Treaty; this may cause distortions of competition and of the proper functioning of the internal market. The result is an evident disparity in even the content of the entitlement to damages guaranteed by EU law. More specifically, a claim under the law of one Member State may lead to full recovery of the claimant’s loss, while a claim for an identical infringement in another Member State may lead to a significantly lower award or even no award at all. This inequality increases if — as is the case at present — only some Member States

11

EN

390  EU Proposal for infringements of the competition law provisions

improve the conditions under which victims of a competition law infringement can claim compensation for the harm suffered. The trans-national dimension of Articles 101 and 102 of the Treaty and their intrinsic link to the functioning of the internal market warrants measures at the EU level. 3.3.

Proportionality principle (Article 5(4) of the Treaty on European Union)

In terms of proportionality, the proposed Directive does not go beyond what is necessary to effectively achieve its objectives, namely to guarantee effective protection of public enforcement of competition law across the EU, and to guarantee access for victims of competition law infringements to a truly effective mechanism for obtaining full compensation for the harm they have suffered, while protecting the legitimate interests of defendants and third parties. Under the proposed Directive, those objectives are also achieved at the lowest possible cost. The potential costs for citizens and businesses are proportionate to the stated objectives. A first step in this direction was taken with the White Paper by excluding more radical measures (e.g. multiple damages, opt-out class actions and wide-ranging discovery rules). The efforts to strike this balance were broadly welcomed during the public consultations. The safeguards built into the proposed Directive strengthen this balance further by reducing potential costs (especially litigation costs) without jeopardising the right to compensation. Furthermore, certain measures suggested in the White Paper, such as collective redress and rules on the fault requirement, have since been discarded for the purposes of this proposal. Finally, the choice of a Directive as the appropriate instrument is in line with the principle that there should be as little intervention as possible, so long as the objectives are achieved. 3.4.

A Directive as the most appropriate legally binding instrument

The objectives of the present proposal can best be pursued through a Directive. This is the most appropriate legal instrument to make the measures work effectively and to facilitate smooth adaptation into the legal systems of the Member States: –

A Directive requires Member States to achieve the objectives and implement the measures into their national substantive and procedural law systems. This approach gives the Member States more freedom when implementing an EU measure than does a Regulation, in that Member States are left the choice of the most appropriate means of implementing the measures in the Directive. This allows Member States to ensure that these new rules are consistent with their existing substantive and procedural legal framework.



Furthermore, a Directive is a flexible tool for introducing common rules in areas of national law that are crucial for the functioning of the internal market and the effectiveness of damages actions, and for ensuring adequate guarantees throughout the EU, while leaving room for individual Member States to go further, should they so wish.



Finally, a Directive avoids unnecessary action wherever the domestic provisions in the Member States are already in line with the proposed measures.

4.

DETAILED EXPLANATION OF THE PROPOSAL

4.1.

Scope and definitions (Chapter I: Articles 1 – 4)

The proposed Directive seeks to improve the conditions under which compensation can be obtained for harm caused by (a) infringements of the EU competition rules, and (b) infringements of national competition law provisions, where the latter are applied by a

EN

12

EN

EU Proposal for infringements of the competition law provisions  

391  

national competition authority or a national court in the same case in parallel to the EU competition rules. Such parallel application has its basis in the way in which Regulation No 1/2003 regulates the relationship between Articles 101 and 102 of the Treaty and national competition laws. Regulation No 1/2003 provides that where national competition authorities or national courts apply national competition law to agreements within the meaning of Article 101 which may affect trade between Member States, they must also apply Article 101. Similarly, where they apply national competition law to any abuse prohibited by Article 102, they must also apply Article 10240. In cases where compensation is sought for a violation of both EU and national competition law, it is appropriate for the same substantive and procedural rules to apply to those damages actions. Applying diverging rules on civil liability for a single specific instance of anticompetitive behaviour would not only make it unworkable for judges to handle the case, it would also imply legal uncertainty for all parties involved, and it could lead to conflicting results depending on whether the national court considers the case as an infringement of EU or of national competition law, thus hampering the effective application of those rules. The proposed Directive therefore refers to damages actions for ‘infringements of national or EU competition law’ or jointly ‘infringements of competition law’, whereby ‘national competition law’ is defined narrowly so as to cover only cases where it is applied in parallel to EU competition law. The proposed Directive sets out rules (i) ensuring that any natural or legal persons harmed by infringements of the competition rules are granted equivalent protection throughout the Union and can effectively enforce their EU right to full compensation through damages actions before national courts; and (ii) optimising the interaction between such damages actions and the public enforcement of the competition rules. Article 2 recalls the acquis communautaire on the EU right to full compensation. The proposed Directive thus embraces a compensatory approach: its aim is to allow those who have suffered harm caused by an infringement of the competition rules to obtain compensation for that harm from the undertaking(s) that infringed the law. Article 2 also recalls the acquis communautaire on standing and on the definition of damage to be compensated. The notion of actual loss referred to in this provision is taken from the case-law of the Court of Justice, and does not exclude any type of damage (material or immaterial) that might have been caused by an infringement of the competition rules. Article 3 recalls the principles of effectiveness and equivalence which must be complied with by national rules and procedures relating to actions for damages. 4.2.

Disclosure of evidence (Chapter II: Articles 5 – 8)

Establishing an infringement of the competition rules, quantifying antitrust damages, and establishing causality between the infringement and the harm suffered typically require a complex factual and economic analysis. Much of the relevant evidence a claimant will need to prove his case is in the possession of the defendant or of third persons and is often not sufficiently known or accessible to the claimants (‘information asymmetry’). It is widely recognised that the difficulty a claimant encounters in obtaining all necessary evidence constitutes in many Member States one of the key obstacles to damages actions in competition cases. In so far as the burden of proof falls on the (allegedly) infringing undertaking41, it too may need to have access to evidence in the hands of the claimant and/or

40 41

EN

Article 3(1) of Regulation No 1/2003. e.g. with regard to the passing-on defence, see section 4.4 below.

13

EN

392  EU Proposal for infringements of the competition law provisions

of a third party. The opportunity to ask the judge to order disclosure of information is therefore available to both parties to the proceedings. The disclosure regime in the proposed Directive builds on the approach adopted in Directive 2004/48/EC on the enforcement of intellectual property rights42. Its aim is to ensure that in all Member States there is a minimum level of effective access to the evidence needed by claimants and/or defendants to prove their antitrust damages claim and/or a related defence. At the same time, the proposed Directive avoids overly broad and costly disclosure obligations that could create undue burdens for the parties involved and risks of abuses. The Commission has also paid particular attention to ensuring that the proposal is compatible with the different legal orders of the Member States. To this end, the proposal follows the tradition of the great majority of Member States and relies on the central function of the court seized with an action for damages: disclosure of evidence held by the opposing party or a third party can only be ordered by judges and is subject to strict and active judicial control as to its necessity, scope and proportionality. National courts should have at their disposal effective measures to protect any business secrets or otherwise confidential information disclosed during the proceedings. Furthermore, disclosure should not be allowed where it would be contrary to certain rights and obligations such as the obligation of professional secrecy. Courts must also be able to impose sanctions which are sufficiently deterrent to prevent destruction of relevant evidence or refusal to comply with a disclosure order. To prevent that the disclosure of evidence jeopardises the public enforcement of the competition rules by a competition authority, the proposed Directive also establishes common EU-wide limits to disclosure of evidence held in the file of a competition authority: (a)

First, it provides for absolute protection for two types of documents which are considered to be crucial for the effectiveness of public enforcement tools. The documents referred to are the leniency corporate statements and settlement submissions. The disclosure of these documents risks seriously affecting the effectiveness of the leniency programme and of settlements procedures. Under the proposed Directive, a national court can never order disclosure of such documents in an action for damages.

(b)

Second, it provides for temporary protection for documents that the parties have specifically prepared for the purpose of public enforcement proceedings (e.g. the party’s replies to the authority’s request for information) or that the competition authority has drawn up in the course of its proceedings (e.g. a statement of objections). Those documents can be disclosed for the purpose of an antitrust damages action only after the competition authority has closed its proceedings.

(c)

Apart from limiting the national court’s ability to order disclosure, the above protective measures should also come into play if and when the protected documents have been obtained in the context of public enforcement proceedings (e.g. in the exercise of one of the parties’ right of defence). Therefore, where one of the parties in the action for damages had obtained those documents from the file of a competition authority, such documents are not admissible as evidence in an action for damages (documents of category (a) above) or are admissible only when the authority has closed its proceedings (documents of category (b) above).

42

EN

Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights, OJ L 157, 30.04.2004, p. 45.

14

EN

EU Proposal for infringements of the competition law provisions  

393  

(d)

Documents which fall outside the above categories can be disclosed by court order at any moment in time. However, when doing so, national courts should refrain from ordering the disclosure of evidence by reference to information supplied to a competition authority for the purpose of its proceedings43. While the investigation is on-going, such disclosure could hinder public enforcement proceedings, since it would reveal what information is in the file of a competition authority and could thus be used to unravel the authority’s investigation strategy. However, the selection of pre-existing documents that are submitted to a competition authority for the purposes of the proceedings is in itself relevant, as undertakings are invited to supply targeted evidence in view of their cooperation. The willingness of undertakings to supply such evidence exhaustively or selectively when cooperating with competition authorities may be hindered by disclosure requests that identify a category of documents by reference to their presence in the file of a competition authority rather than their type, nature or object (e.g. requests for all documents in the file of a competition authority or all documents submitted thereto by a party). Therefore, such global disclosure requests for documents should normally be deemed by the court as disproportionate and not complying with the requesting party's duty to specify categories of evidence as precisely and narrowly as possible.

(e)

Finally, to prevent documents obtained through access to a competition authority’s file becoming an object of trade, only the person who obtained access to the file (or his legal successor in the rights related to the claim) should be able to use those documents as evidence in an action for damages.

To achieve coherence regarding the rules on disclosure and the use of certain documents from the file of a competition authority, it is necessary to also amend existing rules on the conduct of Commission's proceedings laid down in Commission Regulation 773/200444, notably as regards access to the Commission's file and use of documents obtained therefrom, and the explanatory Notices published by the Commission45. The Commission intends doing so once the present Directive is adopted by the European Parliament and Council. 4.3.

Effect of national decisions, limitation periods and joint and several liability (Chapter III: Articles 9 – 11)

4.3.1.

Probative effect of national decisions

Pursuant to Article 16(1) of Regulation No 1/2003, a Commission decision relating to proceedings under Article 101 or 102 of the Treaty has a probative effect in subsequent actions for damages, as a national court cannot take a decision running counter to such Commission decision46. It is appropriate to give final infringement decisions by national competition authorities (or by a national review court) similar effect. If an infringement 43 44 45

46

EN

Obviously, the same limitation applies when national courts order the disclosure of documents of category (b) above, once a competition authority has closed its proceedings. Commission Regulation (EC) No 773/2004 of 7 April 2004 relating to the conduct of proceedings by the Commission pursuant to Articles 81 and 82 of the EC Treaty, OJ L123, 27.04.2004, p. 18. Commission Notice on the cooperation between the Commission and the courts of the EU Member States in the application of Articles 81 and 82 EC, OJ C101, 27.4.2004, p. 54; Commission Notice on the rules for access to the Commission file in cases pursuant to Articles 81 and 82 of the EC Treaty, Articles 53, 54 and 57 of the EEA Agreement and Council Regulation (EC) No 139/2004, OJ C 325, 22.12.2005, p. 7; and Commission Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel cases, OJ C 167, 2.7.2008, p. 1. Case C-199/11 European Community v Otis and others, [2012] ECR I-0000.

15

EN

394  EU Proposal for infringements of the competition law provisions

decision has already been taken and has become final, the possibility for the infringing undertaking to re-litigate the same issues in subsequent damages actions would be inefficient, cause legal uncertainty and lead to unnecessary costs for all parties involved and for the judiciary. The proposed probative effect of final infringement decisions of national competition authorities does not entail any lessening of judicial protection for the undertakings concerned, as infringement decisions by national competition authorities are still subject to judicial review. Moreover, throughout the EU, undertakings enjoy a comparable level of protection of their rights of defence, as enshrined in Article 48(2) of the EU Charter on Fundamental Rights. Finally, the rights and obligations of national courts under Article 267 of the Treaty remain unaffected by this rule. 4.3.2.

Limitation periods

To give victims of a competition law infringement a reasonable opportunity to bring a damages action, while ensuring an appropriate level of legal certainty for all parties involved, the Commission proposes that the national rules on limitation periods for a damages action: –

allow victims sufficient time (at least five years) to bring an action after they became aware of the infringement, the harm it caused and the identity of the infringer;



prevent a limitation period from starting to run before the day on which a continuous or repeated infringement ceases; and



in case a competition authority opens proceedings into a suspected infringement, the limitation period to bring an action for damages relating to such infringement is suspended until at least one year after a decision is final or proceedings are otherwise terminated.

4.3.3.

Joint and several liability

Where several undertakings infringe the competition rules jointly — typically in the case of a cartel — it is appropriate that they be jointly and severally liable for the entire harm caused by the infringement. While the proposed Directive builds on this general rule, it introduces certain modifications with regard to the liability regime of immunity recipients. The objective of these modifications is to safeguard the attractiveness of the leniency programmes of the Commission and of the NCAs, which are key instruments in detecting cartels and thus of crucial importance for the effective public enforcement of the competition rules. Indeed, as leniency recipients are less likely to appeal an infringement decision, this decision often becomes final for them earlier than for other members of the same cartel. This may make leniency recipients the primary targets of damages actions. To limit the disadvantageous consequences of such exposure, while not unduly limiting the possibilities for injured parties to obtain full compensation for the loss suffered, it is proposed to limit the immunity recipient’s liability, as well as his contribution owed to co-infringers under joint and several liability, to the harm he caused to his own direct or indirect purchasers or, in the case of a buying cartel, his direct or indirect providers. Where a cartel has caused harm only to others than the customers/providers of the infringing undertakings, the immunity recipient would be responsible only for his share of the harm caused by the cartel. How that share is determined (e.g. turnover, market share, role in the cartel, etc.), is left to the discretion of the Member States, as long as the principles of effectiveness and equivalence are respected. The protection of immunity recipients cannot, however, interfere with the victims’ EU right to full compensation. The proposed limitation on the immunity recipient’s liability cannot therefore be absolute: the immunity recipient remains fully liable as a last-resort debtor if the

EN

16

EN

EU Proposal for infringements of the competition law provisions  

395  

injured parties are unable to obtain full compensation from the other infringers. To guarantee the effet utile of this exception, Member States have to make sure that injured parties can still claim compensation from the immunity recipient at the time they have become aware that they cannot obtain full compensation from the co-cartelists. 4.4.

Passing-on of overcharges (Chapter IV: Articles 12 – 15)

Persons who have suffered harm caused by an infringement of the competition rules are entitled to compensation, regardless of whether they are direct or indirect purchasers. Injured parties are entitled to compensation for actual loss (overcharge harm) and for loss of profit. When an injured party has reduced his actual loss by passing it on, partly or entirely, to his own purchasers, the loss thus passed on no longer constitutes harm for which the party that passed it on has to be compensated. However, where a loss is passed on, the price increase by the direct purchaser is likely to lead to a reduction in the volume sold. That loss of profit, as well as the actual loss that was not passed on (in the case of partial passing-on) remains antitrust harm for which the injured party can claim compensation. If the harm is suffered as a result of an infringement relating to a supply to the infringing undertaking, passing-on could also take place in an upwards direction on the supply chain. This would, for example, be the case when, as a result of a buying cartel, the suppliers of the cartelists charge lower prices, and those suppliers then in turn require lower prices from their own suppliers. To ensure that only the direct and indirect purchasers that actually suffered overcharge harm can effectively claim compensation, the proposed Directive explicitly recognises the possibility for the infringing undertaking to invoke the passing-on defence. However, in situations where the overcharge was passed on to natural or legal persons at the next level of the supply chain for whom it is legally impossible to claim compensation, the passing-on defence cannot be invoked. Indirect purchasers may be faced with the legal impossibility of claiming compensation because of national rules on causality (including rules on foreseeability and remoteness). Allowing the passing-on defence when it is legally impossible for the party to whom the overcharge was allegedly passed on to claim compensation would be unjustified, since it would mean that the infringing undertaking is unduly freed from liability for the harm he caused. The burden of proving the passing-on always lies with the infringing undertaking. In the case of an action for damages brought by an indirect purchaser, this implies a rebuttable presumption pursuant to which, subject to certain conditions, a passing-on to that indirect purchaser occurred. As regards the quantification of the passing-on, the national court should have the power to estimate which share of the overcharge has been passed on to the level of indirect purchasers in the dispute pending before it. Where injured parties from different levels of the supply chain bring separate actions for damages that are related to the same competition law infringement, national courts should take due account, as far as allowed under applicable national or EU law, of parallel or preceding actions (or judgments resulting from such actions) in order to avoid under- and over-compensation of the harm caused by that infringement and to foster consistency between judgments resulting from such linked proceedings. Actions that are pending before the courts of different Member States may be considered as related within the meaning of Article 30 of Regulation No 1215/201247, meaning that they are so closely connected that it is expedient to hear and determine them together to avoid the risk of 47

EN

Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, OJ L 351, 20.12.2012, p. 1.

17

EN

396  EU Proposal for infringements of the competition law provisions

irreconcilable judgments resulting from separate proceedings. As a consequence, any court other than the court first seized may stay its proceedings or decline jurisdiction if the court first seized has jurisdiction over the actions in question and its law permits the consolidation of the actions. Both Regulation No 1215/2012 and this proposed Directive thus seek to encourage consistency between judgments resulting from related actions. To achieve that, the proposed Directive has an even wider scope than Regulation No 1215/2012, as it also covers the situation of subsequent actions for damages relating to the same competition law infringement, brought by injured parties at different levels of the supply chain. These actions can be brought in the same court, in different courts in the same Member State or in different courts of different Member States. In all instances, the proposed Directive encourages the consistency of linked proceedings and judgments. 4.5.

Quantification of harm (Chapter V: Article 16)

Proving and quantifying antitrust harm is generally very fact-intensive and costly, as it may require the application of complex economic models. To assist victims of a cartel in quantifying the harm caused by the competition law infringement, this proposed Directive provides for a rebuttable presumption with regard to the existence of harm resulting from a cartel. Based on the finding that more than 9 out of 10 cartels indeed cause an illegal overcharge48, this alleviates the injured party’s difficulties and costs related to proving that the cartel caused higher prices to be charged than if the cartel had not existed. The infringing undertaking could rebut this presumption and use the evidence at its disposal to prove that the cartel did not cause harm. The burden of proof is thus placed on the party which already has in its possession the necessary evidence to meet this burden of proof. The costs of disclosure, which would most likely be necessary for the injured parties to prove the existence of harm, are thus avoided. Apart from the above presumption, antitrust harm is quantified on the basis of national rules and procedures. These must, however, be in line with the principles of equivalence and of effectiveness. The latter, in particular, dictates that the burden and the level of proof may not render the injured party’s right to damages practically impossible or excessively difficult. In terms of quantifying antitrust harm, where the actual situation needs to be compared with a hypothetical one, this means that judges must be able to estimate the amount of harm. This increases the likelihood that victims will actually obtain an adequate amount of compensation for the harm they have suffered. To make it easier for national courts to quantify harm, the Commission is also providing nonbinding guidance on this topic in its Communication on quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty on the Functioning of the European Union49. The Communication is accompanied by a Commission Staff Working Paper taking the form of a Practical Guide on quantifying harm in actions for damages based on breaches of EU competition law. This Practical Guide explains the strengths and weaknesses of various methods and techniques available to quantify antitrust harm. It also presents and discusses a range of practical examples, which illustrate the typical effects that infringements of the EU competition rules tend to have and how the available methods and techniques can be applied in practice. 48 49

EN

‘Quantifying antitrust damages — Towards non-binding guidance for courts’, accessible at: http://ec.europa.eu/competition/antitrust/actionsdamages/quantification_study.pdf, p. 91. Communication from the Commission on quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty on the Functioning of the European Union, C(2013) 3440.

18

EN

EU Proposal for infringements of the competition law provisions  

4.6.

397  

Consensual Dispute Resolution (Chapter VI: Articles 17-18)

One of the primary objectives of the proposed Directive is to enable victims of a competition law infringement to obtain full compensation for the harm suffered. That objective can be achieved either through a damages action in court or through a consensual out-of-court settlement between the parties. To incentivise parties to settle their dispute consensually, the proposed Directive aims at optimising the balance between out-of-court settlements and actions for damages. It therefore contains the following provisions: (i)

suspension of limitation periods for bringing actions for damages as long as the infringing undertaking and the injured party are engaged in consensual dispute resolution;

(ii)

suspension of pending proceedings for the duration of consensual dispute resolution;

(iii)

reduction of the settling injured party’s claim by the settling infringer’s share of harm. For the remainder of the claim, the settling infringer could only be required to pay damages if the non-settling co-infringers were unable to fully compensate the injured party; and

(iv)

damages paid through consensual settlements to be taken into account when determining the contribution that a settling infringer needs to pay following a subsequent order to pay damages. In this context, ‘contribution’ refers to the situation where the settling infringer was not a defendant in the action for damages, but is asked by co-infringers who were ordered to pay damages to contribute under the rules of joint and several liability.

5.

BUDGETARY IMPLICATIONS

The proposed Directive does not have any budgetary implications. 6.

ADDITIONAL INFORMATION

6.1.

Repeal of existing legislation

No previous legislative act is repealed through this present proposal. 6.2.

Review

Article 21 of the proposed Directive requires the Commission to report to the European Parliament and the Council on its effects at the latest five years after the deadline for transposition into national law. Once the proposed Directive has been adopted, the Commission will continue to monitor the legal framework for antitrust damages actions in the Member States, focusing primarily on the achievement of the two main objectives of the proposed Directive, i.e. (i)

optimising the interaction between the public and private enforcement of competition law; and

(ii)

ensuring that victims of infringements of the EU competition rules can obtain full compensation for the harm they have suffered.

The Commission will assess whether the Directive is successful in removing inefficiencies and obstacles preventing full compensation of victims of antitrust infringements and whether the interaction between public and private enforcement of competition law is functioning

EN

19

EN

398  EU Proposal for infringements of the competition law provisions

smoothly, to guarantee the optimal overall enforcement of EU competition law. As part of this monitoring process, the Commission will continue its dialogue with all relevant stakeholders. Finally, an ex-post evaluation as regards the need for further modifications will be made once the measures put forward by the Directive are being fully implemented in the Member States, i.e. at least five years after the deadline for transposition of the Directive. 6.3.

Explanatory documents

The proposed Directive sets out specific measures to approximate substantive and procedural national rules governing actions for damages for infringements of the competition law provisions of the Member States and of the European Union. There are several legal obligations stemming from the proposed Directive. Its effective transposition will therefore require that specific and targeted amendments are made to the relevant national rules. In order for the Commission to monitor the correct implementation, it is thus not sufficient for Member States to transmit the text of the implementing provisions, as an overall assessment of the resulting regime under national law may be necessary. For these reasons, Member States should also transmit to the Commission explanatory documents showing which existing or new provisions under national law are meant to implement the individual measures set out in the proposed Directive. 6.4.

European Economic Area

The proposed Directive relates to the effective enforcement of Articles 101 and 102 of the Treaty, by optimising the interaction between the public and private enforcement of these provisions as well as by improving the conditions under which victims of competition law infringements can claim damages. The proposed Directive contributes to the proper functioning of the internal market as it creates a more level playing field both for the undertakings that infringe the competition rules and for the victims of this illegal behaviour. Due to these objectives in the fields of competition and the internal market, which form part of the EEA legal rules, the proposal is relevant for the EEA.

EN

20

EN

EU Proposal for infringements of the competition law provisions  

399  

2013/0185 (COD) Proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL 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 (Text with EEA relevance) THE EUROPEAN PARLIAMENT AND THE COUNCIL OF THE EUROPEAN UNION, Having regard to the Treaty on the Functioning of the European Union, and in particular Articles 103 and 114 thereof, Having regard to the proposal from the European Commission50, After transmission of the draft legislative act to the national Parliaments, Having regard to the opinion of the European Economic and Social Committee51, Acting in accordance with the ordinary legislative procedure, Whereas: (1)

Articles 101 and 102 of the Treaty on the Functioning of the European Union (hereinafter referred to as the Treaty) are a matter of public policy and must be applied effectively throughout the Union to ensure that competition in the internal market is not distorted.

(2)

The public enforcement of those Treaty provisions is carried out by the Commission using the powers provided by Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty establishing the European Community52 (hereinafter: Regulation No 1/2003). Public enforcement is also carried out by national competition authorities, which may take the decisions listed in Article 5 of Regulation No 1/2003.

(3)

Articles 101 and 102 of the Treaty produce direct effects in relations between individuals and create, for the individuals concerned, rights and obligations which national courts must enforce. National courts thus have an equally essential part to play in applying the competition rules (private enforcement). When ruling on disputes between private individuals, they protect subjective rights under Union law, for example by awarding damages to the victims of infringements. The full effectiveness of Articles 101 and 102 of the Treaty, and in particular the practical effect of the prohibitions laid down therein, requires that anyone — be they an individual, including consumers and undertakings, or a public authority — can claim compensation before national courts for the harm caused to them by an infringement

50

OJ C , , p. . OJ C , , p. . OJ L 1, 4.1.2003, p. 1. With effect from 1 December 2009, Articles 81 and 82 of the EC Treaty have become respectively Articles 101 and 102 TFEU. The two sets of provisions are identical in substance.

51 52

EN

21

EN

400  EU Proposal for infringements of the competition law provisions

of those provisions. This Union right to compensation applies equally to breaches of Articles 101 and 102 by public undertakings or undertakings entrusted with special or exclusive rights by Member States within the meaning of Article 106 of the Treaty.

EN

(4)

The Union right to compensation for antitrust harm requires each Member State to have procedural rules ensuring the effective exercise of that right. The need for effective procedural remedies also follows from the right to effective judicial protection as laid down in Article 47, first paragraph, of the Charter of Fundamental Rights of the European Union53 and in Article 19(1), second subparagraph of the Treaty on European Union.

(5)

To ensure effective public and private enforcement of the competition rules, it is necessary to regulate the way the two forms of enforcement are coordinated, for instance the arrangements for access to documents held by competition authorities. Such coordination at Union level will also avoid divergence of applicable rules, which could jeopardise the proper functioning of the internal market.

(6)

In accordance with Article 26(2) of the Treaty, the internal market comprises an area without internal frontiers in which the free movement of goods, persons, services and capital is ensured. There exist marked differences between the rules in the Member States governing actions for damages for infringements of national or Union competition law. Those differences lead to uncertainty concerning the conditions under which injured parties can exercise the right to compensation they derive from the Treaty, and affect the substantive effectiveness of such right. As injured parties often choose the forum of their Member State of establishment to claim damages, the discrepancies between the national rules lead to an uneven playing field as regards actions for damages and may affect competition on the markets on which these injured parties, as well as the infringing undertakings, operate.

(7)

Undertakings established and operating in different Member States are subject to procedural rules that significantly affect the extent to which they can be held liable for infringements of competition law. This uneven enforcement of the Union right to compensation may result in a competitive advantage for some undertakings which have breached Articles 101 or 102 of the Treaty, and a disincentive to the exercise of the rights of establishment and provision of goods or services in those Member States where the right to compensation is more effectively enforced. As such, the differences in the liability regimes applicable in the Member States may negatively affect both competition and the proper functioning of the internal market.

(8)

It is therefore necessary to ensure a more level playing field for undertakings operating in the internal market and to improve the conditions for consumers to exercise the rights they derive from the internal market. It is also appropriate to increase legal certainty and to reduce the differences between the Member States as to the national rules governing actions for damages for infringements of European competition law and, when applied in parallel to the latter, national competition law. An approximation of these rules will also help to prevent the emergence of wider differences between the Member States’ rules governing actions for damages in competition cases.

(9)

Article 3(1) of Regulation (EC) No 1/2003 provides that ‘where the competition authorities of the Member States or national courts apply national competition law to agreements, decisions by associations of undertakings or concerted practices within

53

OJ C 326, 26.10.2012, p. 391.

22

EN

EU Proposal for infringements of the competition law provisions  

401  

the meaning of Article [101(1)] of the Treaty which may affect trade between Member States within the meaning of that provision, they shall also apply Article [101] of the Treaty to such agreements, decisions or concerted practices. Where the competition authorities of the Member States or national courts apply national competition law to any abuse prohibited by Article [102] of the Treaty, they shall also apply Article [102] of the Treaty.’ In the interest of the proper functioning of the internal market and with a view to greater legal certainty and a more level playing field for undertakings and consumers, it is appropriate that the scope of this Directive should extend to actions for damages based on the infringement of national competition law where it is applied pursuant to Article 3(1) of Regulation (EC) No 1/2003. Applying diverging rules on civil liability for infringements of Articles 101 and 102 of the Treaty and for infringements of rules of national competition law which must be applied in the same case and in parallel to Union competition law would otherwise adversely affect the position of claimants in the same case and the scope of their claims, and constitute an obstacle to the proper functioning of the internal market.

EN

(10)

In the absence of Union law, actions for damages are governed by the national rules and procedures of the Member States. All national rules governing the exercise of the right to compensation for harm resulting from an infringement of Article 101 or 102 of the Treaty, including those concerning aspects not dealt with in this Directive such as the notion of causal relationship between the infringement and the harm, must observe the principles of effectiveness and equivalence. This means that they may not be formulated or applied in a way that makes it excessively difficult or practically impossible to exercise the right to compensation guaranteed by the Treaty, and they may not be formulated or applied less favourably than those applicable to similar domestic actions.

(11)

This Directive reaffirms the acquis communautaire on the Union right to compensation for harm caused by infringements of Union competition law, particularly regarding standing and the definition of damage, as it has been stated in the case-law of the Court of Justice of the European Union, and does not pre-empt any further development thereof. Anyone who has suffered harm caused by an infringement can claim compensation for the actual loss (damnum emergens), for the gain of which he has been deprived (loss of profit or lucrum cessans) and payment of interest accruing from the time the harm occurred until compensation is paid. This right is recognised for any natural or legal person — consumers, undertakings and public authorities alike — irrespective of the existence of a direct contractual relationship with the infringing undertaking, and regardless of whether or not there has been a prior finding of an infringement by a competition authority. This Directive should not require Member States to introduce collective redress mechanisms for the enforcement of Articles 101 and 102 of the Treaty.

(12)

Actions for damages for infringements of national or Union competition law typically require a complex factual and economic analysis. The evidence necessary to prove a claim for damages is often held exclusively by the opposing party or by third parties, and is not sufficiently known by and accessible to the claimant. In such circumstances, strict legal requirements for claimants to assert in detail all the facts of their case at the beginning of an action and to proffer precisely specified pieces of supporting evidence can unduly impede the effective exercise of the right to compensation guaranteed by the Treaty.

(13)

Evidence is an important element for bringing actions for damages for infringement of national or Union competition law. However, as antitrust litigation is characterised by

23

EN

402  EU Proposal for infringements of the competition law provisions

an information asymmetry, it is appropriate to ensure that injured parties are afforded the right to obtain the disclosure of evidence relevant to their claim, without it being necessary for them to specify individual items of evidence. In order to ensure equality of arms, those means should also be available to defendants in actions for damages, so that they can request the disclosure of evidence by those injured parties. National courts can also order evidence to be disclosed by third parties. Where the national court wishes to order disclosure of evidence by the Commission, the principle of sincere cooperation between the European Union and the Member States (Article 4(3) TEU) and Article 15(1) of Regulation No 1/2003 as regards requests for information are applicable.

EN

(14)

Relevant evidence should be disclosed upon decision of the court and under its strict control, especially as regards the necessity and proportionality of the disclosure measure. It follows from the requirement of proportionality that disclosure requests can only be triggered once an injured party has made it plausible, on the basis of facts which are reasonably available to him, that the party has suffered harm that was caused by the defendant. The request for disclosure should refer to categories of evidence which are as precise and narrow as possible on the basis of reasonably available facts.

(15)

The requirement of proportionality should also be carefully assessed when disclosure risks unravelling the investigation strategy of a competition authority by revealing which documents are part of the file or causing a negative bearing on the way in which companies cooperate with the competition authority. The disclosure request should therefore not be deemed proportionate when it refers to the generic disclosure of documents in the file of a competition authority relating to a certain case, or of documents submitted by a party in the context of a certain case. Such wide disclosure requests would also not be compatible with the requesting party's duty to specify categories of evidence as precisely and narrowly as possible.

(16)

Where the court requests a competent court of another Member State to take evidence or requests evidence to be taken directly in another Member State, the provisions of Council Regulation (EC) No 1206/2001 of 28 May 2001 on cooperation between the courts of the Member States in the taking of evidence in civil or commercial matters54 apply.

(17)

While relevant evidence containing business secrets or otherwise confidential information should in principle be available in actions for damages, such confidential information needs to be appropriately protected. National courts should therefore have at their disposal a range of measures to protect such confidential information from being disclosed during the proceedings. These may include the possibility of hearings in private, restricting the circle of persons entitled to see the evidence, and instruction of experts to produce summaries of the information in an aggregated or otherwise nonconfidential form. Measures protecting business secrets and other confidential information should not practically impede the exercise of the right to compensation.

(18)

The effectiveness and consistency of the application of Articles 101 and 102 of the Treaty by the Commission and the national competition authorities require a common approach across the Union regarding the interaction of rules on disclosure of evidence and the way these Articles are enforced by a competition authority. Disclosure of evidence should not unduly detract from the effectiveness of enforcement of

54

OJ L 174, 27.6.2001, p. 1.

24

EN

EU Proposal for infringements of the competition law provisions  

403  

competition law by a competition authority. The limitations on the disclosure of evidence should not prevent competition authorities from publishing their decisions in accordance with applicable Union or national rules.

EN

(19)

Leniency programmes and settlement procedures are important tools for the public enforcement of Union competition law as they contribute to the detection, efficient prosecution and sanctioning of the most serious competition law infringements. Undertakings may be deterred from co-operating in this context if disclosure of documents they solely produce to this end were to expose them to civil liability under worse conditions than the co-infringers that do not co-operate with competition authorities. To ensure that undertakings are willing to produce voluntary statements acknowledging their participation in an infringement of Union or national competition law to a competition authority under a leniency programme or a settlement procedure, such statements should be excepted from disclosure of evidence.

(20)

In addition, an exception to disclosure should apply to any disclosure measure that would unduly interfere with an ongoing investigation by a competition authority concerning an infringement of national or Union competition law. Information that was prepared by a competition authority in the course of its proceedings for the enforcement of national or Union competition law (such as a Statement of Objections) or by a party to those proceedings (such as replies to requests for information of the competition authority) should therefore be disclosable in actions for damages only after the competition authority has found an infringement of the national or Union competition rules or has otherwise closed its proceedings.

(21)

Apart from the evidence referred to in recitals (19) and(20), national courts should be able to order, in the context of an action for damages, disclosure of evidence that exists irrespective of the proceedings of a competition authority (‘pre-existing information’).

(22)

Any natural or legal person who obtains evidence through access to the file of a competition authority in exercising his rights of defence in relation to investigations by a competition authority can use that evidence for the purposes of an action for damages to which he is a party. Such use should also be allowed for the natural or legal person that succeeded in his rights and obligations, including through the acquisition of his claim. In case the evidence was obtained by a legal person forming part of a corporate group constituting one undertaking for the application of Articles 101 and 102 of the Treaty, the use of such evidence is also allowed for other legal entities belonging to the same undertaking.

(23)

However, the use referred to in the previous recital may not unduly detract from the effective enforcement of competition law by a competition authority. Limitations to disclosure referred to in recitals (19) and (20) should thus equally apply to the use of evidence which is obtained solely through access to the file of a competition authority. Moreover, evidence obtained from a competition authority in the context of exercise of the rights of defence should not become an object of trade. The possibility of using evidence that was obtained solely through access to the file of a competition authority should therefore be limited to the natural or legal person that exercised his rights of defence and his legal successors, as mentioned in the previous recital. This limitation does not, however, prevent a national court from ordering the disclosure of that evidence under the conditions provided for in this Directive.

(24)

Making a claim for damages, or the start of an investigation by a competition authority, entails a risk that the undertakings concerned may destroy or hide evidence that would be useful in substantiating an injured party’s claim for damages. To prevent

25

EN

404  EU Proposal for infringements of the competition law provisions

the destruction of relevant evidence and to ensure that court orders requesting disclosure are complied with, courts should be able to impose sufficiently deterrent sanctions. Insofar as parties to the proceedings are concerned, the risk of adverse inferences being drawn in the proceedings for damages can be a particularly effective sanction and can avoid delays. Sanctions should also be available for non-compliance with obligations to protect confidential information and for abusive use of information obtained through disclosure. Similarly, sanctions should be available if information obtained through access to the file of a competition authority in the exercise of one’s rights of defence in relation to investigations of that competition authority is used abusively in actions for damages.

EN

(25)

Article 16(1) of Regulation (EC) No 1/2003 provides that where national courts rule on agreements, decisions or practices under Article 101 or 102 of the Treaty which are already the subject of a Commission decision, they cannot take decisions which run counter to the decision adopted by the Commission. To enhance legal certainty, to avoid inconsistency in the application of those Treaty provisions, to increase the effectiveness and procedural efficiency of actions for damages and to foster the functioning of the internal market for undertakings and consumers, it should similarly not be possible to call into question a final decision by a national competition authority or a review court finding an infringement of Article 101 or 102 of the Treaty in actions for damages relating to the same infringement, regardless of whether or not the action is brought in the Member State of the authority or review court. The same should apply to a decision in which it has been concluded that provisions of national competition law are infringed in cases where national and Union competition law are applied in the same case and in parallel. This effect of decisions by national competition authorities and review courts finding an infringement of the competition rules should apply to the operative part of the decision and its supporting recitals. This is without prejudice to the rights and obligations of national courts under Article 267 of the Treaty.

(26)

National rules on the beginning, duration, suspension or interruption of limitation periods should not unduly hamper the bringing of actions for damages. This is particularly important in respect of actions that build upon the competition authority's or a review court’s finding of an infringement. To that end, injured parties should still be able to bring an action for damages after proceedings by a competition authority, with a view to enforcing national and Union competition law.

(27)

Where several undertakings infringe the competition rules jointly (as in the case of a cartel) it is appropriate to make provision for these joint infringers to be held jointly and severally liable for the entire harm caused by the infringement. Amongst themselves, the joint infringers should have the right to obtain contribution if one of the infringing undertakings has paid more than its share. The determination of that share as the relative responsibility of a given infringer and the relevant criteria, such as turnover, market share, or role in the cartel, is a matter for the applicable national law, while respecting the principles of effectiveness and equivalence.

(28)

Undertakings which cooperate with competition authorities under a leniency programme play a key role in detecting secret cartel infringements and in bringing these infringements to an end, thereby often mitigating the harm which could have been caused had the infringement continued. It is therefore appropriate to make provision for undertakings which have received immunity from fines from a competition authority under a leniency programme to be protected from undue exposure to damages claims, bearing in mind that the decision of the competition

26

EN

EU Proposal for infringements of the competition law provisions  

405  

authority finding the infringement may become final for the immunity recipient before it becomes final for other undertakings which have not received immunity. It is therefore appropriate that the immunity recipient is relieved in principle from joint and several liability for the entire harm and that its contribution does not exceed the amount of harm caused to his own direct or indirect purchasers or, in case of a buying cartel, his direct or indirect providers. To the extent a cartel has caused harm to others than the customers/providers of the infringing undertakings, the contribution of the immunity recipient should not exceed his relative responsibility for the harm caused by the cartel. This share should be determined in accordance with the same rules used to determine the contributions among infringing undertakings (recital (27) above). The immunity recipient should remain fully liable to the injured parties other than his direct or indirect purchasers or providers only where they are unable to obtain full compensation from the other infringing undertakings.

EN

(29)

Consumers and undertakings which have been harmed by an infringement of national or Union competition law are entitled to compensation for the actual loss and for loss of profit. The actual loss can result from the price difference between what was actually paid and what would have been paid in the absence of the infringement. When an injured party has reduced his actual loss by passing it on, entirely or in part, to his own purchasers, the loss which has been passed on no longer constitutes harm for which the party that passed it on has to be compensated. It is therefore in principle appropriate to allow an infringing undertaking to invoke the passing-on of actual loss as a defence against a claim for damages. It is appropriate to provide that the infringing undertaking, insofar as it invokes the passing-on defence, must prove the existence and extent of pass-on of the overcharge.

(30)

However, in a situation where the overcharge was passed on to persons who are legally unable to claim compensation, it is not appropriate to allow the infringing undertaking to invoke the passing-on defence, as this would render it free of liability for the harm which it has caused. The court seized of the action should therefore assess, when the passing-on defence is invoked in a specific case, whether the persons to whom the overcharge was allegedly passed on are legally able to claim compensation. While indirect purchasers are entitled to claim compensation, national rules of causality (including rules on foreseeability and remoteness), applied in accordance with principles of Union law, may entail that certain persons (for instance at a level of the supply chain which is remote from the infringement) are legally unable to claim compensation in a given case. Only when the court finds that the person to whom the overcharge was allegedly passed on is legally able to claim compensation will it assess the merits of the passing-on defence.

(31)

Consumers or undertakings to whom actual loss has been passed on have suffered harm that has been caused by an infringement of national or Union competition law. While such harm should be compensated by the infringing undertaking, it may be particularly difficult for consumers or undertakings that did not themselves make any purchase from the infringing undertaking to prove the scope of that harm. It is therefore appropriate to provide that, where the existence of a claim for damages or the amount to be awarded depends on whether or to what degree an overcharge paid by the direct purchaser of the infringing undertaking has been passed on to the indirect purchaser, the latter is regarded as having brought the proof that an overcharge paid by that direct purchaser has been passed on to his level, where he is able to show prima facie that such passing-on has occurred. It is furthermore appropriate to define under what conditions the indirect purchaser is to be regarded as having established such

27

EN

406  EU Proposal for infringements of the competition law provisions

prima facie proof. As regards the quantification of passing-on, the national court should have the power to estimate which share of the overcharge has been passed on to the level of indirect purchasers in the dispute pending before it. The infringing undertaking should be allowed to bring proof showing that the actual loss has not been passed on or has not been passed on entirely.

EN

(32)

Infringements of competition law often concern the conditions and the price under which goods or services are sold and lead to an overcharge and other harm for the customers of the infringing undertakings. The infringement may also concern supplies to the infringing undertaking (for example in the case of a buyer’s cartel). The rules of this Directive and in particular the rules on pass-on should apply accordingly.

(33)

Actions for damages can be brought both by injured parties that have purchased goods or services from the infringing undertaking and by purchasers further down the supply chain. In the interest of consistency between judgments resulting from such related proceedings and hence to avoid the harm caused by the infringement of national or Union competition law not being fully compensated or the infringing undertaking being required to pay damages to compensate for harm that has not been suffered, national courts should take due account, as far as allowed under Union and national law, of any related action and of the resulting judgment, particularly where it finds that passing-on has been proven. This should be without prejudice to the fundamental rights of defence and to an effective remedy and a fair trial of those who were not parties to these judicial proceedings. Any such actions pending before the courts of different Member States may be considered as related within the meaning of Article 30 of Regulation No 1215/2012. Under this provision, national courts other than the one first seized may stay proceedings or, under certain circumstances, decline jurisdiction.

(34)

An injured party who has proven having suffered harm as a result of a competition law infringement still needs to prove the extent of the harm in order to obtain damages. Quantifying antitrust harm is a very fact-intensive process and may require the application of complex economic models. This is often very costly and causes difficulties for injured parties in terms of obtaining the necessary data to substantiate their claims. As such, the quantification of antitrust harm can constitute a substantial barrier preventing injured parties from obtaining compensatory damages for harm suffered.

(35)

To remedy the information asymmetry and some of the difficulties associated with quantifying antitrust harm, and to ensure the effectiveness of claims for damages, it is appropriate to presume that in the case of a cartel infringement, the infringement has caused harm, in particular via a price effect. Depending on the facts of the case this means that the cartel has caused a rise in price, or prevented a lowering of prices which would otherwise have occurred but for the infringement. The infringing undertaking should be free to rebut such presumption. It is appropriate to limit this rebuttable presumption to cartels, given the secret nature of a cartel, which increases the said information asymmetry and makes it more difficult for the injured party to obtain the necessary evidence to prove the harm.

(36)

In the absence of Union rules on the quantification of harm caused by a competition law infringement, it is for the domestic legal system of each Member State and for the national courts to determine what requirements the injured party has to meet when proving the amount of the harm suffered, how precisely he has to prove that amount, the methods that can be used in quantifying the amount and the consequences of not being able to fully meet the set requirements. However, these domestic requirements

28

EN

EU Proposal for infringements of the competition law provisions  

407  

should not be less favourable than those governing similar domestic actions (principle of equivalence), nor should they render the exercise of the Union right to damages practically impossible or excessively difficult (principle of effectiveness). Regard should be had in this respect to any information asymmetries between the parties and to the fact that quantifying the harm means assessing how the market in question would have evolved had there been no infringement. This assessment implies a comparison with a situation which is by definition hypothetical and can thus never be made with complete accuracy. It is therefore appropriate to give national courts the power to estimate the amount of the harm caused by the competition law infringement.

EN

(37)

Injured parties and infringing undertakings should be encouraged to agree on compensating the harm caused by a competition law infringement through consensual dispute resolution mechanisms, such as out-of-court settlements, arbitration and mediation. Where possible, such consensual dispute resolution should cover as many injured parties and infringing undertakings as possible. The provisions in this Directive on consensual dispute resolution are therefore meant to facilitate the use of such mechanisms and increase their effectiveness.

(38)

Limitation periods for bringing an action for damages could be such that they prevent injured parties and infringing undertakings from having sufficient time to come to an agreement on the compensation to be paid. In order to provide both with a genuine opportunity to engage in consensual dispute resolution before bringing proceedings before the national court, the limitation period thus needs to be suspended for the duration of the consensual dispute resolution process.

(39)

Furthermore, when parties decide to engage in consensual dispute resolution after an action for damages has been brought before the national court for the same claim, that court may suspend the proceedings before it for the duration of the consensual dispute resolution process. When considering whether to suspend the proceedings, the national court should take into account the interest in an expeditious procedure.

(40)

To encourage consensual settlements, an infringing undertaking that pays damages through consensual dispute resolution should not be placed in a worse position vis-àvis its co-infringers than it would be in without the consensual settlement. This might happen if a settling infringer, even after a consensual settlement, continued to be fully jointly and severally liable for the harm caused by the infringement. A settling infringer should in principle therefore not contribute to his non-settling co-infringers when the latter have paid damages to the injured party with whom the first infringer had previously settled. The correlate to this non-contribution rule is that the claim of the injured party is reduced by the settling infringer’s share of the harm caused to him. This share should be determined in accordance with the same rules used to determine the contributions among infringing undertakings (recital (27) above). Without such reduction, the non-settling infringers would be unduly affected by the settlement to which they were not a party. The settling co-infringer will still have to pay damages where that is the only possibility for the injured party to obtain full compensation.

(41)

When settling co-infringers are asked to contribute to damages subsequently paid by non-settling co-infringers, the national court should take account of the damages already paid under the consensual settlement, bearing in mind that not all co-infringers are necessarily equally involved in the full substantive, temporal and geographical scope of the infringement.

(42)

This Directive respects fundamental rights and observes the principles recognised in the Charter of Fundamental Rights of the European Union.

29

EN

408  EU Proposal for infringements of the competition law provisions

(43)

As it would be impossible, with a disparity of policy choices and legal rules at national level concerning the Union right to compensation in actions for damages for infringement of the Union competition rules, to ensure the full effect of Articles 101 and 102 of the Treaty, and to ensure the proper functioning of the internal market for undertakings and consumers, these objectives cannot be sufficiently achieved by the Member States, and can therefore, by reason of the requisite effectiveness and consistency in the application of Articles 101 and 102 of the Treaty, be better achieved at Union level. The European Parliament and the Council therefore adopt this Directive, in accordance with the principle of subsidiarity as set out in Article 5 of the Treaty on European Union. In accordance with the principle of proportionality, as set out in that Article, this Directive does not go beyond what is necessary in order to achieve those objectives.

(44)

In accordance with the Joint Political Declaration of Member States and the Commission on explanatory documents of 28 September 201155, Member States have undertaken to accompany, in justified cases, the notification of their transposition measures with one or more documents explaining the relationship between the components of a directive and the corresponding parts of national transposition instruments. With regard to this Directive, the legislator considers the transmission of such documents to be justified.

HAVE ADOPTED THIS DIRECTIVE:

CHAPTER I SCOPE AND DEFINITIONS Article 1 Scope of the Directive 1.

This Directive sets out certain rules necessary to ensure that anyone who has suffered harm caused by an infringement of Article 101 or 102 of the Treaty or of national competition law, can effectively exercise the right to full compensation for that harm. It also sets out rules fostering undistorted competition in the internal market and removing obstacles to its proper functioning by ensuring equivalent protection throughout the Union for anyone who has suffered such harm.

2.

This Directive also sets out rules for the coordination between enforcement of the competition rules by competition authorities and enforcement of those rules in damages actions before national courts. Article 2 Right to full compensation

1. 55

EN

Anyone who has suffered harm caused by an infringement of Union or national competition law shall be able to claim full compensation for that harm. OJ C 369, 17.12.2011, p. 14.

30

EN

EU Proposal for infringements of the competition law provisions  

409  

2.

Full compensation shall place anyone who has suffered harm in the position in which that person would have been had the infringement not been committed. It shall therefore include compensation for actual loss and for loss of profit, and payment of interest from the time the harm occurred until the compensation in respect of that harm has actually been paid.

3.

Member States shall ensure that injured parties can effectively exercise their claims for damages. Article 3 Principles of effectiveness and equivalence

Member States shall ensure that all national rules and procedures relating to actions for damages are designed and applied in such a way as to ensure that any injured party can effectively exercise the Union right to full compensation for harm caused by an infringement of competition law. Any national rules and procedures relating to actions for damages resulting from infringements of Article 101 or 102 of the Treaty shall not be less favourable to the injured parties than those governing similar domestic actions. Article 4 Definitions For the purposes of this Directive, the following definitions shall apply:

EN

1.

‘infringement of competition law’ means an infringement of Article 101 or 102 of the Treaty or of national competition law within the meaning of paragraph 2;

2.

‘national competition law’ means provisions of national law that predominantly pursue the same objective as Articles 101 and 102 of the Treaty and that are applied to the same case and in parallel to Union competition law pursuant to Article 3(1) of Regulation (EC) No 1/2003;

3.

‘action for damages’ means an action under national law by which an injured party brings a claim for damages before a national court; it may also cover actions by which someone acting on behalf of one or more injured parties brings a claim for damages before a national court, where national law provides for this possibility;

4.

‘claim for damages’ means a claim for compensation of harm caused by an infringement of competition law;

5.

‘injured party’ means anyone who has a claim for damages;

6.

‘national competition authority’ means an authority designated by a Member State pursuant to Article 35 of Regulation (EC) No 1/2003 as being responsible for the application of Articles 101 and 102 of the Treaty;

7.

‘competition authority’ means the Commission or a national competition authority;

8.

‘national court’ or ‘court’ means any court or tribunal of a Member State within the meaning of Article 267 of the Treaty;

9.

‘review court’ means a national court that is empowered to review decisions of a national competition authority, in which context it may also have the power to find an infringement of Article 101 or 102 of the Treaty;

31

EN

410  EU Proposal for infringements of the competition law provisions

EN

10.

‘infringement decision’ means a decision of a competition authority or review court that finds an infringement of competition law;

11.

‘final’ infringement decision means an infringement decision of a competition authority or review court that can no longer be reviewed;

12.

‘cartel’ means an agreement and/or concerted practice between two or more competitors aimed at coordinating their competitive behaviour on the market and/or influencing the relevant parameters of competition, through practices such as the fixing or coordination of purchase or selling prices or other trading conditions, the allocation of production or sales quotas, the sharing of markets and customers, including bid-rigging, restrictions of imports or exports and/or anti-competitive actions against other competitors;

13.

‘leniency programme’ means a programme on the basis of which a participant in a secret cartel, independently of the other undertakings involved in the cartel, cooperates with an investigation of the competition authority, by voluntarily providing presentations of his knowledge of the cartel and his role therein, in return for which the participant receives immunity from any fine to be imposed for the cartel or a reduction of such fine;

14.

‘leniency corporate statement’ means an oral or written presentation voluntarily provided by, or on behalf of, an undertaking to a competition authority, describing the undertaking’s knowledge of a secret cartel and its role therein, which was drawn up specifically for submission to the authority with a view to obtaining immunity or a reduction of fines under a leniency programme concerning the application of Article 101 of the Treaty or the corresponding provision under national law; this does not include documents or information that exist irrespective of the proceedings of a competition authority (‘pre-existing information’);

15.

‘settlement submission’ means a presentation voluntarily provided by, or on behalf of, an undertaking to a competition authority describing the undertaking’s acknowledgement of its participation in an infringement of Article 101 of the Treaty or a corresponding provision under national law and its liability for this infringement, which was drawn up specifically as a formal request for the authority to apply an expedited procedure;

16.

‘overcharge’ means any positive difference between the price actually paid and the price that would have prevailed in the absence of an infringement of competition law;

17.

‘consensual settlement’ means an agreement whereby damages are paid following a consensual dispute resolution.

32

EN

EU Proposal for infringements of the competition law provisions  

411  

CHAPTER II DISCLOSURE OF EVIDENCE Article 5 Disclosure of evidence 1.

Member States shall ensure that, where a claimant has presented reasonably available facts and evidence showing plausible grounds for suspecting that he, or those he represents, has suffered harm caused by the defendant’s infringement of competition law, national courts can order the defendant or a third party to disclose evidence, regardless of whether or not this evidence is also included in the file of a competition authority, subject to the conditions set out in this Chapter. Member States shall ensure that courts are also able to order the claimant or a third party to disclose evidence on request of the defendant. This provision is without prejudice to the rights and obligations of national courts under Council Regulation (EC) No 1206/2001.

2.

3.

EN

Member States shall ensure that national courts order the disclosure of evidence referred to in paragraph 1 where the party requesting disclosure has (a)

shown that evidence in the control of the other party or a third party is relevant in terms of substantiating his claim or defence; and

(b)

specified either pieces of this evidence or categories of this evidence defined as precisely and narrowly as he can on the basis of reasonably available facts.

Member States shall ensure that national courts limit disclosure of evidence to that which is proportionate. In determining whether any disclosure requested by a party is proportionate, national courts shall consider the legitimate interests of all parties and third parties concerned. They shall, in particular, consider: (a)

the likelihood that the alleged infringement of competition law occurred;

(b)

the scope and cost of disclosure, especially for any third parties concerned;

(c)

whether the evidence to be disclosed contains confidential information, especially concerning any third parties, and the arrangements for protecting such confidential information; and

(d)

in cases where the infringement is being or has been investigated by a competition authority, whether the request has been formulated specifically with regard to the nature, object or content of such documents rather than by a non-specific request concerning documents submitted to a competition authority or held in the file of such competition authority.

4.

Member States shall ensure that national courts have at their disposal effective measures to protect confidential information from improper use to the greatest extent possible whilst also ensuring that relevant evidence containing such information is available in the action for damages.

5.

Member States shall take the necessary measures to give full effect to legal privileges and other rights not to be compelled to disclose evidence.

33

EN

412  EU Proposal for infringements of the competition law provisions

6.

Member States shall ensure that, to the extent that their courts have powers to order disclosure without hearing the person from whom disclosure is sought, no penalty for non-compliance with such an order may be imposed until the addressee of such an order has been heard by the court.

7.

Evidence shall include all types of evidence admissible before the national court seised, in particular documents and all other objects containing information, irrespective of the medium on which the information is stored.

8.

Without prejudice to the obligation laid down in paragraph 4 and the limits laid down in Article 6, this Article shall not prevent the Member States from maintaining or introducing rules which would lead to wider disclosure of evidence. Article 6 Limits on the disclosure of evidence from the file of a competition authority

1.

2.

3.

Member States shall ensure that, for the purpose of actions for damages, national courts cannot at any time order a party or a third party to disclose any of the following categories of evidence: (a)

leniency corporate statements; and

(b)

settlement submissions.

Member States shall ensure that, for the purpose of actions for damages, national courts can order the disclosure of the following categories of evidence only after a competition authority has closed its proceedings or taken a decision referred to in Article 5 of Regulation No 1/2003 or in Chapter III of Regulation No 1/2003: (a)

information that was prepared by a natural or legal person specifically for the proceedings of a competition authority;

(b)

information that was drawn up by a competition authority in the course of its proceedings.

Disclosure of evidence in the file of a competition authority that does not fall into any of the categories listed in paragraphs 1 or 2 of this Article may be ordered in actions for damages at any time. Article 7 Limits on the use of evidence obtained solely through access to the file of a competition authority

EN

1.

Member States shall ensure that evidence falling into one of the categories listed in Article 6(1) which is obtained by a natural or legal person solely through access to the file of a competition authority in exercise of his rights of defence under Article 27 of Regulation No 1/2003 or corresponding provisions of national law is not admissible in actions for damages.

2.

Member States shall ensure that evidence falling within one of the categories listed in Article 6, paragraph 2 which is obtained by a natural or legal person solely through access to the file of a competition authority in exercise of his rights of defence under Article 27 of Regulation No 1/2003 or corresponding provisions of national law is not admissible in actions for damages until that competition authority has closed its 34

EN

EU Proposal for infringements of the competition law provisions  

413  

proceedings or taken a decision referred to in Article 5 of Regulation No 1/2003 or in Chapter III of Regulation No 1/2003. 3.

Member States shall ensure that evidence which is obtained by a natural or legal person solely through access to the file of a competition authority in exercise of his rights of defence under Article 27 of Regulation No 1/2003 or corresponding provisions of national law, and which is not inadmissible pursuant to paragraphs 1 or 2 of this Article, can only be used in an action for damages by that person or by the natural or legal person that succeeded in his rights, including the person that acquired his claim. Article 8 Sanctions

1.

Member States shall ensure that national courts can impose sanctions on parties, third parties and their legal representatives in the event of: (a)

failure or refusal to comply with any court’s disclosure order;

(b)

the destruction of relevant evidence, provided that, at the time of destruction: (i)

the destroying party was or had been a party to the proceedings of a competition authority in relation to the conduct underlying the action for damages; or

(ii)

the destroying party knew or should reasonably have known that an action for damages had been brought before the national court and that the evidence was of relevance in substantiating either the claim for damages or a defence against it; or

(iii) the destroying party knew that the evidence was of relevance to pending or prospective actions for damages brought by it or against it;

2.

EN

(c)

failure or refusal to comply with the obligations imposed by a court order protecting confidential information; or

(d)

abuse of the rights relating to disclosure of evidence provided for in this Chapter, and of the evidence and information obtained thereunder.

Member States shall ensure that the sanctions that can be imposed by national courts are effective, proportionate and dissuasive. The sanctions available to national courts shall include, insofar as the behaviour of a party to damages action proceedings is concerned, the possibility to draw adverse inferences, such as presuming the relevant issue to be proven or dismissing claims and defences in whole or in part, and the possibility to order the payment of costs.

35

EN

414  EU Proposal for infringements of the competition law provisions

CHAPTER III EFFECT OF NATIONAL DECISIONS, LIMITATION PERIODS, JOINT AND SEVERAL LIABILITY Article 9 Effect of national decisions Member States shall ensure that, where national courts rule, in actions for damages under Article 101 or 102 of the Treaty or under national competition law, on agreements, decisions or practices which are already the subject of a final infringement decision by a national competition authority or by a review court, those courts cannot take decisions running counter to such finding of an infringement. This obligation is without prejudice to the rights and obligations under Article 267 of the Treaty. Article 10 Limitation periods 1.

Member States shall lay down the rules applicable to limitation periods for bringing actions for damages in accordance with this Article. Those rules shall determine when the limitation period begins to run, the duration of the period and the circumstances under which the period can be interrupted or suspended.

2.

Member States shall ensure that the limitation period shall not begin to run before an injured party knows, or can reasonably be expected to have knowledge of: (i)

the behaviour constituting the infringement;

(ii)

the qualification of such behaviour as an infringement of Union or national competition law;

(iii) the fact that the infringement caused harm to him; and (iv) the identity of the infringer who caused such harm.

EN

3.

Member States shall ensure that the limitation period does not begin to run before the day on which a continuous or repeated infringement ceases.

4.

Member States shall ensure that the limitation period for bringing an action for damages is at least five years.

5.

Member States shall ensure that the limitation period is suspended if a competition authority takes action for the purpose of the investigation or proceedings in respect of an infringement to which the action for damages relates. The suspension shall end at the earliest one year after the infringement decision has become final or the proceedings are otherwise terminated.

36

EN

EU Proposal for infringements of the competition law provisions  

415  

Article 11 Joint and several liability 1.

Member States shall ensure that undertakings which have infringed competition law through joint behaviour are jointly and severally liable for the damage caused by the infringement: each of the infringing undertakings is bound to compensate for the harm in full, and the injured party may require full compensation from any of them until he has been fully compensated.

2.

Member States shall ensure that an undertaking which has been granted immunity from fines by a competition authority under a leniency programme shall be liable to injured parties other than its direct or indirect purchasers or providers only when such injured parties show that they are unable to obtain full compensation from the other undertakings that were involved in the same infringement of competition law.

3.

Member States shall ensure that an infringing undertaking may recover a contribution from any other infringing undertaking, the amount of which shall be determined in the light of their relative responsibility for the harm caused by the infringement. The amount of contribution of an undertaking which has been granted immunity from fines by a competition authority under a leniency programme shall not exceed the amount of the harm it caused to its own direct or indirect purchasers or providers.

4.

Member States shall ensure that, to the extent the infringement caused harm to injured parties other than the direct or indirect purchasers or providers of the infringing undertakings, the amount of contribution of the immunity recipient shall be determined in the light of its relative responsibility for that harm.

CHAPTER IV PASSING-ON OF OVERCHARGES Article 12 Passing-on defence

EN

1.

Member States shall ensure that the defendant in an action for damages can invoke as a defence against a claim for damages the fact that the claimant passed on the whole or part of the overcharge resulting from the infringement. The burden of proving that the overcharge was passed on shall rest with the defendant.

2.

Insofar as the overcharge has been passed on to persons at the next level of the supply chain for whom it is legally impossible to claim compensation for their harm, the defendant shall not be able to invoke the defence referred to in the preceding paragraph.

37

EN

416  EU Proposal for infringements of the competition law provisions

Article 13 Indirect purchasers 1.

Member States shall ensure that, where in an action for damages the existence of a claim for damages or the amount of compensation to be awarded depends on whether — or to what degree — an overcharge was passed on to the claimant, the burden of proving the existence and scope of such pass-on shall rest with the claimant.

2.

In the situation referred to in paragraph 1 of this Article, the indirect purchaser shall be deemed to have proven that a passing-on to him occurred where he has shown that: (a)

the defendant has committed an infringement of competition law;

(b)

the infringement resulted in an overcharge for the direct purchaser of the defendant; and

(c)

he purchased the goods or services that were the subject of the infringement, or purchased goods or services derived from or containing the goods or services that were the subject of the infringement.

Member States shall ensure that the court has the power to estimate which share of that overcharge was passed on. This paragraph shall be without prejudice to the infringer's right to show that the overcharge was not, or not entirely, passed on to the indirect purchaser. Article 14 Loss of profits and infringement at supply level 1.

The rules laid down in this Chapter shall be without prejudice to the right of an injured party to claim compensation for loss of profits.

2.

Member States shall ensure that the rules laid down in this Chapter apply accordingly where the infringement of competition law relates to supply to the infringing undertaking. Article 15 Actions for damages by claimants from different levels in the supply chain

1.

2.

EN

Member States shall ensure that, in assessing whether the burden of proof resulting from the application of Article 13 is satisfied, national courts seized of an action for damages take due account of (a)

actions for damages that are related to the same infringement of competition law, but are brought by claimants from other levels in the supply chain; or

(b)

judgments resulting from such actions.

This Article shall be without prejudice to the rights and obligations of national courts under Article 30 of Regulation (EU) No 1215/2012.

38

EN

EU Proposal for infringements of the competition law provisions  

417  

CHAPTER V QUANTIFICATION OF HARM Article 16 Quantification of harm 1.

Member States shall ensure that, in the case of a cartel infringement, it shall be presumed that the infringement caused harm. The infringing undertaking shall have the right to rebut this presumption.

2.

Member States shall pleading required for injured party’s right Member States shall amount of harm.

ensure that the burden and the level of proof and of factthe quantification of harm does not render the exercise of the to damages practically impossible or excessively difficult. provide that the court be granted the power to estimate the

CHAPTER VI CONSENSUAL DISPUTE RESOLUTION Article 17 Suspensive effect of consensual dispute resolution 1.

Member States shall ensure that the limitation period for bringing an action for damages is suspended for the duration of the consensual dispute resolution process. The suspension of the limitation period shall apply only with regard to those parties that are or were involved in the consensual dispute resolution.

2.

Member States shall ensure that national courts seized of an action for damages may suspend proceedings where the parties to those proceedings are involved in consensual dispute resolution concerning the claim covered by that action for damages. Article 18 Effect of consensual settlements on subsequent actions for damages

EN

1.

Member States shall ensure that, following a consensual settlement, the claim of the settling injured party is reduced by the settling co-infringer’s share of the harm that the infringement inflicted upon the injured party. Non-settling co-infringers cannot recover contribution from the settling co-infringer for the remaining claim. Only when the non-settling co-infringers are not able to pay the damages that correspond to the remaining claim can the settling co-infringer be held to pay damages to the settling injured party.

2.

When determining the contribution of each co-infringer, national courts shall take due account of any prior consensual settlement involving the relevant co-infringer.

39

EN

418  EU Proposal for infringements of the competition law provisions

CHAPTERVII FINAL PROVISIONS Article 19 Review The Commission shall review this Directive and report to the European Parliament and the Council by [...] at the latest [to be calculated as 5 years after the date set as the deadline for transposition of this Directive.] Article 20 Transposition 1.

Member States shall bring into force the laws, regulations and administrative provisions necessary to comply with this Directive by [to be calculated as 2 years after the date of adoption of this Directive] at the latest. They shall forthwith communicate to the Commission the text of those provisions. When Member States adopt those provisions, they shall contain a reference to this Directive or be accompanied by such a reference on the occasion of their official publication. Member States shall determine how such reference is to be made.

2.

Member States shall communicate to the Commission the text of the main provisions of national law which they adopt in the field covered by this Directive. Article 21 Entry into force

This Directive shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union. Article 22 Addressees This Directive is addressed to the Member States. Done at Strasbourg,

For the European Parliament

EN

For the Council

40

EN

EU Proposal for infringements of the competition law provisions  

419  

LEGISLATIVE FINANCIAL STATEMENT This proposal has no impact on the EU budget.

EN

41

EN

Annex II Commission Staff Working Document Impact Assessment Report: Damages actions for breach of the EU antitrust rules (2013)

EUROPEAN COMMISSION

Strasbourg, 11.6.2013 SWD(2013) 203 final

COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT REPORT Damages actions for breach of the EU antitrust rules Accompanying the proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL 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 (Text with EEA relevance) {COM(2013) 404 final} {SWD(2013) 204 final}

EN

EN

424  Commission Staff Working Document Impact Assessment Report TABLE OF CONTENTS

EN

1.

The subject-matter of the impact assessment............................................................... 6

2.

Procedural issues and consultation of interested parties .............................................. 7

2.1.

Background .................................................................................................................. 7

2.2.

Public consultation on the White Paper ....................................................................... 8

2.2.1.

Civil society ................................................................................................................. 8

2.2.2.

Institutional stakeholders ............................................................................................. 9

2.3.

Public consultation on the quantification of damages.................................................. 9

2.4.

Public consultation on a coherent approach to collective redress.............................. 10

2.5.

Inter-service consultation ........................................................................................... 11

2.6.

The Impact Assessment Board................................................................................... 12

3.

Problem to be addressed............................................................................................. 14

3.1.

The interaction between public and private enforcement remains unclear under the current legal framework ............................................................................................. 14

3.2.

The current legal framework for damages actions in cases of competition law infringements is ineffective as victims experience major difficulties to obtain compensation.............................................................................................................. 17

3.3.

The current legal framework could endanger the proper functioning of the internal market......................................................................................................................... 21

3.4.

Scope and costs of the uncertainty relating to the interaction between public and private enforcement and the ineffectiveness of the legal framework for actions for damages...................................................................................................................... 23

3.4.1.

Scope and cost: ill-regulated interface between public and private enforcement ...... 23

3.4.2.

Scope and cost: ineffective legal framework for compensation ................................ 24

4.

Objectives................................................................................................................... 27

4.1.

General objectives...................................................................................................... 27

4.2.

Specific objectives ..................................................................................................... 27

5.

Contents of the current Antitrust Damages Initiative ................................................ 32

5.1.

The 2008 IAWP and the choice of the 'Preferred Option'.......................................... 32

5.2.

The current Antitrust Damages Initiative - scope of the assessment ......................... 32

5.3.

Description of the Policy Options assessed in the current initiative.......................... 33

5.3.1.

Option 1: the baseline scenario of no EU action in the field of antitrust damages actions ........................................................................................................................ 34

5.3.2.

Option 2: adopting the White Paper's suggested measures through a legislative instrument................................................................................................................... 34

5.3.3.

Option 3: legislative proposal based on the White Paper and additional safeguards. 35

5.3.4.

Option 4: non-regulatory measures............................................................................ 39

2

EN

Damages actions for breach of the EU antitrust rules (2013) 

425  

5.4.

Provision of non-binding assistance for quantification of damages .......................... 39

6.

Impact analysis........................................................................................................... 41

6.1.

Assessment criteria..................................................................................................... 41

6.1.1.

Benefits ...................................................................................................................... 42

6.1.2.

Costs........................................................................................................................... 44

6.2.

Identifying and assessing the impact of each option.................................................. 45

6.2.1.

Option 1...................................................................................................................... 46

6.2.2.

Option 2...................................................................................................................... 48

6.2.3.

Option 3...................................................................................................................... 51

6.2.4.

Option 4...................................................................................................................... 53

7.

Comparing the options ............................................................................................... 56

7.1.

Comparing the policy options and assessment of the preferred option ..................... 56

7.1.1.

Summary comparison of the options and identification of the preferred option ....... 56

7.1.2.

Litigation costs of the preferred option...................................................................... 63

7.1.3.

Transposition costs for the Member States ................................................................ 66

7.2.

Choice of legal instrument ......................................................................................... 68

7.2.1.

The options................................................................................................................. 68

7.2.2.

Regulation .................................................................................................................. 68

7.2.3.

Regulation and Directive............................................................................................ 69

7.2.4.

Directive..................................................................................................................... 69

7.3.

Proportionality and EU added value of the Preferred Option .................................... 69

7.3.1.

Subsidiarity: European added value........................................................................... 70

7.3.2.

Proportionality............................................................................................................ 71

8.

Monitoring and evaluation ......................................................................................... 72

ANNEX 1 ................................................................................................................................. 74 1.

Overview .................................................................................................................... 74

2.

Functioning of the EU Leniency Programme ............................................................ 74

3.

Leniency programmes in the Member States............................................................. 75

ANNEX 2 ................................................................................................................................. 76 1.

Facts and background................................................................................................. 76

2.

The preliminary question ........................................................................................... 76

3.

The judgment of the Court ......................................................................................... 76

ANNEX 3 ................................................................................................................................. 78 1.

EN

AUSTRIA .................................................................................................................. 78

3

EN

426  Commission Staff Working Document Impact Assessment Report 2.

GERMANY................................................................................................................ 78

3.

UNITED KINGDOM................................................................................................. 79

ANNEX 4 ................................................................................................................................. 80 1.

Introduction ................................................................................................................ 80

2.

The IAWP's Policy Option 1...................................................................................... 80

3.

The IAWP's Policy Option 2...................................................................................... 81

4.

The IAWP's Policy Option 3...................................................................................... 81

ANNEX 5 ................................................................................................................................. 83 ANNEX 6 ................................................................................................................................. 84 ANNEX 7 ................................................................................................................................. 86 ANNEX 8 ................................................................................................................................. 90

EN

1.

Main trends among respondents to the White Paper.................................................. 90

2.

Specific suggestions on individual sections of the White Paper................................ 90

4

EN

Damages actions for breach of the EU antitrust rules (2013) 

427  

COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT REPORT Damages actions for breach of the EU antitrust rules Accompanying the proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL 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 (Text with EEA relevance) 1.

THE SUBJECT-MATTER OF THE IMPACT ASSESSMENT

1.

Since 2001 the Court of Justice ("the Court") repeatedly stated that, as a matter of EU law, any individual must be able to claim compensation for the harm suffered as a result of an infringement of the EU competition rules.1 More than 10 years later, most victims of a competition law infringement are still not able to effectively exercise that EU right to compensation. This is largely due to a lack of appropriate national rules governing actions for damages. Even where such rules exist, they are so different from Member State to Member State that it results in an uneven playing field for both infringers and victims of the illegal conduct. More recently, a new issue has arisen, showing that the EU right to compensation can sometimes be at odds with the effectiveness of public enforcement of the EU competition rules. For example there exists legal uncertainty as to whether information that a competition authority had obtained in the course of its enforcement of the EU competition rules, is disclosable in damages actions before national courts. Such disclosure could be particularly detrimental for the effectiveness of the leniency programmes and hence for the effectiveness of the fight against secret cartels.2

2.

To remedy these two gaps in the enforcement of the EU competition rules, the current Antitrust Damages Initiative has two primary objectives:

1 2

EN

(i)

to maintain effective public enforcement of the competition rules by regulating some key aspects of the interaction between public enforcement of competition law by the Commission and national competition authorities and private enforcement of competition law through actions for damages before national courts; and

(ii)

to ensure an effective exercise of the EU right to compensation.

Case C-453/99, Courage and Crehan, [2001] ECR I-6297, and joined cases C-295/04 to C-298/04, Manfredi, [2006] ECR I-6619. Case C-360/09, Pfleiderer AG v Bundeskartellamt, [2011] ECR I-5161.

5

EN

428  Commission Staff Working Document Impact Assessment Report The result of this initiative should be an effective system of public and private enforcement of competition law that contributes to fostering growth and innovation throughout the EU. 2.

PROCEDURAL ISSUES AND CONSULTATION OF INTERESTED PARTIES

2.1.

Background

3.

On 19 December 2005 the Commission adopted a Green Paper on damages actions for breach of the EC antitrust rules, identifying the main obstacles to effective compensation.3 The public consultation met a wide response4 and also institutional stakeholders issued their opinion. The European Parliament ('the Parliament') adopted a Report and a Resolution in 2007,5 in which it called on the Commission to prepare a White Paper with detailed proposals to facilitate actions for damages.

4.

The Commission met a group of experts from the Member States on two occasions in autumn 2007, in preparation of the White Paper. Experts represented the Ministries of Justice, Ministries of Economic Affairs and national competition authorities ('NCAs'). Representatives of the EFTA and its members were present. The Commission also met a delegation of judges of national supreme courts, courts of appeal, courts of first instance and specialist competition tribunals from 12 Member States6 to discuss specific issues related to antitrust damages actions.

5.

Commission staff participated in a large number of events (conferences, expert panels, etc.) to discuss more effective EU antitrust damages actions and their implications. DG Competition, in particular, repeatedly met a wide range of stakeholders and experts, notably consumer associations, business representatives, lawyers and academics.

6.

On 2 April 2008 the Commission adopted a White Paper on damages actions for breach of the EC antitrust rules7, that put forward suggestions for specific measures that would ensure the effective exercise of the right to compensation of antitrust harm. The White Paper was accompanied by an Impact Assessment ('IAWP').8

2.2.

Public consultation on the White Paper

2.2.1.

Civil society

7.

The White Paper was the object of an intense debate. A large number of public authorities and stakeholders from almost all Member States submitted comments.9

3

4 5 6

7 8 9

EN

COM(2005)672. The Green Paper was prepared by a 2004 study on the conditions under which private parties can bring actions for antitrust damages ('the Comparative Study'). Page numbers quoted in this report refer to the electronic version of the Comparative Study: http://ec.europa.eu/comm/competition/antitrust/others/actions_for_damages/study.html. The submissions on the Green Paper are available at: http://ec.europa.eu/comm/competition/antitrust/actionsdamages/green_paper_comments.html. 2006/2207(INI). The Resolution of the European Parliament is available at http://www.europarl.europa.eu/oeil/file.jsp?id=5378362. Invitations were issued via the Association of European Competition Law Judges, the European Network of Councils for the Judiciary and the Network of the Presidents of the Supreme Judicial Courts of the EU. COM(2008)165 and the accompanying Commission Staff Working Paper (SEC(2008)404) ('the 2008 Staff Working Paper'). SEC(2008)405. See also the Executive Summary of the Impact Assessment (SEC(2008)406). The more than 170 submissions on the White Paper are available at http://ec.europa.eu/competition/antitrust/actionsdamages/white_paper_comments.html.

6

EN

Damages actions for breach of the EU antitrust rules (2013) 

429  

Very few respondents questioned the idea underpinning the White Paper that in most cases victims of competition law infringements in practice do not obtain the compensation they are entitled to under EU law. Most agreed that something needs to be done to remedy this situation. Many respondents explicitly welcomed the approach of the White Paper to pursue, as a primary policy objective, the aim of effective compensation of victims (rather than punishment or deterrence of infringers) and to seek solutions that are balanced and rooted in the European legal traditions. Divergent opinions emerged as to the need for an initiative at EU level, the consequences in terms of increased litigation, the interaction with public enforcement, and the means by which the goal of more effective compensation can be most appropriately achieved.

EN

8.

Different groups of stakeholders belonging to civil society expressed different opinions on the White Paper and its suggested measures. Below is a summary of the opinions of the three groups of stakeholders that submitted most replies.

(a)

Businesses and Business Associations

9.

Among civil society, business associations were generally negative both on the need for a Commission initiative and on the substance of the proposals. However, proposed measures which are favourable to the position of defendants, such as the passing-on defence and protection of corporate statements from disclosure in actions for damages, were received positively by the business community.

10.

There are two groups of exceptions to the generally negative opinion of the business associations. The first exception is that of business associations representing only small and medium sized enterprises. These associations generally welcomed the focus on SMEs, which often suffer harm from anticompetitive behaviour that they are not able to recover, and expressed an overall positive view on the White Paper.

11.

The second exception is given by a number of companies that had allegedly been victims of anticompetitive conduct, such as customers of cartelists in the beer, the paper and the elevator cartel. The submissions of these companies are broadly supportive of the White Paper’s proposals, and in some cases even encourage the Commission to advance broader proposals (for instance by loosening the conditions to obtain a disclosure order).

(b)

Consumer Associations

12.

Consumer associations fully endorsed the White Paper’s proposals. They supported the White Paper’s objectives and suggestions and supported the idea of a legislative proposal from the Commission. In some cases, they even encouraged the Commission to do more in order to set up a truly effective framework for antitrust damages actions.

13.

Consumer associations identified the following issues as main obstacles to an effective right to compensation for victims of antitrust violations: (i) lack of collective redress mechanisms and (ii) difficulties in obtaining disclosure orders. Consumer associations argued that the preservation of public enforcement (and leniency programmes in particular) does not justify a restriction on the rights of the victims and were against any form of limitation of liability for immunity recipients. They largely contested the need for protection of corporate statements, while claiming broader rules on access to the files of competition authorities.

(c)

Lawyers, Law firms and Bar associations

7

EN

430  Commission Staff Working Document Impact Assessment Report 14.

The submissions received from law firms and lawyers’ associations presented a broad range of views, often influenced by the legal culture they come from and by whether they generally represent claimants or defendants. As a general trend, there was wide agreement on the need to establish rules on the protection of corporate statements. Respondents also urged the Commission to exclude from disclosure documents covered by legal privilege. Apart from these issues, submissions from the legal community were very heterogeneous.

2.2.2.

Institutional stakeholders

15.

The Parliament10 welcomed the White Paper and stressed that EU competition rules and their enforcement require that victims of breaches of those rules should be able to claim compensation for the damage suffered. It also stressed that individual consumers and small businesses are often deterred from bringing individual actions. In particular the Parliament stressed the "need for the Commission to propose legislation, without watering it down unnecessarily, to facilitate individual and classaction claims for effective compensation for damages resulting from breaches of EU antitrust law."11

16.

The EESC12 also welcomed the White Paper and stressed the need to promote people's access to effective judicial protection as a fundamental right laid down in the European Charter of Fundamental Rights ('the Charter'). It considered that a legal framework was necessary and provided detailed comments on different aspects.

2.3.

Public consultation on the quantification of damages

17.

One of the suggestions made in the White Paper was to "provide pragmatic, nonbinding assistance in the difficult task of quantifying damages in antitrust cases, both for the benefit of national courts and the parties."13 In 2009, an external study on the quantification of damages was prepared for the European Commission.14 The results of the Quantification Study were taken into account in the Draft Guidance Paper on the quantification of damages which the Commission submitted to public consultation in 2011.15 This draft paper set out insights into a range of methods and techniques used to quantify harm in damages actions and explains strengths and weaknesses of these methods.

18.

Institutional and other stakeholders generally welcomed the idea of issuing nonbinding guidance on quantifying harm caused by antitrust infringements.16 The

10

11

12 13 14

15

16

EN

European Parliament resolution of 26 March 2009 on the White Paper on damages actions for breach of the EC antitrust rules (2008/2154(INI)): http://www.europarl.europa.eu/sides/getDoc.do?pubRef=//EP//TEXT+TA+P6-TA-2009-0187+0+DOC+XML+V0//EN. See paragraph 15 of the European Parliament resolution of 20 January 2011 on the Report on Competition Policy 2009 (http://www.europarl.europa.eu/sides/getDoc.do?type=TA&reference=P7TA-2011-0023&language=EN&ring=A7-2010-0374). Opinion of the EESC on the White Paper on damages actions for breach of the EC antitrust rules, OJ C 228 22.9.2009, p. 40. 2008 Staff Working Paper, paragraph 199. 'Quantifying antitrust damages - Towards non-binding guidance for courts' ('the Quantification Study'). Page numbers quoted in this report refer to the electronic version of the Quantification Study: http://ec.europa.eu/competition/antitrust/actionsdamages/quantification_study.pdf. The draft Guidance Paper on quantifying harm in actions for damages based on breaches of the EU antitrust rules (June 2011) can be found at http://ec.europa.eu/competition/consultations/2011_actions_damages/draft_guidance_paper_en.pdf. The 37 written submissions can be found at http://ec.europa.eu/competition/consultations/2011_actions_damages/index_en.html.

8

EN

Damages actions for breach of the EU antitrust rules (2013) 

431  

Commission organised two workshops on the topic with a number of renowned economists17 and sought the advice of specialised judges from the Member States. 2.4.

Public consultation on a coherent approach to collective redress

19.

In response to the Parliament resolution on the White Paper calling for an integrated approach to collective redress, the Commission held in 2011 a public consultation on a coherent approach to collective redress in different areas of EU law.18 The purpose of this consultation was, inter alia, to identify common legal principles on collective redress in all areas of EU law and to examine how such common principles could fit into the EU legal system and into the legal orders of the Member States. The consultation also explored in which areas different forms of collective redress (injunctive and/or compensatory) could improve the enforcement of EU legislation or the protection of victims' rights.

20.

The consultation attracted more than 300 replies from a wide range of stakeholders and over 18,000 replies from citizens supporting the position of consumer organisations. 15 Member State governments replied, of which 1019 favoured a binding EU instrument on collective redress, while 520 preferred a non-binding approach. 621 Member States supported policy-specific legislation at EU level, explicitly mentioning competition policy, while 422 preferred horizontal initiatives. While consumers were strongly in favour of binding EU measures on collective redress, the majority of businesses were opposed.

21.

Almost all stakeholders agreed with the following basic parameters of a collective redress system: it should be capable of (i) effectively resolving a multitude of individual claims which raise the same or common issues and relate to a single infringement of EU law; (ii) delivering legally certain and fair outcomes within a reasonable timeframe, while respecting the rights of all parties involved; and (iii) providing for safeguards against abusive litigation and avoiding any economic incentives to bring abusive claims.

22.

The public consutation showed that there is a consensus across all stakeholder groups that private collective redress and enforcement by public bodies are two different instruments that pursue different objectives. Business tends to put greater emphasis on the role of public enforcement. Other stakeholder groups are generally of the view that both instruments are in principle equally important and that they should be independent and complementary mechanisms.

23.

Most stakeholders agree that in policy fields where public enforcement plays a major role – particularly in competition – specific rules are required to ensure a smooth interplay between public enforcement and private collective redress. Many replies mention rules on the binding effect of infringement decisions by national competition authorities for follow-on collective actions, limitation periods for bringing follow-on

17 18 19 20 21 22

EN

The workshops were held on 26 January 2010 and on 27 September 2011. The contributions can be found at http://ec.europa.eu/competition/antitrust/actionsdamages/economist_workshop.html. For further information on the public consultation "Towards a coherent European approach to collective redress", see http://ec.europa.eu/competition/consultations/2011_collective_redress/index_en.html. Bulgaria, Denmark, Greece, Italy, Latvia, the Netherlands, Poland, Portugal, Sweden and the UK. Austria, the Czech Republic, France, Germany and Hungary. Bulgaria, Greece, Poland, Portugal, Sweden and the UK. Austria, the Czech Republic, Denmark and the Netherlands.

9

EN

432  Commission Staff Working Document Impact Assessment Report actions and protection of the effectiveness of public enforcement (specifically leniency programmes). 24.

The Parliament resolution on collective redress23 recognizes the importance of collective redress in ensuring effective compensation for victims of EU law infringements. The resolution favours a separate horizontal EU framework including a common set of principles over a sector-specific approach towards collective redress. Nevertheless, the resolution acknowledges that there can be a need for certain competition law specific provisions on collective redress, which could be laid down in a separate chapter of a horizontal instrument or in a separate legal instrument.

2.5.

Inter-service consultation

25.

The Directorate-General for Competition is the lead service on the current Antitrust Damages Initiative, with the involvement of the Secretariat-General, the Legal Service, DG Economic and Financial Affairs, DG Enterprise and Industry, DG Internal Market and Services, DG Health and Consumers and DG Justice. An Impact Assessment Inter-Service Group was set up on 26 September 2008. It met two times in 2008 to discuss earlier drafts of the Impact Assessment report and met once in 2012 to discuss the present draft, which takes account of the additional public consultations held in 2011, of the views expressed by the Parliament since 2008 and of recent developments at the European Courts.

2.6.

The Impact Assessment Board

26.

A draft of the present Impact Assessment was submitted to the Impact Assessment Board on 21 November 2012. Responding to the resulting recommendation in the Board's first opinion of 20 December 2012, a revised draft was submitted in January 2013. The Board issued a second, positive opinion on the resubmitted report on 28 February 2013, with further suggestions for improvement. The Board expressed the following main recommendations, which were considered in the present report: (1)

Better substantiate the problem definition;

(2)

Report in more detail the stakeholder response to the public consultation on the White Paper;

(3)

Clarify the objectives of the initiative;

(4)

Improve the presentation of the options;

(5)

Clarify the differences between Options 2 and 3;

(6)

Better present the analysis of impacts and the comparison of options;

(7)

Further define the monitoring criteria.

For (1), the first part of the Report has been further elaborated to provide more concrete evidence of the problems targeted. Quantitative data from an external impact study on compensation foregone by victims of a competition law infringement have been quoted. The interaction between public and private enforcement has been further clarified, and its background is explained in three 23

European Parliament resolution of 2 February 2012 on 'Towards a Coherent European Approach to Collective Redress' (2011/2089(INI)): http://www.europarl.europa.eu/sides/getDoc.do?type=TA&language=EN&reference=P7-TA-2012-21

EN

10

EN

Damages actions for breach of the EU antitrust rules (2013) 

433  

annexes on the functioning of the leniency programmes, the Pfleiderer case-law and initiatives currently considered at Member State's level. The problem has further been specified by increased analysis of the baseline scenario of no EU action, also through reference to Member States' legislation and the inefficiencies of the current legal framework, as well as the current lack of comparably efficient alternatives to leniency programmes in the detection of secret cartels. For (2), extensive reference has been made to stakeholders opinions throughout the document. A summary overview on the public consultation on the 2008 White Paper has also been added (Annex 8). For (3), the formulation of the objectives has been made more clear. All the objectives and sub-objectives are currently discussed with the same structure, including an introduction where detailed reference is made to the relevant sections of part 3, to allow the reader to clearly link the targeted objectives to the problems addressed by the initiative. The objective of fostering full compensation while avoiding over-compensation has been clearly defined. For (4), the assessed Options have been re-ordered to start from the base-line scenario of zero EU-Action (Option 1), and then the different forms of action proposed. As regards collective redress, Option 2 builds on the 2008 White Paper and assesses two different collective redress mechanisms (opt-in group actions and representative actions), whereas Option 3 and Option 4 are based on the assumption that collective redress is dealt at EU level in a horizontal framework that is yet to be determined. Therefore, the current impact assessment does not address the issue of collective redress for these two options. The assessment of the role of collective redress in the field of competition law enforcement and its relationship with a possible horizontal initiative on collective redress is separately discussed as part of the assessment of the Policy Options. For (5), the differences between Option 2 and 3 have been spelled out more clearly. In particular, the Report explains why certain safeguards or additional measures compared to the White Paper options have been assessed as a bundle in Option 3. For (6), the analysis of impacts has been explained in more detail. Options are also clearly and independently assessed against the base-line scenario. The pitfalls of the options already excluded in the IAWP, which have not been reconsidered in the current exercise, are recalled in an Annex, which details the risks of a number of solutions chosen in other jurisdictions and not fully in line with European legal traditions. In order to improve readability and comparison of the options, the scoring system has also been simplified. For (7) the chapter on monitoring and evaluation has been enriched with reference to more specific criteria to assess the progress achieved through the initiative. In addition to the above, a glossary of technical terms used in the text is added as Annex 7 to make the report more accessible to non-specialist readers.

EN

11

EN

434  Commission Staff Working Document Impact Assessment Report 3.

PROBLEM TO BE ADDRESSED

27.

The Antitrust Damages Initiative adresses two key issues, namely (a) the interaction between public and private enforcement of the EU competition rules and (b) the difficulties for victims of competition law infringements to obtain compensation. In case the baseline scenario of no action at the EU level is followed (see further the description of the baseline scenario or option 1 in section 5.3 below), the problems described in sections 3.1 until 3.3 will continue to exist or deteriorate, and the costs described in section 3.4 will be incurred.

3.1.

The interaction between public and private enforcement remains unclear under the current legal framework

28.

The EU competition rules are a matter of public policy. They are primarily enforced by the Commission and NCAs, which have a number of investigative and enforcement powers, including the power to impose fines on undertakings for the infringement of these rules.24 This type of enforcement, which is exercised by competition authorities in the public interest, is generally referred to as "public enforcement". In addition, the EU antitrust rules – Articles 101 and 102 TFEU – have direct effect, which means that they create rights for individuals that can be enforced before national courts. This type of enforcement is generally referred to as 'private enforcement'. Among these rights is the EU right for victims of an infringement of EU competition law to be compensated for the harm they have suffered. The means by which the right to compensation is put into practice are civil damages actions brought before national courts. Given the subject-matter of this initiative, the notion of private enforcement in this report is used in the narrower sense of antitrust damages actions.

29.

The two kinds of procedures – private enforcement actions under national civil law and public enforcement by competition authorities – are complementary tools serving the objective of an effective enforcement of the EU competition rules. There is a consensus among all stakeholder groups that responded to the public consultations on the White Paper and on collective redress (consumers, business and business representatives, legal experts and public authorities such as Member States and competition authorities) that public and private enforcement are two different instruments that pursue different objectives. All stakeholder groups apart from business are generally of the view that both instruments are in principle equally important and must hence be independent and complementary mechanisms.

30.

Given that antitrust damages actions are often triggered by a competition authority's investigation and are brought either while an investigation is pending or, more typically, after an infringement decision had been adopted (in the latter case referred to as 'follow-on' actions), the interaction between public and private enforcement can be significant. A smooth interplay is vital to ensure maximum effectiveness of both tools. This can be best achieved by regulating certain key aspects of the interplay, such as access to information held by competition authorities (the disclosure of certain documents from the competition authorities's files could negatively affect the effectiveness of public enforcement), binding effect of infringement decisions (to

24

EN

The Commission has the powers foreseen in Regulation 1/2003 for the application of Articles 101 and 102 TFEU. The NCAs have the powers foreseen in national law and may adopt the decisions listed in Article 5 of Regulation 1/2003.

12

EN

Damages actions for breach of the EU antitrust rules (2013) 

435  

avoid re-litigation of the finding of an infringement) or limitation periods (to ensure that they do not expire while potential follow-on claimants are waiting for an infringement decision). The need to provide for rules ensuring a smooth interplay between the public and private enforcement of competition law was confirmed by most stakeholders responding to the public consultation on collective redress.25 31.

Stakeholders generally regard the need to regulate the public/private interface as particularly acute in the case of documents linked to leniency programmes.26 In the public consultations on the White Paper and on collective redress, business representatives, legal experts and public authorities warned against undue disclosure of leniency related documents for the purpose of antitrust damages actions. They held in particular that the protection of corporate statements is an essential condition for the success of leniency programmes. Only a few submissions (mainly by consumer associations) contested the need to protect leniency corporate statements.

32.

In June 2011, the Court held in Pfleiderer that – in the absence of EU law – it is for a national court to determine on a case-by-case basis and according to national law the conditions under which disclosure of leniency-related information to victims of a competition law infringement must be permitted or refused.27 Therefore, leniency applicants cannot know in advance whether documents submitted to competition authorities in the context of a leniency application might be disclosed to claimants in antitrust damages actions and if so, what categories of documents would be disclosable. Although one cannot bring direct evidence that the Pfleiderer judgment has had or will have a negative impact on the number of leniency applications (as it is impossible to know how many leniency applications would have been received without the judgment), the current legal uncertainty could affect the willingness of cartel participants to cooperate with the Commission and NCAs under the leniency programmes and thus negatively affect the public enforcement of competition law.28 More generally, the lack of legal certainty as to the (non-)disclosability in antitrust damages actions of leniency-related documents and other information from competition authorities' files is detrimental for all involved parties, including claimants and the competition authorities.

33.

It should be noted that in Pfleiderer the Court reached its conclusion on the case-bycase application of divergent national laws "in the absence of binding regulation under European Union law on the subject", i.e. in the absence of any "common rules on leniency or common rules on the right of access to documents relating to a leniency procedure".29 A common EU standard providing for an appropriate protection of leniency documents would remove both the uncertainty for potential leniency applicants and the diversity of national rules on the subject. NCAs agree to this solution and they have therefore been urging the Commission to introduce

The Pfleiderer judgment

25 26 27 28

29

EN

See para 0 above. See Annex 1 for a more detailed description of the functioning of the Commission’s leniency programme. See Annex 2 for a more detailed description of the Pfleiderer case. Case C-360/09, Pfleiderer AG v Bundeskartellamt, [2011] ECR I-5161, 20-23: "the view can reasonably be taken that [cartel participants] would be deterred from taking the opportunity offered by such leniency programmes, particularly when, pursuant to Articles 11 and 12 of Regulation 1/2003, the Commission and the national competition authorities might exchange information which that person has voluntarily provided". Ibidem

13

EN

436  Commission Staff Working Document Impact Assessment Report without delay an EU rule that would protect their leniency programmes. The Heads of the European Competition Authorities have repeated this message in a recent resolution.30 Also European businesses have called for such a rule to ensure legal certainty in this regard. Legal uncertainty and risks for public enforcement 34.

Under the baseline scenario of no EU action, the legal uncertainty and the risk of negative consequences on the public enforcement of EU competition law would continue to exist. This is due to the persisting risk of diverging or inconsistent court practice between different Member States or even within the same Member State with regard to disclosability of leniency related documents from the file of competition authorities. Such divergence in national judgments is already visible to date: whereas in Germany the first instance court protected in Pfleiderer all leniency documents from disclosure31, the Düsseldorf Appeal Court32, in a different case, was not ready to protect the information contained in leniency documents in so far as that information was referred to in the infringement decision. In the UK, the High Court in the National Grid case found that partial disclosure of certain documents (such as a reply to the Statement of Objections and replies to Requests for Information) is justified and some parts of the confidential verions of the Commission decision are to be disclosed, whereas documents specifically prepared for the purpose of the Commission's leniency programme should not be disclosed.33 If more national courts are required to make the case-by-case assessment as described in Pfleiderer, the likelihood of diverging rulings on the disclosability of documents from the file of a competition authoritity increases. As stated in paragraph 33 above, a common EU standard providing for an appropriate protection of such documents would remove both the uncertainty for undertakings potentially involved in proceedings before competition authorities and the diversity of judgments by national courts on the subject.

35.

In the absence of adequate protection of leniency programmes, the negative consequences arising from the uncertain legal framework outlined above cannot be offset by efficient alternatives. In particular, it would not be possible to offset the negative effects of impaired leniency programmes by increased ex-officio investigations into suspected infringements by the Commission or NCAs. Such an attempt would not only be more costly both for public enforcers and undertakings alike, but would not enable public enforcers to uncover comparably useful evidence with a view at proving infringements. On the one hand, leniency programmes allow the Commission and NCAs to pursue a targeted enforcement on conducts where the likelihood to find an infringement is much higher, and free resources for the pursuit of ex officio cases while maintaining an adequate degree of deterrence. On the other hand, divesting resources from leniency cases in order to pursue more ex officio cases would impose a significant administrative burden on businesses, as there would be a higher chance of being subject to an investigation while no infringement is later

30

31 32 33

EN

Resolution of the Meeting of the Heads of the European Competition Authorities of 23 May 2012, Protection of leniency material in the context of civil damages actions, available at http://ec.europa.eu/competition/ecn/leniency_material_protection_en.pdf. Amtsgericht Bonn (Local Court Bonn), decision of 18-January-2012, case No 51 Gs 53/09 (Pfleiderer). Oberlandesgericht Düsseldorf (Düsseldorf Appeal Court), decision of 22 August 2012, case No B-4 Kart 5/11 (OWi) (roasted coffee). High Court of Justice (UK first instance court), judgment of 04 April 2012, case No HC08C03243 (National Grid).

14

EN

Damages actions for breach of the EU antitrust rules (2013) 

437  

services, avoiding those Member States where the right to compensation is most effectively enforced. 40.

Furthermore, the Comparative Study on the conditions of damages claims has shown that the differences between national legal systems cause legal uncertainty at several levels. For a number of important legal issues, e.g. the availability of the 'passing-on defence',36 existing national law is unclear about which rule applies in the specific context of antitrust damages cases. Additional legal uncertainty stems from the significant differences between the procedural and substantive rules governing actions for damages in the different Member States (e.g. with regard to access to evidence, limitation periods and the binding effect of NCA decisions). These are described in the 2004 Comparative Study on the conditions of damages claims and in the 2008 Commission White Paper on antitrust damages actions and its accompanying Impact Assessment.37

41.

The marked differences (described as "astonishing diversity" in the 2004 Comparative Study)38 continue to exist in relation to many of the topics to be adressed in this policy initiative:

36 37 38

EN



As regards the binding effect of an NCA's finding of an EU competition law infringement on national courts in a subsequent antitrust damages action, legislation in only one Member State (Germany) recognises the binding effect of final NCA decisions of both the domestic and other European NCAs. In 10 other Member States only the final decision of the domestic NCA is binding on national courts. In the remaining 16 Member States, NCA decisions have no binding effect. In those Member States their evidential value ranges from a rebuttable evidentiary presumption of an infringement to normal evidential value or even to being regarded as just a view on the facts and the law.



As regards the limitation period for bringing an antitrust damages action, some Member States provide for specific limitation periods allowing for follow-on cases. In the vast majority of Member States, however, there are no specific rules on limitation periods for follow-on cases. In these Member States, expiry of the ordinary limitation period before a competition authority renders an infringement decision can form an important obstacle preventing follow-on action from being instituted. The specific limitation periods in some Member States are either based on a suspension of the ordinary limitation period during an investigation by the authorities, in combination with an ensured time period to bring an action after the suspension finishes or a specific limitation period for follow-on cases. The length of these limitation periods differs from 6 months to 5 years.



As to the passing-on defence, it should be noted that in most Member States there is no legislative provision or case-law on the topic. It is therefore yet to be established if and under which conditions the passing-on defence is available. In those few Member States where the legislator or the judiciary has pronounced on the passing-on defence, judgments vary between explicit acceptance and explicit prohibition of the defence. In those Member States

See Comparative Study, pp. 1-78 to 1-80, 1-111 to 1-112 and 1-127 to 1-129. See Comparative Study, pp. 1-26 to 1-102; White Paper, section 1 and para. 2.2 (access to evidence), 2.3 (binding effect of NCA decisions) and 2.7 (limitation periods); and IAWP, section 2.1. See Comparative Study, Executive Summary, para. I. Introduction, p. 1.

16

EN

438  Commission Staff Working Document Impact Assessment Report where it is allowed, the burden of proving the defence varies. On one side of the spectrum, there are Member States (such as France) where passing-on of overcharge from the direct to the indirect purchasers is presumed to be a common commercial practice and where, as a consequence, the direct purchasers will have to prove that no passing-on has taken place.39 On the other side of the spectrum there are Member States such as Germany, that allow the passing-on defence only under specific conditions and in case the defence does not lead to an unjust benefit for the defendant.40 •

42.

The described differences have even increased since 2004. While the rules applicable to antitrust damages actions have not changed in the majority of Member States, a few Member States have adopted amendments, such as the 2007 amendment to the Danish Competition Act, including a new statute of limitation periods for antitrust damages claims; the 2008 amendment to the Hungarian Competition Act, introducing a rebuttable presumption that a hard-core cartel has caused a 10% price increase on the market; or the 2008 amendment to Bulgaria's Law on the Protection of Competition, making final infringement decisions by NCAs binding on national courts in private actions.

43.

In addition, proposals to modify some of the national rules on antitrust damages actions are currently pending in Austria, Germany and the UK. As regards the topics covered by the current policy initiative, the Austrian proposal contains rules on the binding effect of NCA decisions and on the suspension of the limitation period during an investigation of the competition authorities. The German proposal does not directly concern any of the topics of the policy initiative. As regards the interaction of the public and private enforcement of competition law, the UK Government decided not to intervene because it expects measures to be adopted on this issue at the EU level. An overview of the main features of the Austrian, German and UK proposals are contained in Annex 3.

44.

This diversity and legal uncertainty leads to ineffective enforcement of the competition rules. The diversity in the legal systems is not only a problem for victims. Also defendants may suffer disadvantages in terms of imponderability and costs, mainly because they risk being sued for damages in different Member States or in the Member State where the substantive or procedural rules are most favourable for claimants, which is not necessarily their place of establishment (but for example the domicile of a co-infringer).41 For these reasons, a more level playing field is in the interest of both potential victims and potential defendants.

39 40 41

EN

Finally, as regards disclosure of evidence, the overwhelming majority of Member States' legal systems provide for some form of disclosure, varying between a system whereby the required documents have to be precisely described (such as Germany) to one which allows for the disclosure of classes of documents (such as the UK). Some Member States also limit the possibilities to request disclosure to parties to the proceedings, excluding third parties, or require separate proceedings.

Cour de Cassation (French Court of Cassation), judgment of 15 May 2012, case No 11-18495, (Synthetic lysine cartel - Ajinomoto Eurolysine case). Bundesgerichtshof (German Supreme Court for Civil Matters), judgment of 28 June 2011 case No KZR 75/10(Carbonless paper cartel). See Article 2 of Regulation 44/2001 and Article 6(3)(b) of Regulation 864/2007.

17

EN

Damages actions for breach of the EU antitrust rules (2013) 

439  

45.

The above factors, combined with the fact that antitrust cases, by nature, often require an unusually high level of very costly factual and economic analysis and present specific difficulties for claimants regarding access to crucial pieces of evidence that are often kept secret in the hands of the defendants, deter many victims from bringing actions as they consider the risk/reward balance to be negative.42

46.

In that context, many stakeholders (both in response to the public consultation on the White Paper and in response to the 2011 public consultation on collective redress) have – besides remedying the obstacles to effective private enforcement as identified above – insisted on the importance of encouraging consensual dispute resolution mechanisms.

47.

If the baseline scenario of no EU action applies, this situation would not change. The diversity between the rules applicable to actions for damages in the Member States as described above would persist and even increase should some Member States adopt new legislation. Due to the continued existence of important obstacles which prevent victims of EU competition law infringements from obtaining full compensation, the current situation of undercompensation of victims and their problematic access to justice would equally persist. Since the situation has not significantly changed since the adoption of the Green Paper in 2005, there is no reason to believe that – in the absence of any EU action in the field – significant improvements will be made.

48.

In the baseline scenario, undercompensation of victims of EU competition law infringements is the main problem. Risks of overcompensation have not been observed in any EU Member State. Despite this, stakeholders from the business community responding to public consultations contended that an EU initiative on actions for damages may lead to a risk of unmeritorious or abusive litigation or overcompensation. Any initiative to facilitate actions for damages should contain the necessary safeguards to avoid such undesirable side effects.

3.3.

The current legal framework could endanger the proper functioning of the internal market

49.

The EU competition rules are a matter of public policy, which lie at the heart of the functioning of the internal market.43 Shortcomings in the effective enforcement of Articles 101 and 102 hinder not only the achievement of the goals of workable competition, such as better allocation of resources, greater economic efficiency, increased innovation and lower prices. They also have a direct impact on the functioning of the internal market, which relies on a system of undistorted competition. In this context, Article 3(3) of the Treaty on European Union ('TEU') provides that the internal market shall be based on "a highly competitive social market economy".

50.

Sections 3.1 and 3.2 explain that overall effective enforcement of EU competition law consists of its effective public and its effective private enforcement. It is

42

See the external Impact Study of 21 December 2007, 'Making antitrust damages actions more effective in the EU: welfare impact and potential scenarios' ('the Impact Study', available at http://ec.europa.eu/comm/competition/antitrust/actionsdamages/files_white_paper/impact_study.pdf), Part II, section 1.1 for a general model of incentives to sue and section 3.2 for the specific issue of access to evidence. Deficits regarding reparation of harm resulting from an infringement of directly effective EU rules may also exist in other areas of law. In the field of competition, however, the size of the uncompensated harm and the problems encountered are particularly big. 43 Protocol Nr 27 on competition and internal market, Case C-126/97, Eco Suisse, [1999] ECR I-3055, 36, and joined cases C-295/04 to C-298/04, Manfredi, [2006] ECR I-6619, 31.

EN

18

EN

440  Commission Staff Working Document Impact Assessment Report furthermore explained why private enforcement of competition law is currently not effective and how certain aspects of the private enforcement of competition law risk negatively affecting the effective functioning of public enforcement. As set out in section 3.2, the marked differences between Member States regarding the rules governing actions for damages concerning infringements of national or EU competition law that were already described in the 2004 Comparative Study, in the 2008 White Paper and its accompanying Impact Assessment have further increased due to developments in legislation and jurisprudence in a limited number of Member States, as opposed to the lack of developments in other Member States. 51.

The primary example of divergence, also with an impact on the functioning of the internal market, is given by the different national rules applying to access to evidence. With the exception of a few Member States (most notably the UK), the lack of adequate rules on inter partes disclosure of evidence in court means that there is no effective access to evidence for victims of a competition law infringement seeking antitrust damages. Also the differences concerning access to information held by competition authorities are significant (e.g., in Austria such information is in practice entirely exempted from access). Other examples concern rules on passing-on (differences having major implications for the ability of direct/indirect purchasers to claim damages effectively and, in turn, for the defendant's chances of avoiding the obligation to compensate for harm caused), the binding effect of NCA decisions (only Germany acknowledges the binding effect of infringement decisions by the NCAs from all Member States), or on issues relevant for the quantification of antitrust harm (e.g. the existence of a rebuttable presumption of harm, or the power of judges to estimate the amount of damages).

52.

The fact alone that, according to the Commission's knowledge,44 the vast majority of large antitrust damages actions are currently being brought in 3 European jurisdictions – namely in the UK, followed by Germany and the Netherlands – indicates that the rules applicable in these Member States are considered by claimants to be much more suitable for effectively bringing such claims than in other Member States. Out of the 54 final cartel and antitrust prohibition decisions taken by the Commission in the period 2006-2012, only 15 were followed by one or more follow-on actions for damages in one or more Member States. In total, 52 actions for damages were brought in only 7 Member States. In the 20 other Member States, the Commission is not aware of any follow-on action for damages based on a Commission decision. Among those 7 Member States where actions were brought, the vast majority was brought in the 3 above mentioned Member States. The relative preference for a legal system in order to bring an action, or the relatively higher likelyhood of victims of competition law infringements to claim compensation in the UK, Germany and the Netherlands depend on more effective procedural rules for antitrust damages actions. Thus, successful measures from these jurisdictions such as the binding effect of NCA decisions or mechanisms for evidence disclosure are included in the options considered in the present report.45

53.

These differences lead to inequalities and uncertainty concerning the conditions under which injured parties can exercise the right to compensation deriving from the

44

45

EN

The figures contained in this paragraph are based on evidence gathered by the Commission services. As there is no reliable complete overview available nor does a public register of actions for damages brought in national courts exist, thess data can only be taken as a rough indicator. See Section 5.3.

19

EN

Damages actions for breach of the EU antitrust rules (2013) 

441  

Treaty and may affect the substantive effectiveness of this right. As injured parties often choose the forum of their Member State of establishment to claim damages (one reason being that consumers and smaller businesses cannot afford to choose a possibly more favourable jurisdiction of another Member State), the discrepancies between the rules of the different Member States risk leading to an uneven playing field as regards actions for damages and may affect competition on the markets on which these injured parties operate. The existence of this risk is, by way of example, demonstrated by the fact that the vast majority of the 52 known actions for damages following a Commission decision in the field of competition are brought by large undertakings. 54.

Similarly, these differences in applicable rules mean that undertakings established and operating in different Member States are exposed to significantly different risk of being held liable for infringements of competition law. This uneven enforcement of the EU right of compensation may result in a competitive advantage for some undertakings that breached Articles 101 or 102 TFEU, and a disincentive to the exercise of the rights of establishment and provision of goods or services in those Member States where the right to compensation is more effectively enforced. As such, the differences in the liability regimes applicable in the Member States may negatively affect competition and risk to appreciably distort the proper functioning of the internal market.

55.

It is necessary to ensure a more level playing field for undertakings operating in the internal market and to improve the conditions for injured parties, in particular citizens and small businesses, to make use of the rights they derive from the internal market. This more level playing field will not be realised when applying the baseline scenario of no EU action in the field, as explained in section 3.2. It is therefore appropriate to increase legal certainty and to reduce the differences that exist between the Member States as to the national rules governing actions for damages by harmonising certain relevant key rules applicable to actions for damages such as access to evidence, passing-on and the bindig effect of NCA decisions. These measures would avoid divergence of applicable rules, which risks to appreciably distort the proper functioning of the internal market.

3.4.

Scope and costs of the uncertainty relating to the interaction between public and private enforcement and the ineffectiveness of the legal framework for actions for damages

56.

The scope of both the problem of the uncertainty relating to the interaction between public and private enforcement of competition law and the problem of the ineffectiveness of the legal framework for actions for damages is considerable.

3.4.1.

Scope and cost: ill-regulated interface between public and private enforcement

57.

The potential effect on public enforcement (ongoing as well as future investigations) applies in all Member States: the Commission as well as all Member States46 have cartel leniency programmes, which have proven to be of great importance in the detection and prosecution of cartels. Leniency programmes grant immunity from and reduction of fines to those cartel participants who actively cooperate with the

46

EN

All EU Member states already have a working leniency program, with the exception of Malta which is currently implementing one.

20

EN

442  Commission Staff Working Document Impact Assessment Report competition authority.47 Because of the secret nature of cartels, the vast majority of cartels is discovered on the basis of a leniency application of one of the cartel participants. As regards Commission cases, when looking at the period 2008 to 2011, 21 out of 24 decisions (i.e. 88% of decisions) were based on leniency applications. In these years, a total amount of fines of €7,3 billion was imposed on cartel infringers, of which around 83% was imposed in cases based on leniency applications. When looking at the NCAs represented in the ECN, in 2010 18 out of 30 and in 2011 13 out of 21 cartel decisions, imposing a significant amount of fines, were based on leniency applications.48 58.

The uncertainty concerning the interface between public and private enforcement has been revealed quite recently, through the 2011 judgment of the Court in the Pfleiderer case, and in its various offsprings at national level. If the EU legislator were not to act on this problem, substantial problems for the public cartel enforcement could ensue. Fewer leniency applications would mean more undetected cartel activity and would lead to a welfare loss across the EU.

59.

The importance of settlements in cartel cases is also increasing. From June 2008 (introduction of the Settlement Notice) until 2011, 5 cartel cases have been settled. In broadly the same period (2008-2011) the Commission took 24 cartel decisions.

60.

If the baseline option of no EU action in the field would be followed, there is a risk that less leniency applications are submitted and less cartels are thus discovered and fined. Equally, undertakings may be less willing to cooperate with the competition authorities in settlement procedures. If these negative effects materialize, this would lead to a lower discovery rate for of EU competition law infringements, less effective public enforcement and, overall, less cases being dealt with. The effectiveness of public enforcement of EU competition law would thus be at stake, with a negative knock-on effect on private enforcement by further reducing the possibilities for and the likelihood of follow-on actions for damages.

3.4.2.

Scope and cost: ineffective legal framework for compensation

61.

As to the second problem, the ineffectiveness of the legal framework for antitrust damages actions is observed in every Member State, although to differing degrees as the applicable national rules differ significantly. Infringements of competition law, be they hardcore cartels, other infringements of Article 101 TFEU or abuses under Article 102 TFEU, occur in almost every sector of the economy.49 The problem concerns both actions for damages brought following a decision by a competition authority and actions brought on a stand-alone basis.

62.

The lack of an effective compensation mechanism means that the costs of competition law infringements are currently borne by the victims: in the case of harm in the form of a price overcharge, this means that there is a direct transfer of money to the infringers, which they would keep if they do not have to compensate the harm caused by the infringement. In case of a loss of profit because of the infringement (for instance in the case of a competitor illegally foreclosed from a market), the lawabiding undertaking – to which no compensation is paid – has to bear the price of the

47 48 49

EN

See Annex 1 for a more detailed description of the functioning of the Commission’s leniency programme. These numbers are based on the cases reported to the Commission, or published by NCAs or the press. See, for instance, the Commission’s Annual Reports on Competition Policy at: http://ec.europa.eu/competition/publications/annual_report/index.html.

21

EN

Damages actions for breach of the EU antitrust rules (2013) 

443  

infringement while the infringer benefits from his illegal act. While fines imposed by competition can serve as a deterrent for infringements, they cannot alter these effects. In the baseline scenario of no EU action, these problems would persist. 63.

The ineffective legal framework especially affects SMEs and large groups of consumers: as they are often at the end of the distribution chain, they face particular difficulties in identifying and proving the harm they suffered (causal link and quantum). They often perceive the uncertainties, risks and costs of a damages action as disproportionate to potential benefits.50

64.

Even though the absence of reliable empirical data makes precise quantification impossible, there is general agreement that infringements of competition law cause substantial harm to consumers and undertakings. Looking alone at hardcore cartels with effects across the EU,51 the Impact Study estimates that the annual direct cost to consumers and other victims in the EU ranges from approximately €13 billion (on the most conservative assumptions) to over €37 billion (on the least conservative).52 This estimate comprises both the harm resulting from consumers and other victims having to pay a higher price due to the illegal conduct of the cartelists (the “overcharge”) and also the economic benefits forgone by consumers and other victims who do not purchase, or purchase a smaller quantity, due to the unlawfully inflated price (the “deadweight loss”). It therefore covers the direct costs of cartels to consumers and other victims. It takes no account of more indirect macro-economic effects, such as the absence of greater allocative, productive and dynamic efficiency, which contribute to growth and employment, but are extremely difficult to estimate.

65.

If one adds to the figures on EU-wide cartels the annual cost to consumers and other victims of domestic hardcore cartels, the total annual cost for hardcore cartels in the EU can be estimated to range from approximately €25 billion (on the most conservative assumptions) to approximately €69 billion (on the least conservative).53 Expressed as a proportion of the EU’s gross domestic product, the negative consumer welfare impact of all these hardcore cartels is estimated as ranging from 0.20% to 0.55% of the EU’s GDP in 2011,54 which does not include the harm caused by abusive practices and infringements of Article 101 TFEU other than hardcore cartels. More effective redress would lead to a larger percentage of this harm being

50 51 52

53

54

EN

See Impact Study, Part III, section 2.1. Hardcore cartels are agreements between competitors to fix prices or allocate markets. The estimates do not cover other infringements of Article 101 such as vertical restraints nor abuses under Article 102. This estimate is based on the total amount of fines imposed by the European Commission on cartels (annual average for the period 2002 to 2007) and the finding that, on average, the total overcharge applied by these cartels is approximately 50% of the fine. Assuming a given detection rate of cartels (10% on the least conservative assumptions and 20% on the most conservative), the total overcharge applied by undetected EU-wide cartels can be calculated. Another constituent of harm to consumers is the “deadweight loss”. Assuming a set relation to the overcharge applied (50% on the least conservative assumptions and 10% on the most conservative), the total deadweight loss is calculated and then added to the total overcharge in detected and undetected cartels. For further explanations of the method and for data underlying these assumptions plus extensive references to research in this area, see Impact Study, Part I, section 3.1.2. This estimate is based on the figures for (at least) EU-wide cartels, and on the assumption that domestic cartels imply harm to consumers equivalent to 88.4% of the harm resulting from EU-wide cartels; for details on the methodology and underlying assumptions, see Impact Study, Part I, section 3.2.1. See for these estimates and the underlying analysis Impact Study, Part I, section 3.2.1 (Table 10). See for 2011 GDP figures: http://www.economic-growth.eu/English/updated_data/data2011.html.

22

EN

444  Commission Staff Working Document Impact Assessment Report compensated to the victims, whereas application of the baseline scenario would at best lead to a continuation of the current situation. 66.

However, even in the most effective system of private enforcement, not all the harm to consumers and other victims reflected in the above estimates will be compensated. This is, amongst others, because a considerable number of antitrust infringements will remain undetected. For hardcore cartels, the detection rate is generally assumed to be somewhere between 10% and 20%.55 For other infringements, the detection rate is higher, but the 'conviction' rate (i.e. the rate of successful damages actions) is likely to be much lower, since claimants often find it very difficult to produce proof that the contested conduct produced actual anti-competitive effects.56 It also has to be assumed that some victims do not come forward to claim compensation, for instance because they prefer not to disrupt an ongoing business relationship with the infringer. Moreover, in some instances, victims will find it rather difficult to convince courts of a sufficiently close causal link between the damage and the infringement.57

67.

Any realistic estimate of how much compensation victims could expect under a more effective legal framework for antitrust damages actions is therefore necessarily lower than the total harm to consumers and other victims from detected and undetected infringements estimated (for cartels) above. The Impact Study assesses the potential benefit of a more effective compensation system in the EU by comparing the current ineffective legal framework in Europe with a legal system where private enforcement of competition rules by means of damages actions is very effective, i.e. where victims of antitrust infringements no longer encounter the same obstacles to claiming compensation in court for the damage suffered.58 On this basis, the Impact Study estimates that the total amount of compensation (single damages plus pre-judgment interest) that victims of antitrust infringements are currently forgoing ranges from approximately €5.7 billion (on the most conservative assumptions) to €23.3 billion (on the least conservative) each year across the EU.59 These estimates relate to all types of infringements of Articles 101 and 102 TFEU. They provide an approximate idea of the amount of compensation that victims are currently forgoing and do not provide a precise calculation of the magnitude of future antitrust damage awards.

68.

Effective remedies for private parties would not only increase the likelihood that infringers are held liable, but would also increase the likelihood of detection of illegal restrictions of competition. Therefore, improving the effectiveness of the legal framework for actions for damages would also produce beneficial effects in terms of deterrence of future infringements and greater compliance with EU antitrust rules,60 as the Court emphasised.61 If more effective compensation mechanisms were to lead

55 56 57 58

59 60 61

EN

See Impact Study, Part I, sections 2.1.1 and 3.1.1.1. See Impact Study, Part I, section 4.1. Comparative Study, pp. 1-72 to 1-75 and 1-110. The empirical data used by the Impact Study in this comparison are mostly from the USA, where an enhanced system of antitrust damages actions is available. A range of refinements were, however, made to these data. In particular, the figures mentioned above are not based on treble damages as customary in the USA, but rather single damages with pre-judgment interest, to reflect the predominant legal situation in many Member States (on average, single damages with pre-judgment interest can be said to equate roughly to double damages without pre-judgment interest, see Impact Study, Part I, section 6). See Impact Study, Part I, section 6. See also Impact Study, Part I, section 2.1. Case C-453/99, Courage and Crehan, [2001] ECR I-6297, 27 and case C-360/09, Pfleiderer AG v Bundeskartellamt, [2011] ECR I-5161, 29: "the existence of such a right [to claim damages] strengthens the working of the Community competition rules and discourages agreements or practices, frequently

23

EN

Damages actions for breach of the EU antitrust rules (2013) 

445  

to a reduction of hardcore cartels by, for example, 5%, the negative consumer welfare impact would be reduced by €1.25 to €3.45 billion per year.

covert, which are liable to restrict or distort competition. From that point of view, actions for damages before national courts can make a significant contribution to the maintenance of effective competition in the European Union". The complementary function of private enforcement has been stressed by the Court as early as 1963, see case 26/62, van Gend en Loos, [1963] ECR 1 (Eng. Spec. Ed.): "(…) the vigilance of individuals to protect their rights amounts to an effective supervision in addition to the supervision entrusted to the diligence of the Commission and of the Member States."

EN

24

EN

446  Commission Staff Working Document Impact Assessment Report 4.

OBJECTIVES

69.

This section sets out the general policy objectives pursued, along with several more specific underlying objectives. Section 6.1 sets out and explains a set of specific assessment criteria that make it possible to measure, in qualitative terms, to what extent the various policy options considered are capable of contributing to achieving the general and specific objectives pursued.

4.1.

General objectives

70.

This policy initiative has two primary objectives: (i)

In order to address the problems described in section 3.1 and avoid the costs and effects described in section 3.4.1, this policy initiative aims to ensure that the Commission and NCAs can apply a policy of strong public enforcement of competition law, without this public enforcement being unduly affected by the private enforcement before national courts. One of the key elements is to protect leniency and settlement programmes as well as ongoing investigations of the Commission and NCAs. This requires a strict judge-controlled system for access to evidence, which also protects the files of competition authorities. In order to achieve this purpose, common rules should be established relating to the key aspects of the interaction between public and private enforcement.

(ii)

In order to address the problems described in sections 3.2 and 3.3 and to avoid the costs and effects described in section 3.4.2, this policy initiative aims to ensure that victims of infringements of EU competition law have access to truly effective mechanisms for obtaining full compensation for the harm they suffered. By pursuing this objective, the Commission wishes to guarantee, in every Member State, certain standards allowing victims to effectively obtain full compensation from the infringers of Articles 101 and 102 TFEU, thereby giving full effect to such provisions. More effective compensation will ensure that the costs of infringements of competition law are borne by the infringers, and not by the victims, by compliant businesses and, indirectly, by society as a whole. This is in line with the competitiveness objectives for the EU.

4.2.

Specific objectives

71.

To allow for a more systematic and thorough assessment of whether the general objectives are fulfilled, the latter can be split up in the following specific objectives. •

Protection of effective public enforcement

As described in section 3.1, there is a risk that private enforcement of EU competition law by national courts may unduly affect the currently strong public enforcement of these rules, especially in relation to certain important enfocement tools like the leniency and settlement programmes. Stakeholders have confirmed the existence of this risk and have called for a solution. Therefore, one of the specific objectives of this policy initiative is the protection of effective public enforcement. The following elements shall be taken into account in the context of this specific objective: –

EN

Ensuring effective public enforcement of competition law by the Commission and NCAs by regulating the key aspects of interaction between public enforcement and actions for damages before national courts. Leniency and settlement programmes as well as ongoing investigations of the Commission

25

EN

Damages actions for breach of the EU antitrust rules (2013) 

447  

and NCAs should be appropriately protected in all Member States in order to ensure the effective public enforcement of competition law. –

Providing for specific liability rules for immunity recipients in actions for damages before national courts, in order to foster the leniency programmes of the Commission and NCAs.



Full compensation

As indicated in section 3.2, there is an EU right to full compensation for the harm suffered as a result of infringements of EU competition law. Section 3.2 also states that, currently, a large number of victims of EU competition law infringements remain uncompensated. Nevertheless, as also indicated in section 3.2, stakeholders (predominantly businesses) have equally stated that abusive litigation and overcompensation should be avoided. Overcompensation within the meaning of a systematic burden to pay compensation in excess of the harm caused, can effectively be avoided by excluding punitive or multiple damages awards. However, if overcompensation is defined as a case-specific risk of a damages award higher than the harm actually suffered, it must be borne in mind that a trade-off exists between a higher chance of full compensation and a risk of overcompensation, but that the latter risk is offset by the risk of undercompensation resulting from the lack of action or from the absence of appropriate substantive and procedural rules governing damages actions. It must also be borne in mind that the distinction between full compensation and under- or overcompensation in the latter sense may be difficult if not impossible to establish in a given case: in the absence of an empirical measure of the harm suffered, the actual harm suffered is given by definition through reference to the damages obtained from a court. In conclusion, the objective of full compensation consists of two elements: –

Ensuring an effective system of compensation of harm, thereby allowing full compensation for the entire harm suffered. In particular, the damage awards should include pre-judgment interest in order to compensate the victims for the real value of the harm suffered.



Need to avoid over-compensation: measures put forward as a result of this initiative should not lead to victims systematically receiving damages higher than the entire loss suffered.



Greater awareness of the rules and deterrence, increased enforcement and improved compliance, to the benefit of Europe’s competitiveness

As described in section 3.1, public and private enforcement of the EU competition rules coexist as complementary tools serving the objective of an effective enforcement of these rules. Both mechanisms need to function effectively to achieve optimal deterrence and compliance with the EU competition rules. Currently, private enforcement of competition law is not functioning effectively (as described in section 3.2): strengthening this enforcement strand will have important positive effects on the overall effective enforcement of and compliance with the EU competition rules. Raising awareness is necessary in order to ensure that victims of competition law infringements make use of their right to full compensation and, hence, contribute to achieving the purpose of an effective private enforcement of competition law. Therefore, the following elements are to be taken into account under this objective: –

EN

Increasing victims’ awareness of their entitlement to damages and of the conditions for bringing a claim to court. 26

EN

448  Commission Staff Working Document Impact Assessment Report –

Increasing (potential) wrongdoers’ awareness of the rules governing actions for damages and clarifying the conditions for their liability.



Improving compliance with the EU competition rules and deterrence by increasing the likelyhood of civil suits being brought, thus rendering more credible the risk that infringers will have to compensate the victims of their illegal behaviour.



At the same time, avoiding that this policy leads to over-deterrence, where the risk of claims for damages prevents undertakings from engaging in lawful conduct or where damages are to be paid by undertakings which have engaged in lawful conduct or where such companies face high costs to defend themselves against unmeritorious claims.



Reinforcing European competitiveness by means of greater compliance with the EU competition rules.



Access to justice

As described in section 3.2, the fact that a large number of victims of infringements of competition law currently remain uncompensated and see their right to damages under EU law frustrated, leads to a clear deficit in terms of compensatory justice and cannot be reconciled with the fundamental right to access to effective judicial protection. In order to guarantee the respect of this EU fundamental right, access to justice should be facilitated, while at the same time putting in place safeguards to avoid abusive litigation. Under this specific objective, the following elements shall be taken into account: –

Guaranteeing effective access to the courts and an effective remedy for all victims, as required by Article 47, first paragraph of the Charter and Article 19(1), second subparagraph, TEU, including for those who suffered scattered low-value damage, such as SMEs and consumers.



Ensuring that potentially high costs do not deter victims from bringing their legitimate claims.



Facilitating access to the relevant evidence in a case, thus overcoming information asymmetry.



Ensuring that meritorious claims can effectively be brought, and be successful, despite the complexity of antitrust damages cases.



At the same time, avoiding abusive litigation and ensuring that unmeritorious claims do not lead to the award of damages.



Appropriate and efficient use of the judicial system

As indicated in section 3.2, the high level of very costly factual and economic analysis required in antitrust cases constitutes a factor deterring many victims from bringing actions for damages, as they consider the risk/reward balance to be unfavorable. Next to measures which aim at optimising this risk/reward balance, an effective possibility to engage in consensual dispute resolution would equally reduce these costs and thus serve the effective private enforcement of competition law. In order to ensure optimal effectiveness of the private enforcement of competition law, it is necessary not to overburden courts and to avoid procedural abuses. The following elements relating to the use of the judicial system will be taken into account:

EN

27

EN

Damages actions for breach of the EU antitrust rules (2013) 

449  



Streamlining handling of antitrust damages cases by the courts by means such as joining or grouping identical or similar claims.



Reducing the costs of litigation by improving the conditions for settlements: settlements can be cost-efficient and, when fair and swift, are to be preferred to court actions. However, as settlements are voluntary, attainment of this objective presupposes the existence of a credible court alternative if no settlement is reached. A credible court alternative will also serve as a benchmark leading to improvement of the quality of the settlements.



Limiting procedural abuses: while the victims should have better access to the courts, it is important that law-abiding undertakings do not bear the costs and burden of abusive litigation. It is therefore necessary to have appropriate safeguards to prevent abuses.



Limiting the risk of multiple litigation on identical or similar issues: relitigation of issues already settled should be avoided since they entail unnecessary costs and delays plus the risks of a diverging outcome.



A more level playing field and increased legal certainty for businesses operating throughout Europe

Sections 3.2 and 3.3 describe the currently existing diversity in national legal systems in relation to the conditions under which actions for damages can be brought. This diversity causes an uneven playing field both for victims, who seek to exercise their right to full compensation, and for infringing companies, who should be held liable for their infringing behaviour. Furthermore, this diversity could endanger the proper functioning of the internal market. Section 3.1 describes the current legal uncertainty, resulting from the Pfleiderer judgment, in relation to the discosure of evidence from the file of a competition authority in actions for damages and the risk of negative effects on the strong public enforcement of EU competition law. In order to ensure the proper functioning of the internal market, the following elements shall be taken into account: –

Ensuring a more level playing field in Europe so that businesses across Europe compete on an equal footing and that EU citizens and undertakings can rely on a minimum standard to enforce the rights conferred upon them by the EU Treaties.



At the same time, respecting national legal traditions and values: a more level playing field should not be achieved without taking due account of the national legal systems and the balance struck over time by each Member State in its national rules.



Increasing legal certainty for businesses operating throughout Europe through common standards for their liability for infringements of competition law.



Increasing legal certainty as to which documents from the file of a competition authority can be disclosed as evidence in actions for damages.



Providing benefits for SMEs

The studies conducted on demand of the Commission have widely shown how SMEs often suffer from the difficulties in recovering antitrust damages. As described in section 3.2, this is caused by obstacles to effective compensation such as the lack of legal certainty and an unfavourable balance between benefits expected from an action for damages and costs incurred. Individual SMEs might even be less likely to

EN

28

EN

450  Commission Staff Working Document Impact Assessment Report claim damages if they have to engage in litigation with an actual contractual partner, with whom the business relationship should be continued after the proceedings (fear of retaliation). In order to improve the situation of SMEs and put them in a position where they are realistically able to claim damages for competition law infringements, thus also contributing to the objective of full compensation, the following elements should be taken into account: –

Improving the evidentiary position of SMEs in litigation, especially as regards the quantification of the damages. In the framework of the public consultation, SMEs and their associations have explicitly welcomed such measures.



Improving the possibilities for collective redress.



Stimulating economic growth and innovation

In order to provide for the initiative which maximises the functioning and overall enforcement of the EU competition rules as well as the functioning of the internal market, the policy initiative should maximise its long term benefits on growth, productivity and innovation. The following elements play a role in this context:

EN



Improving the possibilities for SMEs by putting them in a position where they are realistically able to claim damages for competition law infringements.



Providing for a balanced initiative providing for both effective public and effective private enforcement of EU competition law.



Ensuring the proper functioning of the internal market by providing for harmonised minimum standards across the Member States in relation to the essential features of actions for damages.

29

EN

Damages actions for breach of the EU antitrust rules (2013) 

EN

451  

5.

CONTENTS OF THE CURRENT ANTITRUST DAMAGES INITIATIVE

72.

The impact assessment of the current Antitrust Damages Initiative builds on the assessment carried out in the 2008 IAWP and should therefore be read in conjunction with it. Annex 4 to this impact assessment offers a summary of the 2008 IAWP. Four Policy Options are analysed in this Chapter, focusing both on their substantive content and on the more general issue of the type of action required. The appropriate choice of the instrument will be addressed in Chapter 0.

73.

In order to better understand the four Policy Options set out in section 5.3 below, an overview of the main results of the 2008 IAWP is presented in section 5.1, while the scope of the assessment of the current Antitrust Damages Initiative is presented in section 5.2.

5.1.

The 2008 IAWP and the choice of the 'Preferred Option'

74.

The 2008 IAWP assessed different bundles of several specific measures aimed at ensuring the effectiveness of the victims' right to compensation guaranteed by the EU Treaties. These policy options ranged from the baseline option of no action at the EU level to legislative measures maximising facilitation of claims and incentives for victims, as recalled by Table 12 in Annex 5 hereto.

75.

The 2008 IAWP led to the elaboration of a Preferred Option, put forward in the White Paper. Its main elements were as follows: full single damages; disclosure of specified categories of evidence, based on fact-pleading and proportionality; indirect purchasers' standing allowed; passing-on defence allowed, accompanied by a facilitation of proof of passing-on in favour of indirect purchasers; binding effect of NCA decisions across EU; rebuttable presumption of fault (once an infringement is established); collective redress in the form of opt-in collective actions and representative actions; rules on limitation periods concerning follow-on actions; and rules on interaction with leniency, namely protection of corporate statements from disclosure and limitation of immunity recipients' civil liability. For a detailed overview of the 2008 IAWP Preferred Option see Table13 in Annex 6 hereto.

5.2.

The current Antitrust Damages Initiative - scope of the assessment

76.

In order to address the problems identified in Chapter 3 and to achieve the objectives referred to in Chapter 4, it is necessary to assess different policy options that imply a different degree of intervention at EU level. The Commission put forward in the White Paper a policy approach to overcome the current inefficiencies of antitrust damages actions. Stakeholders widely acknowledged and endorsed this general approach.

77.

On the basis of the input of other EU institutions, in particular the Parliament, and after a further round of consultations and studies, the current Antitrust Damages Initiative considers a number of adjustments to the White Paper's policy approach. Those adjustments are meant to address the problems identified in Chapter 3 and to further the objectives referred to in Chapter 4 in a cost effective manner.

78.

In addition, the current Antitrust Damages Initiative aims at clarifying the relationship between private and public enforcement of the EU competition rules, in particular the balancing between the protection of leniency programmes and the need for an appropriate access to evidence to ensure efficient compensatory redress. This is reflected in the options discussed below.

30

EN

452  Commission Staff Working Document Impact Assessment Report 5.3.

Description of the Policy Options assessed in the current initiative

79.

The baseline scenario, presented below as Option 1, considers no EU action at all in the field of antitrust damages actions (neither soft law nor legislation). Option 2 aims at codifying the measures suggested in the White Paper in a legislative instrument. Option 3 builds on Option 2, while taking due account both of the main comments received during the public consultations since the White Paper and of recent legislative and jurisprudential developments in the EU and in the Member States. It therefore includes additional safeguards for access to evidence, provides for a rebuttable presumption as to the existence of harm in cartel cases and provides for an effectiveness requirement in relation to the quantification of harm by the claimant. Option 4 considers the adoption at the EU level of non-regulatory measures only. Options 2 to 4 must be all read together with the provision of a non-binding framework on the quantification of damages, as explained below at 5.4.

80.

When considering an appropriate legal framework for antitrust damages actions, experience in other jurisidictions shows that more far reaching measures are available than those considered in this report. These measures are, in particular: multiple damages, extensive discovery among the parties, automatic fee-shifting, opt-out class actions brought by any individual. These measures have been assessed in the IAWP. The result of that assessment was that they were excluded in view of an unsatisfactory cost/benefit ratio (high costs entailed in exchange for a higher rate of achievement of the objectives), as summarised in Annex 4. For these reasons, the above-mentioned measures have been excluded also from the spectrum of available policy options in the present exercise. As a result, all measures included in policy options 2 to 4 contemplate less costly means means to achieve the set objectives.

Table 1: Overview of the Policy Options assessed in the current Antitrust Damages Initiative Option 1 (baseline scenario) Damages Access evidence

No EU action to

No EU action

Option 2

Option 3

Full single (no multiple damages)

Rules on disclosure of specified categories of evidence, based on factpleading and proportionality Protection from disclosure of corporate statements made in the context of leniency programmes

EN

Option 4

Soft law recommending full single damages

Rules on disclosure of specified categories of evidence, based on factpleading and proportionality

Soft law recommending rules on disclosure of specified categories of evidence, based on factpleading and proportionality

No access to corporate statements and settlement submissions. Access to other categories of information only after closure of public proceedings

Soft law recommending not to give access to corporate statements and settlement submissions and recommending access to other categories of information only after closure of public proceedings

Limitation of civil liability of Immunity recipient(s)

No EU action

limitation of liability of the immunity recipient to claims by its direct and indirect contractual partners

Indirect purchaser

No EU action

Standing allowed

Passing-on

No EU action

A passing-on defence for the infringer that shows that the damages claimant has passed-on part or the whole of the illegal overcharge, as well as a rebuttable passing-on presumption in favour of the indirect purchaser

31

Soft law recommending limitation of liability of the immunity recipient to claims by its direct and indirect contractual partners Soft law recommending standing is allowed

that

Soft law recommending a passingon defence for the infringer that shows that the damages claimant has passed-on part or the whole of

EN

Damages actions for breach of the EU antitrust rules (2013) 

453  

the illegal overcharge, as well as a rebuttable passing-on presumption in favour of the indirect purchaser Rebuttable presumption of harm

No EU action

No presumption

Effect of NCA decisions

No EU action

Binding effect for the final infringement decisions of NCAs

Soft law recommending binding effect for the final infringement decisions of NCAs

Fault (once infringement established)

No EU action

Rebuttable presumption; exoneration for excusable errors.

No EU action

Limitation period

No EU action

A specific limitation period for damages actions that rely on an infringement decision by a competition authority

Soft law recommending a specific limitation period for damages actions that rely on an infringement decision by a competition authority

Cost rule

No EU action

No legislative measure

No EU action

No EU action

Consensual dispute resolution

No EU action

No legislative measure

Rules on suspensive effect of consensual dispute resolution, effect of settlements on other injured parties and on subsequent actions for damages

Soft law recommending rules on suspensive effect of consensual dispute resolution, effect of settlements on subsequent actions for damages

Collective redress

No EU action

Two different collective redress mechanisms: optin group actions and representative actions

No competition-specific EU action, separate horizontal initiative on collective redress which is outside the scope of this impact assessment

No recommendations on this topic in the competition-specific soft law

No EU action

Soft law recommending a rebuttable presumption of existence of harm caused by cartels

5.3.1.

Option 1: the baseline scenario of no EU action in the field of antitrust damages actions

81.

Option 1 contains the baseline scenario, entailing no action at all at EU level regarding antitrust damages actions. Some respondents in the public consultation on the White Paper, as well as in the 2011 public consultation on collective redress have suggested to follow this option on grounds of general inappropriateness for EU action in an area that is mostly governed by national substantive and procedural rules. The assessment of the impact of Option 1 examines the status quo and likely developments in the absence of EU action (prospective analysis).

5.3.2.

Option 2: adopting the White Paper's suggested measures through a legislative instrument

82.

Option 2 envisages a legislative instrument incorporating the measures that the Commission has put forward in its White Paper. Such instrument would include: a confirmation that compensation of full single damages can be obtained; two different collective redress mechanisms (opt-in group actions62 and representative actions);

62

EN

Rebuttable presumption of existence of harm caused by cartels

The White Paper and the IAWP referred to 'opt-in collective actions'. In the framework of the public consultation on collective redress it has been observed that this terminology might lead to confusion between collective actions and collective redress mechanisms in general. Therefore, in the following, 'opt-in collective actions' will be referred to as 'opt-in group actions'.

32

EN

454  Commission Staff Working Document Impact Assessment Report rules on disclosure of specified categories of evidence, based on fact-pleading and proportionality; a limitation of liability of successful immunity applicants; binding effect for the final infringement decisions of NCAs; standing for the indirect purchaser; a passing-on defence for the infringer that shows that the damages claimant has passed-on part or the whole of the illegal overcharge, as well as a rebuttable passing-on presumption in favour of the indirect purchaser; minimum requirements for limitation periods for damages actions, also when they are brought after an infringement decision by a competition authority; and a rebuttable presumption regarding fault once the infringement has been established, unless the infringer can prove that the infringement was due to an excusable error. To avoid the risk that enhanced damages actions might have a negative impact on public enforcement activities, in particular on the functioning of leniency programmes, Option 2 suggests protecting from disclosure corporate statements made in the context of leniency programmes.

EN

5.3.3.

Option 3: legislative proposal based on the White Paper and additional safeguards

83.

Option 3 consists of a legislative instrument that is still based on the White Paper (i.e. Option 2), while optimising the relation between public and private enforcement and introducing a number of other modifications. Option 3 contains the following measures: the right to full single damages; rules on disclosure of specified categories of evidence, based on fact-pleading and proportionality with an enhanced protection for documents from the file of a competition authority; a limitation of liability of successful immunity applicants; binding effect for the final infringement decisions of NCAs; standing for the indirect purchaser; a passing-on defence for the infringer that shows that the damages claimant has passed-on part or the whole of the illegal overcharge, as well as a rebuttable passing-on presumption in favour of the indirect purchaser; rules on the beginning of limitation periods as well as a specific rule for limitation periods after the finding of infringement by a competition authority or a review court; a rebuttable presumption relating to harm in cartel cases; and measures facilitating consensual dispute resolution.

84.

Whereas Option 2 does not include any legislative measures on consensual dispute resolution, Option 3 foresees rules meant to facilitate such outcome and to remove legal uncertainties existing about it, for instance on the effect of settlements on subsequent actions for damages.

(a)

Access to evidence: special rules concerning access to the file of a competition authority

85.

After the 2011 Pfleiderer ruling of the Court, Member States, NCAs and representatives of the business and legal community asked explicitly for an EU-wide clarification of the interaction between public enforcement (in particular through leniency programmes) and private damages actions, where access to evidence plays a key role. Option 3 provides for such clarification by suggesting further safeguards concerning the disclosure of documents in the file of a competition authority. It is suggested never to disclose corporate statements and settlement submissions, because disclosure of those documents would jeopardise enforcement instruments at the disposal of the competitition authorities (particularly the leniency programme and the settlement programme). To protect ongoing public investigations, it is also suggested to allow the disclosure of documents that were especially prepared for the purpose of public enforcement proceedings only once those proceedings are finished. All other information (called 'pre-existing information') would remain disclosable. In

33

EN

Damages actions for breach of the EU antitrust rules (2013) 

455  

doing so, Option 3 provides for an appropriate protection of effective public enforcement, while allowing claimants to obtain the information required to successfully bring a damages action. 86.

An alternative regime would have been to leave it to the discretion of the national court whether or not to disclose corporate statements or settlement submissions in the context of a damages action that is brought after public enforcement proceedings are closed. That option, however, would not have given the undertakings that want to engage in leniency or settlement discussions with the competition authority, the desired upfront legal certainty regarding the disclosibility of the said documents. That absence of legal certainty would be to the detriment of the success of those public enforcement instruments and is thus not offering an appropriate balance between protecting both public and private enforcement of the EU competition rules. This alternative regime is therefore not further considered. The benefits and costs of this status quo option are shown in tables 2 and 3 below.

(b)

Rebuttable presumption as to the existence of harm in cartel cases

87.

The studies and consultations referred to in section 2 have shown that proving harm and the quantification thereof often constitutes an important barrier for effectively obtaining compensation for the harm caused by a competition law infringement. To remedy these evidentiary problems of the claimant associated with the quantification of harm, Option 3 considers the introduction of a rebuttable presumption of the existence of harm in cartel cases. The defendant, who is most likely to possess the evidence that is necessary to prove whether or not the cartel caused harm, is free to rebut the presumption. The presumption would thus respect the principle that an information advantage of one party should lead to that party holding the burden of proof. Option 3 considers to limit the rebuttable presumption to cartels because of their secret nature which makes it more difficult for parties to obtain the necessary evidence to prove harm.

88.

Studies in relation to overcharge in cartel cases support the introduction of such rebuttable presumption as to the existence of harm. The Quantification Study63 and a study by Connor and Lande64 conclude that the average overcharge in cartel cases is respectively 20% and 23%. Moreover, a study conducted by Boyer and Kotchoni65 in which a meta-analysis of cartel overcharge estimates is carried out in reaction to potential errors and biases in the model used by Connor and Lande, came to a corrected mean overcharge in all cartel cases of 17,5% with a median of 14%. The studies hold that there are some cartel cases (around 5%) in which no overcharge harm is caused. However, as the presumption is rebuttable and the evidence of the absence of harm is likely to be in possession of the defendant, this small percentage of cases, where the presumption could be rebutted, do not prevent its introduction.

89.

After the existence of harm has been established, the claimant still has to prove the amount of that harm. The burden of proof for the quantification of harm normally lays with the claimant. To further facilitate this burden, one could consider introducing a rebuttable presumption that cartels lead to an overcharge of X% (with

63 64 65

EN

Quantification Study, p. 91. J.M. Connor and R.H. Lande 'Cartel Overcharges and Optimal Cartel Fines', in S.W. Waller (ed.), Issues in Competition Law and Policy, volume 3, ABA Section of Antitrust Law. Marcel Boyer and Rachidi Kotchoni, The Econometrics of Cartel Overcharges, Scientific Serial, Montréal, 10 August 2011.

34

EN

456  Commission Staff Working Document Impact Assessment Report 'X' being a figure supported by the studies referred to above). However, such a specific presumption risks becoming a disincentive for victims of a cartel to engage in further quantification of the harm suffered. Such presumed overcharge figure may thus paradoxically lead to structural under-compensation of harm, in particular when it is at the lower end. That negative effect is increased by the fact that defendants will invest heavily into rebutting, i.e. defending the presumption, depending on whether they caused a lower or a higher overcharge harm. Therefore, this option would not even reduce litigation costs. 90.

(c)

Introduction of provisions on consensual dispute resolution

91.

Option 3 also foresees the introduction of some provisions removing existing obstacles to effective consensual dispute resolution between injured parties and undertakings having infringed competition law. Such provisions relate to the suspension of limitation periods for bringing actions for damages if parties prove that they are or were engaged in consensual dispute resolution, the suspension of pending proceedings for the duration of consensual dispute resolution. They also require that damages paid through consensual settlements should be taken into account in determining the amount of compensation or contribution that a settling infringer needs to pay following a subsequent action for damages.

92.

Further modalities to facilitate consensual dispute resolution have not been considered. For instance, the requirement of an attempt of consensual dispute resolution before having access to a court has not been considered because it may unduly prolong litigation and it can constitute a violation of the fundamental right of access to a judge.66 Also the possibility to oblige non-settling infringers to contribute to the damages paid in the context of a consensual settlement or the possibility to release settling infringers from contributing to the damages paid in a subsequent damages action, have not been retained, because they would violate the basic freedom of all parties to settle or not, and thus the fundamental right of a party to resolve a dispute via court proceedings.

(d)

Detailed comparison with Option 2 and policy issues underpinning the measures

93.

The third option consists of a binding instrument that partly revises the options put forward in the White Paper in order to keep account of more recent developments at national and EU level in two ways, namely by referring to a separate horizontal EU approach to collective redress instead of regulating a sector-specific mechanism, and by introducing limitations to access to evidence aimed at preserving the effectiveness of public enforcement tools. The provisions on fault, which were particularly criticised by some business respondents within the public consultation, have also been removed. All these broad changes have in common that they reduce to some

66

EN

To ensure the effectiveness of the right to claim damages and to keep litigation costs low, Option 3 considers a rebuttable presumption of harm (without mentioning any figure), combined with the general principle of effectiveness, acoording to which the burden and level of proof required for the quantification of the harm cannot render the exercise of the claimant's rights to damages practically impossible or excessively difficult. In that context, it is also suggested that Member States should enable the national judge to estimate the amount of the harm.

See also paragraph 25 of the European Parliament resolution of 2 February 2012 on 'Towards a Coherent European Approach to Collective Redress (see footnote 23 above)

35

EN

Damages actions for breach of the EU antitrust rules (2013) 

457  

extent the benefits in terms of effective compensation fostered by Option 2 in order to pursue additional policy objectives, i.e. a horizontal approach to collective redress as suggested by some stakeholders and by the European Parliament and a higher protection of public enforcement following the Pfleiderer judgment of the Court of Justice. The option has thus specifically been designed in order to assess whether the loss in benefits as regards effective compensation are counterbalanced by reduced costs of litigation and/or by an optimised balance between public and private enforcement of the EU competition rules. More specifically, Option 3 differs from Option 2 in the following points:

EN



As regards the protection of public enforcement tools, Option 2 only protects leniency corporate statements from disclosure in actions for damages. Option 3 adds to that the protection from disclosure of settlement submissions, in which undertakings acknowledge their participation in a cartel to obtain a simplified procedure and a reduction of the fine. Moreover, Option 3 protects on-going investigations of the competition authorities by limiting disclosure during those investigations. With these measures, Option 3 seeks to provide for a more appropriate protection of effective public enforcement. However, because of its limited scope, claimants could still obtain the information required to successfully bring a damages action. The envisaged protection of public enforcement tools is therefore considered not to make it excessively difficult for victims of a competition law infringement to obtain compensation for the harm they suffered. The protection is thus compatible with the right to effective judicial protection, as it is laid down in the EU Charter of Fundamental Rights.



As regards quantification of antitrust damages, Option 3 - contrary to Option 2 - provides for a rebuttable presumption relating to the existence of harm in cartel cases. This presumption is based on the findings of an external study which concluded that 93% of examined cartels cause harm. This measure has been introduced in order to reduce the adverse impact of the more limited access by claimants to some types of evidence that may nonetheless have been useful in view of proving the harm created by a cartel. For the same reason, Option 3 contains a rule that the exercise of the claimant's right to damages cannot be rendered practically impossible or excessively difficult by the required level of proof. In that context, this option suggests that Member States should enable the judge to estimate the amount of the harm. With these rules, option 3 seeks to enhance the effectiveness of the right to claim damages and to keep litigation costs low.



As regards collective redress, Option 3 does not contain any competitionspecific measures. While acknowledging the specificities of EU competition law enforcement and the possibility of specific rules, this option relies on the possibility of a separate, but horizontal approach to collective redress, through initiatives characterised by a broader scope. Finally, in order to partially counterbalance the absence of a specific collective redress mechanism, and to facilitate other kinds of cost-effective procedural means for the parties, Option 3 contains measures on consensual dispute resolution. This would remove existing disincentives to engage in out-of-court settlements to compensate for harm caused by an EU competition law infringement.

36

EN

458  Commission Staff Working Document Impact Assessment Report

EN

5.3.4.

Option 4: non-regulatory measures

94.

Policy Option 4 entails no legislative measures at EU level but would identify, mainly from the experience in Member States, a range of useful solutions and best practices. These non-binding recommendations would be inspired by the measures as proposed under Option 3, but would – by their very nature – not be formulated with the same level of detail. The impact of this option largely relies on the willingness of the Member States to carry out the suggested actions.

95.

Option 4 thus contains the following measures: a recommendation that Member States ensure the possibility to obtain compensation of full single damages; a recommendation that Member States adopt rules on disclosure of specified categories of evidence, based on fact-pleading and proportionality; a recommendation to exclude certain documents from the file of a competition authority from such disclosure; a recommendation that Member States adopt provisions limiting the liability of successful immunity applicants; a recommendation for Member States to grant binding effect to the final infringement decisions of NCAs; a recommendation that Member States recognize standing of indirect purchasers in actions for damages; a recommendation that Member States recognise the passing-on defence for the infringer that shows that the damages claimant has passed-on part or the whole of the illegal overcharge as well as the suggestion that Member States introduce a rebuttable passing-on presumption in favour of the indirect purchaser; a recommendation as to when limitation periods should start and as to its duration after the finding of infringement by a competition authority; a recommendation that Member States introduce a rebuttable presumption relating to the existence of harm in cartel cases; and a suggestion that Member State facilitate consensual dispute resolution.

5.4.

Provision of non-binding assistance for quantification of damages

96.

The provision of a non-binding legal framework for the quantification of damages has proven to be one of the measures most widely supported by the respondents to the public consultation on the White Paper. Quantification is one of the most complex issues in the framework of actions for damages for breach of antitrust rules. In this respect, the provision of pragmatic guidance, by reviewing available methods for estimating the loss suffered as a result of a competition law infringement, would be a useful tool for parties and judges in antitrust damages cases.

97.

Stakeholders have stressed the need for the non-binding character of such initiative. This stance appears justified, since the exact quantification of the harm largely depends on the specific features of each case (i.e. the structure of the market, the behaviour of players at each level of the distribution chain, etc.). A non-binding measure would therefore ensure that judges can rely on the guidance provided therein, when they deem it appropriate given the circumstances of the case at hand.

98.

The 2009 Quantification Study served as a basis for the Commission's formulation of the non-binding framework for damages quantification on which a public consultation was held in 2011. The responses to the public consultation confirmed the need for a pragmatic non-binding guidance. All Policy Options set out for assessment in this report (except the zero-action at the EU level option) must therefore be read as encompassing the adoption of this guidance.

37

EN

Damages actions for breach of the EU antitrust rules (2013)  6.

IMPACT ANALYSIS

6.1.

Assessment criteria

99.

The below analysis is predominantly qualitative, due to both methodological and factual considerations. From a methodological point of view, the current Antitrust Damages Initiative aims at regulating the interaction between public and private enforcement and at removing the obstacles to effectively exercising the right to compensation guaranteed by the Treaty, also by removing obstacles currently existing under national rules governing antitrust damages actions.

100.

While these changes will secure effective public enforcement and will make it easier to claim compensation, an empirical quantification of the effect of the proposed policy options is not feasible, since it would require access to data that are currently not available. To our knowledge, no jurisdiction in the world has adopted a similar bundle of measures, which makes it impossible to estimate the impact of the proposed policy options in light of another jurisdiction’s experience. Furthermore, the impact of each measure is determined by the interaction with other measures: the assessment of individual measures on the basis of their effects in countries where they are in place is therefore not directly relevant.

101.

An approximative quantitative impact would be expressed in a very wide range of values. At the lower end of the range, there is the theoretical hypothesis that the measures will not see a substantial increase in effectiveness of damages claims. It is, however, highly unlikely that an initiative at EU level would lead to such a zero benefit. At present, it has been regularly reported that potential damages claimants have not sought redress in a number of prominent cases of infringement of the EU antitrust rules (such as the heat stabilisers, bathroom fittings and bananas cartels) because of procedural obstacles (e.g. the absence of effective access to evidence) and uncertainty of the law (e.g. on the passing-on of overcharges). Removing such obstacles and introducing clear rules would undoubtedly result in an increased level of meritorious damages actions and thus of compensation for victims of antitrust violations. Where undertakings (in particular corporations) or public bodies (such as municipalities) are injured parties, the pursuit of meritorious damages claims may be a legal obligation under applicable rules of corporate law (fiduciary duty of the management of a corporation) or of budgetary rules. A more effective legal framework for bringing damages actions will likely allow decision-makers to decide that it is in the best interest of the corporation/public body to pursue compensation.

102.

At the upper end of the range, there is the equally theoretical hypothesis that following the EU initiative all the harm caused by antitrust infringements will be recovered by victims, meaning a recovery of currently foregone compensation that could be as high as €5.7 to €23.3 billion per year.67 The width of such a range, combined with the impossibility to predict the increase for each option of both the willingness of injured parties to actually claim compensation for harm suffered and the likelihood that meritorious claims succeed, does not allow quantifying with a satisfactory degree of preciseness the impact of each policy option in terms of enhanced compensation for injured parties. A qualitative assessment on the basis of multiple criteria is therefore the only feasible way of analysing and comparing the impact of each of the policy options.

67

EN

459  

See paragraph 0.

38

EN

460  Commission Staff Working Document Impact Assessment Report 6.1.1.

Benefits

a.

Ensuring full compensation of the entire harm suffered

103.

According to the Manfredi judgment of the Court, “it follows from the principle of effectiveness and the right of any individual to seek compensation for loss caused by a contract or by conduct liable to restrict or distort competition that injured persons must be able to seek compensation not only for actual loss (damnum emergens) but also for loss of profit (lucrum cessans) plus interest”.68 A legal system which does not allow for an effective exercise of the right to full compensation of the entire harm would therefore violate the EU principle of effectiveness.

104.

Since the objective is full compensation for the harm suffered, options will score lower on this criterion to the extent that they are likely to lead to under- or overcompensation of the injured party. An injured party is under- or over-compensated when it receives less or more compensation than the harm actually suffered. Since full compensation in the greatest possible number of cases is a primary objective, good scores on achieving the goal of full compensation will weigh heavily in the comparative impact analysis and in the identification of the preferred option.

b.

Protection of effective public enforcement

105.

Options will score higher to the extent that they optimise the interaction between public enforcement by the Commission and NCAs and actions for damages before national courts. In addition, options will score higher to the extent that they provide for an appropriate protection of leniency and settlement programmes, as well as of ongoing investigations by competition authorities. Since this is also a primary objective, good scores on achieving the goal of protecting effective public enforcement will weigh heavily in the comparative impact analysis and in the identification of the preferred option.

c.

Increased awareness, enforcement, deterrence and legal certainty

106.

Options will score higher to the extent that they make economic operators more aware of their rights and obligations under EU competition rules. Clear and explicitly formulated rules add to such awareness, just as much as they clarify the conditions applicable to claims for damages and the conditions of liability of offenders.

107.

Likewise, options will score higher to the extent that they widen the scope or increase the intensity of enforcement (by increasing the number of cases for which infringers are held responsible, by addressing different types of infringements or by involving a wider variety of economic operators), the likelihood of detection and of having to bear the financial consequences of anti-competitive behaviour and, assuming the optimum level has not yet been reached, the level of those financial consequences.

108.

A higher degree of awareness of the EU competition rules, combined with better enforcement by means of damages actions, contributes to greater compliance with these rules and, hence, to better achievement of their objective to ensure fair competition on the internal market.

109.

Options will score higher if they are likely to lead, all other things being equal, to an increase in deterrence rates. However, because increasing deterrence is not a primary

68

EN

Joined cases C-295/04 to C-298/04, Manfredi, [2006] ECR I-6619 95.

39

EN

Damages actions for breach of the EU antitrust rules (2013) 

461  

objective, in the final stage of comparing options and determining the preferred option less weight will be given to positive scores on deterrence, particularly compared with good scores on the primary objectives. Options will score lower on this category to the extent that they lead to a situation of over-deterrence, where the risk of damages claims prevents undertakings from engaging in lawful conduct or where damages are paid by undertakings that did not infringe the competition rules.

EN

d.

Access to justice

110.

Options will score higher on this criterion to the extent that they ensure more effective access to justice for all victims of an infringement of competition law, and as such, promote the right to an effective judicial protection as laid down in Article 47, first paragraph of the Charter and in Article 19(1), second subparagraph, TEU. That is particularly relevant for victims who have only small claims and/or have difficulties in gaining access to the evidence necessary to prove their case. Options will score well if they adequately address the reasons why victims who are willing to bring a damages action eventually decide not to do so.

111.

At the root of these reasons is the fact that victims fear the opportunity cost and the financial consequences of losing a meritorious case. More than the former factor, which is largely a matter of personal judgment, the latter is the key to ensuring adequate access to justice, e.g. by reducing the financial risks or the likelihood that meritorious cases will be lost. For instance, options that facilitate access to evidence will score higher. Options that would allow effective and efficient collective redress would also score higher.

e.

Efficient use of the judicial system

112.

Efficient use of the judicial system means avoiding unnecessary delays, multiple proceedings and contradictory outcomes. Options will score higher in so far as they avoid these undesirable situations. As the judicial system should not be overburdened, options will also score higher to the extent that they allow cases to be settled adequately out of court.

113.

Options will score lower on this criterion to the extent that they offer claimants excessive incentives encouraging them to bring damages actions although they have suffered no harm. Options that offer incentives to claimants who suffered minor harm are not considered abusive, even if the harm is outweighed by the litigation costs of the actions.

f.

A more level playing field

114.

Since actions for damages resulting from competition law infringements are put forward by the Court as a remedy stemming directly from EU law, claimants in such cases should all be able to use this remedy effectively. Article 47 of the Charter explicitly guarantees the right to an effective remedy to everyone whose rights and freedoms guaranteed by the EU law are violated. Although the effectiveness of the remedy may allow some divergence between Member States, it does require compliance with certain standards, which would have to be the same across Europe.

115.

Conversely, the dissimilar exposure to antitrust damages claims in the Member States may have an influence on the market behaviour of companies. They could, for instance, decide not to make use of the freedom of establishment, the free movement of goods or the freedom to provide services in those Member States where the right to compensation is most effectively enforced. For companies that have infringed the EU competition rules, there may be a competitive advantage for those that are

40

EN

462  Commission Staff Working Document Impact Assessment Report established or operate in a Member State with a less well functioning system of antitrust damages.

EN

116.

Effective redress, fair competition and the proper functioning of the internal market would therefore require comparable exposure to damages claims, which can be brought about only by similar basic procedural rules governing actions for damages. Options will thus score higher to the extent that they create a level playing field in Europe for claimants and defendants in antitrust damages actions.

g.

Providing benefits for consumers and SMEs

117.

This report also assessed the impact of the different options on consumers and SMEs. Options will score higher in so far as they provide more benefits to the situation of consumers and SMEs by increasing the effectiveness of their right to obtain full compensation if they are victims of an infringement of EU competiton law. Such effectiveness of the right to full compensation will be increased by, among others, reducing the costs consumers and SMEs have to incur in actions for damages and increasing their access to the necessary data to prove their claim.

h.

Stimulating economic growth and innovation

118.

The report assesses the impact of the different options on macro-economic variables, such as competitiveness, innovation, growth and jobs. This impact coincides largely with the expected level of future compliance with the competition rules. In particular, the more undertakings comply with the rules, the more competitive markets will be and the lower any allocative inefficiency. Further, it is considered that a balanced system, fostering both effective public and effective private enforcement of competition law, stimulates growth, productivity and innovation. Options that are more likely to achieve these results are therefore more likely to contribute positively to growth and employment. Such positive overall effects are likely to outweigh certain negative effects in those rare cases where the breach of competition law and the resulting public fines and liability for civil damages pose a financial threat to the survival of the infringing firm.

6.1.2.

Costs

a.

Litigation costs

119.

This broad category of costs covers both the litigation costs for parties to proceedings (both settlement costs and costs incurred when the case is brought to court) and the enforcement costs for public authorities (such as courts and competition authorities). Options will therefore score higher on these costs to the extent that they offer incentives to litigate and/or suggest measures that increase the costs (resources, opportunity cost and money) of litigation for the parties or for public authorities.

b.

Administrative burden

120.

This category includes costs incurred by businesses, consumers and public authorities in order to meet legal obligations to keep information and to provide it, either to public authorities or to private parties. Only the net costs are taken into account, i.e. excluding those that would be incurred anyhow, even without any legal obligation.

121.

Options will score higher to the extent that they require businesses, consumers or public authorities to keep information for a long period of time (storage costs) and that they impose an obligation to provide that information to one of the parties

41

EN

Damages actions for breach of the EU antitrust rules (2013) 

463  

(disclosure costs). Options will score high on disclosure costs to the extent that they have a low threshold triggering disclosure and/or wide scope of disclosure. c.

Error costs

122.

Error costs are costs related to the possibility of courts issuing a mistaken decision. That could take the form either of incorrectly awarding damages (type I error) or of incorrectly rejecting a claim for damages (type II error). Options will score high on error costs to the extent that they suggest measures that increase the likelihood of error and/or measures that amplify the impact of the error.

123.

Errors in quantification of damages are not included in this category because the resulting under- or over-compensation already has a negative impact on the corrective justice objective (see paragraph 104).

d.

Implementation costs

124.

Implementation costs means costs incurred by businesses, consumers and national public authorities to adapt to new rules (e.g. training costs, compliance costs, etc.). Although real, these costs are therefore transitory. Furthermore, implementation costs cover the costs for transposition as well as the costs that are brought about by incoherences in national legislation as a result of an EU legislative initiative. Options will score high on these costs to the extent that they lead to a big change in the regulatory context in which businesses, consumers and public authorities operate. Options will also score lower in so far as transposition costs are limited and the resulting coherence of national legislation is higher.

6.2.

Identifying and assessing the impact of each option

125.

The assessment presented in this report is based on the findings of the Impact Study, of the IAWP, of the Quantification Study and the consultations on quantification of antritrust harm and on a horizontal approach towards collective redress. This section sets out, in the form of tables, the conclusions of the Commission’s assessment of the likely positive and negative impacts that options 1 to 4 would have.

126.

Each option is assessed on its own merits, including Option 1, which is the baseline “no policy change” scenario. Option 3 is furthermore assessed in comparison to Option 2, of which it is a refinement. The reason for choosing this method is that, in the context of claims for damages for breach of the EU antitrust rules, it is important to illustrate the significant costs and limited benefits that taking no action at all would have. This approach makes it possible to compare the options with each other.

127.

The impact of the option against the baseline is summarised in the tables under the following scoring system: + + + / - - - Very positive / negative impact ++/--

Moderate positive / negative impact

+ / - Negligible positive / negative impact 0 128.

EN

No Impact

As the initiative has no significant environmental impact and no appreciable effects on trade with and investments by non-EU countries, these factors will not be further addressed in this impact assessment report. The below tables contain the assessment of the different policy options on the basis of the assessment criteria described in Section 6.1.

42

EN

464  Commission Staff Working Document Impact Assessment Report 6.2.1.

Option 1

129.

Option 1 entails the baseline scenario of zero action at EU level on antitrust damages actions. The assessment of the impacts of Option 1 therefore examines the status quo and any developments considered likely without EU action (prospective analysis). The likelihood of any such developments is difficult to foresee, even for the short and medium term. It should be borne in mind that Member States have been required to guarantee the effectiveness of the EU competition rules since the EC Treaty entered into force. In 2001 the Court explicitly called upon Member States to provide, in the absence of EU rules, for effective remedies under their civil law and procedure to safeguard the right of all victims to compensation for harm suffered as a result of infringements of Article 101 or 102 TFEU.

130.

Notwithstanding these obligations under EU law, to date very few Member States have taken action to improve the effectiveness of their legal framework for antitrust damages actions and there is no clear indication that a significant number of other Member States would adopt the necessary measures without any impetus from EU level. Moreover, in those Member States where measures have been taken or are being contemplated, those measures are rarely covering all existing obstacles to an effective redress system for victims of a competition law infringement.69 The Commission’s analysis of various likely impacts of the option of zero action at EU level is based on the probably still conservative (i.e. optimistic) assumption that some Member States would adopt a number of measures to improve the effectiveness of antitrust damages actions in Europe, whereas currently only the UK envisages a relevant initiative.

131.

In the Pfleiderer case, the Court ruled that it is for a national court to determine on a case-by-case basis and according to national law the conditions under which disclosure of leniency-related information must be permitted or refused. It is thus clear that in the absence of EU rules on this issue, legal uncertainty and a potential uneven playing field between Member States may develop with regard to the balance to be struck between the public and private enforcement of competition law. When choosing Option 1, this legal uncertainty would persist.

Table 2: Benefits of Option 1 Benefits achieved/ problem addressed

Impact compared to base-line (0 to +++)

Explanation of rating and aspects of the policy option most relevant to the benefits

1. Ensuring full compensation of the entire harm suffered

0

Under-compensation of many victims likely to remain: obstacles to bringing actions and to successfully proving the conditions for compensation highly likely to persist in most MS • consequently, no guarantee that all EU citizens will enjoy a certain minimum level of protection of their right to antitrust 70 damages

2. Protection of effective public enforcement

0

Without a legislative measure at EU level, the balancing exercise on the interplay public/private enforcement is left to national courts. There are risks of fragmentation and a lower overall level of protection of public enforcement in the EU. This might endanger the effectiveness of certain public enforcement tools, such as leniency and settlement programmes

69 70

EN

See paragraph 3.2 above for more information on this subject. See, in particular, Impact Study, Part III, section 2.1 and Part II, sections 1.6.1, 2.2.1, 3.2.1 and 4.2.1.

43

EN

Damages actions for breach of the EU antitrust rules (2013) 

465  

3. Increased awareness, enforcement, deterrence and legal certainty

0

Continued under-deterrence and lack of legal certainty: no general increase in deterrence rates and compliance by means of private actions for damages, at best a very small increase in some Member States • clear underdeterrence, especially for cartels, in most MS • uncertainty resulting from the differences between the national legal systems will persist • no increase of awareness at European level, at best only in some MS • without a minimum level of protection, victims of antitrust infringements would have no clear picture of their basic right under EU law to damages in the MS • the lack of a common approach to the interaction with public enforcement (e.g. protection of leniency related information) creates a serious risk that inconsistencies or even loopholes might hamper public enforcement • the uncertainty for leniency applicants would continue to exist

4. Access to justice

0

In most MS, access to justice, in particular for SMEs and consumers, remains problematic: even though some measures may be taken in a few MS (for example on collective redress), access to justice in antitrust damages cases will continue to be difficult, as in many MS significant changes are unlikely, especially in the following areas: access to evidence, collective 71 redress and the passing-on defence

5. Efficient use of the judicial system

0

Current inefficiencies remain in most MS, except maybe improvement as a result of better collective redress in certain MS72

6. A more level playing field

0

Substantial differences remain between MS: highly unlikely that in the absence of any EU action businesses would compete across Europe on an equal footing • highly unlikely that citizens and undertakings can enforce their rights conferred by the Treaty throughout the EU in an similar manner • no indication that a virtuous “mutual learning” process or a “race to the top” in the form of competition between legal orders would stimulate adoption of similar best practices across the EU • the current fragmentation of the legal framework for damages actions could even become wider in a few years if some MS 73 enact significant reforms while others maintain the status quo

7. Positive impact on SMEs and consumers

0

Exercise of right to damages would remain very difficult: SMEs and consumers are likely to suffer particularly from the persisting legal uncertainty74

8. Stimulating economic growth and innovation

0

Negligible contribution

some

75

Table 3: Costs of Option 1 Costs

Impact compared to base-line (0 to - - -)

Explanation of rating and aspects of the policy option most relevant to the costs

1. Litigation costs

0

Litigation costs may be significant in individual cases, but relatively low overall and total costs might increase slightly: although exercising the right to damages is currently costly (especially due to lack of widespread effective collective redress, legal uncertainty, difficult access to accurate evidence, lack of binding force and differences between legal systems), the level of litigation is likely to remain low • number of cases may increase slightly, if some MS facilitate damages actions

71 72 73 74 75

EN

See, in particular, Impact Study, Part III, section 2.1 and Part II, sections 1.6.1, 2.2.1 and 3.2.1. See, in particular, Impact Study, Part III, section 2.1 and Part II, sections 2.2.1 and 3.2.1. See paragraph 3.2 above as well as Annex 3 for more information on this subject. See, in particular, Impact Study, Part III, section 2.1.1. See, in particular, Impact Study, Part III, section 2.1.2.

44

EN

466  Commission Staff Working Document Impact Assessment Report but if certain MS improve collective redress, costs for claimants may decrease as a result of efficiencies (compared with individual actions)76 2. Administrative burden

0

Relatively small burden in most MS, but may increase slightly: relatively small number of companies concerned due to low number of cases • in most MS, currently limited information/disclosure obligations on other party • in the not very likely event that some MS increase the scope of disclosure and limitation periods in antitrust damages cases, administrative burden would slightly increase77

3. Error costs

0

Little litigation, some errors will persist, but total number moderate: some errors likely due to lack of effective access to evidence and, hence, access truth and absence of learning effects • given low level of litigation, total number of errors not high but some MS might possibly introduce rules for greater accuracy in fact-finding (although not very likely) • to the extent that the number of cases increases, courts will grow more familiar with antitrust cases and avoid errors78

4. Implementation costs

0

No implementation costs, in the absence of any provisions at EU level

6.2.2.

Option 2

132.

The second option is to codify the White Paper's suggestions, that were identified as the preferred option in the IAWP.79 In the below tables, the measures proposed by Option 2 will be scored in comparison to the baseline scenario of no EU action in the field (Option 1). For stakeholder views and other evidence on the different measures introduced by Option 2, reference is made to Section 7.1 and Chapter 3 of this report, as well as Annex 8.

Table 4: Benefits of Option 2 Benefits achieved/ problem addressed

Impact compared to base-line (0 to +++)

Explanation of rating and aspects of the policy option most relevant to the benefits

1. Ensuring full compensation of the entire harm suffered

+++

Overall, a more effective compensation mechanism for all victims: clear but measured facilitation of damages claims (especially by means of disclosure of categories of evidence, opt-in group and representative actions and non-binding guidance on the damages quantification) is likely to lead to an increase in the number of victims compensated • in many MS, disclosure of categories of evidence will improve the likelihood of proving liability and quantification of the full actual harm suffered • facilitation, for indirect purchasers, of proof of passing-on makes compensation of such victims more likely • passing-on defence is in line with the compensation objective, insofar as it endeavours the allocation of compensation to the level where the harm has effectively been suffered • availability of both opt-in group and representative actions remove the often unfavourable cost/benefit ratio and make recovery of scattered damage in the field of competition more likely • rules on limitation periods ensure that there is enough time to bring an action for the whole harm in case of repeated or continuous infringements and infringements of which the victim cannot reasonably have knowledge • limitation of liability for immunity recipient does not entail a risk of reduction of the available compensation, e.g. in case of insolvency, as it does not apply if

76 77 78 79

EN

See, in particular, Impact Study, Part III, section 2.1 and Part II, sections 1.6.1, 2.2.1 and 4.2.1. See, in particular, Impact Study, Part III, section 2.1 and Part II, sections 1.6.1 and 4.2.1. Ibidem For a detailed description, see Section 0.

45

EN

Damages actions for breach of the EU antitrust rules (2013) 

467  

the compensation cannot be obtained from other infringers • no overcompensation but some risk of under-compensation where cases are settled (at a lower amount than the actual harm) • single damages and requirements to fulfil in order to obtain a disclosure order might play as disincentives • no binding legislative measures on costs pose a strong risk of lack of effective action at the national level. 2. Protection of effective public enforcement

++

Increase in protection of effective public enforcement: protection of corporate statements safegurds protection of leniency programmes • limitation of civil liability for the immunity recipient keeps the incentives high to apply for leniency but no effective protection of ongoing investigations of the Commission or NCAs

3. Increased awareness, enforcement, deterrence and legal certainty

+++

Overall, increase in deterrence rates, enforcement and legal certainty: increase in number of damages awards, especially by means of opt-in group and representative actions, plus more effective disclosure rules improve enforcement and, thereby, deterrence rate • increase in the number of victims compensated adds to this effect • rules on limitation periods provide for effective enforcement in most cases while avoiding legal uncertainty • no risk of over-deterrence • significant, comprehensive clarification of conditions for exercising the right to damages and for liability of companies • increased legal certainty on burden of proof for passing-on of overcharges but no incentives (e.g. multiple damages) to stimulate market players to monitor and detect infringements • limitation of liability of the immunity recipient might slightly decrease deterrence of actions for damages for the immunity recipient, but increases the deterrence through actions for damages for other infringers and thus the overall deterrence in cartel cases

4. Access to justice

+++

Overall, broader access to justice especially for indirect purchasers: compensation of greater number of victims (including those who suffered scattered damage) • opt-in group actions and representative actions improve access to justice for low-value small claims • facilitation of pass-on makes proof of damage more likely • in terms of access to evidence, significant improvement in several MS, as disclosure of categories of evidence possible and as initial fact-pleading threshold is adapted to circumstances of each case and protection of corporate statements does not negatively affect access to justice as all pre-existing documents are still available for disclosure • rules on limitation periods, especially on limitation periods in follow-on cases, allow proper access to justice but threshold to obtain disclosure could make it difficult to obtain evidence in some cases • limitation of liability of the immunity recipient may marginally affect access to justice, even if injured parties can still claim compensation from immunity recipients if they cannot obtain it in full from the other cartel participants • no measures on costs regimes could result in no action at Member States level, thus not completely removing one of the obstacles to effective access to justice

5. Efficient use of the judicial system

+++

Overall, significant improvement of efficiency: binding effect of NCA decisions and fault presumption enhances efficiency in follow-on cases • opt-in group/representative actions allow some measure of aggregation of small claims • disclosure is not likely to lead to abuses because of ex ante judicial control, especially of proportionality but no rules on costs entails a strong risk of lack of implementation at national level • in the absence of introduction of opt-out actions, Option 1 may still lead to quite a number of individual claims concerning the same infringement.

6. A more level playing field

EN

+++

Similarly effective protection of right to damages across the EU: more level playing field for consumers and businesses alike.

46

EN

468  Commission Staff Working Document Impact Assessment Report 7. Positive impact on SMEs and consumers

+++

SMEs and consumers are likely to benefit from facilitation of damages claims, especially from improvements in their evidentiary position and from the introduction of opt-in group and representative actions but slight risk that the 'loser pays' rule will discourage SMEs and consumers in low-probability cases or, due to the rules on disclosure, claimants who do not possess much evidence at the outset

8. Stimulating economic growth and innovation

++

Second pillar of enforcement likely to emerge • push for more competitive markets with likely positive effects on growth and employment • very low risk of excessive litigation leading to a deteriorating business environment • positive effects on SMEs and hence for economic growth but the safeguards against abusive proceedings and the lack of measures on costs could still discourage victims from claiming compensation

Table 5: Costs of Option 2

EN

Costs

Impact compared to base-line (0 to - - -)

Explanation of rating and aspects of the policy option most relevant to the costs

1. Litigation costs

--

Overall, moderate increase in total litigation costs and per average case: slight increase in number of lawsuits (less than the number of victims compensated due to collective redress mechanisms) • costs per claimant per case may even decrease due to efficiencies produced by collective redress mechanisms, even though they may entail new costs for the courts • in MS where evidence disclosure is currently uncommon, increase in burden on courts • binding effect of NCA decisions across EU allows concentration of damages claims in multi-state cases in one court and avoids re-litigation • presumption of fault reduces parties’ costs in follow-on claims • pass-on defence may have little impact on length of procedure, but costs are borne by defendant invoking it • early disclosure may stimulate cost-efficient early settlements

2. Administrative burden

--

Overall, a moderate impact: slight increase in number of lawsuits and broader disclosure than currently exists in several MS lead to slightly more screening and production of documents • in some MS longer record-keeping obligations due to possibly longer limitation periods than baseline

3. Error costs

-

Number of errors may increase slightly, if at all: as the number of cases increases slightly, so may the total number of errors, but no indication that overall statistical incidence of errors would increase • binding effect of NCA decisions may make errors less likely in follow-on cases • greater accuracy in fact-finding • the 'loser pays' rule stimulates selection of meritorious cases and prevents frivolous suits

4. Implementation costs

--

Moderate implementation costs: some measures under this option (especially disclosure rules, limitation of liability of the immunity recipient, binding effect and collective redress mechanisms) require changes in the law of several MS. Especially the implementation of sectoral collective redress mechanisms leads to costs. • need for training of judges and the legal community • none of the changes required raises major public policy concerns • all one-off costs, except for the training of judges

6.2.3.

Option 3

133.

Compared to Option 2, the legislative initiative envisaged under Option 3 mainly proposes further safeguards to access to evidence, introduces a rebuttable presumption of harm caused by a cartel and proposes measures to facilitate

47

EN

Damages actions for breach of the EU antitrust rules (2013) 

469  

consensual dispute resolution.80 The scores attributed to Option 3 in the below tables are based on a comparison of this option with the baseline option of no EU action in the field (Option 1). However, for the sake of clarity and in order to avoid repeating the same arguments as given in Tables 4 and 5 in relation to Option 2, the explanation of the rating is presented also in comparison to Option 2. For stakeholder views and other evidence on the different measures introduced by Option 2, reference is made to Section 7.1 and Chapter 3 of this report. Table 6: Benefits of Option 3 Benefits achieved/ problem addressed

Impact compared to base-line (0 to +++)

Explanation of rating and aspects of the policy option most relevant to the benefits

1. Ensuring full compensation of the entire harm suffered

++

Lower than under Option 2: similar incentives and measures aimed at facilitating damages actions • the rebuttable presumption on the existence of harm and the introduction of the principle of effectiveness as regards quantification increase the possibility to obtain compensation • the strong protection of public enforcement may lead to more cartel cases and thus to more follow-on damages actions • the facilitation of consensual dispute resolution increases the possiblities for out-of-court settlements, but additional safeguards for evidence disclosure might have an adverse impact on the use of such procedural tools • contrary to Option 2, no competition specific collective redress is introduced

2. Protection of effective public enforcement

+++

Higher than under Option 2: strong protection of the file of the competition authority regarding corporate statements and settlement submissions. • access to other documents in the file of the competition authority only after the closure of the proceedings provides increased protection of ongoing investigations of the Commission and NCAs

3. Increased awareness, enforcement, deterrence and legal certainty

+++

The same as under Option 2: the rebuttable presumption on the existence of harm and the introduction of the principle of effectiveness as regards quantification increases the chance that victims can succesfully claim damages

4. Access to justice

+++

but the absence of collective redress mechanisms in Option 3 makes that the increased damages awards as expected under Option 2 through the introduction of collective redress would not be realized. The same as under Option 2: • the rebuttable presumption on the existence of harm and the introduction of the principle of effectiveness as regards quantification lowers victims' hesitance to initiate actions for damages • the facilitation of consensual dispute resolution increases the possibilites of a outof-court settlement and thus of obtaining fair compensation within a short time period, but the additional safeguards for evidence disclosure might have an adverse impact on the use of such procedural tools • contrary to Option 2, no competition specific collective redress mechanism is introduced

5. Efficient use of the judicial system

++

Lower than under Option 2: • no risk of burdening the courts with unfounded or otherwise abusive claims because no sectoral collective redress mechanism as under Option 2 is introduced• stimulating fair out-of-court settlements reduces the burden on the judiciary but in the absence of introduction of collective redress mechanisms Option 3 may lead to more individual claims concerning the same infringement

80

EN

For a detailed description, see section 0.

48

EN

470  Commission Staff Working Document Impact Assessment Report 6. A more level playing field

+++

The same as under Option 2: more level playing field for both infringing undertakings and injured parties

7. Positive impact on SMEs and consumers

++

Lower than under Option 2: in comparison to Option 2, the evidentiary position of SMEs and consumers is better under Option 3, due to the introduction of the rebuttable presumption of harm in cartel cases. but the protection of the file of competition authorities may lead to more difficulties to obtain access to evidence • no collective redress mechanisms

8. Stimulating economic growth and innovation

++

The same as under Option 2

Table 7: Costs of Option 3 Costs

Impact compared to base-line (0 to - - -)

Explanation of rating and aspects of the policy option most relevant to the costs

1. Litigation costs

-

Overall, slight increase in total litigation costs and per average case; though less costs than in Option 2: the fact that there is no collective redress under Option 3 leads to less actions being brought and thus to lower litigation costs • rebuttable presumption of harm in cartel cases saves costs • the facilitation of quantification of the harm reduces the administrative burden for courts and increases the likelihood of out-of-court settlements

2. Administrative burden

-

Lower than under Option 2: the protection of public enforcement will reduce the administrative burden that damages actions could cause on competition authorities

3. Error costs

0/-

Lower than under Option 2: due to the fact that no collective redress mechanisms are introduced, less actions will be brought and error costs will be lower, if any.

4. Implementation costs

-

Lower than under Option 2: implementation costs are expected as a result of introducting collective redress mechanisms; as those mechanisms are absent under Option 3, implementation costs are expected to be lower.

6.2.4.

Option 4

134.

Option 4, the 'non-regulatory approach', identifies a range of useful solutions and best practices, mainly from the experience in Member States, which the Commission would recommend to all Member States for implementation in their legal systems. The impact assessment of this non-regulatory approach is based on the assumption that the specific measures that in Option 3 are suggested as legislative measures would in Option 4 only be recommended to the Member States.81

135.

Predicting the impact of mere recommendations is a delicate exercise, as it will to a large extent depend on the Member States. The likelihood that they will implement legislative changes on the basis of recommendations is difficult to foresee and would mainly depend on a persuasive effect of the recommendations, next to a more general awareness raising effect. Although some literature advocates the use of soft law and recommendation of best practices to achieve broad policy objectives at the EU level, there is no compelling evidence that the persuasive effect has ever been strong enough to create a high likelihood that EU Member States would change their

81

EN

See paragraph 5.3.3 above for a description of those measures.

49

EN

Damages actions for breach of the EU antitrust rules (2013) 

471  

existing national laws, particularly not in areas of law such as civil liability and civil procedure.82 Also, the publication of the 2005 Green Paper and the 2008 White Paper and the discussions around these and other policy documents of the Commission, while having some impact on the debate in the Member States, has so far not led to significant changes in the statutory rules prevalent in the Member States. Further, the Impact Study confirms that there is little likelihood that the effects of Option 4 would add up to much more than those of the baseline scenario of Option 1, under which no EU action is foreseen.83 136.

The shortcomings of a non-regulatory initiative could neither be avoided if in parallel the Commission were to start a series of infringement proceedings under Article 258 TFEU to ensure compliance with the acquis communautaire on this subject. Such actions would remove existing obstacles on a case by case basis and would only foster change of provisions that do not comply with the principle of effectiveness. They would not indicate which measures can better ensure effective compensation.

137.

The rating of various impacts of Option 4 set out below reflects both effective and rather ineffective persuasive effect. The rating is put in a range, from the hypothesis that all Member States follow the recommendations to the hypothesis that a few or no Member State will follow the recommendations. The scores attributed as such entail an independent comparison of Option 4 with the baseline scenario of no action at EU level (Option 1). However, Option 4 is necessarily interconnected with both the baseline scenario and with Option 3, to which its measures are identical. If few or no Member States decide to follow the measures recommended under Option 4, its effects coincide with that of the baseline scenario of no EU action (Option 1). Therefore, the lower end of the range of scores attributed to Option 4 corresponds with the likely impacts of Option 1. For the explanation of these scores, reference is made to Tables 2 and 3 above. As explained in the previous paragraphs, the adoption of Option 4 is more likely to lead to the effects at the lower end of the range rather than those at the higher end of the range.

138.

If, on the other hand, all Member States decide to follow the recommendations, the effects of Option 4 could be close to those of Option 3, where the same measures are laid down in a legislative proposal. However, account should be taken of the fact that even if all Member States implement the recommendations, divergent choices will be made among the Member States as to the exact measures to be taken, leading to reduced legal certainty, lower efficiency of judicial systems and access to justice as well as a potential increase in litigation costs.84 Therefore, the upper end of the range will not necessarily equal the score assigned to Option 3. In order to avoid a repetition of arguments, reference is made to tables 6 and 7 on the assessment of option 3: in the Tables 8 and 9 below it is explained why the upper end of the range is lower than the score of Option 3, where applicable.

Table 8: Benefits of Option 4

82 83 84

EN

See, e.g., Impact Study, Part III, section 2.5. See Impact Study, Part III, section 2.5.1 and Table 79. The risks related to fragmented national legislations are explained in Sections 3.2 and 3.3 above.

50

EN

472  Commission Staff Working Document Impact Assessment Report Benefits achieved/ problem addressed

Impact compared to base-line (0 to +++)

Explanation of rating and aspects of the policy option most relevant to the benefits

1. Ensuring full compensation of the entire harm suffered

0/+

The upper end of the range is lower than under Option 3 because if the implementation is left to the Member States on a voluntary basis, the resulting measures are likely to diverge to some extent. This divergence may affect the willingness of victims to seek redress and may thus reduce the likelihood that full compensation is obtained in all Member States

2. Protection of effective public enforcement

0/+

Without a legislative measure at EU level, the balancing exercise is left to national courts on the basis of national law: this risks leading to fragmentation of the applicable rules. Given that the Commission and NCAs jointly enforce the EU antitrust rules through the ECN, a soft law approach would fail to achieve the legal certainty necessary to safeguard effective public enforcement

3. Increased awareness, enforcement, deterrence and legal certainty

0/+

The upper end of the range is lower than under Option 3 because of lack of legal certainty and awareness in cross border cases, caused by likely divergence between Member States

4. Access to justice

0/+

The upper end of the range is lower than under Option 3 because the likely divergent options in the different Member States may discourage victims to seek redress in other Member States than their own

5. Efficient use of the judicial system

0/+

The upper end of the range is lower than under Option 3 because of the likely divergence in rules between Member States

6. A more playing field

level

0/+

Differences between Member States may even become wider if they act differently on the recommendations85

7. Positive impact on SMEs and consumers

0/+

The upper end of the range is lower than under Option 3 because diverging rules in the Member States may cause additional difficulties for victims to initiate proceedings in other Member States

8. Stimulating economic growth and innovation

0

The persistent divergence of national rules does not entail significant benefits for the internal market.

Table 9: Costs of Option 4 Costs

Impact compared to base-line (0 to - - -)

Explanation of rating and aspects of the policy option most relevant to the costs

1. Litigation costs

0/--

The upper range is higher than under Option 3 because divergent measures in the Member States produce higher litigation costs in cross-border cases

2. Administra-tive burden

0/-

See explanations in the tables on Options 1 and 3 for the ratings in this table

3. Error costs

0/-

See explanations in the tables on Options 1 and 3 for the ratings in this table

85

EN

The experiences since the White Paper, as described in Section 3.2 and Annex 3 of this report show that Member States are expected to act differently on recommendations.

51

EN

Damages actions for breach of the EU antitrust rules (2013)  4. Implementation costs

EN

0/-

473  

See explanations in the tables on Options 1 and 3 for the ratings in this table

52

EN

474  Commission Staff Working Document Impact Assessment Report 7.

COMPARING THE OPTIONS

7.1.

Comparing the policy options and assessment of the preferred option

7.1.1.

Summary comparison of the options and identification of the preferred option

139.

In Chapter 6, the four policy options have been scored on the basis of a number of different criteria. In order to determine the preferred option, this section will provide a short comparison between the options. A preference for binding EU action

140.

It is clear from the above assessments that options envisaging EU legislative action (Options 2 and 3) are preferred over options that do not envisage EU legislative action (Options 1 and 4). In this respect, whereas Option 4 (recommendations and good practices) envisages some EU-action and Option 1 entails the baseline scenario of zero action at EU level, the Impact Study confirms that there is little likelihood that the effects of Option 4 would add up to much more than those of Option 1.86 Even if under Option 4 some progress is made towards achieving the policy objectives of this initiative, the shortcomings of Options 1 and 4 in comparison to Options 2 and 3 largely coincide.

141.

The main issue related to Options 1 and 4 is that the current broad divergence between national legislations relating to actions for damages would persist.87 This would be problematic in terms of the effectiveness of damages actions due to divergent national legislations. As to Option 1, this is confirmed in the Impact Study, which finds that the level of corrective justice would be very low, while deterrence may increase only very slightly over the next few years.

142.

The broad divergence in legislation is described in more detail in Section 3.2 above. It is equally explained why no significant change in the situation of divergent national provisions is to be expected if no EU action in the field of antitrust damages actions is taken (the baseline scenario, Option 1). Furthermore, on the basis of the experience after the White Paper (also further described in Section 3.2), it cannot be expected that soft law at the EU level would significantly change the current situation. After the White Paper, very few Member States have taken measures to improve the effectiveness of antitrust damages actions; the vast majority of Member States have not taken any action whatsoever. Moreover, these measures concerned only specific elements of the White Paper's suggestions and varied largely among Member States. There are no indications that adoption of yet another act of EU soft law on the subject would significantly change this situation.

143.

As a result, both Options 1 and 4 would cause the internal market to remain fragmented in terms of the level of judicial protection due to the uneven playing field and, as a result, prone to forum-shopping (this problem is further described in Section 3.3 above). Also, divergence between national legislations would cause significant burdens to be borne by SMEs and consumers, who would continue to

86 87

EN

See Impact Study, Part III, section 2.5.1 and Table 79. Whereas the recommendations of Option 4 might lead to more Member States taking certain measures than would be the case under Option 1, it would most likely still lead to a broad divergence in national legislations as each Member State would be free to choose in relation to which issues it would like to take measures and the content of such measures.

53

EN

Damages actions for breach of the EU antitrust rules (2013) 

475  

suffer from inadequate access to justice, in particular in cross-border cases.88 Finally, there would be little or no positive macro-economic impact. 144.

Furthermore, when taking no EU action as foreseen under the baseline scenario (Option 1), the uncertainty following the Pfleiderer ruling of the Court as to the use of documents from the file of a competition authority in actions for damages would persist. This uncertainty and its potential consequences are described in Section 3.1 and Annex 2 of this report. The legal uncertainty may be a negative factor having an impact on the public enforcement of EU competition law, in particular on the competition authorities' leniency programmes.

145.

As described above, experience shows that EU soft law in the field of antitrust damages actions has not had the effect that Member States take adequate measures. Quite on the contrary, hardly any action is taken and where measures were adopted, these varied widely from one Member State to another. Such inconsistencies are also emerging in the case-law on disclosure of documents from the file of a competition authority in actions for damages. There is therefore no indication that soft law would be sufficient for all Member States to take adequate measures to protect effective public enforcement. The need to regulate the interaction between the public and private enforcement of competition law, thus ensuring the effectiveness of both systems, would therefore not be fulfilled when choosing Options 1 or 4.

146.

In conclusion, even though under Options 1 and 4, the administrative burden, litigation costs and error costs would be likely to remain at a relatively low level (all due to the quasi-absence of damages actions), the achievement of the policy objectives would be very limited and important problems, with regard to the protection of effective public enforcement as well as in relation to the effective private enforcement, would persist. A preference for a separate, but horizontal approach to collective redress

147.

The preference for an option including EU legislative action points to Option 2 or Option 3. Before discussing and comparing the benefits of these Options, it is necessary to assess the main difference among the measures they contain, namely the provision of a competition-specific system of collective redress.

148.

In the public consultation on the White Paper, consumer associations have expressed themselves in favour of both opt-in group actions and respresentative actions, in order to enhance the opportunities to effectively obtain compensation. Collective redress mechanisms have been seen as essential mechanisms in this respect. Businesses and business associations, however, generally fear that collective actions will only increase litigation to the detriment of business. The results of the 2011 public consultation of collective redress show the same image: consumers are in favour and urgently call for EU action in the field, whereas business generally holds that there is no need/justification for EU action on collective redress. Among Member States, the views are divided, whereas the five national competition authorities that responded are all in favour of introducing collective redress.

149.

Option 2 contains two complementary means of collective redress which would be specific to actions for damages for infringements of EU competition law. Option 3 contains no such means, referring instead to horizontal initiatives on collective redress. While Option 2 is clearly more advantageous in providing immediate means

88

EN

See Impact Study, Part III, section 2.1.

54

EN

476  Commission Staff Working Document Impact Assessment Report of redress to victims such as consumers and SMEs, it is also necessary to assess this measure against the more general objective of the Commission to foster a coherent EU policy on collective redress, 150.

Taking into account the results of the public consultation and in particular the position taken by the Parliament in its Resolution of 2 February 2012,89 a separate horizontal approach appears currently more appropriate than a competition-specific solution. The main reason is that the field of competition law is not the only field of EU law in which scattered harm frequently occurs and in which it is difficult for consumers and SMEs to obtain damages for the harm they suffered. Similar problems (e.g. high litigation costs in comparison to the individual damage) exist also in other fields of law, such as consumer law or environmental law. Therefore, the basic principles applicable to collective redress can, to a large extent, be common to all these fields of law.

151.

A horizontal instrument may also avoid unnecessary fragmentation of national civil laws and provide consumers and SMEs with a mechanism to obtain effective redress through collective actions in all fields where this is considered necessary. As such, a horizontal intiative would ensure coherence on collective redress across all fields of law at the EU level. However, this does not affect the need for the Commission to take characteristics specific to a certain field of law into account. In so far as specific provisions are considered necessary in relation to – for example – the field of competition law, these provisions could be laid down in a separate chapter of such horizontal instrument or in subsequent separate legal instruments. A preference for a more balanced system of public and private enforcement

152.

While the preference for a coordinated approach to collective redress points at Option 3, it is also necessary to assess the benefits and costs of the other specific measures of the options that include legislative action. As explained in section 5.3.2. above, Option 3 is to a large extent based on Option 2, but in addition optimises the interplay between public and private enforcement by regulating access to evidence in the file of a competition authority. Option 3 also considers to facilitate the claimant's burden of quantifying antitrust harm by introducing a rebuttable presumption of harm in cartel cases. It also introduces measures to further stimulate consensual dispute resolution. On the basis of the above, it must be concluded that both Option 2 and Option 3 fulfil to a large extent the policy objectives. Both options effectively address the main obstacles that currently hinder effective redress for victims of antitrust infringements, building on European legal traditions. In comparison to the baseline scenario of no EU action in the field (Option 1), both Option 2 and Option 3 take measures to improve the effectiveness of the private enforcement of competition law. These measures will be discussed below.

153.

First of all, both Option 2 and Option 3 improve the possibilities for claimants to obtain access to the evidence they need to substantiate their claim, by providing for rules on disclosure of specified categories of evidence, based on fact-pleading and proportionality. In the public consultation on the White Paper, respondents from the consumer side, a number of law-firms and even business respondents that had allegedly been victims of anticompetitive conduct, identified the limited possibilities as regards disclosure as one of the main obstacles standing in the way of effectively exercising the EU right to full compensation. These respondents encouraged the

89

EN

See footnote 24.

55

EN

Damages actions for breach of the EU antitrust rules (2013) 

477  

Commission to go further than the system provided for in the White Paper (and proposed by Options 2 and 3), by loosening the conditions to obtain a disclosure order. Nevertheless, respondents from the business side considered the rules from the White Paper not to be restrictive enough and asked for further measures to avoid 'fishing expeditions' and to protect confidential information. Options 2 and 3 choose the middle ground between the opinions from the consumer-side and those of the business-side, providing for a system that significantly improves the possibilities for victims to obtain the necessary evidence, without being overburdensome or costly for the defendants.

EN

154.

The disclosure rules proposed under Option 2 and 3 will constitute a significant improvement for the effectiveness of private enforcement in almost all Member States. Nevertheless, as the rules in Options 2 and 3 would be minimum standards, those Member States having a more far reaching system could maintain this system. For further information on the rules applicable to disclosure in the Member States, reference is made to section 3.2.

155.

Options 2 and 3 both allow standing for indirect purchasers, in order to make sure compensation is awarded to the entities who actually suffered damage. This rule is endorsed by almost all stakeholders; only 8 of the respondents to the public consultation of the White Paper suggested to limit standing to direct purchasers. Allowing standing of indirect purchasers is in conformity with the applicable law in the vast majority of Member States.

156.

In connection with this, the passing-on defence is available for the infringer that shows that the damages claimant has passed-on part or the whole of the illegal overcharge. Nevertheless, as the proof of passing on is hard for both parties, it is considered fair to put the burden of proof on the party infringing competition law by introducing a rebuttable passing-on presumption in favour of the indirect purchaser.

157.

In the public consultation on the White Paper, the availability of the passing-on defence has generally been supported by all respondents. Furthermore, as to the facilitation of the burden of proof of passing on, consumer association and other nonbusiness respondents held that whereas it is difficult for both the claimant and the defendant to prove the pass on, and this is thus an important obstacle for indirect purchasers to obtain compensation, the statutory simplification of the burden of proof should benefit the victims, otherwise it would result in the infringer taking advantage of its illegal conduct, and retaining the illegal overcharge. Businesses and business associations have generally expressed reservations against such facilitation in the burden of proof, because of the risk of double or multiple compensation. In view of the fact that the risk of undercompensation is currently deemed more important than the risk of overcompensation, as well as the fact that indirect purchasers would have significant difficulties to prove passing-on, Options 2 and 3 introduce a facilitation on the burden of proof.

158.

As discussed in section 3.2 above, the status of the passing-on defence is unclear in most Member States due to lack of legislation and case-law on the topic, even though there are no legal obstacles to such defence. As to the burden of proof of the passingon defence in those Member States where it is allowed, different national systems exist (see section 3.2 above). The rules provided for in Options 2 and 3 therefore constitute a necessary harmonisation of the national rules, and a facilitation of actions for damages.

56

EN

478  Commission Staff Working Document Impact Assessment Report

EN

159.

Option 2 and 3 both facilitate follow-on actions by providing for rules on limitation periods after the finding of infringement by a competition authority or a review court and for binding effect for the final infringement decisions of NCAs.

160.

In the public consultation on the White Paper, potential claimants have expressed themselves in favour of the introduction of such rules on limitation periods, whereas some business associations and law-firms argued that a new limitation period for follow-on actions would jeopardize legal certainty. However, none of the respondents have contested that clarifying rules on limitation periods would have a positive effect on the possibilities to obtain full compensation.

161.

Currently, only 6 Member States provide for specific limitation periods for follow-on cases. These limitation periods differ in system (suspension of ordinary limitation period or separate limitation period) and in duration. The other 21 Member States, however, do not provide for such limitation periods at all, which may hinder the possibility to bring follow-on actions for damages because the limitation period has already expired. In other Member States this is corrected by case-law, which sets the starting point of the ordinary limitation period at the moment the infringement decision has become final. Therefore, the rules proposed by Options 2 and 3 as regards limitation periods would constitute a significant improvement in most Member States.

162.

The binding effect of NCA decisions has been one of the more controversial issues in the public consultation on the White Paper. The main criticism is that such rule is running counter to the independence of the judiciary or that it constitutes a potential violation of the rights of defence. As to decisions of NCAs from other Member States, some stakeholders pointed at different standards among authorities across the EU. Nevertheless, it should be noted that a significant number of respondents (among which consumer associations, law firms and other respondents from jurisdictions where a similar provision already exists) recognize the usefulness of this measure not only for damages claimants, but also for the efficient and consistent application of competition rules.

163.

National legislation on binding effect of NCA decisions is at the moment very divergent: there is one Member State where both its own and other NCA decisions have binding effect, 10 Member States in which the final decisions of their own NCA have binding effect but those of other NCAs do not and 16 Member States where NCA decision do not have binding effect. In those Member States their evidential value ranges from a rebuttable evidentiary presumtion to constituting merely "a view" on the facts and the law. The rule on binding effect of NCA decisions proposed by Options 2 and 3 would thus constitute an important means of harmonising national legislation and making follow-on actions for damages more effective.

164.

Both options provide for measures improving the interplay between public and private enforcement in comparison to the baseline scenario (Option 1). Such measures include an enhanced protection against disclosure in actions for damages for documents from the file of a competition authority and a limitation of liability of successful immunity applicants. In the public consultation on the White Paper, respondents (apart from consumer associations) generally deemed the protection of corporate statements from disclosure to be an essential conditions for the success of leniency programmes and thus an important guarantee for effective public enforcement of EU competition law. The limitation of liability of immunity

57

EN

Damages actions for breach of the EU antitrust rules (2013) 

479  

recipients is a debated topic in the public consultation on the White Paper. Nevertheless, it is generally considered that the measure is a positive incentive to apply for leniency, and thus to have a positive effect on public enforcement. However, some respondents thought that the measure would provide an excessive benefit for the immunity recipient.

EN

165.

Both options provide safeguards avoiding abuse of litigation and unmeritorious claims. As such, they have a direct positive impact on the fundamental right to effective judicial protection laid down in Article 47, first paragraph of the Charter and in Article 19(1), second subparagraph, TEU.

166.

When comparing the two options, Option 2 is a bit stronger as regards achieving the policy objective of ensuring full compensation of the entire harm suffered. This is due not only to the introduction of a competition specific system of collective redress discussed above, but also to fewer limitations on disclosure of evidence in relation to documents from the file of a competition authority (i.e. settlement submissions are not protected under Option 2; Option 2 limits disclosure during investigations of competition authorities).

167.

Option 2 is stronger as regards efficient use of the judicial system and the positive impact on SMEs and consumers. The objectives of access to justice, a more level playing field and the macro-economic impact are achieved alike under Options 2 and 3.

168.

However, Option 3 generally provides for a more balanced system. It contains an overall improvement of the possiblity to obtain access to evidence, while offering a stronger protection of effective public enforcement, by protecting more documents from the file of the competition authorities from disclosure in actions for damages. As to settlement submissions, stakeholders responding to the question on the interplay between public and private enforcement of competition law in the 2011 public consultation on collective redress confirmed that the settlement programme could be weakened by the stimulation of private enforcement in broadly the same way as the leniency programme. In a similar manner, some respondents to the public consultation on the White Paper indicated that the attractiveness of NCA and EU settlement procedures could be adversely affected by the strengthening of the private enforcement of competition law.

169.

Because of its limited scope, such additional protection is considered not to make it impossible or excessivly difficult for victims of a competition law infrirgement to obtain compensation for the harm they suffered. The protection is therefore considered to be compatible with the right to effective judicial protection laid down in Article 47, first paragraph of the Charter and in Article 19(1), second subparagraph, TEU. Furthermore, the introduction of a rebuttable presumption in relation to the existence of harm in cartel cases and the introduction of a rule that the burden and level of proof and of fact-pleading required for the quantification of the harm cannot render the exercise of the claimant's rights to damages practically impossible or excessively difficult, make it more likely that compensation is awarded. A balanced system fostering both effective public and private enforcement of competition law brings about benefits to growth, productivity and innovation.

58

EN

480  Commission Staff Working Document Impact Assessment Report 170.

In terms of costs, Option 3 scores better than Option 2. Litigation costs90 are reduced by the introduction of the rebuttable presumption in relation to quantification of harm. Also, error costs and implementation costs are lower under Option 3, because Option 3 does not foresee the introduction of a sector-specific framework of collective redress. The administrative burden is lower under Option 3 as the interests of public enforcement are better protected. The below table provides for an overview of the scores of the four options.

Table 10: Summary of impacts of Policy Options 1-4 Benefits achieved/problem addressed

Impact compared to base-line (0 to +++) Option 1

Option 2

Option 3

Option 4

1. Full compensation

0

+++

++

0/+

2. Protection of effective public enforcement

0

++

+++

0/+

3. Increased awareness, deterrence, enforcement and legal certainty

0

+++

+++

0/+

4. Access to justice

0

+++

+++

0/+

5. Efficient use of judicial system

0

+++

++

0/+

6. A more level playing field

0

+++

+++

0/+

7. Positive impact on SMEs and consumers

0

+++

++

0/+

8. Stimulating economic growth and innovation

0

++

++

0

Costs

Impact compared to base-line (0 to - - -) Option 1

Option 2

Option 3

Option 4

1. Litigation costs

0

--

-

0/--

2. Administrative burden

0

--

-

0/-

3. Error costs

0

-

0/-

0/-

4. Implementation costs

0

--

-

0/-

171.

90

EN

The comparative assessment of the four options shows that Option 3, in combination with the non-binding guidance on damages quantification as identified in Section 5.4, is the most suitable option. It achieves better than any other option the objectives that have been set out, producing the benefits outlined in this impact assessment at the lowest comparative cost, and taking into account the concerns and some of the suggestions that have been put forward during the public consultations. See Section 7.1.2 for a more detailed assessment of the litigation costs of the preferred option.

59

EN

Damages actions for breach of the EU antitrust rules (2013) 

481  

172.

The potential benefits of the preferred option in the attainment of the objective of full compensation can be estimated by reference to the data on foregone compensation by year from the impact study cited above.91 It was estimated that foregone compensation in the EU ranges from €5.7 to €23.3 billion per year. The improvement of the procedural framework under the preferred option does not include competition specific systems of collective redress available throughout the EU to facilitate consumer claims. In this situation, although it is not possible to predict the quantitative chance of damages recovery, it is safe to conclude that the improved procedural framework will still produce an increased compensation benefit comprised within the range indicated, which embodies a conservative estimate and thus also takes into account the reduced benefits due to the absence of competition specific collective redress mechanisms. Maintaining the assumption of a potential reduction of cartels by 5% due to increased deterrence, the negative consumer welfare impact of such infringements could be additionally reduced by € 1.32 to 3.45 billion per year.

173.

There are two different reasons to predict that the impact of the preferred option will remain within the ranges indicated by the impact study. The lower-bound of the broad range of foregone compensation already incorporates a number of restrictive assumptions, which compensate for the possible lack of mass consumer claims in those countries where this is not currently provided by national law. Moreover, the data from the impact study on foregone compensation and welfare impact are based on an average annual amount of cartel fines, which is then elaborated through the application of a detection rate, an overcharge to fine ratio, and refined by other criteria. In recent years, cartel fines have for a variety or reasons increased (e.g. longer duration and/or wider geographic scope of the cartel, bigger companies involved, etc): from the average annual fines considered by the study, amounting to €1.248 billion, in 2012 the cartels found by the Commission have been sanctioned with a total of € 1.875 billion in fines (with an average over the years 2008-2012 increased to €1.832 billion per year). Thus, the same methodology would produce higher impacts on the levels of compensation to be expected from an initiative on private enforcement.

7.1.2.

Litigation costs of the preferred option

174.

The fact that it is not possible to quantify the litigation costs of the preferred option in detail, explains the essentially qualitative assessment. However, on the basis of the available data, drawing in particular on the Impact Study92, the following can be said as to the potential magnitude of litigation costs. As most of the available quantitative information on litigation costs originates from the US, those figures are taken as a starting point. On litigation and litigation costs

175.

91 92 93

EN

An often quoted source for litigation cost estimates in US private antitrust actions is the Georgetown Project, which provided estimates of litigation costs for both settlements and trials93. As part of this detailed empirical exercise, it was found that in the US, claimant's lawyers's fees in private antitrust actions were in the range of 10

See Section 0. Impact Study, pages 85-89 and pages 150-152. The Georgetown Private Antitrust Litigation Project and Study (1973-83).

60

EN

482  Commission Staff Working Document Impact Assessment Report to 20% of the awards.94 Further, opportunity costs for parties, such as the time spent by their executives on these matters, has been estimated to range between approximately 50 and 75% of the lawyers' fees, while the court costs are estimated at a few percent of the awards.95 On this basis, one could conclude that between 3575% of the awards is ‘dissipated’ through litigation costs in the US.96 176.

The US system is of course characterized by a set of measures that affect the magnitude of litigation costs and that are neither present nor proposed in Europe. For example, in the US, claimant's lawyers' fees are often awarded on the basis of a percentage of the award, while in Europe pacta de quota litis are generally not allowed.97 Also different rules regarding evidence98 and lower court costs99 have an impact on litigation costs. Taking these differences into consideration, it is safe to assume that even after the introduction of the measures suggested in the preferred option, litigation costs in Europe will not go beyond 10 to 20% of the awards.100

177.

Considering now the overall level of litigation, a US-style system, where litigation is encouraged through e.g. multiple damages awards,101 asymmetric fee-shifting, extensive discovery and opt-out class actions, scores high on the deterrence scale as it very strongly encourages litigation. However, this high level of litigation is a costly way to achieve the corrective justice objective. The various measures and legal mechanisms retained in the preferred option differ in its significant aspects very strongly from the US model, thereby limiting, if not excluding unmeritorious actions. As a result, the preferred option will lead to a significantly lower level of litigation and litigation costs than what is witnessed in the US.

178.

The impact of the four policy options on litigation costs have been assessed in Section 6.2, taking the said differences with the US-system as a starting point. Of the options involving EU-legislative action, Option 3 scores the best, due to the fact that the litigation costs of individual claimants are reduced as a result of the shift of the burden of proof with regard to the existence of harm in cartel cases, the facilitation of consensual dispute resolution and the fact that Option 3 does not foresee the introduction of sector-specific measures on collective redress.

179.

However, in relation to the base-line of no EU action is foreseen, the preferred option (Option 3) will lead to some increase in litigation costs due to an increase in the

94 95 96

97 98 99

100

101

EN

On this basis Salop and White (1986) estimated that the total annual costs of private antitrust litigation in the US amounted to approximately US$250 million for 1973-1983 (in 1984 dollars). Court costs in the U.S. have been estimated at 5.5% of the (untrebled) award (Impact Study, page 89). The lower end of this range is calculated on the basis of lawyers’ fees amounting to 10% of the award for each party, an opportunity cost equal to 50% of these fees. The upper end is obtained assuming that lawyers' fees are 20% of the award for each party, that the opportunity cost amounts to 75% of these fees. For both, court costs account for 5% of the award. For exceptions, see the Impact Study, page 88. Ibidem For instance, the Impact Study indicates on page 89 that "in Europe, court costs are often expressed as a percentage of the damage award, which can range – depending on the country and on the value of the claim – from 2% to 6%." See the Comparative Study, paragraph I(viii) on p.96, which suggests that the cost of bringing an antitrust claim is lower than 10% of the value of the claim in most EU countries, although it is higher in the UK and Ireland. In systems with multiple damages (like the US), it may be profitable for the claimant to file a suit, even if the probability of winning the case is small. Indeed, if it prevails, the award would be very high. While multiple damages thus increase deterrence, they encourage cases with little merit and hence increase litigation costs, see e.g. the Impact Study, page 88.

61

EN

Damages actions for breach of the EU antitrust rules (2013) 

483  

number of lawsuits. The structure and amount of fees for litigation (and out-of court activity connected to it) vary across Member States, so it is difficult to provide an exact estimate. However, as a proxy, it is interesting to consider the expected impacts of reforming the system of private enforcement in the UK, as currently envisaged in a consultation paper.102 The option of improving the framework for private actions by introducing rules on the relationship with public enforcement and encouraging ADR (without introducing public opt-out collective actions) is estimated to lead to an increase in average annual costs for participants of £17,7 million (€21,35 million) against total benefits of £61,4 million (€74 million) in increased deterrence103 and cartel prevention, and £296 000 (€357 000) in public sector benefits. Impact of the preferred option on the number of damages actions 180.

Given the complexity of establishing a competition law infringement, it is realistic to assume that antitrust damages actions will usually be follow-on actions. Under that assumption, the upper ceiling of the number of actions for damages that can be initiated equals the number of cases in which the Commission or the NCAs take a decision establishing an infringement. Since 2007, NCAs took on average 55 infringement decisions per year and the Commission 8. However, only very few of those cases have so far led to an antitrust damages action. In relation to NCA cases, we are aware of damages actions being brought in less than 10% of these cases. As regards Commission cases, it seems that no follow-on damages actions have been brought in relation to more than two thirds of the infringements found by the Commission.104 Furthermore, as far as the latter category of follow-on actions is concerned, they rarely cover all victims: indeed, whereas the Commission decision usually finds an infringement for the whole EEA (or at least a substantial part of it), damages actions are brough by or on behalf of victims from only one or more Member States. Given these data, the most extreme theoretical effect of the current Antitrust Damages Initiative would be that it results in actions for damages in all remaining cases, covering all victims of those infringements.

181.

Such an increase is, however, not to be expected. In Member States which currently have a regime favourable to actions for damages (like the UK where there are favourable regimes for disclosure of evidence and collective actions), experience shows that the number of cases in which actions for damages are initiated does by no means coincide with the total number of decisions taken by the Commission and the NCA. In fact, the recently published impact assessment by the UK government on private actions in competition law indicates that only few cases are brought before the national courts.105 Out of 21 findings of infringement by the OFT between 2000 and 2007, only 2 led to follow-on cases brought by injured parties.106 Out of 45

102

103 104

105 106

EN

See BIS – Department for Business Innovation & Skills – "Private actions in competition law: a consultation on options for reform" and the Impact Assessment thereof, both available at: http://www.bis.gov.uk/Consultations/consultation-private-actions-in-competition-law. In the UK impact assessment's methodology, increased deterrence is calculated as a multiplier of damages, and particularly in a benefit of 5:1 in stand alone cases, and 1:1 in follow-on cases. In the absence of any EU-wide obligation of parties and/or national courts to inform national or European authorities of damages actions brought, this figure is only an approximation, based on anecdotal information. Moreover, this figure does not take account of informal settlements, which rarely are made public. Department for Business, Innovation & Skills, Impact Assessment, Private actions in competition law: a consultation on options for reform, April 2012, paragraph 22. Ibid, paragraph 25. It should be noted, however, that settlement may have been reached in other cases.

62

EN

484  Commission Staff Working Document Impact Assessment Report companies harmed by a breach of competition law, only five companies finally decided to bring an action, most commonly because the expected costs of litigation did not outweigh the benefits.107 182.

The same reasoning can be applied by victims in relation to the current Antitrust Damages Initiative. Furthermore, smaller sized victims, such as SMEs, might fear for their good business relationship with infringers if they sue them in actions for damages. Therefore, the implementation of the preferred option is expected to have the effect of increasing actions for damages, but not causing mass litigation such as in the US. Finally, the fact that out-of-court settlements are fostered by the current Antitrust Damages Initiative supports the conclusion that the increase in litigation and related costs will be moderate.

7.1.3.

Transposition costs for the Member States

183.

"Transposition costs" refers to the likely costs incurred by Member States in adapting national law in order to implement the measures proposed under the preferred option. For the purposes of this analysis: •

‘High transposition costs’ indicates that significant changes to national laws, regulations or administrative practices or structural changes, such as setting up an agency or making recurrent budgetary costs, are required in order to implement the proposed measures in the Member States’ domestic legal system;



"Low transposition costs" indicates that a measure would only require minor adaptations in the national legal systems because of its limited impact or because of already existing similar laws, regulations or administrative practices under national law;



No transposition costs are envisaged where the requirements of a specific measure are already met in the Member State(s), where the envisaged instrument is of a non-binding nature or where the measure in question reproduces the acquis communautaire, and should therefore be already complied with by the Member States;



The intermediate indicators of "medium" (between high and low) transposition costs and "very low" (between low and no) transposition costs will also be used.

As shown in table below, the current Antitrust Damages Initative does not require Member States to make significant or structural changes. Hence, all transposition costs in relation to the current Antitrust Damages Initiative are rather limited. Where the table indicates that differences may exist among Member States, reference is made to section 3.2 above, where these differences are described. Table 11: Overview of transposition costs Proposed Measures Assessment of Transposition Costs under the Preferred Option

107

EN

Ibid, paragraph 24.

63

EN

Damages actions for breach of the EU antitrust rules (2013) 

485  

Standing of direct and indirect purchasers

Acquis communautaire: no new transposition costs.

Joint and several liability and limitation thereof for immunity recipients

Depending on the existing legal framework in the Member States, the introduction of joint and several liability for co-infringers will entail low or no transposition costs. Transposition costs of measures concerning the limitation of the civil liability of the immunity recipient depends on the existing situation under national law. For Member States that do not provide any form of limitation of civil liability: medium transposition costs. Member States that already provide for such mechanisms would only have to apply these to immunity recipients: low transposition costs.

Rules on access to evidence (inter partes disclosure)

Some form of disclosure is or should already be available under the legal system of the Member States in some areas of law (i.e. intellectual property rights enforcement, for which rules on disclosure are part of the acquis communautaire under Directive 2004/48/EC). In these cases, should disclosure rules not yet be available in antitrust damages actions, Member States would only need to expand the scope of application of rules on disclosure, also adapting them to the specificities of the envisaged measures: medium transposition costs. Where Member States already have disclosure rules for antitrust damages actions along the lines of the envisaged measures, the transposition costs will even be lower.

Binding effect of final decisions of NCAs

If a Member State already grants the binding effect before national courts of final decisions by NCAs: no transposition costs. In Member States that recognise binding effect only to the decisions adopted by domestic NCAs, the transposition costs of extending that effect also to final decisions of other NCAs are considered very low. In those Member States where no such effect is provided for at all, transposition costs are considered low because a decision by a national competition authority already bears a strong persuasive effect on national courts.

Full single damages Acquis communautaire: no new transposition costs.

EN

Passing-on defence and rebuttable presumption of passing on of the illegal overcharge to the indirect purchaser

In most Member States it is not clear whether or not a passing-on defence can be invoked in antitrust damages actions. However, where not explicitly excluded, this defence could probably be allowed pursuant to the general principles underlying national tort rules according to which the burden of proof lies with the party invoking the defence, hence very low transposition costs. If a Member State directly or indirectly excludes such defence: medium transposition costs. Introducing a rebuttable presumption that the illegal overcharge has been passed-on is something yet absent in most national legal systems. However, this is in essence a rebut of the burden of proof, that would shift from the injured party to the infringer. The application of this technique (rebuttable presumption) to antitrust damages actions brought by indirect purchasers would not imply any major change in the legal systems of the Member States: medium transposition costs.

Rules on limitation periods

To the extent that the envisaged measures implement the EU principle of effectiveness, that is part of the acquis communautaire and thus entails no new transposition costs. For what concerns the possibility to claim compensation after a final decision by the Commission or a NCA, the provision of a new limitation period would raise some issues only in those cases where the limitation period has expired before the start of public proceedings (in other cases, a suspension pending an investigation of an NCA would allow for a similar effect). As this is rarely the case: low transposition costs.

Rebuttable presumption of harm caused by cartels

Introducing a rebuttable presumption that in cartel cases harm exists shifts the burden of proof to the party which most likely has the relevant information. The application of this technique (rebuttable presumption) would not imply any major change in the legal systems of the Member States: low transposition costs.

Non-binding framework on damages quantification

The provision of non-binding guidance on the quantification of damages by its very nature does not require any implementation activity on the part of the Member States: no transposition costs.

64

EN

486  Commission Staff Working Document Impact Assessment Report Rules on Consensual dispute resolution mechanisms exist in all Member States. The introduction Consensual Dispute of rules to stimulate recourse to consensual settlements, and their co-ordination with the Resolution rules on joint and several liability, would thus not entail significant modifications in national legal orders: low transposition costs.

EN

7.2.

Choice of legal instrument

7.2.1.

The options

184.

Having identified the adoption of a legislative proposal at the EU level as the preferred option, it is necessary to assess which particular instrument would be most suitable in the area of antitrust damages actions: a regulation, a directive or a combination of the two. The assessment of these options takes account of which one most effectively achievemes the objectives set out above, the costs (in particular the implementation costs for the Member States) and the effects in terms of simplification or complication of the existing legal framework.

7.2.2.

Regulation

185.

The advantage of introducing the preferred option through a regulation is clearly its direct applicability that would ensure the elimination of most obstacles in the legal framework as of the day of its entry into force. A regulation would also create a set of provisions that are equally available to all victims of antitrust infringements in the Member States.

186.

However, a regulation could require a detailed elaboration of all the aspects that are necessary for the concrete functioning of the envisaged redress system. It would thus constitute a separate body of law, next to the existing national rules on civil liability and civil procedure. This approach is quite intrusive, and could be pursued only if proven necessary, especially if less intrusive measures, such a directive, would not be able to effectively achieve the objectives.

7.2.3.

Regulation and Directive

187.

A mix of instruments would allow the inclusion in a regulation of some of the measures of the preferred option for which there exists little or no political discretion for Member States when implementing (e.g. binding effect of decisions of NCAs and the protection of corporate statements), therefore benefiting from direct applicability of the legislative text of reference throughout all the Member States. At the same time, other measures that only provide for minimum requirements and allow for more political discretion for Member States when implementing (e.g. disclosure and limitation periods) would be included in a directive, thus allowing Member States to properly adapting them into their own legal systems.

188.

The major pit-fall of this option is that it implies an increased heterogeneity of the legal framework for damages actions, which would oblige parties and judges in damages actions to rely on a set of multiple legal sources.

7.2.4.

Directive

189.

A directive appears to be the most appropriate legal instrument to effectively bring forward the current Antitrust Damages Initiative, to make the measures work effectively and to allow for a smooth adaptation into the legal systems of the Member States.

190.

A directive would allow the Member States to achieve the objectives and implement the measures of the preferred option into their national substantive and procedural

65

EN

Damages actions for breach of the EU antitrust rules (2013) 

487  

law systems. This approach is less intrusive than any of the previous options, to the extent that Member States are left the choice of the most appropriate technical and regulatory tools to implement the measures that are contained in the preferred option. It allows greater flexibility to take account of specificities of national legal systems. A directive would furthermore be a flexible tool to introduce a minimum harmonisation in those areas of national law that are crucial for the functioning of damages actions, ensuring a common minimum standard all across the EU, but leaving room for further reaching measures to the individual Member States.

EN

191.

Finally, a directive would avoid regulatory intervention in all those cases where the domestic provisions in the Member States are already in line with the proposed measures: in other words a directive would allow a targeted implementation.

7.3.

Proportionality and EU added value of the Preferred Option

192.

The Commission considers that action at EU level along the lines of the preferred option would respect the principle of subsidiarity since there is a clear need for and added value in such action. The preferred option is also fully in line with the principle of proportionality, both as regards its general approach and the content of the individual measures envisaged.

7.3.1.

Subsidiarity: European added value

193.

The preferred option would have European added value for the following reasons: •

There is a significant risk that the lack of EU-wide regulation of the interaction between public and private enforcement would jeopardise effective public enforcement. The Pfleiderer ruling leaves a considerable discretionary power to the national courts, which could lead to important discrepancies between the Member States and even within Member States regarding the disclosure of evidence from the files of competition authorities. This causes appreciable legal uncertainty, not only for claimants of damages before national courts, but also for defendants and competition authorities and can potentially cause a disincentive for cartel participants to cooperate with the competition authorities in the context of their leniency programme, thereby weakening public cartel enforcement. As leniency applicants can and may have to apply for leniency in several jurisdictions (given the flexible case allocation system within the European Competition Network), any loophole in the protection of corporate statements across the EU would be highly detrimental to public enforcement of competition rules in Europe as a whole. Moreover, discrepancies between Member States might also have a negative impact on the exchange of information between competition authorities within the framework of the ECN. It is thus necessary to establish certain harmonised rules setting a common interaction between public and private enforcement applicable in all Member States: this can only be achieved at the EU level.



Experience shows that in the absence of EU legislation, most Member States do not on their own initiative provide for an effective framework for compensation for victims of infringements of Articles 101 and 102 of the Treaty, as required by the Court in its Courage and Manfredi judgments. Since the publication of the Commission’s Green and White Paper, only few Member States have enacted legislation aimed at facilitating antitrust damages cases. The actions at national level are also limited to specific issues and do not cover the whole range of measures envisaged by the current Antitrust Damages

66

EN

488  Commission Staff Working Document Impact Assessment Report Initiative. Despite the few steps taken individually by some Member States, there is thus still a lack of effective compensation of victims of infringements of the EU antitrust rules. Only further incentives at European level can create a legal framework that adequately ensures effective redress and guarantees the right of effective judicial protection as laid down in Article 47 of the Charter. •

EN

There is currently marked inequality between Member States in the level of judicial protection of individual rights guaranteed by the Treaty, which may cause distortions of competition and hamper the proper functioning of the internal market. The result is an evident disparity in the very content of the entitlement to damages guaranteed by EU law. More specifically, a claim under the law of one Member State may lead to full recovery of the claimant’s loss, while a claim for an identical infringement in another Member State may lead to a significantly lower award or even no award at all. This inequality between Member States increases if few Member States take measures in limited fields and other Member States do not take action, as it currently happens. The trans-national dimension of Articles 101 and 102 TFEU warrants measures at the EU level.

7.3.2.

Proportionality

194.

In terms of proportionality, the preferred option strikes a careful balance between maintaining the effectiveness of public enforcement of competition law, stimulating effective protection of victims’ rights to compensation, the legitimate interests of potential defendants and third parties and important interests of Member States. The preferred option is the minimum necessary to effectively achieve its objectives: to guarantee effective protection of public enforcement of competition law across the EU, as well as access for victims of competition law infringements to truly effective remedies to obtain full compensation for the harm they suffered.

195.

Those objectives are also achieved at the lowest possible costs. The costs imposed on citizens and businesses are proportionate to the stated objectives. A first step in this direction was already taken with the White Paper by excluding more radical measures (for instance multiple damages, opt-out class actions and wide discovery rules). The efforts to strike this balance were widely welcomed during the public consultations. The additional safeguards included in the preferred option further strengthen this balance by reducing potential costs (especially in terms of litigation) without jeopardising the effectiveness of the right to compensation. Finally, also the choice for a Directive as the instrument is in line with the principle that there should be as little intervention as possible, while attaining the objectives pursued.

67

EN

Damages actions for breach of the EU antitrust rules (2013) 

EN

489  

8.

MONITORING AND EVALUATION

196.

Extensive research and consultation preceded the adoption of the Green Paper, the White Paper and the current Antitrust Damages Initiative. Three wide-ranging studies, four public consultations with a high number of responses and a series of additional consultations with stakeholders at Member State and EU levels, including public authorities, prominent scholars and practitioners from the private sector, greatly contributed to the analysis and evaluation of the relevant issues.

197.

Following the adoption of the current Antitrust Damages Initiative, the Commission will continue monitoring the applicable legal framework for antitrust damages claims in Europe, as much as it will monitor the implementation of the Directive in the Member States once adopted and the effects it produces in terms of improved framework for damages actions for breach of the EU antitrust rules.

198.

The monitoring exercise will be focussing primarily on the achievement of the objectives set out in Chapter 4. Where the preferred policy option targets existing inefficiencies and obstacles preventing full compensation of victims of antitrust infringements, the monitoring process will aim at identifying which of these obstacles have successfully been removed and to what extent. This analysis should be carried out through the monitoring of national legislation and practice in the domains of civil and procedural rules and through continuous contact with the stakeholders.

199.

Apart from the applicable legal framework for antitrust damages actions, other monitoring indicators will include the extent to which victims of competition law infringements effectively obtain compensation and the optimisation of the interaction between private and public enforcement. The degree of effective compensation should encompass both successful damages actions before national courts and out-of court settlements to the extent that the information is made available.

200.

The evaluation of the successful removal of obstacles to effective compensation on the basis of the increased instances in which infringers of the competition rules pay damages to injured parties can also be performed with reference to a number of proxies. These include: the increase in overall number of actions for damages brought before national courts and the increase in the number of Member States in which actions for damages are brought. The optimisation of the interaction between public and private enforcement will be evaluated through the number of judgments of national courts providing adequate protection to public enforcement instruments and the number of cases in which immunity recipients are not primarily targeted in actions for damages.

201.

Dialogue with stakeholders will also be necessary to monitor the effects in terms of increased litigation costs, particularly as regards false positives, i.e. the situation in which undertakings that did not infringe competition law are obliged to pay damages. Although the safeguards in the preferred policy option are meant to avoid such false positives, we should be vigilant for any significant increase, since that would be an important indication of inefficient features of the devised system.

202.

The monitoring and evaluation process will provide useful information for potential future modifications. However, before carrying out a meaningful ex-post evaluation exercise it will be necessary to wait until the measures put forward at the European level are fully implemented and functioning in the Member States. Therefore, the ex-

68

EN

490  Commission Staff Working Document Impact Assessment Report post evaluation will have to take place at least five years after the Directive has been adopted.

EN

203.

The implementation process in the Member States will also be monitored on the basis of certain provisions included in the legislative initiative. These provisions require the Member States to communicate to the Commission the text og the provisions that implement the proposal, as well as explanatory documents, such as correlation tables between those provisions and the proposal. When Member States adopt those provisions, they shall contain a reference to the proposal or be accompanied by such a reference on the occasion of their official publication. Furthermore, Member States shall communicate to the Commission the text of the main provisions of national law which they adopt in the field covered by the proposal.

204.

Finally, implementation by Member States will be monitored through analysis of the relevant case-law of national courts. The parameters of the analysis should cover, in particular: the type of harm for which compensation has been awarded (actual loss and/or loss of profit); the type of victims reached by awards (businesses and consumers); the type of action brought (stand alone or follow-on); reasons for which damages are not awarded by the court. The analysis of the relevant case-law should be facilitated by Member States by increasing compliance with the legal obligation under Article 15(2) of Regulation 1/2003, under which they should transmit to the Commission any judgment applying Articles 101 and 102 TFEU.

69

EN

Damages actions for breach of the EU antitrust rules (2013) 

491  

ANNEX 1 Overview of the functioning of the Commission's leniency programme 1.

OVERVIEW

Article 101 TFEU prohibits agreements between undertakings having as their object or effect the prevention, restriction or distortion of competition. Secret cartels are one of the most harmful infringements of this provision, as they have as their object the very distortions prohibited under Article 101, such as price-fixing, market-sharing and output-limiting agreements. The Commission enforces EU competition rules, together with National Competition Authorities, under Council Regulation 1/2003. This legal instrument, further specified by Commission Regulation 773/2004, also grants the Commission investigative powers (such as the power to conduct inspections and request informations) and the authority to impose administrative fines and penalty payments on infringers. The fines imposed by the Commission on any infringing undertaking can amount to a maximum of 10 per cent of its turnover. Despite its investigative powers, it may be very difficult for the Commission to detect harmful cartels in view of their secret nature. In order to encourage participants to cartels to cooperate with the Commission bring infringements to an end, the Commission launched its leniency programme in a 1996 Notice.108 Drawing inspiration from other jurisdictions, and building on its discretionary fine-setting powers, the Commission comitted not to impose fines, or to reduce them, on undertakings that cooperate by giving notice of the Cartel and providing evidence for its successful prosecution. Thanks to this incentive, several high-profile cartel cases have been detected and prosecuted by the Commission, which also imposed fines to the other cartel members amounting to several billion euros. The 1996 Notice has been replaced in 2002109 and later reviewed with the adoption of the most recent Commission Leniency Notice in 2006 ('Leniency Notice').110 2.

FUNCTIONING OF THE EU LENIENCY PROGRAMME

The EU leniency programme as set out in the 2006 Leniency Notice rewards undertakings that co-operate with the Commission by disclosing their participation in an alleged cartel, also providing information and evidence and fulfilling other requests that can be made by the Commission. The first undertaking coming forward may thus obtain a total immunity from the fine, provided that the Commission did not already have enough evidence to find an infringement as regards the alleged cartel, and that it is the first to provide contemporaneous incriminated evidence of the alleged cartel as well as a corporate statement containing information about it. When an undertaking does not qualify for total immunity (e.g. it is not the first to submit a request to the Commission, or it provides insufficient evidence), it could still obtain a reduction of the fine in view of its contribution to the investigation of the infringement. 108 109 110

EN

18 July 1996, Commission Notice on the non-imposition or reduction of fines in cartel cases, [1996] OJ C 207, p. 4. 19 February 2002, Commission Notice on Immunity from fines and reduction of fines in cartel cases, [2002] OJ C 45 p. 3. 8 December 2006, Commission Notice on Immunity from fines and reduction of fines in cartel cases, [2006] OJ C 297 p. 17.

70

EN

492  Commission Staff Working Document Impact Assessment Report The Notice also sets out the procedure for the participation to the EU leniency programme. The undertaking must approach the Commission's Directorate-General for Competition with a formal application for immunity. When an undertaking is willing to cooperate and needs time to collect the information and evidence required by the Leniency Notice, but does not want to lose its ‘place in the queue’ (i.e. it wants to avoid that other undertakings may issue a formal application and possibly obtain immunity in its place), it can request a marker. If the Commission grants the marker for a certain period of time, the formal application by the undertaking will be regarded as having been made at the time it first approached the Commission. When submitting formal applications for immunity, undertakings must provide evidence and information on the cartel. Other than the evidence in its possession, the applicant must produce a corporate statement, i.e. a voluntary submission containing the information specifically required by the Notice. This information includes a detailed description of the alleged infringement, the identity of the participant undertakings and of individuals who have been involved in the cartel. Corporate statements are sensitive documents not only because they contain confidential information, but also because they contain the acknowledgement of an undertaking's participation to serious infringements of the EU competition rules. The disclosure of corporate statements could thus expose their authors to liability. In order to protect the information in its possession, also from the perspective of its disclosure in other jurisdictions, and not to discourage undertakings from co-operating, the Notice envisages certain forms of protection. For instance, the corporate statement can be provided orally by the undertakings and recorded at the Commission's premises. Access to the corporate statements is also restricted by the Notice to the addressees of a Statement of Objections (i.e. alleged co-infringers), that cannot obtain copies of the statements, and can only use the information for the purposes of judicial and administrative proceedings for the application of the EU Competition Rules. 3.

LENIENCY PROGRAMMES IN THE MEMBER STATES

Alongside the leniency programme provided for at EU level by the 2006 Notice, leniency programmes are currently operated at national level by the National Competition Authorities of all Member States except Malta. In order to approximate the existing provisions on leniency at EU and national level, to avoid excessive discrepancies among legislation that could discourage applicants, and to ensure coordination in the enforcement of the Treaty provisions, the European Competition Network ('ECN’) has developed an ‘ECN Leniency model programme’,111 which sets out a framework against which national programmes should progressively be aligned, which concerns both the conditions for the granting of immunity and the provisional requirements thereof.

111

EN

Available at http://ec.europa.eu/competition/ecn/mlp_revised_2012_en.pdf .

71

EN

Damages actions for breach of the EU antitrust rules (2013) 

493  

ANNEX 2 Summary of the Pfleiderer judgment of the Court of Justice 1.

FACTS AND BACKGROUND

On 21 January 2008, the German Competition Authority (the Bundeskartellamt) imposed fines amounting to EUR 62 million on three European manufacturers of decor paper and on five individuals who were personally liable for agreements on prices and capacity closure. The decisions of the German Competition Authority became final. Pfleiderer is a purchaser of decor paper and thus alledgely suffered harm as a result of the cartel. Following the Bundeskartellamt’s decision, Pfleiderer submitted an application to the competition authority seeking full access to the file, with a view to preparing civil actions for damages. This request expressly included all document relating to the leniency application which had been voluntarily submitted and the evidence seized. The competition authority rejected Pfleiderer's request and restricted access to the file to a version from which confidential business information, internal documents and leniency documents had been removed and refused access to the evidence which had been seized by the Authority. Pfleiderer brought an action before the District Court (Amtsgericht) in Bonn challenging this partial refusal. The Bonn District Court ordered access both to the leniency documents in the file and to the incriminating material and evidence collected. The enforcement of this decision and the proceedings were, however, stayed by the District Court, as it referred a question to the European Court of Justice (the Court) for preliminary ruling. 2.

THE PRELIMINARY QUESTION

The preliminary question referred by the Bonn District Court was whether EU competition law has to be interpreted as meaning that victims of cartels may not, for the purpose of bringing an action for damages, be given access to leniency documents from the file of a national competition authority, within the framework of public enforcement proceedings. 3.

THE JUDGMENT OF THE COURT

The Court held that neither the provisions of the Treaty on competition nor Regulation 1/2003 lay down common rules on leniency or on the right of access to leniency documents. As regards the leniency notice and the ECN model leniency programme, the Court held that these document are not binding on Member States. The Court then established that it is for the Member States to lay down and apply national rules on the right of access to leniency documents by victims of a cartel. In exercising this competence, Member States have to act in accordance with EU law; in particular they may not render the implementation of EU law impossible or excessively difficult or jeopardize the effective application of Articles 101 and 102 TFEU. Leniency programmes were recognised as useful tools to uncover and bring to an end infringements of the competition rules, thus serving the objective of effective application of Articles 101 and 102 TFEU. On the question whether the effectiveness of those programmes could be compromised if leniency documents were disclosed to persons seeking to bring an actions for damages, the Court held that the view can reasonably be taken that a cartel participant would be deterred from cooperation with the competition authorities in the context of the leniency programme if these leniency documents were disclosed.

EN

72

EN

494  Commission Staff Working Document Impact Assessment Report Nevertheless, the Court recalled that it is settled case-law that any individual has the right to claim damages for loss caused to him by violations of the EU competition rules. This right constitutes a significant contribution to the maintenance of effective competition in the EU. As such, the applicable national rules for disclosure cannot be less favourable than those governing similar domestic claims and cannot operate in such a way as to make it practically impossible or excessively difficult to obtain compensation. National courts thus have to weigh the diverging interests in favour of disclosure of the information and in favour of the protection of that information provided voluntarily by the leniency applicant on a case-by-case basis, according to national law and taking into account all the relevant factors in the case.

EN

73

EN

Damages actions for breach of the EU antitrust rules (2013) 

495  

ANNEX 3 Overview of the most important measures for antitrust damages actions included in the legislative proposals in Austria, Germany and the United Kingdom 1.

AUSTRIA

In early 2012, the Austrian government held a public consultation on proposed changes to the Austrian competition law (Kartellgesetz und Wettbewerbsgesetz). The proposal contains among others provisions aiming to facilitate both 'follow-on' and 'stand-alone' antitrust damages actions by:

2.

(a)

requiring the cartel court to publish its decisions (which is so far not the case and one of the major obstacles for bringing follow-on actions)

(b)

introducing antitrust specific provisions for damages actions that foresee, inter alia:



a binding effect of Commission and NCAs decisions on civil courts as regards the finding of an antitrust infringement;



the possibility for civil courts to suspend proceedings pending proceedings before the Commission or NCAs;



the suspension of the limitation period for damages actions until six months after a related public enforcement decision has become final; and



the payment of interest from the time the harm occurred (to exclude that interest may be due only from a later moment, e.g., when the damages action was brought).

(c)

allowing parties to seek in stand-alone cases a declaratory judgment by the cartel court, finding a competition law infringement, in view of bringing an antitrust damages action before the competent civil court.

GERMANY

In March 2012, the German government put forward a legislative proposal for an 8th Amendment to the GWB (the German act against restraints on Competition). This Amendment foresees a specific right of action for consumer associations in the field of private enforcement. This right of action, which builds on an existing right for business associations, has the following features:

EN

(a)

It is specific to competition law (EU and national law is covered). While it adopts an approach which has previously been used in other areas (notably in the law of unfair competition), the proposed rule is limited to an infringement of antitrust rules.

(b)

It empowers consumer associations to bring the action. The legislative proposal makes explicit reference in this respect to the Consumer injunctions Directive (2009/22/EC) as well as to a similar national law on injunctions in consumer law matters. The GWB already now empowers business associations in a similar way and the legislative proposal contains in fact a clarification as to the scope of the empowerment of business associations.

74

EN

496  Commission Staff Working Document Impact Assessment Report

3.

(c)

The aim of the action is the skimming-off of illicit profits. The defendant can raise as a defence that he already has paid compensation to private parties or that the illicit profit which he obtained through the infringement has already been skimmed off through an action of the NCA (who has the power to make such an order).

(d)

The proceeds of the action are to be paid to the budget of the federal government. The consumer and/or business association can nevertheless obtain reimbursement of their costs.

(e)

In its explanatory memorandum for the legislative proposal, the federal government explicitly mentioned the ongoing debate on collective redress at EU level as a reason for making such proposal.

UNITED KINGDOM

On 24 April 2012, the UK Government (Department for Business, Innovation and Skills) launched a public consultation on ‘Private Actions in Competition Law: A Consultation on Options for Reform’. The consultation document, as well as the January 2013 Government's response, put forward the UK Government's views on a number of issues relating to the private enfocement of competition law, in particular: (a)

Competition-specific proposals on collective redress

The UK Government believes there is a robust case for collective actions in the competition field, and does not favour generic 'horizontal' collective redress. This reiterates the position expressed by the UK Government in their reply to the Commission's public consultation on collective redress last year. The UK Government thus decided to introduce a limited opt-out collective actions regime for competition law infringements, particularly due to the dismal record of the pure opt-in model currently available to consumers in the UK. The Government also envisages the introduction of a new opt-out settlement regime for competition law in the CAT, similar to the model of the Netherlands, and to require that any opt-out settlement must be judicially approved. (b)

Complementarity of public and private enforcement

While the UK Government is fully committed to maintaining the public enforcement by competition authorities at the heart of the enforcement regime, it believes that private-sector led challenges to anticompetitive behaviour should be encouraged to complement public enforcement. As regards the interplay of public and private enforcement, the UK Government puts emphasis on the need to protect the effectiveness of the leniency programme. To this end, it considers protecting certain categories of leniency documents, andmiting the joint and several liability of immunity recipients. In its response to the public consultation, while stressing the need to preserve incentives for potential leniency applicants to cooperate with antitrust investigations, the Government has announced its intention not to propose action at national level because it expects the Commission to take an initiative on this issue. Absent such an initiative, or in case of a significant delay, the Government will consider specific proposals.

EN

75

EN

Damages actions for breach of the EU antitrust rules (2013) 

497  

ANNEX 4 Detailed Summary of the Impact Assessment report for the White Paper on actions for damages for breach of the EU antitrust rules (IAWP)112 1.

INTRODUCTION

In 2005, the Commission published a Green Paper on actions for damages for breach of the EU antitrust rules, by which it consulted stakeholders on a number of possible measure to remove existing obstacles to private enforcement of Articles 101 and 102 TFEU. These options concerned procedural and substantive requirements governing actions for damages. Following the public consultation on the Green Paper, the Commission services impactassessed these options in order to select the most cost-efficient bundle of suggestions to overcome current obstacles to private enforcement, as put forward in the 2008 White Paper on actions for damages for breach of the EU antitrust rules. The present Impact Assessment Report has not reviewed options that had already been discarded because of ther excessive costs, insufficient benefits or unsatisfactory impact on the attainement of the objectives. The conclusions drawn in the IAWP are shortly summarised below, with the exception of the baseline option (zero action at EU level) and the option on non-regulatory measures, which have been reassessed during the current exercise. 2.

THE IAWP'S POLICY OPTION 1

The first policy option assessed in the IAWP concerned legislative measures maximising the facilitation of claims and the incentives for victims. This option was meant not only to remove all the identified existing obstacles to private enforcement, but was also aimed at providing significant incentives for potential claimants to enforce their rights, as experimented in other jurisdictions. Thus, the option envisaged fully-fledged opt-out class actions, by which some victims were allowed to claim damages on behalf of all injured parties except those that expressly opted not to be represented. As a further incentive to claimants, the option suggested that they should never be liable for the defendant's costs even if they lost the case (unilateral fee-shifting), except for frivolous or vexatious claims. As regards access to evidence, the option allowed for broad disclosure, by which the claimant could request relevant documents with a low threshold to obtain an order from the judge, i.e. only having to demonstrate a plausible case. The passing-on defence was not accepted under this option, allowing direct purchasers to always claim damages, but claims by indirect purchasers were also allowed as they are part of the acquis. The liability of the infringers did not require that they acted at fault. Finally, limitation periods for the claim were set at 20 years, and the decisions of NCAs were binding on civil courts. The assessment of this policy option resulted in undoubtedly high benefits in terms of full compensation and access to justice for victims, and significant benefits in terms of deterrence of anticompetitive conduct. However, due to the potential for abuses and the lack of significant safeguards, the option was not deemed satisfactory in view of its sub-optimal use of the judicial system, but most notably for its high administrative and litigation costs, which were deemed to be excessive. The option also entailed very high implementation costs because measures such as full-fledged opt-out by any group of victim, the unilateral fee112

EN

2 April 2008, Commission Staff Working Document – Impact Assessment Report accompanying the White Paper on Damages actions for breach of the EC antitrust rules, SEC (2008) 405.

76

EN

498  Commission Staff Working Document Impact Assessment Report shifting and the disclosure of evidence not tied to fact pleading are far from European legal traditions, thus requiring significant changes to the law of all Member States. 3.

THE IAWP'S POLICY OPTION 2

The IAWP's Policy Option 2 was the basis upon which the White Paper was drafted, because it was identified as the bundle which could foster the attainment of objectives at the lowest possible costs. Policy Option 2 allowed 2 complementary forms of collective redress, i.e. an opt-in collective actions, and a representative actions that could be brought on behalf of the victims by qualified representative bodies such as consumer associations. The disclosure of evidence was based on an initial exchange of lists in which the parties indicdated the relevant evidence, only after the claimant had presented reasonably available facts and evidence in support of its allegations. Incentives for the claimant were also provided in the softer form of rebuttable presumptions on fault and on the passing-on in case of claims by indirect purchasers. The decisions of national competition authorities were meant to be binding on civil courts throughout the EU. Limitation periods were set at five years from the date on which the claimant should reasonably have been aware of the harm, with a new limitation period of two years in case of public proceedings on the infringement. A peculiarity of this policy option was that it provided recovery of single damages (full compensation) except in case of cartels, when double damages would be allowed. As anticipated, the assessment of this option showed a more balanced cost/benefit ratio. Litigation and administrative costs were significantly lower than in option 1 thanks to the provision of safeguards on collective redress, through the presumption of pass-on in case of claims by indirect purchasers while allowing the passing on defence in case of claims by direct purchasers, and by maintaining a ‘loser pays’ cost rule, with a discretionary power for the judge to shift fees on the defendant. Error costs and implementation costs of the option, however, remained high due to double damages in cartel cases, which are not consistent with the compensatory nature of tort litigation in Europe. 4.

THE IAWP'S POLICY OPTION 3

The third policy option targeted the adoption of legislative measures of a more limited scope by comparison with Option 2. In particular, Option 3 envisaged full single damages rather than double damages in cartel cases; disclosure of specified categories of evidence based on fact-pleading. Different from Option 2, it did not establish a passing-on presumption in claims by indirect purchasers and it limited the binding effect of decisions of National Competition authorities to the domestic civil courts only. Limitation periods were similarly envisaged as in Option 2, except for the elimination of a new 2-year period in case of public enforcement. The collective redress measures maintained the representative actions, but did not retain the complementarity with opt-in actions. On balance, the costs entailed by Option 3 were still lower, due to the further elimination of incentives for claimants, even in the form of rebuttable presumptions (e.g. as regards the passing-on presumption for indirect purchasers). However, the same reasons led to a lower estimation of the benefits of the option, which would not significantly foster a better legal framework for all victims of competition law infringements and an insufficient level of access to justice. The efficient use of the judicial system was also not enhanced because of the limited scope of the binding effect of decisions adopted by national competition authorities. However, some elements of Option 3 were estimated as worth retaining to correct the pitfalls of Option 2 and further limit its costs (e.g. evidence disclosure based on fact-pleading; no

EN

77

EN

Damages actions for breach of the EU antitrust rules (2013) 

499  

double damages). Only with regard to costs the assessment led to the identification of a preferred option more similar to one of the other scenarios (i.e. non-regulatory intervention).

EN

78

EN

500  Commission Staff Working Document Impact Assessment Report ANNEX 5 Table 12: Overview of the Policy Options which were analysed in the 2008 IAWP Option 1:

Option 2:

Option 3:

Damages

Double, all types of damage, including interest

Double for cartels; for rest full single (i.e. all types of damage, including interest)

Full single

Access to evidence

Broad disclosure, low threshold

Initial provision of lists + broad Disclosure of specified categories, based on disclosure, both based on fact-pleading and fact-pleading threshold proportionality

Indirect purchaser

Effect of NCA decisions

EN

Option 5:

Standing allowed

Passing-on Defence excluded

Defence allowed; facilitation of proof of pass-on in favour of indirect purchaser Binding across EU

Defence allowed

Binding in own Member State

Fault (once Strict liability infringement established)

Rebuttable presumption; exoneration for excusable errors

Strong probative value of finding of infringement

Collective redress

Opt-out class actions

Opt-in collective + representative actions

Representative actions

Limitation period

20 years as of damage or 5 years as of reasonable knowledge

Minimum 5 years as of reasonable knowledge + restart + two years

Minimum 5 years as of reasonable knowledge + suspension

Cost rule

One-way shifting

Loser pays, but judge may shift all costs

Loser pays, but judge may shift part of costs

Interaction with leniency

Option 4:

No legislative measures, only identification and No EU recommen- action at dation of all good practices in line with Option 3

Protection of corporate statements from disclosure; limitation of liability on the part of immunity recipient

79

EN

Damages actions for breach of the EU antitrust rules (2013) 

501  

ANNEX 6 Table 13: Overview of the Preferred Option in the 2008 IAWP Preferred Option

Main objective(s)

Further description

Damages

Full single

Full compensation, legal Entitlement to damages for actual loss, loss of certainty, greater awareness profit plus interest. of the rules.

Access to evidence

Disclosure of specified categories, based on factpleading and proportionality

Access to justice, Upon request by one of the parties, the national appropriate and efficient use court can order the disclosure of evidence held by of the judicial system. the other party or by a third party. Based on factpleading, under the control by the judge on proportionality and necessity.

Indirect purchaser

Standing allowed

Full compensation, access to Any individual can claim compensation for harm justice, legal certainty suffered, hence standing for both direct and indirect purchasers.

Passing-on Defence allowed facilitation of proof of passing-on in favour of indirect purchaser

Effect of NCA decisions

Binding across EU

Full compensation, legal If the direct purchaser has passed on the illegal certainty, access to justice. overcharge resulting from an infringement to his own customers, the infringer can invoke it as a defence against a damages action by the direct purchaser. Conversely, when indirect purchasers claim compensation from the infringer they should benefit from a rebuttable presumption that the illegal overcharge has been passed on to themin its entirety. Greater awareness of the National courts cannot take a decision running rules, increased enforcement counter to a final finding of an infringement by and improved compliance, an EU national competition authority. efficient use of the judicial system, increased legal certainty

presumption Full compensation, access to Fault (once Rebuttable exoneration for excusable justice infringement esta- errors blished)

Collective redress

Opt-in collective representative actions

+ Full compensation, access to justice, appropriate and efficient use of the judicial system

Limitation Minimum of 5 years as of Access to reasonable knowledge + certainty period restart + 2 years

Cost rule

EN

justice,

In those Member States that provide for a fault requirement in damages cases, once an infringements of Article 101 or 102 TFEU has been found, fault is presumed unless the infringer can show that the infringement was the consequence of an excusable error. Complementary collective redress mechanisms: collective actions by victims on an opt-in basis or representative actions brought by qualified entities designated in advance or authorised on a case by case basis by a Member State.

legal Limitation periods cannot start to run before the infringement ceases or is brought to an end, nor before the victim can reasonably be expected to have knowledge of the infringement and the harm it caused. If a competition authority finds an infringement of Article 101 or 102 TFEU, possibility to bring an action within two year of the final decision.

No legislative measure, Full compensation, access to Member States should reflect upon their cost only identification and justice, efficient use of the rules and consider procedural rules fostering recommendation of good judicial system settlements as well as design means to avoid that

80

EN

502  Commission Staff Working Document Impact Assessment Report practices in line with Option 3: loser pays, but judge may shift all or part of the costs Interaction Protection of corporate statements from disclosure; with limitation of liability on the leniency part of immunity recipient

EN

the loser-pays rule discourages victims from bringing a meritorious action.

Greater awareness of the rules, increased enforcement and improved compliance, to the benefit of Europe's competitiveness.

81

Corporate statements submitted by leniency applicants should not be disclosed in a damages action; further reflection on the possibility to limit liability of the immunity recipient only to its own direct and indirect purchasers.

EN

Damages actions for breach of the EU antitrust rules (2013) 

503  

ANNEX 7 Glossary of terms Action for damages Action by which an injured party or someone acting on behalf of one or more injured parties brings a claim for damages before a national court. Antitrust Field of competition law and policy. In the EU context, ‘antitrust’ refers both to the rules governing anti-competitive agreements and practices (cartels, other cooperation agreements, distribution agreements, etc.) based on Article 101 TFEU and to the rules prohibiting abuses of (existing) dominant positions based on Article 102 TFEU. Cartel Agreement and/or concerted practice between two or more competitors aimed at coordinating their competitive behaviour on the market and/or influencing the relevant parameters of competition, through practices such as the fixing or coordination of purchase or selling prices or other trading conditions, the allocation of production or sales quotas, the sharing of markets and customers including bid-rigging, restrictions of imports or exports and/or anti-competitive actions against other competitors. Collective redress A legal mechanism whereby claims (such as claims for damages) of a group of persons are brought collectively before a national court or tribunal (private enforcement of EU antitrust rules). Collective redress can take several forms, such as representative actions, group actions, etc. Representative actions are brought by a natural or legal person (e.g. consumer organisations, business associations or ombudsmen) on behalf of a group of persons who claim to be affected by an antitrust law infringement. Group actions are actions in which several victims combine their individual claims into one single legal action. Consensual dispute resolution Out-of-court resolution of a dispute between private parties (in this case the victims of a competition law infringers and the persons having infringed competition law). Consensual dispute resolution covers any form of out-of-court resolutions of disputes, including settlement, mediation, arbitration and other forms of out-of-court resolution of disputes. Direct purchaser Natural or legal person having purchased the goods or services which were subject of an infringement of competition law directly from one or more of the undertakings having infringed competition law. Disclosure of evidence Process by which the national court can order a party to the proceedings or a third party to submit to it and to the other party or parties to the proceedings pieces of evidence the former party has in its possession. The order can be issued on request of a party or ex officio. European Competition Network (ECN) The competition authorities of the Member States (NCAs) and the Commission form a network of public authorities: they act in the public interest and cooperate closely to protect competition. This network is a forum for discussion and cooperation in the application and

EN

82

EN

504  Commission Staff Working Document Impact Assessment Report enforcement of EU competition policy. It provides a framework for European competition authorities to cooperate in cases where Articles 101 and 102 TFEU are applied, and for flexible rules on the allocation of cases between the authorities. It is the basis for creating and maintaining a common competition culture in Europe. The European Competition Network was created on the basis of Regulation No 1/2003. ECN Model Leniency Programme A document endorsed by the ECN members. It aligns the key elements of leniency policies within the ECN to increase the effectiveness of leniency programmes in the EU and to simplify the burden for applicants and authorities in case of multiple filings. The Model Programme sets out the essential procedural and substantive elements that ECN members believe every leniency programme should contain. The ECN authorities made a political commitment to use their best efforts to align their leniency programmes with the ECN Model Leniency Programme or to introduce aligned programmes. However, this document is not a programme as such under which applicants could apply for leniency and it does not raise legitimate expectations. Fine A monetary penalty imposed by a Commission or NCA decision on an undertaking, for a violation of EU competition rules. Immunity recipient An undertaking receiving immunity from any fine to be imposed in application of a leniency programme. Indirect purchaser Natural or legal person which did not purchase the goods or services which were subject of a competition law infringement directly from one or more of the undertakings having infringed competition law, but who purchased those goods from the direct customer. An indirect purchaser is a (direct or indirect) customer of the direct purchaser who is further down the distribution or supply chain. (Leniency) corporate statement A voluntary presentation by, or on behalf of, an undertaking to a competition authority, describing the undertaking’s knowledge of a secret cartel and its role therein, which was drawn up specifically for submission to the authority with a view to obtaining immunity or a reduction of fines under a leniency programme concerning the application of Article 101 of the Treaty or the corresponding provision under national law. Leniency programme (See also Annex 1) A programme on the pasis of which a participant to a secret cartel, independently from the other undertakings involved in the cartel, co-operates with the investigation of the competition authority, by providing voluntary presentations of its knowledge of the cartel and its role therein, in return for which the participant receives immunity from any fine to be imposed for the cartel or a reduction of such fine. The Commission's current leniency programme is set out in the 2006 notice on immunity from fines and reduction of fines in cartel cases. Limitation period The period during which an action can be brought before a national court. In case of this impact assessment, limitation periods concern the period during which injured parties of an infringement of competition law can bring an action for damages to obtain full compensation for the harm they suffered as a result of such infringement.

EN

83

EN

Damages actions for breach of the EU antitrust rules (2013) 

505  

National Competition Authority (NCA) National competition authorities (NCAs) are the authorities designated by the Member States pursuant to Article 35 of Regulation 1/2003 as responsible for the application of Article 101 and 102 TFEU in their territories. EU law obliges Member States to ensure that NCAs are set up and equipped in such a way that the provisions of Regulation No 1/2003 are effectively complied with. Together with the Commission, the competition authorities from Member States form the European Competition Network (ECN). National court A court or tribunal of a Member State that is authorised to seek a preliminary ruling from the Court of Justice of the European Union pursuant to Article 267 of the Treaty. Overcharge Any positive difference between the price actually paid and the price that would have prevailed in the absence of an infringement of competition law. Passing-on defence A defence against a direct purchaser's damages claim, relying on evidence showing that the overcharge resulting from a cartel was passed on – fully or partially - by the direct purchaser to its own customers further down the distribution chain (the indirect purchasers). Quantification of harm The process by which the amount of harm suffered by the injured party is determined by the national court, and normally expressed in monetary terms. Rebuttable presumption A means of evidence by which the existence or non-existence of a fact is deemed to be proven, unless evidence to the contrary is brought. Regulation No 1/2003 Council Regulation setting out the main rules for the enforcement of EU antitrust rules (Articles 101 and 102 TFEU). This Regulation, which took effect on 1 May 2004, has modernised the procedural rules governing how EU antitrust rules are enforced. It put an end to the notification system under which companies notified agreements to the Commission for approval under the antitrust rules. Regulation 1/2003 entrusts, in parallel with the Commission, competition authorities of the Member States (NCAs) and national courts with the role of applying Article 101 and 102 TFEU. This means there is wide-spread enforcement of the same set of rules to prosecute cartels and other anti-competitive practices throughout Europe. Regulation 1/2003 also forms the basis for the European Competition Network (ECN) in the framework of which the Commission and NCAs coordinate the application of EU antitrust rules. Regulation 1/2003 replaced Regulation No 17/1962. Settlement procedure The settlement procedure is a simplified procedure in cartel investigations, which results in a faster handling of the case and in a reduction of the fines. In order to participate in this procedure, the undertakings involved have to acknowledge their participation in the cartel. Settlement submission A voluntary presentation by, or on behalf of, an undertaking to a competition authority describing the undertaking’s acknowledgement of its participation in an infringement of Article 101 of the Treaty or a corresponding provision under national law and its liability for

EN

84

EN

506  Commission Staff Working Document Impact Assessment Report this infringement, which was drawn up specifically as a formal request for the authority to apply a simplified procedure and a reduction of the fine.

EN

85

EN

Damages actions for breach of the EU antitrust rules (2013) 

507  

ANNEX 8 Overview of the results of the Public Consultation on the White Paper on Antitrust Damages Actions (2008) On 2 April 2008, the Commission published for public consultation a White Paper on damages actions for breach of the EU antitrust rules. This overview summarises (I) the main trends in the submissions and (II) the specific suggestions received on the individual sections of the White Paper. 1.

MAIN TRENDS AMONG RESPONDENTS TO THE WHITE PAPER

The White Paper triggered a large number of submissions (more than 170) by governments of several Member States, national parliaments, national competition authorities, judges, businesses and business associations, law firms, consumer associations, academics and individuals. In terms of geographical spread, comments were issued by public authorities or stakeholders from almost all Member States. All the submissions are published on the website of DG Competition.113 Of all the respondents, very few questioned the underpinning idea of the White Paper, namely that in most cases victims of competition law infringements in practice do not obtain the compensation they are entitled to under the Treaty. The majority of respondents agreed that something needs to be done to address this issue, while divergent opinions emerged as to the specific measures to be adopted. There was unanimous support for the approach of the White Paper to pursue, as primary policy objectives, the effective compensation of victims (not punishment of infringers) and the preservation of strong public enforcement. 2.

SPECIFIC SUGGESTIONS ON INDIVIDUAL SECTIONS OF THE WHITE PAPER

(a)

Standing: indirect purchasers and collective redress

Standing for indirect purchasers, already recognized in the case-law of the Court of Justice, did not give rise to many comments. As regards collective redress, a majority of respondents from the business community observed that the mechanisms proposed in the White Paper could increase litigation costs for business and that safeguards are necessary to avoid abuses. Consumer associations emphasised the need for collective redress mechanisms, possibly optout representative actions, in order to provide effective compensation also for consumers, who typically suffer low-value and scattered damage. There was also a wide support for a consistent approach to collective redress across different EU policy areas. (b)

Access to evidence

Many respondents from the business community warned that rules on inter partes disclosure should equally work for claimants and defendants in order not to unbalance their procedural position, and contended that safeguards should be put in place to avoid 'fishing expeditions' and to protect confidential information. On the latter issue, some suggested protective mechanisms such as confidentiality rings or confidentiality orders. Other respondents (e.g. consumer associations and several law firms) argued that the conditions for disclosure suggested in the White Paper are too strict, as they would still make it difficult for a claimant to bring a case, especially when he cannot rely on a prior finding of 113

EN

http://ec.europa.eu/comm/competition/antitrust/actionsdamages/white_paper_comments.html.

86

EN

508  Commission Staff Working Document Impact Assessment Report an infringement by a national competition authority. Victims often are not in a position to know which precise categories of documents among those held by the infringer are needed to help proving the case. (c)

Binding effect of final NCA decisions

Many welcomed that the measure would foster consistent application of competition rules and legal certainty, and avoid unnecessary costs and delays due to re-litigation of the same issue. Other commentators criticized that a rebuttable presumption would be sufficient and more in line with the role of the judge. Some stakeholders recommended that the rights of defence should be fully taken into account when considering this measure, especially in view of possible different standards of proof for civil courts and competition authorities. (d)

Fault requirement

Some stakeholders, especially among business respondents, argued that the fault requirement should not be specifically regulated in the antitrust sector, but should follow the applicable general provisions of civil law. It was also argued that the proposed test of 'excusable error' could amount to a strict liability test, in particular in situations where self-assessment is not easy for firms. Consumers, some Member States and legal practitioners, on the other hand, welcomed the Commission's proposal. Some of them suggested that an infringer should never be allowed to escape liability on grounds of fault (referring to ECJ case-law). Many respondents called for a better definition of the excusable error test. (e)

Damages

All respondents welcomed the choice of compensation and single damages as the guiding principle of the White Paper. Some within the business community asked the Commission to clearly rule out punitive damages also under national law. Others argued that Member States should be left free to introduce them in their national legal order. The vast majority of respondents welcomed the Commission's commitment to produce guidance on the quantification of damages, as long as it is not binding and leaves judges to take account of the particularities of individual cases. Regarding quantification methods, for instance calculation of damages in the aggregate or based on the illicit gain was opposed by some companies but held necessary by consumer associations for the effective functioning of representative actions. (f)

Passing-on of overcharges

The passing-on of overcharges as a defence against damages claims was broadly welcomed, especially by business respondents. The White Paper proposal that indirect purchasers should benefit from a rebuttable presumption that the illegal overcharge has been passed on to them in its entirety has been contested in submissions by companies and business associations. It has been argued that such presumption would expose defendants to the risk of double or multiple compensation. Others stressed that, if the passing-on defence is allowed, a statutory facilitation of the proof for the victims is necessary, while others stressed that, absent a facilitation, the infringer would on one side benefit from the defence and then not compensate anyone. To avoid multiple compensation, some suggested consolidating cases so that the arguments of direct and indirect purchasers could be heard in the same proceedings. Others proposed that the first claimant (either direct or indirect purchaser) is awarded the entire compensation and that the second claimant should sue the first to obtain his share. (g)

EN

Limitation periods

87

EN

Damages actions for breach of the EU antitrust rules (2013) 

509  

Several respondents welcomed the White Paper's suggestions as they would give claimants the necessary time to file a lawsuit after an NCA/Commission decision. Some preferred a suspension of the limitation period until the NCA/Commission decision is taken over a new limitation period starting then. Businesses and some law-firms argued that the provision of a new limitation period in follow-on cases may not be in line with needs of legal certainty. Others suggested introducing, at European level, an objective limitation period running from the occurrence of the damage to ensure that at some stage claims would be time-barred in the interest of legal certainty. (h)

Costs of damages actions

Potential claimants agreed that Member States should assess their cost regimes in order to address the obstacles often faced by victims. They recalled that the cost risk of antitrust damages actions is often the main disincentive preventing victims from seeking redress. A few stakeholders also made the case for contingency fees, insurance schemes, purchase of claims by third parties and public funding. Nevertheless, most respondents from industry and law practice stressed their opposition to any revision of the loser-pays principle, seeing this general rule as a useful safeguard against abuses. Some argued for a derogation from the general rule in special circumstances based on the judge's discretion. Upfront cost orders were, however, widely rejected as an unjust burden on companies. There was broad support for early resolution of cases, although consumers insist that fair settlements can only be reached if an effective framework for antitrust damages actions is in place. (i)

Interaction between leniency programmes and actions for damages

The protection of corporate statements from disclosure was generally deemed important for the success of leniency programmes. Some suggested extending this protection to all documents submitted by leniency applicants. Other submissions, instead, suggested that private and public enforcement are two separate issues and that there are no grounds to protect leniency submissions from disclosure. A similar reasoning led several respondents to reject a limitation of liability for the immunity recipient, if it is formulated as a limitation of the right to compensation. Some submissions from businesses welcomed this proposal as an incentive to apply for leniency.

EN

88

EN

Annex III Proposal for a Council Directive on rules governing damages actions for infringements of Articles 81 and 82 of the Treaty (2009)

COMMISSION INTERNAL

EN ref number COMMISSION OF THE EUROPEAN COMMUNITIES

Brussels, XX.X.2009 Ref institutionnelle

Proposal for a COUNCIL DIRECTIVE on rules governing damages actions for infringements of Articles 81 and 82 of the Treaty

EXPLANATORY MEMORANDUM 1.

CONTEXT OF THE PROPOSAL

1.1.

Grounds for and objectives of the proposal

The present legislative proposal gives effect to Articles 81 and 82 (hereafter also: "the EC competition rules") of the Treaty establishing the European Community (hereafter: "the Treaty") by rendering more effective the right of victims of infringements of these Articles to obtain compensation for the harm they have suffered. As the Court of Justice (hereafter: "the ECJ") has ruled, this right to compensation is guaranteed by the Treaty itself1; its exercise is, however, in practice often impeded. The objective of the proposed Directive is to ensure, through the introduction of a set of rules designed to address the most important obstacles

1

EN

ECJ, Case C-453/99, Courage and Crehan, [2001] ECR I-6297, and Joined Cases C-295 – 298/04, Manfredi, [2006] ECR I-6619.

1

EN

514  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty currently standing in the way of effective reparation, that all victims are in a position to obtain full compensation of the damage caused by an infringement of the EC competition rules. 1.2.

General context

Regulation 1/2003 gives effect to the EC competition rules by setting out the conditions under which the Commission and the competition authorities of the Member States apply Articles 81 and 82 of the Treaty in individual cases. The Commission and national competition authorities can make findings of an infringement of Articles 81 or 82 of the Treaty. Article 23 of Regulation 1/2003 gives the Commission the power to impose fines on undertakings which have – intentionally or negligently – infringed Articles 81 or 82 of the Treaty. The application of the Treaty rules by the Commission and national competition authorities is commonly referred to as "public enforcement". Articles 81 and 82 of the Treaty are directly effective; they create rights and obligations for individuals which must be protected by the courts of the Member States.2 National courts have the power to apply Articles 81 and 82 of the Treaty.3 It follows from this direct effect of the prohibitions laid down in Articles 81 and 82 of the Treaty that any individual can claim compensation for the harm suffered where there is a causal relationship between that harm and an infringement of the EC competition rules.4 Injured persons must be able to seek compensation not only for actual loss (damnum emergens) but also for loss of profits (lucrum cessans) plus interest.5 Compensation cannot be achieved through public enforcement of the competition rules; it is a specific function which is of the domain of courts and of civil law and procedure. While the right to compensation is guaranteed by the Treaty itself and is part of the acquis communautaire, the practical exercise of this right is often rendered difficult or impossible. Although there have recently been some signs of improvement in certain Member States, to date in practice victims of infringements of the EC competition rules only rarely obtain reparation of the harm suffered. It is estimated that these victims forego several billions of Euros each year in compensation; this means that the costs associated with infringements of the EC competition rules are borne by the victims themselves. For example, monies which have illegally been transferred to the infringers in the form of a price overcharge are not paid back to their rightful owners; there is thus a deficit in compensatory justice in the enforcement of the EC competition rules. In its 2005 Green Paper6 the Commission identified the main obstacles to a more effective system of damages claims and put forward, for further reflection and debate, a number of options designed to improve the effectiveness of antitrust damages actions. On the basis of the public consultation on the Green Paper, the 2008 White Paper7 put forward proposals for policy choices and specific measures that would ensure more than is the case 2 3 4 5 6 7

EN

ECJ, Case 127/73, BRT and SABAM, [1974] ECR 51, para. 16; Case C-282/95 P, Guérin Automobiles v Commission [1997] ECR I-1503, para. 39. Article 6, Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, OJ [2003] L 1/1. ECJ, Joined Cases C-295 – 298/04, Manfredi, [2006] ECR I-6619, para. 61. ECJ, Joined Cases C-295 – 298/04, Manfredi, [2006] ECR I-6619, para. 95. COM(2005) 672final; see also Commission Staff Working Paper annexed to the Green Paper, SEC(2005) 1732. COM(2008) 165final – with annexed Commission Staff Working Paper, SEC(2008) 404.

2

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

515  

today that all victims of infringements of EC competition law have access to effective redress mechanisms so that they can be fully compensated for the harm they suffered. The proposed Directive is based on the above-mentioned White Paper and takes into account the results of the public consultation on the White Paper and the opinions of the European Parliament8 and the European Economic and Social Committee9. The Directive is designed to – Guarantee that in all Member States there is access to effective redress mechanisms by alleviating major obstacles standing in the way of the victims’ right to obtain compensation, thus giving full effect to Articles 81 and 82 of the Treaty; – Establish common minimum guarantees on antitrust damages actions, thereby creating a more level playing field for undertakings and consumers, which in turn enhances predictability and legal certainty and reduces forum shopping. The proposed Directive takes a compensatory approach: its aim is to allow those who have suffered damage caused by an infringement of the EC competition rules to recuperate that loss from the undertaking(s) which infringed the law. The Commission recognises at the same time that more effective antitrust damages actions will have a deterrent effect, which it welcomes. 1.3.

Existing provisions in the area of the proposal

Article 2 of Regulation 1/2003 contains rules on the burden of proving an infringement of Article 81(1) or 82 of the Treaty (and for proving that the conditions of Article 81(3) of the Treaty are fulfilled). These rules are applicable in actions for the compensation of damages caused by an infringement of Articles 81 or 82 of the Treaty. Regulation 44/2001 contains rules on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters.10 Under the conditions set out in that Regulation courts of the Member States have jurisdiction to hear antitrust damages actions and judgments in such actions are recognized and enforced. Article 6(3) of Regulation 864/2007 contains rules on the law applicable in antitrust damages actions.11 1.4.

Consistency with the other policies and objectives of the Union

The proposed Directive is consistent with the other policies and objectives of the Union, and notably with the wider competition policy of the Union.

8 9 10 11

EN

Insert reference. Insert reference. Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, OJ [2001] L 12/1. Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II), OJ [2007] L 199/4.

3

EN

516  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty 2.

CONSULTATION ASSESSMENT

OF

INTERESTED

2.1.

Consultation of interested parties

PARTIES

AND

IMPACT

Both the Green Paper and the White Paper triggered a broad debate among stakeholders, and a large number of comments were submitted within the public consultation.12 The public consultation resulted in an almost unanimous approval for the general approach to antitrust damages actions pursued by the Commission, namely the guiding principle of compensation. Thus respondents welcomed the choice not to suggest measures such as opt-out class actions and pre-trial discovery, as well as multiple damages, which would pursue primarily an objective of deterrence through private damages actions. There was a wide acknowledgement of the existence of obstacles that prevent effective redress for victims of infringements of the competition rules. Different opinions were voiced as to the substantive measures suggested in order to remedy the problems at hand. 2.2.

Collection and use of external expertise

The 2005 Green Paper was prepared by an external study on the conditions for antitrust damages actions in the – then – 25 Member States.13 The Impact Assessment for the White Paper was prepared by an external study.14 2.3.

Impact assessment

The proposed Directive was preceded by an Impact Assessment, which built largely on the findings of the Impact Assessment on the White Paper. In particular measures that had been excluded in the White Paper because of their likely ineffectiveness or excessive costs have not been reconsidered. The impact assessment report focused on four options for a follow-up initiative aimed at ensuring a more effective legal framework for damages actions for breach of Articles 81 and 82 EC across Europe. The draft Directive is the most effective (and notably) cost effective way of achieving the objectives of the Commission. It builds on the measures put forward in the White Paper, with the exception of limitation of liability for the immunity recipient, and embodying additional safeguards concerning representative actions and access to evidence, as suggested by stakeholders in the framework of the public consultation.

12

13 14

EN

The written comments received by the Commission during the public consultation can be accessed here: http://ec.europa.eu/competition/antitrust/actionsdamages/green_paper_comments.html (Green Paper consultation) and here http://ec.europa.eu/competition/antitrust/actionsdamages/white_paper_comments.html (White Paper consultation). Study on the conditions of claims for damages in case of infringement of EC competition rules, accessible at: http://ec.europa.eu/competition/antitrust/actionsdamages/study.html. Making antitrust damages actions more effective in the EU: welfare impact and potential scenarios, accessible at: http://ec.europa.eu/competition/antitrust/actionsdamages/files_white_paper/impact_study.pdf.

4

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty   3.

LEGAL ELEMENTS OF THE PROPOSAL

3.1.

Legal basis of the proposal

517  

The proposal is based on Article 83 of the Treaty, because its provisions and its main objective is to give effect to the principles set out in Articles 81 and 82 of the Treaty. Indeed, the proposal aims at defining the relationship between national rules on civil liability and the application of Articles 81 and 82 of the Treaty. 3.2.

Subsidiarity principle

The proposed Directive respects the subsidiarity principle (Article 5(2) of the Treaty). The objectives of the proposed measure cannot be sufficiently achieved by the Member States and therefore require action at Community level: a legislative act at Community level can best guarantee that full effect is given across the EU to Articles 81 and 82 of the Treaty in damages actions. In the absence of Community legislation, a disparity of policy choices at national level concerning damages actions for infringements of the EC competition rules would not ensure full effect of Articles 81 and 82 of the Treaty: – The proposed Directive is designed to render effective a right which is part of the acquis communautaire and which directly stems from Articles 81 and 82 of the Treaty: this in itself justifies a heightened Community interest in that consistent policy choices be made with respect to the implementation of this right. There is currently a marked inequality between the Member States in the level of judicial protection of this individual right guaranteed by the Treaty. This is particularly unsatisfying as Articles 81 and 82 of the Treaty are public order rules which by their nature concern cases that affect trade between Member States, and have a specific role for the internal market; – Damages actions for infringements of Articles 81 and 82 of the Treaty are part of the overall enforcement of these rules. Consistent policy choices on the basic underlying principles will better ensure that damages actions can be coordinated with the public enforcement of the competition rules. For example, common rules (on such issues as protection of corporate statements under the immunity/leniency programmes of the Commission or national competition authorities, binding effect of decisions by national competition authorities) are appropriate to ensure adequate coordination between public and private enforcement; – The sometimes large differences in the level of legal protection of the rights also distort the competitive environment for businesses, as the likelihood and scope of claims for damages against undertakings directly affect their competitive position. Isolated initiatives by Member States are not capable of producing a more level playing field for businesses and to reduce the uncertainty created by the currently big differences between the national legal systems. On the contrary, individual initiatives may even widen the gaps and increase the risk of negative impacts resulting from forum shopping. The need for a level competitive playing field is implicit in the concept of the internal market, which relies on a system ensuring that competition is not distorted; – Similarly, disparate policy choices by the Member States would lead to unequal protection for victims in an area where the acquis communautaire itself guarantees the victims right to effective compensation: a claim under the law of one Member State should not lead to

EN

5

EN

518  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty full recovery of the claimant’s loss, while at the same time a claim for an identical infringement in another Member State may lead to a significantly lower award or even no award at all. 3.3.

Proportionality principle

The proposed Directive respects the principle of proportionality as it includes only the minimum necessary to effectively achieve its objective, namely to guarantee that across the EU victims of infringements of EC competition law have access to a truly effective mechanism for obtaining full compensation for the harm they suffered. Furthermore, the costs imposed on citizens and businesses are proportionate to the stated objective. The proposed Directive excludes more radical, cost-intensive measures (for instance multiple damages, optout collective actions and wide pre-trial discovery rules). 3.4.

Choice of instruments

The objectives of the present legislative proposal can best be pursued through a Directive. This is the most appropriate legal instrument to make the measures work effectively and to allow for a smooth adaptation into the legal systems of the Member States: – A Directive allows the Member States to achieve the objectives and implement the measures into their national substantive and procedural law systems. This approach limits the freedom of the Member States less than other options (such as a Regulation), to the extent that Member States are left the choice of the most appropriate tools to implement the measures that are contained in the Directive. This allows Member States to ensure consistency of these rules with their wider substantive and procedural law. A Directive furthermore is a flexible tool to introduce a minimum standard in those areas of national law that are crucial for the functioning of damages actions, ensuring common minimum guarantees all across the EU, while leaving room to the individual Member States for further reaching measures, if they elect to adopt them. – Finally, a Directive avoids intervention in all those cases where the domestic provisions in the Member States are already in line with the proposed measures. 4.

DETAILED EXPLANATION OF THE PROPOSAL

4.1.

Purpose, Subject matter and definitions (Articles 1 – 2)

As the ECJ has held, the full effectiveness of Articles 81 and 82 of the Treaty requires that any individual can claim compensation for the harm caused by an infringement of these rules. The purpose of the proposed directive is to ensure that all individuals (consumers and undertakings) can effectively enforce this right to compensation guaranteed by the Treaty. The proposed Directive recalls the acquis communautaire on standing and on the definition of damage, as well as the principles of effectiveness and equivalence, that must be complied with by national legislation. To this extent, Article 1 of the proposed Directive is purely declarative. 4.2.

Right to damages/Collective redress (Articles 3 – 6)

Harm caused by infringements of Articles 81 and 82 of the Treaty is often scattered among a multitude of injured parties, typically consumers or small and medium-sized enterprises.

EN

6

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

519  

Therefore, availability of suitable collective redress mechanisms is crucial for such injured parties to have a realistic and effective chance to obtain compensation. Two complementary mechanisms, namely group and representative actions, are designed to enable injured parties to use the mechanism that is most appropriate to the circumstances of each particular case. By enabling courts to adjudicate on a number of similar individual damages claims in one proceeding, collective redress mechanisms also improve the overall efficiency in the administration of justice and reduce the risk of contradictory decisions. The proposed Directive sets out the main features of group and representative actions, including safeguards against abusive litigation, whilst leaving considerable scope to Member States to implement detailed rules that best fit into their existing legal systems. Whereas the proposed mechanisms of collective redress are considered necessary to guarantee a minimum level of protection of injured parties' right to damages, Member States are free to maintain or introduce also other forms of collective redress, including alternative dispute resolution mechanisms. 4.3.

Disclosure of evidence (Articles 7 – 9)

The finding of an infringement of Article 81 and 82 of the Treaty, the quantification of damages and the establishment of causality all typically require a complex factual and economic analysis. Many of the relevant facts are in the possession of the defendant or of third persons and are often not sufficiently known or accessible to the claimants ("information asymmetry"). It is widely recognized that the difficulty for a claimant of obtaining all the necessary evidence constitutes in many Member States one of the major obstacles to damages actions in competition cases. The proposed Directive, building on the approach adopted in Directive 2004/48/EC on the enforcement of intellectual property rights,15 will ensure that in all Member States there is a minimum level of effective access in antitrust damages actions to the evidence needed by injured parties to prove their claim. At the same time, the Directive avoids overly broad and costly disclosure obligations that would create undue burdens for the parties involved and risks of abuses. The Commission also paid particular attention to ensure that the proposal is compatible with the different legal orders of the Member States. To this end, the proposal follows the tradition of the great majority of Member States and relies on the central function of the court seised with an action for damages: disclosure of evidence held by the opposing party or a third party can only be ordered by judges and is subject to strict and active judicial control as to its necessity, scope and proportionality. National courts should have at their disposal effective measures to protect business secrets or otherwise confidential information disclosed during the proceedings. It is also necessary that courts can impose sufficiently deterrent sanctions to prevent destruction of relevant evidence or refusal to comply with a disclosure order. 4.4.

Passing-on of overcharges (Articles 10 – 11)

To ensure that those who suffered a loss caused by an infringement of the EC competition rules are compensated for that loss, while avoiding that those who infringed the law or those who were able to pass on their loss would be unjustly enriched, the Commission proposes:

15

EN

Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights, OJ 2004 L 195/16.

7

EN

520  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty – to allow the infringer to invoke as a defence the fact that the overcharge was (partially or in its entirety) passed on along the supply chain; – to put the burden of proof for this defence on the infringer; and – to provide for the compensation of the claimant for the loss resulting from a passing-on of the overcharge unless the infringing undertaking can show that the overcharge was not passed on to the claimant. Where claimants from different levels in the supply chain bring separate actions for damages that are related to the same competition law infringement, national courts are to be encouraged to take due account of the parallel or preceding actions in order to avoid underand over-compensation of the harm caused by that infringement. 4.5.

Effect of national decisions (Article 12)

According to Article 16(1) of Regulation 1/2003, decisions by the Commission have a binding effect on national courts (under the conditions set out in this Article). It is appropriate to give final infringement decisions by national competition authorities (or by a review court) a similar effect. If an infringement decision has already been taken, it would be inefficient to re-litigate the question whether there has been an infringement in subsequent damages actions. This does not entail a lessening of judicial protection for the undertakings concerned, as infringement decisions by national competition authorities are subject to judicial review. Also, the rights of defence of the undertakings concerned are protected by EU law. The rights and obligations of national courts under Article 234 of the Treaty are unaltered by this rule. 4.6.

Fault (Article 13)

Some Member States require, beyond the proof of an infringement of a statutory duty (such as Articles 81 and 82 of the Treaty), causation and damage, that an element of fault (intent or negligence) be proven for a damages action to succeed. Other Member States do not know this requirement ("strict liability"). The proposed Directive does not have an incident on the existence of this requirement as such: Member States which require fault to be proven as a condition for a damages claim are free to keep this requirement; Member States which do not have it are not required to introduce it. Where a Member State does require the proof of fault, the burden of proof should fall on the infringer: once an infringement of Articles 81 or 82 of the Treaty has been proven, the infringer will be – refutably – presumed to have acted at fault. This is an appropriate allocation of risks and is also justified by the consideration that usually the infringer has control over the relevant facts and evidence. It should also be recalled that undertakings have a very high duty of care with regard to their compliance with Articles 81 or 82 of the Treaty. 4.7.

Limitation periods (Article 14)

To give victims of a competition law infringement a reasonable opportunity to bring a damages action, while ensuring an appropriate level of legal certainty for all parties involved, the Commission proposes that the national rules on limitation periods for a damages action – allow victims sufficient time to bring an action after they became aware of the infringement and the harm it caused;

EN

8

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

521  

– prevent a limitation period from starting to run before the day on which a continuous or repeated infringement ceases and – allow victims at least two years to bring an action after a final finding of the infringement by a competition authority or a review court. 5.

BUDGETARY IMPLICATIONS

The present proposal does not have implications for the Community budget. 6.

ADDITIONAL INFORMATION

6.1.

Repeal of existing legislation

No previous legislative act is repealed through this present proposal. 6.2.

Review/revision/sunset clause

According to Article 16 of the proposed Directive, the Commission will review it and report to the European Parliament and the Council at the latest five years after the deadline for transposition into national law has expired. 6.3.

European Economic Area

The proposed Directive is relevant for the EEA because it relates to the effective enforcement of Articles 81 and 82 of the Treaty, by giving effect to these rules through private damages actions.

EN

9

EN

522  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty Proposal for a COUNCIL DIRECTIVE on rules governing damages actions for infringements of Articles 81 and 82 of the Treaty

[text with EEA relevance]

THE COUNCIL OF THE EUROPEAN UNION, Having regard to the Treaty establishing the European Community, and in particular Article 83 thereof, Having regard to the proposal from the Commission16, Having regard to the opinion of the European Parliament17, Having regard to the opinion of the European Economic and Social Committee18,

Whereas: (1)

In order to ensure that competition in the internal market is not distorted, Articles 81 and 82 of the Treaty must be applied effectively throughout the Community. These Treaty provisions are a matter of public policy, produce direct effects in relations between individuals and create, for the individuals concerned, rights and obligations which the national courts must enforce.

(2)

The full effectiveness of Articles 81 and 82 of the Treaty, and in particular the practical effect of the prohibitions laid down therein, requires that any individual can claim compensation before national courts for the harm caused to him by an infringement of these provisions.

(3)

The exercise of this Community right to compensation must be effectively ensured in all the Member States through a set of common standards. Such standards also increase legal certainty for consumers and undertakings alike, and strengthen the practical effect of competition rules for all the undertakings operating in the internal market.

16

OJ C […], […], p. […]. OJ C […], […], p. […]. OJ C […], […], p. […].

17 18

EN

10

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

EN

523  

(4)

This Directive sets out the rules that must be provided for in national law to enable injured parties to effectively exercise their right to compensation guaranteed by the Treaty. National provisions on damages actions remain unaffected as long as they meet the standards laid down in this Directive and do not hamper the achievement of its objectives.

(5)

All national rules governing the exercise of the right to compensation for harm resulting from an infringement of Article 81 or 82 of the Treaty, including those concerning aspects not dealt with in this Directive, must observe the principles of effectiveness and equivalence. This means that they may not be formulated or applied in a fashion that makes it excessively difficult or practically impossible to exercise the right to compensation guaranteed by the Treaty, and they may not be formulated or applied less favourably than those applicable to similar domestic actions.

(6)

It is appropriate to recall in this Directive the acquis communautaire regarding standing and the definition of damage. Any individual that has suffered harm can claim compensation for the actual loss (damnum emergens) and loss of profit (lucrum cessans), as well as payment of interest accruing from the time the harm occurred until it has actually been compensated. This right is recognised for consumers and undertakings alike, irrespective of the existence of a direct contractual relationship with the infringing undertaking, and regardless of whether or not there has been a prior finding of an infringement by a competition authority.

(7)

Harm caused by infringements of Articles 81 and 82 of the Treaty is often scattered among a multitude of injured parties, typically consumers or small and medium-sized enterprises. These injured parties are likely to be dissuaded from bringing an individual action for damages given the costs, complexities, legal uncertainties and the resulting risks associated with such action, in particular compared to the value of their individual claim. It is therefore important that effective mechanisms of collective redress are available for the parties injured by infringements of Articles 81 and 82 of the Treaty. A combination of two complementary mechanisms, namely group and representative actions, will enable injured parties to use the collective redress mechanism that is most appropriate to the circumstances of each particular case, thus increasing the likelihood that they will actually obtain compensation and making the administration of justice more efficient overall.

(8)

Qualified entities, such as consumer organisations, trade associations or state bodies, should be able to bring representative actions for damages on behalf of defined groups of injured parties. To ensure the effectiveness of such actions, qualified entities should not be required to individually identify all the injured parties belonging to the represented group. Qualified entities should fulfil eligibility conditions designed to ensure that representative actions are brought in the best interests of the injured parties and not in an abusive manner. Such eligibility should be subject to regular monitoring by the Member States.

(9)

Qualified entities should take appropriate measures to make the representative action known to the injured parties concerned so that they can exercise their right not to be represented in such action. It follows from the nature of a representative action that any decision on the merits must be binding on all the injured parties that are represented by the qualified entity at the time when such decision is adopted.

11

EN

524  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty

EN

(10)

Given that alternative dispute resolution mechanisms can serve as a useful means of resolving disputes concerning damages claims in the field of competition law, it is appropriate to call on the Member States to ensure that such mechanisms can be used effectively also by parties to court proceedings on representative actions.

(11)

Damages actions for infringements of Articles 81 and 82 of the Treaty typically require a complex factual and economic analysis. The evidence necessary for proving a damages claim is often held by the defendant or by third parties, and not sufficiently known and accessible to the claimant. In such circumstances, strict legal requirements for claimants to assert in detail all the facts of their case at the beginning of an action and to proffer exactly specified pieces of supporting evidence can unduly impede the effective exercise of the right to compensation guaranteed by the Treaty.

(12)

Given that evidence is an important element for bringing a damages action for infringement of Article 81 or 82 of the Treaty, it is appropriate to ensure that in all Member States effective means to access relevant evidence are available to injured parties, particularly in situations where information asymmetry leads to their being unable to specify individual pieces of evidence. In order to ensure equality of arms, these means should also be available to defendants in damages actions.

(13)

Disclosure of relevant evidence should take place upon decision of the court and under its strict control, especially as regards the necessity and proportionality of the disclosure measure. From the requirement of proportionality it follows, inter alia, that the request for disclosure should refer to as precise and narrow categories of evidence as is possible on the basis of reasonably available facts.

(14)

To ensure that relevant evidence containing business secrets or otherwise confidential information is in principle available in actions for damages, national courts should have at their disposal a range of suitable measures to protect such confidential information disclosed during the proceedings. These measures may include the possibility of hearings in private, judicial orders limiting the circle of persons entitled to take note of the evidence, and the instruction of experts to produce summaries of the information in an aggregated or otherwise non-confidential form. Measures protecting business secrets and other confidential information must not practically impede the exercise of the right to compensation.

(15)

The effectiveness and consistency of the application of Articles 81 and 82 of the Treaty by the Commission and national competition authorities require a common approach across the Community regarding the interaction of rules on disclosure and public enforcement of these Articles. In order to ensure willingness on the part of undertakings to produce voluntary statements acknowledging their participation in an infringement of Article 81 of the Treaty to a competition authority in the context of a request for application of a leniency programme or a settlement procedure, such statements should be excepted from disclosure of evidence. In addition, an exception to disclosure should apply to any disclosure measure that would unduly interfere with an ongoing investigation by a competition authority concerning an infringement of Article 81 or 82 of the Treaty.

(16)

The invoking of a claim for damages, or the start of an investigation by a competition authority, entails a risk that the undertakings concerned may destroy or hide evidence that is relevant for injured parties to substantiate a claim for damages. To prevent the

12

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

525  

destruction of relevant evidence and to ensure that court orders requesting disclosure are complied with, courts must be able to impose sufficiently deterrent sanctions. Insofar as parties to the proceedings are concerned, the sanction of drawing adverse inferences in the proceedings for damages can be particularly effective and can avoid time delays. Sanctions should also be available for non-compliance with obligations to protect confidential information and for abusive use of information obtained through disclosure.

EN

(17)

Consumers and undertakings that are harmed by an infringement of Article 81 or 82 of the Treaty are entitled to compensation for the actual loss and any loss of profit suffered. The actual loss results from the price difference between what was actually paid and what would have been paid in the absence of the infringement. When an injured party has reduced its actual loss by passing it on, entirely or in part, to its own purchasers (or its own suppliers, in the case of an infringement that relates to the supply of the infringing undertaking), the loss thus passed on no longer constitutes harm for which the party that passed it on has to be compensated. It is therefore appropriate to allow an infringing undertaking to invoke the passing-on of actual loss as a defence against a claim for damages.

(18)

Consumers or undertakings, to which actual loss has been passed on, suffer harm that has been caused by an infringement of Article 81 or 82 of the Treaty. While such harm should be compensated by the infringing undertaking, it may be particularly difficult for consumers or undertakings that did not themselves make any purchase from the infringing undertaking, to prove the scope of that harm. To alleviate the burden of proof for a claimant showing himself to be a purchaser of goods or services that are affected by the infringement, it is appropriate to presume that the actual loss initially suffered by the one who made the purchase from the infringing undertaking has been fully passed on to the claimant. The infringing undertaking should be allowed to rebut this presumption by showing that the actual loss has not been passed on. Where the infringing undertaking can only show that part of the actual loss has not been passed on, it should still compensate the claimant for the remaining part.

(19)

Actions for damages can be brought both by injured parties that purchased goods or services from the infringing undertaking and by purchasers further down the supply chain. These actions can be considered related within the meaning of Article 28 of Regulation 44/2001. Under this provision, where related actions are pending in the courts of different Member States, national courts other than the one first seized may stay proceedings or decline jurisdiction. Furthermore, to foster consistency between judgments resulting from parallel or consecutive proceedings, national courts are encouraged to take due account of any related action and of the resulting judgment, particularly where it finds that the passing-on has been proven.

(20)

Where national courts rule on agreements, decisions or practices under Article 81 or 82 of the Treaty which are already the subject of a Commission decision, they cannot take decisions running counter to the decision adopted by the Commission.19 To enhance legal certainty, to avoid inconsistency in the application of these Treaty provisions and to increase the effectiveness and procedural efficiency of actions for damages, also the finding of an infringement of Article 81 or 82 of the Treaty in a final

19

Council Regulation (EC) No 1/2003 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, OJ [2003] L 1/1, Article 16(1).

13

EN

526  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty decision of a national competition authority or a review court should not be called into question in actions for damages relating to the same infringement, regardless of whether or not the action is brought in the Member State of the authority or review court. This effect of decisions by national competition authorities and review courts shall be without prejudice to the rights and obligations of national courts under Article 234 of the Treaty. (21)

The laws of some Member States require, for damages actions, that an element of fault (intent or negligence) be proven in addition to the proof of an infringement of Article 81 or 82 of the Treaty. As the evidence needed to prove fault is rarely available to injured parties, it can be excessively difficult or even impossible for them to bring this proof. Once it is established that an undertaking has infringed Article 81 or 82 of the Treaty, it is no longer appropriate that the injured parties bear the burden of proof. It should be for the infringing undertaking to demonstrate that it did not act intentionally or negligently, taking into account the high standard of care that flows from the fact that Articles 81 and 82 of the Treaty are a matter of public policy.

(22)

Injured parties should be able to bring an action for damages before, in parallel to or after proceedings by a competition authority, aimed at enforcing Article 81 or 82 of the Treaty. To this end, national rules on the beginning, duration, suspension or interruption of limitation periods should not unduly hamper the bringing of damages actions. That holds particularly for actions that build upon the finding of an infringement by a competition authority or by a review court.

(23)

In accordance with the principles of subsidiarity and proportionality as set out in Article 5 of the Treaty, this Directive pursues objectives that can be better achieved at Community level for the reasons set out above, and does not go beyond what is necessary to achieve those objectives.

(24)

This Directive respects the fundamental rights and observes the principles recognised in particular by the Charter of Fundamental Rights of the European Union. Accordingly, this Directive should be interpreted and applied with respect to those rights and principles.

HAS ADOPTED THIS DIRECTIVE:

CHAPTER I PURPOSE, SUBJECT MATTER AND DEFINITIONS Article 1 Right to compensation 1.

EN

The purpose of this Directive is to ensure that consumers and undertakings can effectively enforce their right, guaranteed by the Treaty, to full compensation for harm caused to them by an infringement of Article 81 or 82 of the Treaty.

14

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

527  

2.

Full compensation includes compensation for actual loss (damnum emergens), loss of profit (lucrum cessans) and payment of interest from the time the harm occurred until it has actually been compensated.

3.

All national rules and procedures concerning actions for damages shall be designed and applied in a manner that ensures effective exercise of the right to full compensation for harm caused by an infringement of Article 81 or 82 of the Treaty. Further, any rules and procedures on actions for damages resulting from infringements of Article 81 or 82 of the Treaty must not be less favourable to the injured parties than those governing actions for damages resulting from infringements of national competition law. Article 2 Definitions

For the purposes of this Directive, the following definitions shall apply:

EN

1.

"claim for damages" means a claim for compensation of harm caused by an infringement of Article 81 or 82 of the Treaty;

2.

"action for damages" means an action by which a claim for damages is brought before a national court;

3.

"injured party" means anyone who claims compensation for harm caused by an infringement of Article 81 or 82 of the Treaty;

4.

"national competition authority" means an authority designated by a Member State pursuant to Article 35 of Regulation 1/2003 as being responsible for the application of Articles 81 and 82 of the Treaty;

5.

"competition authority" means the Commission of the European Communities or a national competition authority;

6.

"national court" or "court" means a court or tribunal within a Member State that can apply Articles 81 and 82 of the Treaty and that is authorised to seek a preliminary ruling from the Court of Justice of the European Communities pursuant to Article 234 of the Treaty;

7.

"infringement decision" means either a decision of a competition authority or review court that finds an infringement of Article 81 or 82 of the Treaty or a decision of a review court that upholds such finding of an infringement;

8.

"final" infringement decision means a decision of a competition authority or review court that can no longer be reviewed because it was not appealed within the applicable time limit or was confirmed upon appeal;

9.

"corporate statement" means a voluntary presentation by, or on behalf of, an undertaking to a competition authority, describing the undertaking’s knowledge of an infringement of Article 81 of the Treaty and its role therein, which was drawn up

15

EN

528  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty specially for submission to the authority with a view to obtaining immunity or a reduction of fines under a leniency programme concerning the application of Article 81 of the Treaty; and 10.

"settlement submission" means a voluntary presentation by, or on behalf of, an undertaking to a competition authority describing the undertaking’s acknowledgement of its participation in an infringement of Article 81 of the Treaty and its liability for this infringement, which was drawn up specially as part of a formal request for the authority to apply an expedited procedure.

CHAPTER II ACTIONS FOR DAMAGES Article 3 Actions for damages 1.

Member States shall ensure that any injured party can effectively claim full compensation by bringing an action for damages either individually or by means of a group action pursuant to Article 4.

2.

Member States shall further ensure that qualified entities within the meaning of Article 6 can effectively bring representative actions pursuant to Article 5 on behalf of injured parties.

3.

Member States shall not be prevented from maintaining or introducing other forms of actions for damages. Article 4 Group action

EN

1.

Member States shall ensure that two or more injured parties who suffered harm caused by the same infringement of Article 81 or 82 of the Treaty can jointly bring an action for damages ("group action").

2.

National courts shall treat a group action as one single action. However, the defendant shall not be prohibited from raising any defence that might be relevant with regard only to one or several of the claimants.

3.

Member States shall ensure that the court, having heard the parties, can allow other injured parties who suffered harm caused by the same infringement to join a group action after the court has been seised of such action, provided that this does not impair the sound administration of justice.

16

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

529  

Article 5 Representative action 1.

Member States shall ensure that qualified entities within the meaning of Article 6 can bring an action for damages on behalf of injured parties who suffered harm caused by the same infringement of Article 81 or 82 of the Treaty ("representative action").

2.

The qualified entity shall define the group of injured parties on behalf of which it brings the representative action. It shall not be required to individually identify all injured parties belonging to this group.

3.

The qualified entity shall take the appropriate means to make the representative action known to all those who may have claims for damages within the scope of the representative action. It shall inform the court of the means by which it has done so or intends to do so. The court shall either approve such means, if it is satisfied that they are suited to the particular circumstances of the case and provide those concerned with a reasonable opportunity to learn of the existence of the representative action, or request appropriate amendments.

4.

Member States shall ensure that any injured party can exercise its right not to be represented by the qualified entity in the representative action. Any decision by the court on the merits of the case shall be binding on all injured parties represented by the qualified entity at the time when such decision is adopted.

5.

Damages awarded in a representative action shall be distributed, to the largest possible extent, to the injured parties represented. Member States may allow that a part of the damages award is used to cover expenses reasonably incurred by the qualified entity in connection with the representative action.

6.

Member States shall ensure that effective mechanisms are available for the parties in representative actions to agree on a fair and efficient consensual resolution of the case. Article 6 Qualified entities

1.

EN

For the purposes of this Directive, “qualified entity” shall mean any body which is: (a)

representing legitimate interests and is designated by a Member State as an entity generally entitled to bring representative actions on behalf of injured parties who have suffered harm caused by infringements of Article 81 or 82 of the Treaty; or

(b)

representing legitimate interests of its members and is authorised by a Member State, in relation to a specific infringement of Article 81 or 82 of the Treaty, to bring a representative action on behalf of all or some of its members who have suffered harm caused by such infringement.

17

EN

530  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty 2.

Member States shall define the eligibility conditions to be fulfilled in order for a body to be designated or authorised as a qualified entity. Such conditions shall in particular ensure that the body has the capability to effectively bring a representative action and that it acts, or can be expected to act, in the best interests of those it represents.

3.

Member States shall take the measures necessary to ensure effective monitoring of the continued eligibility of the qualified entities they designate to bring representative actions. They shall further establish procedures allowing for withdrawal of the designation, in particular if there is evidence that a designated qualified entity abuses the right to bring representative actions or does not act in the best interests of those it represents.

4.

Member States shall notify the Commission of qualified entities that have been designated by them pursuant to paragraph 1(a) and of any designation withdrawals pursuant to paragraph 3. The Commission shall ensure that a list of all designated qualified entities is publicly available.

5.

Qualified entities which have been designated in one Member State in accordance with the conditions set out in this Article shall, without any additional requirements, have the right to bring representative actions in all other Member States according to the procedural rules applicable in these Member States. The same shall apply to authorised qualified entities with regard to the specific infringement.

CHAPTER III DISCLOSURE OF EVIDENCE Article 7 Disclosure of evidence

EN

1.

Where a claimant has presented reasonably available facts and evidence showing plausible grounds to suspect that it, or those it represents, suffered harm through infringement of Article 81 or 82 of the Treaty by the defendant, Member States shall ensure that national courts order evidence to be disclosed by the other party or by a third party, subject to the conditions set out in this Directive. Member States shall ensure that the right to disclosure of evidence is available also to defendants in actions for damages.

2.

National courts shall order the disclosure of evidence referred to in paragraph 1 where the party requesting disclosure has (a)

shown that evidence lying in the control of the other party or a third party is relevant to substantiate its claim or defence;

(b)

specified either pieces of this evidence or as precise and narrow categories of this evidence as it can on the basis of reasonably available facts; and

18

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty   (c) 3.

531  

shown that it is unable, applying reasonable efforts, to produce these.

Disclosure pursuant to this Directive shall not be ordered to the extent that it would be disproportionate. In assessing this, national courts shall consider the legitimate interests of all parties and third parties concerned. They shall, in particular, consider: (a)

the value of the claim for damages;

(b)

the likelihood that the addressee of the disclosure order infringed Article 81 or 82 of the Treaty;

(c)

the scope and cost of disclosure, especially for any third parties concerned; and

(d)

the fact that the evidence to be disclosed may contain confidential information, especially that of any third parties. Member States shall provide for effective measures so that national courts can protect confidential information to the greatest extent possible whilst ensuring that relevant evidence containing such information is available in the action for damages.

4.

Member States shall ensure that national courts, before adopting a disclosure order and except in cases of particular urgency, hear the addressee of such an order in accordance with the principles of fair trial.

5.

Evidence within the meaning of this Directive includes all types of evidence admissible before the national court seised, in particular documents and all other objects containing information irrespective of the medium on which the information is stored. It also includes, to the extent and in the form permitted under applicable national rules, the hearing of witnesses, parties, experts and other third parties.

6.

Member States shall not be prevented from maintaining or introducing rules which provide for wider disclosure of evidence. Article 8 Exceptions from disclosure of evidence

EN

1.

Member States shall ensure that national courts at no point in time order the disclosure of corporate statements or settlement submissions.

2.

Member States shall ensure that national courts refrain, to the extent necessary, from ordering disclosure upon application by a competition authority that has shown to the court that disclosure would undermine an ongoing investigation concerning a suspected infringement of Article 81 or 82 of the Treaty.

3.

Member States shall take the necessary measures to give full effect to all legal privileges and other rights not to be compelled to disclose evidence that exist under the law of the European Union.

19

EN

532  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty Article 9 Sanctions 1.

2.

Member States shall ensure that national courts can impose sanctions on parties, their legal representatives and third parties, in the event of (a)

failure or refusal to comply with a court's disclosure order;

(b)

the destruction of relevant evidence by a party or a third party, provided that at the time of destruction the party or third party knew, or could reasonably have known, that the evidence was relevant to substantiate a claim for damages or a related defence, and that either such claim had been invoked by an injured party, or a competition authority had started an investigation in relation to the infringement underlying such a claim for damages;

(c)

failure or refusal to comply with the obligations imposed by a court order protecting confidential information; and

(d)

abuse of the rights relating to disclosure of evidence provided for in this Directive, and of the evidence and information obtained thereunder. Abuse in this context shall include use for a purpose other than substantiating a claim for compensation of harm caused by an infringement of Article 81 or 82 of the Treaty, or substantiating a defence against such a claim.

Member States shall ensure that the sanctions imposed by national courts are effective, proportionate and dissuasive. The sanctions available to national courts shall include, insofar as the behaviour of a party to damages action proceedings is concerned, the possibility to draw adverse inferences such as presuming the relevant issue to be proven or dismissing claims and defences in whole or in part, and the possibility to order the payment of costs.

CHAPTER IV PASSING-ON OF OVERCHARGES Article 10 Passing-on of overcharges 1.

EN

Member States shall ensure that the defendant in an action for damages can invoke as a defence against a claim for damages the fact that the claimant passed on the whole or part of the overcharge imposed upon him. For the purposes of this Article, any positive difference between the price actually paid and the price that would have prevailed in the absence of infringement of Article 81 or 82 of the Treaty, is referred to as an overcharge. The burden of proving this passing-on defence shall rest with the defendant.

20

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty   2.

533  

Member States shall ensure that, without prejudice to claims for loss of profit, the claimant in an action for damages is compensated for the harm resulting from a passing-on of the total overcharge where he has shown that (a)

the defendant has infringed Article 81 or 82 of the Treaty;

(b)

this infringement resulted in an overcharge for the customer(s) of the defendant; and

(c)

the claimant has purchased the goods or services that were the subject of the infringement or has purchased goods or services that are derived from or contain the goods or services that were the subject of the infringement.

This provision is without prejudice to the defendant's right to show that the overcharge was not passed on to the claimant. 3.

Member States shall ensure that the rules laid down in this Article apply also where the infringement of Article 81 or 82 of the Treaty relates to the supply of the infringing undertaking. Article 11 Parallel and consecutive actions for damages Member States shall enable national courts that are seised with an action for damages to take due account of parallel or preceding actions for damages that are related to the same infringement of Article 81 or 82 of the Treaty, but are brought by a claimant from another level in the supply chain.

CHAPTER V EFFECT OF NATIONAL DECISIONS, FAULT AND LIMITATION PERIODS Article 12 Effect of national decisions Where national courts rule, in actions for damages, on agreements, decisions or practices under Article 81 or 82 of the Treaty which are already the subject of a final infringement decision by a national competition authority or by a review court, Member States shall ensure that the national courts cannot take decisions running counter to such infringement decision. This obligation is without prejudice to the rights and obligations under Article 234 of the Treaty.

EN

21

EN

534  EU Proposal regarding infringements of Articles 81 and 82 of the Treaty Article 13 Fault To the extent that Member States require, in addition to proof of an infringement of Article 81 or 82 of the Treaty, fault to be proven in actions for damages, they shall ensure that, once an injured party has established such an infringement, the infringer is liable for damages unless he shows that he could not reasonably have been aware that his conduct distorted competition. Article 14 Limitation periods 1.

Member States shall lay down the rules stipulating when the limitation period for seeking compensation for harm caused by an infringement of Article 81 or 82 of the Treaty begins to run, the duration of that period and the circumstances under which that period can be interrupted or suspended. Member States shall ensure that the applicable national rules do not prevent an injured party from effectively seeking compensation once it knows, or can reasonably be expected to have knowledge, of the infringement and of the fact that the infringement caused harm to it.

2.

Member States shall ensure that the limitation period does not begin to run before the day on which a continuous or repeated infringement ceases.

3.

Where compensation is sought for harm caused by an infringement that was found by a competition authority or a review court, Member States shall ensure that the injured party can bring an action for damages during a period of at least two years after the infringement decision has become final.

CHAPTER VI FINAL PROVISIONS Article 15 Transposition

EN

1.

Member States shall bring into force the laws, regulations and administrative provisions necessary to comply with this Directive by [...] at the latest. They shall forthwith communicate to the Commission the text of those provisions and a correlation table between those provisions and this Directive.

2.

When Member States adopt those provisions, they shall contain a reference to this Directive or be accompanied by such a reference on the occasion of their official publication. Member States shall determine how such reference is made.

22

EN

EU Proposal regarding infringements of Articles 81 and 82 of the Treaty  

535  

Article 16 Review The Commission shall review this Directive and report to the European Parliament and the Council by [...] at the latest [to be calculated as 5 years after the date set as the deadline for transposition of this Directive]. Article 17 Entry into force This Directive shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union. Article 18 Addressees This Directive is addressed to the Member States. Done at Brussels, […]

For the Council The President […]

EN

23

EN

References

I. Publication, speeches and presentations Melanie Aitken, ‘Remarks’, Cambridge, Ontario, 24 February 2011, http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03377.html Almunia, Joaquín, ‘Common Standards for Group Claims Across the EU’, speech, Valladolid, 15 October 2010, http://europa.eu/rapid/pressReleasesAction.do? reference=SPEECH/10/554&format=PDF&aged=1&language=EN&guiLan guage=en Areeda, Philip and Herbert Hovenkamp, Antitrust Law, 3rd edn, Wolters Kluwer, 2007 Arnull, Anthony, The European Union and its Court of Justice, OUP, 1999 Ashton, David and David Henry, eds, Competition Damages Actions in the EU: Law and Practice, Edward Elgar, forthcoming Azizi, Josef, ‘Unveiling the EU Courts’ Internal Decision-Making Process: A Case for Dissenting Opinions?’, 12 ERA Forum 49 (2011) Baker, Donald I, ‘The Use of Criminal Law Remedies to Deter and Punish Cartels and Bid-Rigging’, 69 George Washington Law Review 693 (2001) Baker, Donald I, ‘The UK marine hoses prosecutions: beachhead or aberration?’ Global Competition Review, August/September 2008 Bane, Charles A, The Electrical Equipment Conspiracy: The Treble Damage Actions, New York Legal Publications, 1973 Blanke, Gordon and Phillip Landolt, eds, EU and US Antitrust Arbitration, Wolters Kluwer, 2011 Bloom, Anne Bloom, ‘From Justice to Global Peace: A (Brief) Genealogy of the Class Action Crisis’, 39 Loyola of Los Angeles Law Review 719 (2006) Bos, Iwan and Erik Pot, ‘On the Possibility of Welfare-Enhancing Hard Core Cartels,’ 107 Journal of Economics 199 (2012) Bourgeois, Jacques, ‘EC Competition Law and Member State Courts’, in Barry Hawk, ed, Antitrust in a Global Economy 1993, Annual Proceedings of the Fordham Corporate Law Institute, Transnational/Juris Publications, 1994, chapter 21 Bourgeois, Jacques and Stéphanie Strievi, ‘EU Competition Remedies in Consumer cases: Thinking Out of the Shopping Bag’, 33 World Competition 241 (2010)

538  References Braakman, August, ‘The Application of the Modernised Rules Implementing Articles 81 and 82 EC Treaty in Injunction Proceedings: Problems and Possible Solutions’, in Barry Hawk, ed, International Antitrust Law and Policy 1999, Annual Proceedings of the Fordham Corporate Law Institute, Juris Publishing, 2000, chapter 1 Brent, Richard, ‘The Binding of Leviathan? The Changing Role of the European Commission in Competition Cases’, 44 International and Comparative Law Quarterly 255 (1995) Buccirossi, Paolo and Michele Carpagnano, ‘Is it Time for the European Union to Legislate in the Field of Collective Redress in Antitrust (and how)?’, 4(1) Journal of European Competition Law and Practice 3 (2013) Calkins, Stephen, ‘An Enforcement Official’s Reflections on Antitrust Class Actions’, 39 Arizona Law Review 413 (1997) Calvani, Terry and Torello Calvani, ‘Custodial sanctions for cartel offences: An appropriate sanction in Australia?’, 17 Competition and Consumer Law Journal 119 (2009) Canivet, Guy, ‘The Responsibility of the Judiciary in the Implementation of Competition Policy’, in OECD, ‘Judicial Enforcement of Competition Law’, 1997 Canivet, Guy, Laurence Idot and Denys Simon, eds, Lamy procédures communautaires, Wolters Kluwer, 2005 Cavanagh, Edward D, ‘The Private Antitrust Remedy: Lessons from the American Experience’, 41 Loyola University of Chicago Law Journal 629 (2010) Charleton, Peter and Marguerite Bolger, ‘The Competition (Amendment) Act, 1996: Extending the Criminal Law’, 3 The Bar Review 214 (1998) Coffee, John C, Jr, ‘The Regulation of Entrepreneurial Litigation: Balancing Fairness and Efficiency in the Large Class Action’, 54 University of Chicago Law Review 877 (1987) Collins, Philip, ‘New decade, new Government—Reflections on possible evolution of the UK’s competition and consumer regimes’, speech, London, 20 May 2010 Craig, Paul, ‘Once upon a Time in the West: Direct Effect and the Federalization of EEC Law’, 12 Oxford Journal of Legal Studies 453 (1992) Craig, Paul and Gráinne De Búrca, EU Law: Text, Cases and Materials, 4th edition, OUP, 2008 Dalheimer, Dorothe Christoph Feddersen and Gerald Miersch, eds, EUKartellverfahrensverordnung, Kommentar zur VO 1/2003 (Munich, 2005), Dashwood, Alan and Angus Johnston, eds, The Future of the Judicial System of the European Union, Hart Publishing, 2011

References   539   Di Stefano, Gianni, ‘Access of damage claimants to evidence arising out of EU cartel investigations: a fast evolving scenario’, [2012/3] Global Competition Litigation Review 95 Drahos, Michaela, Convergence of Competition Laws and Policies in the European Community, Kluwer Law International, 2001 Ehlermann, Claus-Dieter, ‘Deregulation and Enforcement of Competition Laws’, in Proceedings of the Second Seminar on European Union/Japan Competition Policy, Luxembourg, 1995, pp. 22 et seq. Ehlermann, Claus-Dieter and Isabela Atanasiu, eds, European Competition Law Annual 2006: Enforcement of Prohibition of Cartels, Hart Publishing, 2007 Ehlermann, Claus-Dieter and Mel Marquis, eds, European Competition Law Annual 2008: Antitrust Settlements under EC Competition Law, Hart Publishing, 2010 Ehlermann, Claus-Dieter and Mel Marquis, eds, European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011 Epstein, Richard, ‘Class Actions: The Need for a Hard Second Look’, Civil Justice Report No. 4 (March 2002), http://www.manhattan-institute.org/html/ cjr_4.htm Flynn, John J, ‘Criminal Sanctions Under State and Federal Antitrust Laws’, 45 Texas Law Review 1301 (1967) Fölster, Stefan and Sam Peltzman, The Social Cost of Regulation and Lack of Competition in Sweden: A Summary in The Welfare State in Transition: Reforming the Swedish Model, University of Chicago Press, 1997 Forrester, Ian S, ‘Due Process in EC competition cases: a distinguished institution with flawed procedures’, 34 European Law Review 817 (2009) Fox, Eleanor M, ‘Antitrust and Institutions: Design and Change’, 41 Loyola University of Chicago Law Journal 473 (2010) Gavil, Andrew I, ‘Antitrust Remedy Wars Episode I: Illinois Brick from Inside the Supreme Court’, 79 St. John’s Law Review 553 (2005) Gavil, Andrew I, ‘The Challenges of Economic Proof in a Decentralized and Privatized European Competition Policy System: Lessons from the American Experience’, 4 Journal of Competition Law and Economics 177 (2008) Gavil, Andrew I, ‘Burden of Proof in U.S. Antitrust Law’, in Wayne Dale Collines, ed, Issues in Competition Law and Policy, Vol. I, pp. 125 et seq., ABA Antitrust Section, 2008 Gavil, Andrew I, ‘Thinking Outside the Illinois Brick Box: A Proposal for Reform’, 76 Antitrust Law Journal 167 (2009)

540  References Ginsburg, Douglas H and Leah Brannon, ‘Determinants of Private Antitrust Enforcement in the United States’, 1 Competition Policy International 29 (Autumn 2005) Gippini-Fournier, Eric, ‘Institutional Report: The Modernisation of European Competition Law: First Experiences with Regulation 1/2003’, in Franz Koeck and Margit Maria Karollus, eds, The Modernisation of European Competition Law, Initial Experiences with Regulation 1/2003, Vienna, 2008 Goddin, Gaëtane, ‘Recent Judgments Regarding Transparency and Access to Documents in the Field of Competition Law: Where Does the Court of Justice of the EU Strike the Balance?’, 2 Journal of European Competition Law & Practice 10 (2011) González-Diaz, Enrique, ‘Minority Shareholdings and Creeping Acquisitions: The European Union Approach’, in Barry Hawk, ed, International Antitrust Law & Policy: Fordham Competition Law Institute 2011, Juris Publishing, 2012 Harding, Christopher and Julian Joshua, Regulating Cartels in Europe, A Study of Legal Control of Economic Delinquency, OUP, 2003 Hodges, Christopher, ‘European Competition Enforcement Policy: Integrating Restitution and Behaviour Control. An Integrated Enforcement Policy, Involving Public and Private Enforcement with ADR’, in 34 World Competition 383 (2011) Hodges, Christopher, ‘New Modes of Redress for Consumers and Competition Law’, in Competition & Regulation Review, December 2012, pp. 225 et seq. Hofstetter, Karl and Melanie Ludescher, ‘Fines against Parent Companies in EU Antitrust Law: Setting Incentives for ‘Best Practice Compliance’’, 33 World Competition 55 (2010) István Nagy, Csongor ‘The new Hungarian rules on damages caused by horizontal cartels: presumed price increase and limited protection for whistleblowers – an analytical introduction’, 32 European Competition Law Review 63 (2011) Italianer, Alexander, ‘Public and Private Enforcement of Competition Law’, speech, 17 February 2012, Brussels, http://ec.europa.eu/competition/speeches/ text/sp2012_02_en.pdf Jeffries, Andrew, ‘EU Proposals on Collective Redress – Lost in Consultation?’, CPI Antitrust Chronicle (April 2011) Jones, Clifford, ‘A New Dawn for Private Competition Law Remedies in Europe? Reflections from the US’, in Claus-Ehlermann and Isabela Atanasiu, eds, European Competition Law Annual 2001: Effective Private Enforcement of EC Antitrust Law, Hart Publishing, 2003 Jones, Clifford, Private Antitrust Litigation in the European Union: A Comparative Analysis, 2nd edition, Oxford University Press, forthcoming

References   541   Joshua, Julian, ‘Shooting the Messenger: Does the UK Criminal Offense Have a Future?’, The Antitrust Source 1 (August 2010) Kaplan, Benjamin, ‘A Prefatory Note’, 10 Boston College Industrial and Commercial Law Review 497 (1968-1969) Klees, Andrea, Europäisches Kartellverfahrensrecht mit Fusionskontrollverfahren (Cologne/Berlin/Munich, 2005) Komninos, Assimakis, ‘New Prospects for Private Enforcement of EC Competition Law: Courage v Crehan and the Community Right in Damages’, 39 Common Market Law Review 447 (2002) Komninos, Assimakis P, ‘Public and Private Antitrust Enforcement in Europe: Complement? Overlap?’, 3(1) Competition Law Review 5 (2006) Komninos, Assimakis P, ‘Effect of Commission Decisions on Private Antitrust Litigation: Setting the Story Straight’, 44 Common Market Law Review 1387 (2007) Komninos, Assimakis P, EC Private Antitrust Enforcement, Decentralised Application of EC Competition Law by National Courts, Hart Publishing, 2008 Komninos, Assimakis P, ‘The Road to the Commission’s White Paper for Damages Actions: Where We Came From’, 4(2) Competition Policy International 80 (2008) Kovacic, William E, ‘Institutional Foundations for Economic Legal Reform in Transition Economies: The Case of Competition Policy and Antitrust Enforcement’, 77 Chicago-Kent Law Review 265 (2001) Kovacic, William E, ‘Private Participation in the Enforcement of Public Competition Laws’ (15 May 2003), http://www.ftc.gov/speeches/other/030514biicl.shtm Kovacic, William E, ‘Rating the Competition Agencies: What Constitutes Good Performance?’, 16 George Mason Law Review 903 (2009) Kovacic, William, Hugh Hollman and Patricia Grant, ‘How Does Your Competition Agency Measure Up?’, 7 European Competition Journal 25 (2011) Kroes, Neelie, ‘Collective Redress – Delivering Justice for Victims’, speech, 4 March 2009, http://europa.eu/rapid/pressReleasesAction.do?reference=SPEE CH/09/88&format=PDF&aged=1&language=EN&guiLanguage=en Lande, Robert, ‘New Options for State Indirect Purchaser Legislation: Protecting the Real Victims of Antitrust Violations’, 61 Alabama Law Review 447 (2010) Lasserre, Bruno, ‘Towards the ECN’s Second Decade’, in Barry Hawk, ed, International Antitrust Law & Policy Fordham Competition Law 2011, Juris Publishing, 2012, pp. 19 et seq. Lianos, Ioannis and Ioannis Kokkoris, eds, The Reform of EC Competition Law: New Challenges, Kluwer, 2010

542  References Marquis, Mel, ‘Cartel Settlements and Commitment Decisions’, in Claus-Dieter Ehlermann and Mel Marquis, eds, European Competition Law Annual 2008: Antitrust Settlements under EC Competition Law, Hart Publishing, 2010, pages xxix et seq. Micklitz, Hans, ‘Transborder Law Enforcement – Does it Exist?’, in Stephen Weatherill and Ulf Bernitz, eds, The Regulation of Unfair Commercial Practices under EC Directive 2005/29, New Rules and New Techniques, Hart Publishing, 2007 Milutinović, Veljko, ‘Private Enforcement: Upcoming Issues’, in Giuliano Amato and Claus-Dieter Ehlermann, eds, EC Competition Law: A Critical Assessment, Hart Publishing, 2007 Milutinović, Veljko, The Right to Damages under EU Competition Law: from Courage v. Crehan to the White Paper and Beyond, Kluwer, 2010 Montag, Frank, ‘The Case for a Radical Reform of the Infringement Procedure’, 8 European Competition Law Review 28 (1996) Monti, Giorgio, ‘Anticompetitive Agreements: the Innocent Party’s Right to Damages’, 27 European Law Review 282 (2002) Monti, Mario, ‘Effective Private Enforcement of EC Antitrust Law’, in ClausDieter Ehlermann and Isabela Atanasiu, eds, European Competition Law Annual 2001: Effective Private Enforcement of EC Antitrust Law, Hart Publishing, 2003 Morais, Luís D S and Francisco Costa Cabral, ‘Mitigation of fine and competition law: An overview of EU and national case law’, 19 July 2012, e-Competitions, No. 48098, www.concurrences.com Morais, Luís Silva, Joint Ventures and EU Competition Law, Hart Publishing, 2013 Motta, Massimo, Competition Policy – Theory and Practice, Cambridge University Press, 2004 Neal, Phil C and Perry Goldberg, ‘The Electrical Equipment Antitrust Cases: Novel Judicial Administration’, 50 A.B.A. Journal 621 (1964) Nebbia, Paolisa and Ioannis Lianos, Damages Claims for the Infringement of Competition Law, Oxford University Press, forthcoming O’Brien, Daniel and Steven Salop, ‘Competitive effects of partial ownership: financial interest and corporate control’, 67 Antitrust Law Journal 559 (2000) Paulis, Emil and Céline Gauer, ‘La réforme des règles d’application des articles 81 et 82 du Traité’, 11 Journal des tribunaux – Droit européen 65 (2003) Pietrini, Silvia, L’action collective en droit des pratiques anticoncurrentielles. Perspectives nationale, européenne et internationale, Bruylant, 2012

References   543   Posner, Richard, Antitrust Law, 2nd edition, University of Chicago Press, 2011 Prosperetti, Luigi, Il danno antitrust, Bologna, 2009 Randall, Jaynie, ‘Does De-Trebling Sacrifice Recoverability of Antitrust Awards?’, 23 Yale Journal on Regulation 101 (2006) Rodger, Barry, ‘The Big Chill for National Courts: Reflections on Market Foreclosure and Freezer Exclusivity under Article 81’, 11 IJEL 77 (2004) Rodger, Barry, ‘Editorial – Private enforcement and collective redress: the benefits of empirical research and comparative approaches’, 8 Competition Law Review 1 (2012) Rossi, Leonor and Miguel Ferro, ‘Private Enforcement of Competition Law in Portugal (1) An Overview of Case Law’, Competition & Regulation Review, April/June 2012, pp. 91 et seq. Segal, Ilya and Michael Whinston, ‘Public vs. Private Enforcement of Antitrust Law: A Survey’, University of Chicago, John M Olin Program in Law and Economics, Working Paper No. 335 (December, 2006) Shaw, Jo, ‘Decentralization and Law Enforcement in EC Competition Law’, 15 Legal Studies 128 (1995) Shaw, Malcolm N, International Law, 6th edition, Cambridge University Press, 2008 Sherman, Edward F, ‘American Class Actions: Significant Features and Developing Alternatives in Foreign Legal Systems’, 215 F.R.D. 130 (2003) Sherman, Edward, ‘Complex Litigation: Plagued By Concerns Over Federalism, Jurisdiction and Fairness’, 37 Akron Law Review 589 (2004) Siragusa, Mario and Cesare Rizza, eds, EU Competition Law, Volume III: Cartel Law, Claeys & Casteels, 2007 Sokol, D Daniel, ‘Antitrust, Institutions, and Merger Control’, 17 George Mason Law Review 1055 (2010) Sproul, Michael F, ‘Antitrust and Prices’, 101 Journal of Political Economy 742 (1993) Struijlaart, Robin A, ‘Minority share acquisitions below the control threshold of the EC Merger Control Regulation: An economic and legal analysis’, in 25 World Competition 173 (2002) Temple Lang, John, ‘EEC Competition Actions in Member States’ Courts – Claims for Damages, Declarations and Injunctions for Breach of Community Antitrust Law’, 7 Fordham International Law Journal 389 (1983) Van Bael & Bellis, Competition Law of the European Community, 5th Edition, Kluwer Law International, 2010

544  References Van Boom, Willem, ‘Collective settlement of mass claims in the Netherlands’, in Matthias Casper, André Janssen, Petra Pohlmann and Reiner Schulze, eds, Auf dem Weg zu einer europäischen Sammelklage?, Sellier, 2009 Vande Walle, Simon, Private Antitrust Litigation in the European Union and Japan, Maklu, 2013 Vauchez, Antoine, ‘‘Integration-through-law’: Contribution to a Socio-history of EU Political Commonsense’, EUI Working Paper (RSCAS) 2008/10 Venit, James, ‘Human All Too Human: The Gathering and Assessment of Evidence and the Appropriate Standard of Proof and Judicial Review in Commission Enforcement Proceedings Applying Articles 81 and 82’, in Claus-Dieter Ehlermann and Mel Marquis, eds, European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011 Weiser, Philip J, ‘Towards an International Dialogue on the Institutional Side of Antitrust’, 66 New York Annual Survey of American Law 445 (2011) II. Legislation, proposed legislation and soft law A. The European Union European legislation Commission Regulation (EC) No 622/2008 of 30 June 2008 amending Regulation (EC) No 773/2004, as regards the conduct of settlement procedures in cartel cases, 2008 OJ L171/3 Council Regulation (EC) No. 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, 2003 OJ L1/1 Regulation (EC) No 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, 2001 OJ L145/43 Regulation (EU) 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, 2012 OJ L351/1 European soft law and consultation documents Commission, Green Paper - Damages actions for breach of the EC antitrust rules, SEC(2005) 1732 of 19 December 2005 Commission, Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, 2009 OJ C45/7

References   545   Commission, Joint Information Note: Towards a Coherent European Approach to Collective Redress: Next Steps, SEC(2010) 1192 Commission Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel cases, 2008 OJ C167/1 Commission Notice on Immunity from fines and reduction of fines in cartel cases, 2006 OJ C298/11 Commission Notice on the rules for access to the Commission file in cases pursuant to Articles 81 and 82 of the EC Treaty, Articles 53, 54 and 57 of the EEA Agreement and Council Regulation (EC) No 139/2004, 2005 OJ C325/07 Commission, Towards a Coherent European Approach to Collective Redress, SEC(2011) 173 Commission, White Paper on Modernisation of the Rules Implementing Articles 85 and 86 of the EC Treaty, 1999 OJ C132/1 Commission, White Paper - Damages actions for breach of the EC antitrust rules, SEC(2008) 404/ SEC(2008) 405/ SEC(2008) 406 of 2 April 2008 Commission, Notice on Case Referral in respect of concentrations, 2005 OJ C56/02 B. United States Antitrust Criminal Penalty Enhancement and Reform Act of 2004, Pub. L. 108237 (2004), 118 Stat. 665 (2004) USA PATRIOT Improvement and Reauthorization Act of 2005 § 113(g)(3), Pub. L. No. 109-177, 120 Stat. 192, 210 (2006) U.S. Department of Justice, Corporate Leniency Policy, http://www.justice.gov/ atr/public/guidelines/0091.htm III. Official and unofficial reports, newsletters, studies, papers, etc. A. National jurisdictions United Kingdom Anthony Hammond and Roy Penrose, Proposed criminalisation of cartels in the UK, OFT365 (November 2001), http://www.oft.gov.uk/shared_oft/reports/ comp_policy/oft365.pdf Department for Business Innovation & Skills, A Competition Regime for Growth: A Consultation on Options for Reform (March 2011), https://www.gov.uk/ government/uploads/system/uploads/attachment_data/file/31411/11-657competition-regime-for-growth-consultation.pdf

546  References Department for Business, Innovation and Skills, ‘Private Actions in Competition Law: a Consultation on Options for Reform’ (April 2012), https://www.gov. uk/government/uploads/system/uploads/attachment_data/file/31528/12-742private-actions-in-competition-law-consultation.pdf OFT, Response to the European Commission’s White Paper, Damages actions for breach of the EC antitrust rules (July 2008), OFT 1006 United States Antitrust Modernization Commission, Report and Recommendations, April 2007, http://govinfo.library.unt.edu/amc/report_recommendation/amc_final_report.pdf B. International organizations International Competition Network ICN Cartels Working Group Report, ‘Interaction of Public and Private Enforcement in Cartel Cases’ (May 2007) Organisation for Economic Cooperation and Development OECD, ‘Judicial Enforcement of Competition Law’ (1997) OECD, ‘Private Remedies’ (2007)

Table of Cases

I. Judgments of the Courts of the European Union Case T-31/99, ABB v Commission [2002] ECR II-1884 71, 312 Case 41/69, ACF Chemieforma [1970] ECR 61 65 Case 107/82, AEG v Commission [1983] ECR 3151 349 Case C-477/10, Agrofert, judgment of the ECJ of 28 June 2012, not yet reported xlix, 280 Case 53/85, Akzo [1986] ECR 1965 249, 280 T-164/12 R, Alstom v Commission, Order of 29 November 2012, not yet reported 288, 311, 312 Case 106/77, Amministrazione delle finanze dello Stato v Simmenthal [1978] ECR 629 347 Case 199/82, Amministrazione delle Finanze dello Stato v SpA San Giorgio ECR [1983] 3595 lvi, 372 Case T-348/08, Aragonesas Industrias y Energia, SAU v Commission [2011] ECR II-7583 306 Case T-224/00, Archer Daniels Midland v Commission [2003] ECR II-2597 71 Case T-24/90, Automec Srl v Commission (II) [1992] ECR II-2223 142, 346 Case C-128/92, Banks v British Coal [1994] ECR 1-1209 345, 348, 349 Case 65/86, Bayer AG and Maschinenfabrik Hennecke GmbH v Heinz Süllhöfer [1988] ECR 5249 144 Case 23/67, Brasserie de Haecht v Wilkin [1967] ECR 127 297 Case 48/72, Brasserie de Haecht II [1973] ECR 77 297 Joined Cases C-46 and C-48/93, Brasserie du Pêcheur v Germany and R v Sec. of State for Transport ex parte Factortame [1996] ECR I-1029 lxxvii, 345 Case C-76/06 P, Britannia Alloys & Chemicals Ltd v Commission [2007] ECR Ι-4405 143 Case 127/73, BRT v SABAM [1974] ECR 313 xl, lxxvi, lxxvii, 295, 341–76, 380, 514 Case C-536/11, Bundeswettbewerbsbehörde v Donau Chemie et al., judgment of the ECJ of 6 June 2013, not yet reported xliii, 126, 261, 369 Case 792/79 R, Camera Care Ltd. v Commission [1980] ECR 119 296 Case T-437/08, CDC Hydrogene Peroxide, judgment of the GC of 15 December 2011, not yet reported xxii, 36, 103, 279, 288, 289, 362 Case 283/81, CILFIT v Ministry of Health [1982] ECR 3415 348 Case C-59/89, Commission v Germany [1991] ECR I-2607 267 Cases 56 and 58/64, Consten and Grundig v Commission [1966] ECR (Eng Spec Ed.) 299 346

548  Table of Cases Case C-453/99, Courage Ltd v Bernard Crehan [2001] ECR I-6297 8, 28, 111, 144, 180, 260, 282, 297, 315 Case C-234/89, Delimitis v Henninger Bräu [1991] ECR I-935 xlii, lxxvi, lxxvii, 297, 298, 305, 343, 344, 346, 349, 351, 355, 369 Case C-126/97, Eco Swiss v Benetton [1999] ECR I-3055 xlvii, xlviii, 116, 134 Case T-344/08, EnBW, judgment of the GC of 22 May 2012, not yet reported xix, xlix, 23, 279, 280, 287, 288, 289 Case T-348/94, Enso Española v Commission [1998] ECR II-1875 359 Case C-199/11, European Union v Otis et al., judgment of the ECJ of 6 November 2012, not yet reported 353 Case C-68/93, Fiona Shevill, Ixora Trading Inc., Chequepoint SARL and Chequepoint International Ltd v Presse Alliance SA [1995] ECR I-415 105 Case T-28/02, First Data Corp et al v Commission [2005] ECR II-4119 354, 356 Joined Cases C-6 and C-9/90, Francovich and Bonifaci v Italy [1991] ECR I-5357 345 Case C-429/07, Inspecteur van de Belastingdienst v X BV [2009] ECR I-4833 315 Case 155/73, Italy v Sacchi [1974] ECR 409 348 Joined Cases C-430 and C-431/93, Jeroen van Schijndel and Johannes Nicolaas Cornelis van Veen v Stichting Pensioenfonds voor Fysiotherapeuten [1995] ECR I-4705 xlvii, 116 Case T-44/07, Kaučuk a.s. v Commission [2011] ECR II-4601 306 Case C-52/09, Konkurrensverket v TeliaSonera Sverige AB [2011] ECR I-527 88 Case C-453/00, Kühne & Heinz v Produkschap voor Pluimvee en Eiren [2004] ECR I-837 348 Case 258/78, L.C. Nungesser KG and Kurt Eisele v Commission (Maize Seed) [1982] ECR 2015 144 Joined Cases C-295 to C-298/04, Manfredi v Lloyd Adriatico Assicurazioni [2006] ECR I-6619 xliii, xliv, lv, lvi, lix, 8, 18, 28, 111, 125, 144, 150, 152, 154, 180, 260, 282, 283, 304, 341, 342, 363, 364–65, 371, 372, 373, 380, 427, 439, 460, 487, 513, 514 Case C-344/98, Masterfoods Ltd v HB Ice Cream Ltd [2000] ECR I-11369 xxxviii, xl–xlii, lxxiv, lxxvi, 24, 90, 120, 146, 273, 289, 305, 320, 343, 344, 351, 352, 353, 354, 355, 356, 369 Case 26/76, Metro v Commission [1977] ECR 1875 349 Joined Cases 100-103/80, Musique Diffusion Française and Others v Commission [1983] ECR I-11005 65, 359 Case T-403/05, MyTravel Group v Commission [2008] ECR II-2027, partially annulled, Case C-506/08, Sweden v Commission and MyTravel Group plc [2011] ECR I-6237 122 Case C-239/99, Nachi Europe v Hautpzollamt Krefeld [2001] ECR I-1197 356

Table of Cases  

549  

Case C-404/10 P, Odile Jacobs, judgment of the ECJ of 28 June 2012, not yet reported xix, xlix, 23, 280 Case C-360/09, Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161 xxiii, 102, 124, 145, 261, 281, 321, 360, 380, 381, 427, 435, 444 Case T-72/09, Pilkington, 2009 OJ C102/24 99, 100, 310 Case T-353/94, Postbank [1996] ECR II-921 280 Joined Cases C-241/91P and C-242/91 P, Radio Telefis Eireann (RTE) and Independent Television Publications Ltd (ITP) v Commission [1995] ECR I-743 297 Case 2/74, Reyners v Belgium [1974] ECR 631 349, 350 Case T-11/06, Romana Tabacchi srl v Commission [2011] ECR II-6681 306 Case T-73/09, Saint-Gobain, 2009 OJ C102/25 99, 100, 310 Case C-308/04 Ρ, SGL Carbon AG v Commission [2006] ECR Ι-5977 143 Case 319/82, Société de Vente de Ciments et Bétons de l’Est SA v Kerpen & Kerpen GmbH und co KG [1983] ECR 4173 297, 350 Case 56/65, Société Technique Minière v Maschinenbau Ulm GmbH [1966] (‘STM’) ECR Eng Spec Ed 235 350 Case T-68/09, Soliver, 2009 OJ C90/32 99, 100 Joined Cases T-22/02 and T-23/02, Sumitomo Chemical Co Ltd and Sumika Fine Chemicals Co Ltd v Commission [2005] ECR II-4065 145 Case C-406/92, Tatry [1994] ECR I-5439 248 Case C-52/09, TeliaSonera Sverige AB [2011] ECR I-527 88, 119 Case C-8/08, T-Mobile Netherlands BV et al v Raad van bestuur van de Nederlandse Mededingingsautoriteit [2009] ECR I-4529 156 Joined Cases T-236/01, T-239/01, T-244/01 to T-246/01, T-251/01 and T‑252/01, Tokai Carbon Co Ltd et al v Commission [2004] ECR II-1181 68 Case C-188/92, TWD Textilwerke v Germany [1994] ECR I-833 356 Case T-349/08, Uralita SA v Commission [2011] ECR II-373* 306 Case 10/86, VAG France v Établissements Magne [1986] ECR 4071 350 Case 41/74, Van Duyn v Home Office [1974] ECR 1337 350 Case 26/62, Van Gend & Loos v Netherlands Inland Revenue Administration [1963] Eng Spec Ed 1 17, 345, 346, 348, 349, 350, 352, 445 Joined Cases 209/78, etc Van Landewyck v Commission (‘Fedetab’) [1980] ECR 3125 359 Case 8/72, Vereeniging van Cementhandelaren v Commission [1972] ECR 977 348 Case C-73/95 P, Viho Europe v Commission [1996] ECR I-5457 349 Case T-2/03, VKI [2005] ECR II-1121 279 Case 5/69, Völk v Vervaecke [1969] ECR 295 348, 349 Case C-534/07 P, William Prym GmbH & Co KG and Prym Consumer GmbH & Co KG v Commission [2009] ECR I-7415 156

550  Table of Cases II. Decisions of the European Commission Commission Decision of 18 April 2007 in Case COMP/37.766 – Dutch beer market, 2008 OJ C122/1 66 Commission Decision of 12 November 2008 in Case COMP/39125 – Carglass, 2009 OJ C173/13 xxxvi, 67, 99 Commission Decision of 3 December 2003 in Case COMP/38.359 – Electrical and mechanical carbon and graphite products, 2004 OJ L125/45 71 Commission Decision 2006/903/EC of 3 May 2006 in Case COMP/38620 –Hydrogen Peroxide 106 Commission Decision of 13 May 2009 in Case COMP/37.990 – Intel, 2009 OJ C227/13 67, 75, 80 Commission Decision 89/205/EEC of 21 December 1988 in Case IV/31851, Magill TV Guide/ITP, BBC and RTE, 1989 OJ L78/43 297 Commission Decision of 27 February 2008 in Case COMP/37.792 – Microsoft, 2009 OJ C166/20 67 Commission Decision 2003/569/EC of 5 December 2001 in Case IV/37.614 – PO/Interbrew and Alken-Mars, 2003 OJ L200/1 68 III. US cases (courts and the agencies) Am Cent Eastern Tex Gas Co v Union Pacific Res Group, Inc., 93 Fed Appx 1 (5th Cir 2004) 172 (In re) Am Express Merchants’ Litig., 634 F.3d 187 (2d Cir 2011) 167 (In re) America Online, Inc., 131 FTC 829 (2001) 175, 176 American Safety Equipment Corp v J P Maguire & Co., 391 F.2d 821 (2d Cir 1968) l, 159 Anderson v Liberty Lobby, Inc., 477 US 242 (1986) 187, 189 Ashcroft v Iqbal, 556 US 662 (2009) 188 Associated Gen’l Contractors of Cal., Inc v Cal State Council of Carpenters, 459 US 519 (1983) 7 AT&T Mobility LLC v Concepcion, 2011 WL 1561956 (US) l, li, lii, lxii, 169, 171 Atlantic Richfield Co v USA Petroleum, 495 US 328 (1990) 318 (In re) Baycol Products Litigation, MDL No 1431, pretrial order No 77 (US Dist Minn 2003) 221 Baxter Intern, Inc v Abbott Laboratories, 315 F3d 829 (2003) 161 Bell Atlantic Corp v Twombly, 550 US 544 (2007) 12, 188, 199 Blackie v Kushner, 524 F2d 891 (9th Cir 1975) 185 Blue Cross Blue Shield of Virginia v McCready, 457 US 465 (1982) 7, 194 Brooke Group Ltd v Brown & Williamson Tobacco Corp, 509 US 209 (1993) 12 Brunswick Corp v Pueblo Bowl-O-Mat, Inc, 429 US 477 (1977) 7, 365 Byram Concretanks, Inc v Warren Concrete Prods Co, 374 F2d 649 (3d Cir 1967) 182

Table of Cases  

551  

Celotex Corp v Catrett, 477 US 317 (1986) 187, 189 Chevron USA v Natural Resources Defense Council, 467 US 837 (1984) 146 City of Roseville Employees’ Retirement Sys v Horizon Lines, Inc et al, No 08-cv-969 (D Del), affirmed, Nos 10-2788 & 10-3815 (3rd Cir 2011) 79 Clark v Universal Builders, Inc, 501 F2d 324 (7th Cir 1974) 185 Comedy Club, Inc v Improv West Associates, 553 F3d 1277 (9th Cir 2009) 175 (In re) Cotton Yarn Antitrust Litigation, 505 F3d 274 (4th Cir 2007) 164 Cox v Am Cast Iron Pipe Co, 784 F2d 1546 (11th Cir 1986) 185 Credit Suisse Securities (USA) LLC v Billing, 551 US 264 (2007) liii, 188, 189 (In re) Crompton Corp Sec Litig, No 03-cv-1293 (D Conn) 78 Daubert v Merrell Dow Pharmaceuticals, Inc, 509 US 579 (1993) 12 Dellums v Powell, 566 F2d 167 (DC Cir 1977) 185 Discover Bank v Superior Court, 36 Cal 4th 148 (2005) 169, 170 Dowdy & Dowdy Partnership v Arbitron Inc, No 2:09-cv-253 KS-MTP, 2010 US Dist LEXIS 108798 (SD Miss 30 September 2010) 193 Ervin v OS Rest Servs, Inc, 632 F3d 971 (7th Cir 2011) 185 (In re) Exxon Corporation and Mobil Corporation, FTC No C-3907 (30 November 1999) 176 Fortner Enterprises v US Steel Corp, 394 US 495 (1969) 318 FTC v Ticor Title Ins Co, 504 US 621 (1992) 190 Gilmer v Interstate/Johnson Lane Corp, 895 F2d 195 (4th Cir 1990), affirmed, 500 US 20 (1991) 173 Gilmer v Interstate/Johnson Lane Corp, 500 US 20 (1991) l, 160, 173 Green Tree Fin Corp v Bazzle, 539 US 444 (2003) 164 Hanover Shoe, Inc v United Machinery Corp, 392 US 481 (1968) lxiii, 182, 210, 233 Hawaii v Standard Oil Co, 405 US 251 (1972) 182, 183 Hoffman-La Roche, Inc v Sperling, 493 US 165 (1989) 67, 193 (In re) Hydrogen Peroxide Antitrust Litigation, 552 F3d 305 (3d Cir 2008) lxii, 12 Illinois v Sangamo Constr Co, 657 F2d 855 (7th Cir 1981) 194 Illinois Brick Co v Illinois, 431 US 720 (1977) lxiii, lxiv, 7, 8, 9, 12, 182, 183, 192, 198, 199, 206, 207, 208, 210, 233, 318, 365, 372 (In re) Ill Petition to Inspect & Copy Grand Jury Materials, 659 F2d 800 (7th Cir 1981) 192 (In re) Intel Corp Derivative Litigation (MDL), Case 1:09-00867-JJF (D Del 23 July 2010) 80 In re Intel Corp Derivative Litigation, “Stipulation of Settlement” Case 1:09-cv-00867-JJF (D Del 25 May 2010) 80 Investment Partners, LP v Glamour Shots Licensing, Inc, 298 F3d 314 (5th Cir 2002) 172 Janus Capital Group, Inc v First Derivative Traders, No 09-525, 2011 WL 2297762 (US 13 June 2011) 79

552  Table of Cases Jaroslawicz v Krass et al, No CV 980331117S (Conn) 80 JLM Indus, Inc v Stolt-Nielsen SA, 387 F3d 163 (2d Cir 2004) 163, 171 Kristian v Comcast Corp, 446 F3d 25 (1st Cir 2006) 167, 168, 172 Leegin Creative Leather Products, Inc v PSKS, Inc, 551 US 877 (2007) 12 (In re) LG Philips LCD Co, Ltd Sec Litig, 1:07-cv-909 (SDNY) 79 In re Linerboard Antitrust Litigation, 333 F Supp 2d 33 (East Penn Dist Ct, 2004) 221 Matsushita Elec Indus Co, Ltd v Zenith Radio Corp, 475 US 574 (1986) 187, 188, 189 Menkes v Stolt-Nielsen SA, No 3:03-cv-409 (D Conn) 78, 79 Menkes v Stolt-Nielsen SA, 2006 WL 1699603 (D Conn 19 June 2006) 78 Menkes v Stolt-Nielsen SA, 2006 US Dist LEXIS 42644, No 3:03-cv-409 (D Conn 19 July 2006) 78 Menkes v Stolt-Nielsen SA, No 3:03-cv-409, 2011 US Dist LEXIS 7066 (D Conn 25 January 2011) 79 (In re) Micron Techs, Inc Sec Litig, No 06-cv-085 (D Idaho) 78 Mitsubishi Motors Corp v Soler Chrysler-Plymouth, Inc, 473 US 614 (1985) l, li, 159, 160, 161, 162, 168, 172 Montgomery Ward & Co v Langer, 168 F2d 182 (8th Cir 1948) 197 Perma Life Mufflers, Inc v Int’l Parts Corp, 392 US 134 (1968) 182 Perot v Federal Election Commission, 97 F3d 553 (DC Cir 1996), cert denied (1997) 175 Phillips Petroleum Co v Shutts, 472 US 797 (1985) 186, 189, 190 (In re) Real Estate Title & Settlement Servs Antitrust Litig, 1986-1 Trade Cases ¶ 67,149 (ED Pa 1986), affirmed without opinion, 815 F2d 695 (3d Cir 1987), cert denied, 485 US 909 (1988) 190 Reazin v Blue Cross & Blue Shield, Inc, 899 F2d 951 (10th Cir 1990) 194 Reiter v Sonotone Corp, 442 US 330 (1979) 7 RH Johnson & Co v SEC, 198 F2d 690 (2d Cir), cert denied 344 US 855 (1952) 175 Ross v American Express Co, 547 F3d 137 (2d Cir 2008) 171 Securities and Exchange Commission v Dell Inc et al, Case: 1:10-cv-01245 (DCC 22 July 2010) 75 Simula, Inc v Autoliv, Inc, 175 F3d 716 (9th Cir 1999) 162 Son et al v LG Display Co Ltd et al, No 10-cv-4380 (SDNY) 79 (In re) Sotheby’s Holdings, Inc Sec Litig, 2000 US Dist LEXIS 12504, No 00 Civ 1041 (DLC) (SDNY 31 August 2000) 78 (In re) Sotheby’s Holdings, Inc Securities Litig, 2000 WL 1234601 (SDNY 31 August 2000) 78 (In re) Sotheby’s Holdings, Inc Sec Litig, No 00-cv-1041 (SDNY) 78 Stolt-Nielsen SA v AnimalFeeds International Corp, 559 US 662 (2010) 161, 162, 164, 167, 168 Sullivan v DB Investments, Inc, 613 F3d 134 (3d Cir), rehearing granted and vacated, 2010 US App LEXIS 18088 (3d Cir 27 August 2010) 192

Table of Cases  

553  

Swanson v Citibank, NA, 614 F3d 400 (7th Cir 2010) 184, 185 Swensen’s Ice Cream Co v Corsair Corp, 942 F2d 1307 (1991) 162 Tamburo v Dworkin, No 04 C 3317, 2010 US Dist LEXIS 121510 (ND Ill 17 November 2010) 193 Texas Indus, Inc v Radcliff Materials, Inc, 451 US 630 (1981) liv, 184, 213 (In re) TFT– LCD (Flat Panel) Antitrust Litigation, 2011 WL 1753784 (NDCal 2011) 165 Thorogood v Sears, Roebuck & Co, 624 F3d 842 (7th Cir), rehearing denied, 627 F3d 289 (7th Cir 2010), petition for cert filed, No 10-1087 (US 2 March 2011) 184 Ticor Title Insurance Co v Brown, 511 US 117 (1994) liv, 189, 190, 191 Tucker v Scrushy, CV-02-5212 AEH, “Memorandum Opinion Regarding Final Judgment Order Against Defendant Richard M Scrushy Under Alabama Rules of Procedure Rule 54(b)” (Ct of Jeff Cty Ala 18 June 2009) 81 United States v Cargill, Inc, 2000–2 Trade Cases (CCH) (DDC 30 June 2000) 174 (In re) Ciba-Geigy Ltd 123 FTC 842 (1997) 175 United States v Data Card Corp, 1987-1 Trade Cases (CCH) (DDC, 29 January 1987) 174, 175 United States v The Dow Chemical Co, 1987-2 Trade Cases (CCH) (ND M, 12 August 1987) 174 (In re) DTE Energy Co, 131 FTC 962 (2001) 175 United States v El Paso Natural Gas Co, 1995-2 Trade Cases (CCH) (DDC, 4 August 1995) 174 (In re) El Paso Energy Corp, No C-3915, 2000 FTC LEXIS 7 (FTC, 6 January 2000) 175 (In re) Evanston Northwestern Healthcare Corp, No 9315, 2008 FTC LEXIS 47 (FTC, 28 April 2008) 175, 176 United States v Imetal SA, 2000–1 Trade Cases (CCH) (DDC 25 May 2000) 174, 175 44 US v Foley, 598 F2d 1323 (4th Cir 1979) United States v Morton Plant Health Sys, 1994-2 Trade Cases (CCH) (MD Fla, 28 June 1994) 174 United States v Schiff, 602 F3d 152 (3d Cir 2010) 79 Verizon Communications, Inc v Law Offices of Curtis V Trinko, LLP, 540 US 398 (2004) 12, 189 Western Intern Media Corp v Johnson, 754 F Supp 871 (SDFla 1991) 162 IV. Other national jurisdictions Canada Attis v Ontario Minister of Health, 2010 ONSC 4508 Axiom Plastics Inc v EI DuPont Canada Co, 2008 CanLII 23490 (Ont Div Ct)

223 207

554  Table of Cases Bellefontaine v Purdue Frederick Inc, 2010 NSCA 58 225, 226, Bona Foods Ltd v Ajinomoto USA Inc, [2004] OJ No 908 (SCJ) 228 British Columbia v Canadian Forest Products Ltd, [2004] 2 SCR 74 207 Brito v Pfizer Inc, [2008] JQ no 4642 (Sup Ct) 222 Caputo v Imperial Tobacco, [2005] OJ No 842 (SCJ) (costs) [Caputo]; (2004), 74 OR (3d) 728 (SCJ) 223 Carom v Bre-X Minerals Ltd (1999), 43 OR (3d) 441 (Gen Div) 225 Central Trust Co v Rafuse, [1986] 2 SCR 147 211 Chadha v Bayer, (2003), 63 OR (3d) 22 (CA), leave to appeal to SCC refused, [2003] SCCA No 106 lxiii, 207, 208, 212, 213 Coleman v Bayer Inc, [2004] OJ No 1974 (SCJ) 221 Crosslink Technology, Inc v BASF Canada (November 30, 2007), London, 5030CP (Ont SCJ) (unreported) 213 Currie v McDonald’s Restaurants of Canada Ltd (2004), 4 CPC (6th) 299 (Ont SCJ) 228, 229, 230 Currie v McDonald’s Restaurants of Canada, (2005), 74 OR (3d) 321 (CA) 228 Dabbs v Sun Life Assurance Co of Canada (1998), 40 OR (3d) 429 (Gen Div) 228 Dugal v Manulife Financial Corporation, [2011] OJ No 1239 (SCJ) 224 Dumoulin v Ontario, [2005] OJ No 3961 (SCJ) 216 Eli Lilly and Co v Apotex Inc, 2009 FC 991, affirmed 2010 FCA 240, leave to appeal to SCC refused, [2010] SCCA No 434 211 Fischer v IG Investment Management Ltd, 2010 ONSC 5132 227 Ford v F Hoffman-LaRoche Ltd (2005), 74 OR (3d) 758 (SCJ) 227 Garford Pty Ltd v Dywidag, 2010 FC 996 211 Goyette c GlaxoSmithKline Inc, 2009 QCCS 3745, affirmed, 2010 QCCA 2054 226 Harrington v Dow Corning Corp (1997), 29 BCLR (3d) 88 (SC), affirmed, [2000] 11 WWR 201 (CA), leave to appeal to SCC refused, [2001] SCCA No 21 226 Hobshawn v Atco Gas and Pipelines Ltd (14 May 2009), Action 0101-04999 (Alta QB) (unreported) 224 Hocking v Haziza, 2008 QCCA 800 222, 226 Hollick v Toronto (City), [2001] 3 SCR 158 216, 217, 218 Hunt v Carey Canada Inc, [1990] 2 SCR 959 217 Kamloops (City) v Nielson, [1984] 2 SCR 2 211 Kelman v Goodyear Tire and Rubber Company, [2005] OJ No 175 (SCJ) 230 Kerr v Danier Leather Inc, 2007 SCC 44 222 Kingstreet Investments Ltd v New Brunswick (Finance), [2007] 1 SCR 3 207, 209 Knight v Imperial Tobacco Canada Limited, 2006 BCCA 235 212 Laboratoires Servier v Apotex Inc, 2008 FC 825, affirmed 2009 FCA 222, leave to appeal to SCC refused, [2009] SCCA No 403 211 Lupsor Estate v Middlesex Mutual Insurance Co, [2003] OJ No 1038 (SCJ), leave to appeal granted, [2003] OJ No 3745 (Div Ct) 225 M(K) v M(H), [1992] 3 SCR 6 211

Table of Cases  

555  

MacQueen v Sydney Steel Corporation (19 October 2010), Action 218010 (NSSC) (unreported) 224 Main v Cadbury Schweppes plc, 2010 BCSC 816, affirmed, 2011 BCCA 21, leave to appeal to SCC refused, 2011 CanLII 40927 213 Mandeville v Manufacturers Life Insurance Co, [2002] OJ No 5387 (SCJ) 227 Maritime Travel Inc v Go Travel DirectCom Inc, 2008 NSSC 163, affirmed, 2009 NSCA 42 213 Markson v MBNA Canada (21 October 2009) (Ont SCJ) (unreported) 219 Mazzona c Daimler Chrysler, 2008 JQ no 13987 (SCJ) 219 McKenna v Gammon Gold, [2010] OJ No 1057 (SCJ), leave to appeal granted, 2010 ONSC 4068 (Div Ct), varied, 2011 ONSC 3782 (Div Ct) 227 Morguard Investments Ltd v De Savoye, [1990] 3 SCR 1077 214, 228 Option Consommateurs v Novopharm Ltd, [2006] QCCS 118 218, 219 Osmun v Cadbury Adams Canada Inc, 2010 ONSC 2643, affirmed, 2010 ONCA 841, leave to appeal to SCC refused, 2011 CanLII 40927 213 Peixeiro v Haberman, [1997] 3 SCR 549 211 Peter v Medtronic, Inc, [2009] OJ No 4364 (SCJ), affirmed, 2010 ONSC 3777 (Div Ct) 219 Pharmascience Inc v Option Consommateurs, [2005] QCCA 437 219 Pro-Sys Consultants Ltd v Infineon Technologies AG, 2009 BCCA 503, leave to appeal to SCC refused, [2010] SCCA No 32 212 Punit v Wawanesa Mutual Insurance Co, [2005] OJ No 1928 (SCJ) 222 Quizno’s Canada Restaurant Corporation v 2038724 Ontario Ltd, 2010 ONCA 466, leave to appeal to SCC refused, 2011 CanLII 4632 213, 216 R v Oakes, 1986] 1 SCR 103 213 R v Roddy (1877) 41 UCQB 291 at 297, 1 Cart BNA 709 (Ont CA) 208 Robinson v Medtronic, Inc, [2009] OJ No 4366 (SCJ), affirmed, 2010 ONSC 3777 (Div Ct) 219 Rumley v British Columbia, [2001] 3 SCR 184 218 Ryan v Moore, [2005] 2 SCR 53 211 Serhan Estate v Johnson and Johnson, (2006), 269 DLR (4th) 279 (Div Ct), leave to appeal to SCC refused, [2006] SCCA No 494 212, 217 Sharma v Timminco, [2009] OJ No 4511 (SCJ) 232 Sierra Club of Canada v Canada (Minister of Finance), 2002 SCC 41 220 Smith v Inco, 2010 ONSC 3790 211 Smith Estate v National Money Mart Company, 2011 ONCA 233 224 Société canadienne des postes v Lépine, [2007] JQ no 8498 (SCJ), affirmed, [2009] 1 SCR 549 222 Sun-Rype Products Ltd v Archer Daniels Midland Company, 2008 BCCA 278, leave to appeal refused, [2008] SCCA No 416 207, 209, 210, 211 Unifund Assurance Co v Insurance Corp of British Columbia, 2003 SCC 40 225 Vitapharm Canada Ltd v Hoffman-Laroche Ltd, [2005] OJ No 1117 (SCJ) 232 Vitapharm Canada Ltd v F Hoffmann-La Roche Ltd, [2005] OJ No 1118 (SCJ) 228

556  Table of Cases Western Canadian Shopping Centres Inc v Dutton, [2001] 2 SCR 534 205, 217 Wilson v Servier Canada Inc (2002), 59 OR (3d) 656 (Sup Ct) and Wilson v Servier Canada Inc (2000), 50 OR (3d) 219 (Sup Ct), leave to appeal to Div Ct refused, 52 OR (3d) 20, leave to appeal to SCC refused, [2001] SCCA No 88 225 Wong v Sony of Canada, [2001] OJ No 1707 (SCJ) 212 Young v Noranda Inc, [2005] JQ no 16113 (Sup Ct) 219 Germany Local Court Bonn, decision of 18 January 2012, WuW/E DE-R 3499 – Pfleiderer II 103, 436 Landgericht Dortmund, reference 13 O 23/09 (Kart) xxxvii, 105, 261, 436 Landgericht Düsseldorf, reference 34 O 147/05 (Kart) xxv, 99, 106 Higher Regional Court of Berlin (KG), judgment of 1 October 2009, WuW/E DE-R 2773 - Berliner Transportbeton 97 Higher Regional Court of Düsseldorf, court order of 3 May 2006, Ref No VI-W (Kart) 6/06 36 Higher Regional Court of Düsseldorf, judgment of 3 May 2006, WuW/E DE-R 1755 106 Higher Regional Court of Düsseldorf, judgment of 14 May 2008, Case No VI U (Kart) 14/07 37 Higher Regional Court of Düsseldorf, judgment of 26 June 2009, reference VI-2a Kart 2 - 6/08 OWi, BeckRS2010, 04805 106 Higher Regional Court of Düsseldorf, judgment of 5 October 2012, WuW/E DE-R 3662 – Kaffeeröster 103 BGH, judgment of 7 April 2009, Case No KZR 42/08 37, 300 BGH, decision of 28 June 2005, WuW/E DE-R 1567, 1569 – Berliner Transportbeton I 97 BGH, decision of 10 April 2008, WuW/E DE-R 2225, 2229 – Papiergroßhandel 97 BGH, decision of 28 June 2011, WuW/E DE-R 3431 – ORWI 104 Italy Eni, Milan Tribunal, 5 May 2009 Inaz Paghe, Milan Court of Appeal, 10 December 2004 Indaba v Juventus, Turin Court of Appeal, 6 July 2000 International Broker, Rome Court of Appeal, 31 March 2008 Unimare v Geasar, Cagliari Court of Appeal, 23 January 1999 Viaggi del Ventaglio, Milan Court of Appeal, 30 April 2003

249 251 246, 248 251 247 251

Table of Cases  

557  

United Kingdom Bernard Crehan v Inntrepreneur Pub Company Limited and Brewman Group Limited [2003] EWHC 1510 (Ch) 298 Bernard Crehan v Inntrepreneur Pub Company CPC [2004] EWCA 637 298 Cooper Tire, [2009] EWHC 2609 (Comm), on appeal: [2010] EWCA Civ 864 275, 302, 313 Crehan v Inntrepreneur Pub Company Ltd and another [2007] EWHC 90081 298 Devenish Nutrition Ltd & Ors v Sanofi-Aventis SA (France) & Others [2007] EWHC 2394 (Ch), [2008] EWCA Civ 1086 xxx, 323 Enron Coal Services Ltd v English Welsh & Scottish Railway Ltd, [2011] EWCA Civ 2 xxxvi, 148, 289, 305, 320 Garden Cottage Foods v Milk Marketing Board, [1983] 2 All ER 770 (HL) 296 Inntrepreneur Pub Company (CPC) et al v Crehan, [2006] UKHL 38 146, 298, 307, 353 National Grid Electricity Transmissions Plc v ABB Limited & Others [2012] EWHC 869 (Ch) xxiv, 269, 286 Norris v Government of the United States of America [2008] 2 WLR 673 (HL) 44 Norris v Government of the United States of America and Others [2008] UKHL 16 46 Provimi [2003] EWHC 961 (Comm) 275 Regina v Ghosh [1982] 2 ALL ER 689 45 Regina v Whittle [2008] EWCA Crim 2560 46