European Competition Law Annual 2012: Competition, Regulation and Public Policies 9781474201803, 9781849465823

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] Cleary, Gottlieb, Steen & Hamilton Contact: Mario Siragusa 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] CRA International Contact: Cristina Caffarra 99 Bishopsgate London EC2M 3XD UK Tel.: +44 20 7664 3700 Fax: +44 20 7664 3998 E-mail: [email protected]

vi  Sponsors Hengeler Müller Contact: Jochen Burrichter Benrater Strasse 18-20 D-40213 Düsseldorf Germany Fax: +49 211 83 04 222 E-mail: [email protected] Latham & Watkins LLP Contact: Sven Völcker Boulevard du Régent, 43-44 B-1000 Brussels Belgium Tel.: +32 2 788 62 42 Fax: +32 2 788 60 60 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 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 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  



17th ANNUAL EU C OMP E T IT ION L AW AND P OLIC Y W OR K S HOP :

COMPETITION, REGULATION AND PUBLIC POLICIES

Organizers Philip Lowe and Mel Marquis Contributing authors Rafael Allendesalazar

Martínez Lage, Allendesalazar & Brokelmann, Madrid Robert D. Anderson World Trade Organization, Geneva University of Perugia Marco Boccaccio Ginevra Bruzzone Assonime, Rome Cristina Caffarra CRAI International, Brussels and London Alexandre de Streel Universities of Namur and Louvain Ian Forrester White & Case, Brussels Douglas Ginsburg U.S. Ct. of Appeals for the District of Columbia Circuit and George Mason University Law School, Arlington Shearman & Sterling, Brussels Geert Goeteyn Calvin Goldman Goodmans, Toronto Antitrust Division, Dept. of Justice, Washington, D.C. Daniel Haar European Court of Justice, Luxembourg Küllike Jürimäe University College Dublin Suzanne Kingston Lars Kjølbye Latham & Watkins, Brussels Philip Lowe U.K. Competition and Markets, Authority, London Baker Botts, Brussels Paul Lugard Mel Marquis European University Institute, Florence Giorgio Monti European University Institute, Florence Anna Caroline Müller World Trade Organization, Geneva Rosa Perna Tribunale Amministativo Regionale, Rome Ofgem, London Anthony Pygram Pierre Régibeau CRA Inernational and Imperial College, London Jon Stern Centre for Competition and Regulatory Policy, City University, London Latham & Watkins, Brussels Sven Völcker

Mel Marquis and Veljko Milutinovič*

Competition, Regulation and Public Policy Issues: An Overview

This volume presents contributions prepared for the 17th edition of the Annual EU Competition Law and Policy Workshop, held on 13–14 July 2012 at the European University Institute in Florence. The 2012 Workshop focused on a rather wide-ranging cluster of topics relating generally to matters of public policy. Public policy, in this sense, is loosely defined and covers competition law itself, which is understood legally as part of the ordre public of the European Union.1 But public policy is also expressed through regulation, which often establishes an institutional setting in which market transactions occur and market relationships exist and develop. Regulation can serve as a means through which, for example, a sector experiences a controlled transition from a paradigm of legal monopoly to one of – at least de jure – open competition. Such regulation may also embody a variety of other policies in addition to the policy of ‘more competition’; for example, it may be concerned with consumer protection, health and safety, security and continuity of supply, and so on. The Workshop participants addressed these subjects, but they also discussed the occasionally delicate matter of how conflicts may be resolved, in law and policy, when principles of competition raise tensions, of varying intensity, with other legitimate public objectives. To cite just two examples, how is the relationship between competition law and environmental protection concerns to be framed, defined and managed? Or again: how should competition law be applied in contexts where innovation policy is at stake large given possible tradeoffs between foundational and follow-on innovation? We do not mean to suggest that competition law objectives inevitably come into conflict with other * Mel Marquis is Part-Time Professor at the European University Institute in Florence; Professore a contratto at LUMSA University in Rome; and Visiting Professor at Central University of Finance and Economics in Beijing. He is Co-Director of the EU Competition Law and Policy Workshop and Co-Director of the Rome Antitrust Forum. Veljko Milutinovič is Assistant Professor at the Graduate School of Business Studies, Megatrend University, Belgrade. On specific points the authors have benefited from comments kindly provided by Luca Rubini. In preparing the transcripts in this book, Mel Marquis was kindly assisted by Emma Linklater. 1 More prosaically, the EU competition rules constitute public policy within the meaning of the New York Convention on the recognition and enforcement of foreign arbitral awards. See Case C-126/97 Eco Swiss China Time Ltd. v Benetton International NV [1999] ECR I-3055, paras 39-40 (Article 101); and Joined Cases C-295/04 to C-298/04 Vicenzo Manfredi v Lloyd Adriatico Assicurazioni SpA [2006] ECR I-6619, para 31 (Articles 101 and 102). Consistent with their public policy character, the competition law prohibitions are of a ‘mandatory’ nature: private parties are precluded from using private law to avoid their constraints. See, e.g., Case T-128/98 Aéroports de Paris v Commission [2000] ECR II-3929, para 241; Case 102/81 Nordsee Deutsche Hochsee Fischerei GmbH v Reederei Mond Hochseefischerei Nordstern AG & Co. KG [1982] ECR 1095.

xiv  Mel Marquis and Veljko Milutinovič public interest goals. But it is nonetheless useful to explore instances in which there might be a conflict, and what to do about it.2 In the background to these discussions there was an additional contextual issue, namely the financial and economic crisis which continued in 2013 to plague the economies of Europe, and which served as a point of departure for our debates on matters of public policy. 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.3 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 competition-related 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. This chapter briefly presents the various contributions provided by the authors that participated in the 2012 Workshop. It is divided into the following sections: (1) Between competition, regulation and other public policies; (2) Unilateral conduct in post-liberalization markets (or markets at some stage of liberalization); (3) Technology and standard-essential patents; (4) The role of courts. Reconciling competition, non-competition and constitutional imperatives. While the remainder of this chapter discusses the written contributions of the authors, the standard caveat applies: our remarks cannot substitute for the chapters themselves. Furthermore, we do not attempt to synthesize the transcriptions of the oral debates. The reader is invited to consult the transcriptions, also included in this book, for a more complete picture of the proceedings.

2 There are of course legal constraints – above all those inscribed in primary law – which may have a decisive effect in cases of conflict. See, e.g., Heike Schweitzer, ‘Competition Law and Public Policy: Reconsidering an Uneasy Relationship. The Example of Art. 81’, EUI Working Paper 2007/30, for example at 11 (distinguishing between permissible instances where objectives other than competition may be taken into account and impermissible policy choices that subvert the fundamental principle of competition given expression in the Treaty). 3 In this series, published by Hart of Oxford (see http://www.hartpub.co.uk/SeriesDetails.aspx?Se riesName=European+Competition+Law+Annual), we have covered (to name only a few topics) the application of competition rules in settings involving intellectual property, the prohibition against cartels, cartel settlements and commitment decisions, judicial review by the EU Courts, and the legislative initiative to enhance the ‘private enforcement’ of competition law in Europe. Among other forthcoming works, the next volume will address the difficult task of enforcing competition law in a way that is both effective and legitimate.

Competition, Regulation and Public Policy Issues: An Overview 

xv  

1. Between competition, regulation and other public policies Giorgio Monti delivered a paper entitled ‘Regulatory Holidays in Utilities Regulation and EU Competition Law: A Case Study on the Role of Efficiency Considerations in Economic Law’. In it he examines the concept of so-called regulatory holidays, i.e., the disapplication of certain regulatory and/or antitrust obligations that would otherwise require a telecoms or energy incumbent carrying out certain types of investment to grant access to essential infrastructure. An important premise of the paper seems to be the economic cycle. Regulatory holidays are not as necessary in ‘times of plenty’, when subsidies can be provided by the state; on the other hand, for most Member States these are not times of plenty. Noting the relative under-investment in infrastructure that pervades much of Europe, Monti considers regulatory holidays to be one possible tool that might alleviate the problem. As he points out, schemes do exist whereby the sectoral regulator allows the incumbent to raise access prices to what are known in the UK (Ofgen) as RIIO, or ‘Revenue set to deliver strong Incentives, Innovation and Outputs’. Regulatory holidays do not exist as such at EU level, in the sense that the law does not explicitly allow an undertaking to trade off short term losses in allocative efficiencies against gains in dynamic efficiencies. However, the Commission does allow for ‘risk premiums’, for example, in the case of nextgeneration networks in the telecoms sector. The problem with such schemes is that there are costs entailed in determining the right level of the risk premium and the right duration for it (e.g. through constant periodic reviews). In energy, the Third Energy Package does allow something akin to a proper regulatory holiday (in the form of an exemption from third party access in the case of new investments). However, as Monti shows, the requirements applied by the Commission are very similar to Article 101(3) exemption criteria: the focus is still on maintaining competition in the structural sense, without sufficient consideration of dynamic efficiency. Indeed, the Commission seems to be open to exemptions primarily in cases where application of the regulatory package would be counterproductive to investment. An actual balancing act between risks to competition and increases in investment is, however, still in its infancy. One particular constraining factor that Monti identifies is the primacy of EU antitrust rules over sectoral regulation. From the standpoint of an investor, this feature of EU law may create a substantial degree of legal uncertainty. Even if the Commission itself is consistent in its policy and does not prosecute cases where it has approved an exemption for investment, NCAs could still in principle intervene to attack ostensibly anticompetitive conduct. The most significant constraint imposed by EU law – for ideological reasons but with practical consequences – is that, regardless of any actual or potential efficiencies, the competition rules always insist on the maintenance of residual competition in the static, structural sense. Monti does not propose that competition should be allowed to be eliminated: instead, his concern is that, given the way EU competition

xvi  Mel Marquis and Veljko Milutinovič law is interpreted, competition must (re)emerge within a time frame that is too short or unpredictable in the context of the large-scale, risky investments that characterize network industries. However, Monti adds that more space may be given to dynamic efficiency considerations under the new energy chapter of the TFEU; to some degree this would loosen up the insistence on (ostensibly dynamic but operationally static) ‘competition as a process’ considerations. While the jury is still out on the net impact of post-1990s pro-competitive intervention on the rate of infrastructure investment in Europe, waiting for that verdict without simultaneously reconsidering the competition-versus-investment balance may be a costly tactic in an increasingly competitive world, where other continents may outpace the ‘old’ one. Anthony Pygram’s chapter is entitled ‘Competition, Regulation and Other Public Policies in the UK Energy Sector’. This chapter highlights the duty of the UK’s Ofgem to take account of competition policy in its activities, and to function as an ex post competition law enforcer in some instances. As Pygram notes, competition policy is one of several interests that have to be considered within a larger framework guided by consumer interests in quite broad terms: i.e., the interests of (present and future) consumers ‘taken as a whole’, insofar as the provision of gas and electricity is concerned. This formulation would naturally extend not only to supervision of anticompetitive conduct but also to vigilance against market failures, and above all to safeguarding security of supply and environmental objectives. With regard to the ‘concurrency’ of competition jurisdiction, Pygram notes both the advantages (expertise) and disadvantages (lack of pan-sectoral perspective) of endowing a regulator with competence to handle competition law enforcement.4 As Pygram points out, however, the UK legislator has sought to minimize the risk of inappropriate enforcement by the regulator by providing, in the Enterprise and Regulatory Reform Act 2013, that: (i) Ofgem and the other concurrent regulators in the UK must coordinate with the Competition and Markets Authority; (ii) the CMA, apart from its normal concurrent power to initiate a case, can ultimately commandeer proceedings and relieve the regulator of its competence in the given case;5 and (iii) the UK government can withdraw concurrent enforcement powers from a sectoral regulator.6 Pygram does not comment on these latter scenarios, but it may well be that any problems of under-enforcement can be addressed simply in the shadow of the new provisions. In any case, one may readily accept 4 These strengths and weaknesses are of course precisely what creates the potential for synergies between a sector regulator and a competition policy if their action is rationally coordinated. 5 This provision is inspired to some extent by Article 11(6) of Regulation 1/2003 and, like the latter provision, it is expected and hoped that it will never have to be used. See David Currie, ‘The new Competition and Markets Authority: how will it promote competition?’, speech, London, 7 November 2013, https://www.gov.uk/government/speeches/the-new-competition-and-markets-authority-howwill-it-promote-competition, specifically section 3 (‘The Concurrency Regime and UKCN’). 6 For discussion, see Niamh Dunne, ‘Recasting Competition Concurrency Under the Enterprise and Regulatory Reform Act 2013’, 77 Modern Law Review 254 (2014).

Competition, Regulation and Public Policy Issues: An Overview 

xvii  

Pygram’s point that competition law enforcement alone is insufficient to protect consumer interests, and that a dutiful regulator may have to take proactive (and proportionate) steps to ensure that ‘the incentives of suppliers align with those of society as a whole’.7 Geert Goeteyn’s chapter, called ‘Regulation and Competition in the Aviation Sector: Focus on Airline Content and Global Distribution Systems’, looks back at the enforcement priorities set by the Commission over a decade ago in the civil aviation sector, specifically in relation to mergers and other types of horizontal co-operation agreements between air carriers, unilateral abusive behaviour, and traditional industry-wide restrictive practices and agreements. Goeteyn then raises the question ‘whether the current competition policy and enforcement, which is almost exclusively aimed at airlines, should be broadened to reflect the fact that the airlines themselves are looking at the bigger picture (beyond consolidation) and that there are wider issues to be addressed in the sector’.8 Relatively early it became apparent that computerised reservation systems (CRS) were potential vehicles for collusion, as they permitted the exchange of pricing information. The Commission thus intervened to regulate such systems, with non-discrimination being the key concern. But resort to non-discrimination paradoxically stifled competition and shifted rents to travel agencies – at the expense of airlines and consumers. In any event, airline-owned/controlled CRS have become largely a thing of the past due to the expansion of the internet. Turning to more current problems, General Distribution Systems (GDS) may pose special challenges. On the one hand they are two-sided markets. They are also often vertically integrated with travel agents, who generally operate online and create a market force by themselves. As GDS increasingly hold a grip on the relevant information, airlines need to accomodate them. This leaves room for new forms of potentially anticompetitive conduct, not least through the use of Most Favoured Nation clauses. Goeteyn concludes that the time may be right to switch from ex ante regulation to ex post antitrust control in this field and to shift the focus from airlines as such to the operation of GDS. He contends that DG Competition has the methods and the tools it needs to fulfill this new role. Jon Stern’s chapter is entitled ‘A Market Design for Water Trading in England and Wales: The Role of Water Rights’. In this chapter Stern discusses challenges associated with the introduction of competition in the water sector.9 His purpose is to encourage reforms to ensure that pricing on the basis of scarcity will enhance trade and innovation. This would be achieved first and foremost through increased trading of abstraction licenses, with follow-on effects in physical trade of (raw or Page 29. Page 86. 9 In May of 2014, the UK adopted the new Water Act, which bears on a number of the license and abstraction issues arising in this context. The Act and associated explanatory notes are available at http://services.parliament.uk/bills/2013-14/water/documents.html. 7 8

xviii  Mel Marquis and Veljko Milutinovič treated) bulk water. As Stern explains, in order to develop (upstream) competition in water supply, it is necessary to secure greater trading and investment opportunities, which in turn necessitates a clear definition of enforceable water property rights (which may take the form of collective ownership rights and not necessarily individual rights). The need for clearly defined property rights is problematic because water (other than underground water sources) has traditionally been treated as part of a ‘commons’, in the same way as, for example, the air one breathes; the rights to water have thus been riparian in nature, allowing for reasonable use. In the long term this is unsustainable in the UK, where the limits of supply will become gradually but increasingly acute. Environmental impacts from current levels of abstraction, which arise partly from mismatches of supply and demand, are already causing concern. Greater recourse to efficient trading within and between regions could help to respond to these pressures. In view of this objective, Stern suggests a redefinition of abstraction licenses on the basis of percentage shares of available water, allowing those shares to be traded, as a means to achieve scarcity-based abstraction prices. He further provides several reasons why this approach, forms of which have been tested in other jurisdictions, would be superior to the simple introduction of abstraction prices in areas of relative water shortage. The commodification of water ownership and use (while retaining exceptional riparian rights for certain uses such as fishing) would follow the pattern observed in other areas of scarce resources, such as forests and grazing land. In a final part of the paper, Stern explains why a market-based property rights regime is preferable to quantity-based water rationing. Rafael Allendesalazar asks: ‘Does Merger to Monopoly Become Efficient When it Refers to Sector Regulators and the Competition Authority?’ That question, which is a reference to the establishment in Spain of a ‘Super-Regulator’ in October 2013, seems to telegraph an inevitably negative reply. And indeed, the chapter does express skepticism with regard to the merits of the merger, though in a somewhat agnostic vein. More forthright criticism is aimed at the rushed and opaque way the initial reform proposals were pushed through. The SuperRegulator, or to use the proper name, the Comisión Nacional de los Mercados y de la Competencia (CNMC), is an amalgamation of the former competition authority (the CNC) and numerous sectoral authorities, such as those regulating the energy, telecoms and postal sectors. Allendesalazar’s chapter initially reacts to the 2012 draft ‘merger’ legislation but it also discusses the final form of the legislation.10 Along the way, Allendesalazar questions whether the Spanish 10 For a review of the first four months of the CNMC’s activities, see Ana Valiente, ‘What is it Like to have a Single, Multi-sector Regulator – the Spanish Experience’, 5 Journal of Competition Law and Practice 373 (2014). A brief account is also given in Andrew Ward, ‘Spain’s CNMC: The Story So Far’, Competition Policy International (2014). For a broad account by the regulator, taking account of the historical development of the relevant Spanish institutions, see José Maria Marin Quemada, ‘The Spanish National Authority of Markets and Competition (CNMC): A New Institutional Model of Competition and Regulatory Supervision’, in Philip Lowe, Mel Marquis and Giorgio Monti, eds., Institutional Change and Competition Authorities, Hart Publishing, forthcoming.

Competition, Regulation and Public Policy Issues: An Overview 

xix  

Government had put forward a persuasive case for reform. For example, while reorganizations of this kind are often driven by the prospect of cost savings, the math in this case seemed to yield hardly any savings at all, whereas definite costs could be expected in the form of the time and adjustment factors and coordination problems, among others. More generally, the objectives of the reform had not been articulated with sufficient clarity. Some of Allendesalazar’s concerns have been alleviated by the final form of the legislation, as he explains in his post script. However, a short passage in the main part of the chapter raises important questions about the wisdom of combining competition enforcement and sectorspecific regulatory functions. Suzanne Kingston’s chapter raises another rhetorical question: ‘Competition and Environmental Protection: A Case of Ne’er the Twain Shall Meet?’ The arguments advanced are summarized in part 5 of her chapter. The central idea is that, for a variety of (mostly but not exclusively legal) reasons, competition law in the EU must be applied in such a way as to synthesize or balance the need for competitive markets and the need for well-tailored measures protecting the environment (to the extent that these objectives are at variance, which cannot be assumed a priori). The chapter serves as a concise survey of this subject, which is handled more extensively in Kingston’s monograph.11 Here, Kingston presents a number of persuasive arguments in favor of an approach based on ‘substantive integration’. If an integrated approach is necessary then the question becomes how to implement this concept. The answer may depend on which competition law instruments are being deployed. To shed some light on this, Kingston reviews the relevance of environmental policies in the application of a series of Treaty provisions: Article 101 (where the issue is complicated, in our view, by the Article’s unsatisfactory structure and wording12) as well as Articles 102, 106 and 107. Overall, the chapter is astute and pertinent. Perhaps one premise that may be challenged is the idea that the ‘public/private divide’ with regard to environmental policies is becoming increasingly blurred. We would agree that this blurring may be seen insofar as questions of agency and instrumentality are concerned, and that this may result, as Kingston claims, in the increasing frequency of cases where environmental issues arise in antitrust cases. But at the ‘upstream’ level of deciding and shaping policy 11 Suzanne Kingston, Greening EU Competition Law and Policy, Cambridge University Press, 2012. For an earlier treatment, see also Hans Vedder, Competition Law and Environmental Protection in Europe: Towards Sustainability?, Europa Law Publishing, 2003. 12 For further discussion of the application of Article 101 in ways that take account of non-economic objectives (with reference to the implications of the direct applicability of Article 101(3) and, in the case of the more recent works, with reference to the ‘post-Lisbon’ Treaty rules), see, among others, Schweitzer, ‘Uneasy Relationship’, cited above note 2; Constanze Semmelmann, ‘The future role of the non-competition goals in the interpretation of Article 81 EC’, 1 Global Antitrust Review 15 (2008); Ben Van Rompuy, Economic Efficiency: The Sole Concern of Modern Antitrust Policy? Non-efficiency Considerations under Article 101 TFEU, Kluwer Law International, 2012; Christopher Townley, Article 81 EC and Public Policy, Hart Publishing, 2009; Christopher Townley, ‘Is There (Still) Room for Non-Economic Arguments in Article 101 TFEU Cases?’, in Caroline Heide-Jorgensen, Christian Bergqvist, Ulla Neergard and Sune Troels Poulsen, eds., Aims and Values in Competition Law, DJøF Publishing, 2013, pp. 115 et seq.

xx  Mel Marquis and Veljko Milutinovič and regulation, the respective roles of public and private actors can generally still be distinguished. This might have consequences, for example, in scenarios where it must be determined whether a measure is to be attributed to undertakings or to public authorities; or where undertakings claim exceptions to antitrust rules on grounds of public policy. Robert D. Anderson and Anna Caroline Müller discuss ‘Competition Policy and the Multilateral Trading System: Three Propositions, an Observation and Some Questions for Reflection’. Their first proposition is that effective national competition policies are essential to the success of the world trade system. The authors stress the importance, for example, of structural reforms in key sectors (especially those affecting input costs) and, more generally, the importance of policies lubricating entry, exit and dynamism. The need for adequate means to control international collusion and bidrigging is, understandably, another point of emphasis, since these types of misconduct can reduce, negate or perhaps dwarf gains from the relaxation of trade barriers, and since the harm done by cartels often falls disproportionately on the poor.13 The second proposition is that the WTO Agreements already contain certain limited competition rules, and a basic inventory is given. The third proposition is that there are ‘additional issues at the interface of trade and competition policy which (arguably) call for further joint application of the two disciplines’.14 Here the authors return to the theme of international cartels, with specific emphasis on export cartels. The point that export cartels continue to pose a problem for the international community is well taken. Eleanor Fox has often likened such cartels to environmental pollution, which is to say that, for reasons of both efficiency and fairness, the external costs should be internalized. And yet the regulation of export cartels is entirely unsatisfactory. The authors do not explore the possibility of controlling some conduct of this kind through the use of the either Article 11.1(b) of the Agreement on Subsidies and Countervailing Measures or (more tenuously) Article 11:3 of the Agreement on Safeguards. However, these possibilities remain untested and would not offer easy solutions. For example, with regard to the latter Agreement, explicit exemptions (although most are implicit) could perhaps conceivably be challenged; but given the stringent conditions established under that Agreement (and under Article XIX GATT) as it has been interpreted by the WTO Appellate Body (serious injury, unforeseen circumstances, causation, etc.), its utility is doubtful. In the absence of a coherent approach, the fight against harmful export cartels (i.e., we exclude the subset of beneficial, pro-trade export associations involving, for example, small businesses) is thus conducted merely through the 13 Somewhat paradoxically, this point seems to attract broad agreement and yet the related jurisdictional and institutional questions of how to achieve satisfactory global control remains contested. The essential mismatch, as scholars such as Fox and others have long noted, is the global capacity of cartels and the fragmentary mosaic of antitrust regulation. It does not necessarily follow that the WTO is a proper forum in which to devise solutions, but arguably the WTO could potentially make a contribution to what we feel requires a more comprehensive, multi-level effort. 14 Page 135.

Competition, Regulation and Public Policy Issues: An Overview 

xxi  

use of extraterritorial application of antitrust, a notoriously incomplete answer. In the context of their third proposition the authors also discuss the challenges presented by the participation of state-owned enterprises in international trade. Subsequently the authors turn to the ‘Observation’ referenced in their title: specifically, they point out that although some countries pulled the plug on the efforts made in the mid-1990s up until the Cancún Ministerial meeting in 2003 to develop a limited regime of ‘world antitrust’, many of the same countries later voluntarily entered into competition policy commitments in bilateral and regional trade agreements (some going much further than what was contemplated in the WTO context in the period of around 2001-2003). Does this perhaps suggest a change in the wind? We do not think so, and one has to note here the utter failure of the Doha Round – the anticlimactic Bali agreement notwithstanding. Given the abysmal state of the Doha negotiations it is unsurprising to see hardly any momentum toward WTO-level competition law initiatives, although the debate has not been snuffed out entirely. The authors are nevertheless quite right to urge further dialogue (including a trans-epistemic dialogue between the trade and competition communities), as silence on these issues is unlikely to be productive. A major difficulty is the existential challenge posed to the WTO itself, caught as it is in an anachronistic yet unalterable mold. With regard to competition law problems, the limitations of WTO remedies may present difficulties, although they could be overcome to some degree if realistic obligations are imposed on Members to ensure effective enforcement by domestic authorities including, in particular, courts and competition authorities. This may be challenging, however, for a number of WTO Members. A possible way forward could be a plurilateral agreement which focuses on very specific issues, and which synergizes with national capacities. Be that as it may, we come finally to the authors’ ‘Questions for Reflection’, which relate back to some of the issues described above. The questions posed may be translated into the following summarized points: – Some benefits might be achieved if the WTO Agreements more explicitly recognized the interdependencies of trade law and competition policy; – It may be desirable to require WTO Members to enforce a cartel prohibition, and to supplement such an obligation with commitments to avoid policies that encourage cartel behavior; – Renewed dialogue could be useful to ensure that the competition-oriented provisions already embedded within the WTO Agreements are applied appropriately; – Competition policy provisions routinely embraced in bilateral and regional agreements might usefully be considered for possible inclusion in multilateral agreements; – The inclusion within the WTO Agreements of a reference to competition policy as a ‘cornerstone’ principle could strengthen the status and legitimacy of competition authorities, particularly in times where economic pressure and political expedience threaten to weaken enforcement;

xxii  Mel Marquis and Veljko Milutinovič – Stronger coordination mechanisms in the WTO context, if based on careful study, might help to ensure the transparent and non-discriminatory application of competition laws.

2. Unilateral conduct in post-liberalization markets (or markets at some stage of liberalization) Alexandre de Streel discusses ‘The Antitrust Activism of the European Commission in the Telecommunications Sector’, and critically comments on the Commission’s continued antitrust interventionism after the initial stages of liberalization. De Streel takes issue primarily with the approach of the European Court of Justice (and of the Commission) in relation to margin squeeze, a practice central to the key decisions in Telefónica15 and Deutsche Telekom.16 Unlike the legal approach in the U.S., under EU law, neither an Oscar Bronner-type17 duty to supply the margin squeezed product nor predatory pricing had to be established as a condition for finding an abuse of dominance. In addition, although EU law accepts that when margins are assessed under EU law for their reasonableness it is the dominant undertaking’s retail costs that are taken as the default position, it nevertheless allows the retail costs of a competitor to be taken into account – a competitor that may or may not be ‘as efficient as’ the incumbent. Thus, De Streel concludes, the test is fairly ‘lax’. De Streel seems to hint that margin squeeze is, in essence, predation cloaked in different garments. Proof of predation under EU law is not normally accompanied by an Oscar Bronner test and it is therefore indeed logical to wonder what the distinction is between the two types of infringement. The essence of De Streel’s disagreement with the case law seems to be an economic one and it may well go to the very core of the concept of dominance, ultimately raising the question: how do we justify penalizing dominant undertakings when the dominated input is not essential? De Streel does not seek to address this (admittedly endless) political economy argument directly and remains focused on the practical and constitutional problems created by the case law. One fundamental critique that he levels at the Court is that it seems to apply the same criteria for margin squeeze in regulated and non-regulated industries. This is a fair point, as the Court’s approach undermines the rationale for having price regulators in the first place – the crux of the problems that arose in Telefónica and Deutsche Telekom. The Commission seems to take a different view in its Guidance Paper, in which it allows for a relaxation of the criteria applied to margin squeeze in 15 Case T-336/07 Telefónica, SA and Telefónica de España, SA v Commission [2012] ECR II-000; upheld on appeal: Case C-295/12 P, judgment of the ECJ of 10 July 2014 16 Case C-280/08 P Deutsche Telekom AG v Commission [2010] ECR I-9555. See also Case C-52/09 Konkurrensverket v TeliaSonera [2011] ECR I-527. 17 Case C-7/97 Oscar Bronner GmbH v Mediaprint Zeitungs-und Zeitschriftenverlag GmbH [1998] ECR I-7791.

Competition, Regulation and Public Policy Issues: An Overview 

xxiii  

regulated industries. Presumably, this is part of a broader liberalization policy that deliberately places a heavier burden on these industries. These problems are distinct from but intimately connected to the multiplication of legal proceedings, whereby regulatory approval may be followed by an antitrust challenge. As De Streel points out, this problem increases transaction costs and legal uncertainty. In his conclusion, De Streel points to two possible paradoxes of great import. One is the development of sector-specific competition law and the other, ‘worse still’ as he says, is the infiltration of sectoral reasoning into the general competition rules. From the arguments he sets out in the paper, it seems that the second risk is more likely. If the ECJ applies the same criteria across the board, whereas the Commission insists on laxer criteria for margin squeeze in regulated industries and is not, fundamentally, being overruled by the Court, then it may be more likely that these softer criteria will ‘leak’ into general antitrust cases. Returning to telecoms, De Streel points to the need to essentially let sectoral regulators do their job. In this context he is optimistic in the sense that the Commission is boosting the capacities of national regulators, reviewing their decisions ex ante and trying to intervene in a timely manner. De Streel submits finally that – to return to the proper province of antitrust as defined by the ECJ – the Commission should use its antitrust powers cautiously, once a regulatory decision has been made, in order not to ‘abuse’ them. Lars Kjølbye analyses some key abuse of dominance cases in the energy sector and asks whether the issues they raise constitute competition policy or energy policy. Going beyond the traditional essential facilities doctrine, Kjølbye looks at the ENI case,18 where the Commission objected to the ‘failure’ of the Italian gas operator to expand its existing pipeline capacity, despite repeated offers from third parties. Similarly, GDF and E.ON Ruhrgas were obliged to make a portion of their reserved capacity available to third parties, even though they were not hoarding but actually using that capacity. In terms of unilateral capacity reductions, in the E.ON case,19 the company allayed the Commission’s concerns by divesting some of its production capacity. Concluding that the Commission’s intervention in such cases goes beyond the general notion of abuse of dominance, Kjølbye argues that, in energy cases, it seems that the mere exercise of market power is becoming an abuse. He sees the decisions as being conceived as complements to the Third Energy Package. Yet despite the logic of having complementary policies under the same ‘internal market’ roof, Kjølbye argues that ‘the existence of regulatory obligations does not provide a basis for derogating from the normal conditions for finding an abuse and that the EU competition rules serve to protect competition on the market – not to enforce regulatory obligations’.20 To that extent, the paper complements those of De Streel, who warns of possible ‘spillover’ of regulatory thinking into antitrust, and of Ginsburg and Haar (see below), who perceive a Commission Decision of 29 September 2010 in Case COMP/39.315 – ENI, 2010 OJ C352/8. Commission Decision of 26 November 2008 in Case COMP/39.388 – German electricity wholesale market. 20 Page 213. 18 19

xxiv  Mel Marquis and Veljko Milutinovič certain ‘twisting’ of antitrust in order to avoid collision with regulation. It also sits well beside Monti’s paper, which notes, as explained above, how blunt-edged competition law thinking is ill-suited for sectors that may require a more nuanced approach if investment incentives are to be preserved. Ginevra Bruzzone and Marco Boccaccio write about ‘The Boundaries of Legitimate Conduct in Post-Liberalization Markets’. The general aims of the chapter are to consider appropriate antitrust approaches, in a post-liberalization environment, to: (i) certain pricing practices applied by multi-product dominant firms, bearing in mind that it is socially useful for such firms to be able to compete aggressively; and (ii) dominant firms’ investment choices in circumstances where a decision not to invest may contribute to a bottleneck constricting the supply operations of (potential) rivals. In their introduction, Bruzzone and Boccaccio observe that, quite often in markets formerly protected by exclusive rights, the market power of incumbents results not from a competitive winnowing process but from the legal impossibility of challenges from would-be contenders for success in the market. In EU competition law, there is a principle of competitive neutrality, meaning that while public enterprises should not be treated better than other undertakings, they should not be treated worse either.21 Article 345 TFEU is often cited as legal authority for this neutrality principle, although the principle may in fact be said to be embedded in the Treaty as a whole. (Furthermore, the principle understood more broadly would encompass not just public undertakings but the broader category of privileged undertakings.) In light of the competitive neutrality principle, the authors point out that discrimination against public undertakings would thus be precluded. On the other hand, they also note that the Court of Justice has explicitly stated that, where a de facto dominant position originates from a legal monopoly and not from competition ‘on the merits’, this must be ‘taken into account’.22 At first sight these positions may seem difficult to reconcile; but in reality the ECJ’s approach, which calls for a case-by-case assessment, does not require any discrimination. The Court is in effect saying that, when a dominant position has been achieved off the merits, so to speak, it may be that the situation is no longer equivalent to that of a firm that has won a market by virtue of a superior product or superior business acumen. Discrimination in such circumstances would not be an issue. Nevertheless, the fact remains that Article 102 can only apply where there has been abusive conduct; privileged firms whose conduct conforms to the requirements of the law as interpreted by the ECJ should in principle have nothing to fear (although the jurisprudence on the combined application of Article 102 and Article 106 is perhaps now calling this into question). In the second section of the chapter the authors discuss price competition by multi-product dominant firms. 21 For an aging but still relevant discussion, see, e.g., Giuliano Marenco, ‘Public Sector and Community Law’, 20 Common Market Law Review 495 (1983). 22 On this point the authors cite Case C-209/10 Post Danmark A/S v Konkorrencerådet [2012] ECR I-000, para 23.

Competition, Regulation and Public Policy Issues: An Overview 

xxv  

Several cases at EU and national level are discussed. Beginning with the Commission’s Deutsche Post decision,23 where the vexed problem of common costs came into focus, the authors note that where common costs are particularly significant, incremental costs may prove to be an inadequate benchmark (indeed, most price-cost tests seem to have their limits, not least in scenarios involving two-sided or multi-sided markets). This leads to a detailed discussion of the ECJ’s Post Danmark judgment,24 which has been construed variously as a watershed by some (including Bruzzone and Boccaccio) and as something of a non-event by others. In this case, without providing sharp clarity on the issue, the ECJ accepted the allocation by the Danish authorities of part of Post Danmark’s common costs for purposes of determining incremental costs. The authors understandably worry that such an approach could engender significant uncertainty if clear limiting principles are not developed – although they take comfort in the additional criterion under Post Danmark that actual or likely anticompetitive effects must in any case be shown. The authors proceed to discuss such limiting principles, underlining for example that account should be taken of the need to allow dominant firms to capitalize on economies of scope, and of the need to incentivize competitors to expand their own product range. The remainder of the second section provides a round-up of administrative and judicial practice at the level of the Member States where issues similar to those in Post Danmark have arisen. In the third section Bruzzone and Boccaccio review the controversial subject of strategic underinvestment (or ‘duty to invest’), with a focus on both national (TTPC) and EU-level (Eni foreclosure) cases concerning the Italian natural gas incumbent, ENI. For several reasons the authors recommend against any further imposition of affirmative investment decisions under an abuse of dominance theory, as punishing a failure to invest surely lies at the outer limit of the scope of Article 102, if it does not slip over the edge. In some sectors, such as energy or telecoms, the matter of a duty to invest may be subject to regulatory obligations in line with EU Directives.

3. Technology and standard-essential patents Calvin Goldman, Julie Soloway and David Dueck delivered a paper entitled ‘Competition Policy and Technology Markets: A Canadian Perspective’. The paper also adopts a Schumpeterian perspective and its central message is that positive Schumpeterian effects – disruptive innovation, dynamic efficiency and 23 Commission Decision of 20 March 2001 in Case COMP/35.141 – Deutsche Post AG, 2001 OJ L25/27. 24 Cited above note 22. The Court of Justice will rule on Post Danmark’s pricing practices again in Case C-23/14 but the issues this time concern the granting of rebates. In light of the reasoning in the General Court’s Intel judgment it will be very interesting indeed to see what analytical framework the ECJ decides to adopt in Post Danmark II.

xxvi  Mel Marquis and Veljko Milutinovič technological paradigm shifts – need to be taken into account when competition law concepts are applied to particular industries. The authors do not speak in terms of error costs, but their arguments can be understood in this way. Type I errors in the application of merger control or antitrust rules will be particularly harmful where potential dynamic efficiencies are concerned, often in the context of the exercise of intellectual property rights. From this perspective the authors give examples of how Canadian (and U.S.) antitrust law has taken account of the balance between short-term allocative and long-term dynamic effects. Needless to say, depending on the circumstances in particular cases, the tradeoffs can be difficult. One can understand how probabilities can be assessed, and how imperfect predictions can be made, when industry trends are discernible. It is less clear how Schumpeterian competition can be taken into account in the absence of manageable data. In these instances it does not seem inappropriate to resort to the usual legal mechanism, i.e., the burden of proof. If it is claimed, for example, that dynamic efficiencies justify a merger with short-term anticompetitive effects, the onus is properly on the merging parties to show why any doubts should be put aside. This approach should not imply inflexible skepticism but it does imply that even in high-tech settings, the actualization of Schumpeterian benefits should not be taken on faith. Paul Lugard’s chapter is melodically entitled: ‘Innovation in the Domain of EU Competition Law Enforcement: Blowing in the Wind?’. In this chapter Lugard discusses how innovation is (not) taken into account in EU competition law. For this purpose, he posits two questions: ‘whether under EU competition law a comprehensive vision on the nature and the importance of innovation has been developed, and [...] which analytical framework, if any, applies to protect innovative business activities’.25 He identifies, in this context, three ‘pillars’ of innovative activities: intellectual property rights, technology platforms and technical standards. Lugard notes that, despite the myriad ‘innovation cases’ decided thus far by the Commission and despite the various political strategies and agendas, there is still no clear framework under EU competition law concerning the ‘nature and the importance of innovation and the way in which innovation should be incorporated in the competitive assessment of business transactions’.26 He identifies three possible causes for the lack of such a framework: the unpredictability of markets, the tight legal constraints of the Merger Regulation and of Article 101(3), and the mostly reactive nature of the Commission’s action – responding to cases as they occur. In contrast to this enforcement reality, Lugard observes the significant advances economic science has made in isolating factors that favor or disfavor innovation in various settings. He also notes two fundamentally different types of innovation: gradual and breakthrough (disruptive) innovation. In its decision practice, the Commission seems mostly intent on removing impediments to gradual innovation, with little experience in the field of breakthrough innovation. Attention is then drawn to the 25 26

Page 269. Page 272.

Competition, Regulation and Public Policy Issues: An Overview 

xxvii  

Commission’s toughening stance against the abuse of intellectual property rights in recent years, citing cases involving ‘pay for delay’ deals in pharmaceuticals, the greater insistence on FRAND in standardization agreements, and the removal from the Technology Transfer Block Exemption of a clause that exempted (from Article 101) licensors who terminate licensees if they challenge the validity of their IP. Under Article 102, particular problems have arisen with FRAND: the Commission finds that Article 102 essentially forces standard-essential patent holders to try to negotiate with patent infringers on FRAND terms first, before suing. In conclusion, Lugard points out that, while there is still no comprehensive framework that accounts for innovation, innovation is playing an increasing role in many competition cases, including those that raise issues of standardization, access to technology platforms and (essential) intellectual property rights. Sven Völcker’s contribution is called ‘Promises, Promises … Law and Policy Considerations with Respect to the Application of Article 102 TFEU to FRANDEncumbered Standard-Essential Patents’. As Völcker says in his summary, this paper argues that the Commission is justified in investigating attempts by dominant firms to secure injunctions against the use of SEPs by a willing licensee where a FRAND commitment has been made.27 As he maintains, the Commission is better placed than national courts (and ADR fora) to develop guiding principles for the meaning of FRAND, since for example specialized patent courts may not give sufficient attention to competition law principles, or because the Commission may be in a better position to consider related litigation worldwide. Courts and arbitrators may be involved, however, when a specific FRAND rate needs to be determined on the basis of such guiding principles. Völcker argues further that applications for injunctions in such circumstances should not be evaluated exclusively under the strict standard developed in the ‘vexatious litigation’ case law. Rather, a more flexible standard should be applied, thus opening the door to intervention under Article 102 TFEU. In particular, and as explained in part 3 of his chapter, Völcker considers such conduct to infringe that provision in three ways. First, he argues that seeking an injunction in this context amounts to an abusive refusal to deal. Second, it can be understood as a means to extract unfairly high royalties and/or to impose other unfair terms. And third, it may also constitute anticompetitive discrimination. According to this approach, once it is established that a dominant firm has sought (or has enforced) an injunction against a willing licensee despite having given a FRAND commitment, the burden of proof would shift to that firm to demonstrate that its actions are objectively justified. In this context, Völcker suggests that the objective justification concept should be construed narrowly, and he underlines the relevance of paragraph 82 of the Commission’s Guidance Paper on exclusionary conduct under Article 102. He advances a series of reasons why claimed justifications based on the supposed 27 Many recent articles discuss SEPs, injunctions, hold-up problems, opportunistic behaviour by IP holders and the like. See, e.g., Urska Petrovcic, ‘Patent Hold-Up and the Limits of Competition Law: A Trans-Atlantic Perspective’, 50 Common Market Law Review 1363 (2013).

xxviii  Mel Marquis and Veljko Milutinovič diminished investment incentives of the defendant are likely to be suspect, and suggests that in any case an injunction will tend to be a disproportionate measure given the possibility to file claims for damages. Following the completion of Völcker’s chapter, the Commission finalized its investigations in the Motorola Mobility case,28 which resulted in an infringement decision (but no fines), and in the Samsung case,29 resolved by way of commitments. Although the Motorola Mobility decision is not yet available as of this writing, the Commission’s press release seems to indicate that the theory of liability is linked to the unfair distortion of license negotiations, with adverse effects for consumers. The Commission also found that it was abusive for Motorola to insist, under threat of enforcement of an injunction, that the implementer, Apple, surrender its rights to challenge the validity of, or the allegation that Apple’s mobile devices infringe, Motorola’s SEPs. This suggests that the Commission will broadly construe the concept of a ‘willing’ licensee. In Samsung, the commitments accepted by the Commission, which will expire in 2019, establish a ‘safe harbor’ for willing licensees that sign up to a particular licensing framework, with 12 months to strike a deal. This framework provides for dispute settlement (by a court or arbitrator) where disputes arise with regard to the meaning of FRAND. In the future, the Court of Justice is expected to weigh in on the issue of what distinguishes unwilling from willing licensees.30 From an economic perspective, Cristina Caffarra and Pierre Régibeau also address SEP-related issues in their paper, ‘Patent Explosion and Patent Wars: Hold-Up, Royalties and Misunderstandings over “Market Value”’. By way of introduction, Caffarra and Régibeau present a general overview of the ongoing global conflicts among major IP holders. Many of these conflicts involve SEPs and applications for injunctions but, as the authors point out, these suits may be seen as part of a broader struggle between platforms in a converged mobile environment. But the heart of their chapter is section 3, where a series of arguments are presented in support of the need for (foundational) innovators to be able to secure adequate royalties, including in the FRAND context, to maintain investment incentives. Much of this section is aimed at countering mistrust in royalties that seem to ‘overreach’ insofar as they are partly based on the final price of the product in which a technology is embedded (thus appearing to extend to unrelated functionalities). The authors put great emphasis on a body of economic literature finding it appropriate for follow-on innovators to share rents with those innovators that blazed a trail for them. The argument that it is not the base itself that is decisive but the amount of compensation that matters is intuitively appealing. As Caffarra and Régibeau explain, ‘the level of compensation for the licensor must be linked to the economic value that its technology creates. This See Commission Press Release IP/14/489 of 29 April 2014. See Commission Press Release IP/14/490 of 29 April 2014. The decision is available at http:// ec.europa.eu/competition/antitrust/cases/dec_docs/39939/39939_1501_5.pdf. 30 See Case C-170/13 Huawei Technologies Co. Ltd. v ZTE Corp, ZTE Deutschland GmbH, OJ C215/5, not yet decided. 28 29

Competition, Regulation and Public Policy Issues: An Overview 

xxix  

economic value will typically include not only the standalone value of the features enabled by the technology but also the increase in the value of features enhanced by the technology (synergies) and the extra profit on the whole device arising because of the additional sales which are induced by the increased value of the product incorporating the infringed technology.’31 Taking that logic a step further, increased value from enhanced sales by the licensee of complementary products would also be taken into account. The authors’ royalty-friendly approach might draw objections on the ground that future innovation might be inhibited, but the authors dismiss this concern. As they claim, for example, the licensor and the licensee have generally aligned interests in maximizing the total surplus ‘pie’, and thus in promoting future innovation that grows a bigger pie. In a final part of the chapter, the authors consider the possibility that decision makers might be focusing on the royalty base as one tactic (another would be ex post interventions carving out fields of use) to ‘redesign’ patent rights, given the general discontent with patent systems which confer overly broad (and/or lengthy) protection from competition and which may produce, in some industries, a foul swampland of patent claims.

4. The role of courts. Reconciling, competition, non-competition and constitutional imperatives Küllike Jürimäe delivered a chapter entitled: 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 this chapter Jürimäe discusses the legally and politically complex issue of third party access to documents generated in antitrust proceedings before the Commission and the delicate balance between effective public enforcement, transparency and victims’ rights, especially the right to damages. (An ‘epilogue’ to the chapter was added after the Commission published its draft ‘Damages Directive’ of June 2013 but before the final form of the Directive was clear.) She emphasizes two particular aspects of this issue. The first is institutional: who should strike the balance – EU courts, national courts, the EU legislator or the national legislator? The second aspect is substantive and looks at existing case law and legislation where attempts to strike such a balance were made. Having called attention to these two aspects, Jürimäe proceeds to consider the need for, and eventual content of, EU legislation in this field. The analysis takes us through two of the three main channels that third parties use to try to obtain access to documents: (i) Transparency Regulation actions, and (ii) Article 267 preliminary references. Interventions in Article 263 actions for annulment are left out, as documents obtained by interveners cannot generally be used in subsequent damages actions. The first avenue has evolved, through the case law, 31

Page 326 (emphasis in original).

xxx  Mel Marquis and Veljko Milutinovič from wide access combined with individual examination of documents, to general presumptions of protection for some categories of documents. With regard to competition, the General Court, in CDC32 and other cases, has refused to allow the Commission to exclude entire categories in antitrust cases a priori. For its part, the Court of Justice, when dealing with national competition authority documents to be disclosed under national law, stated the EU position to be, essentially, that national courts had to perform the balancing exercise between public and private enforcement (in Pfleiderer33). As regards national law solutions, while all Member States offer some protection to competition documents, the level of protection varies greatly. Likewise, post-Pfleiderer, Jürimäe notes a significant divergence among national courts’ performance of the balancing exercise. Jürimäe’s analysis confirms the fears by some commentators in the aftermath of Pfleiderer,34 and confirms moreover that EU legislative intervention in this area is essential. She cites three main factors necessitating such intervention, consisting of the three-fold need: to provide guidance for national judges; to achieve closer harmonization of rules; and to prevent forum-shopping. It is suggested further that the peculiarities of antitrust enforcement that make access to documents under the Transparency Regulation controversial must be taken into account, and this can be done in the form of lex specialis. Such lex specialis should set out the main principles by which transparency, public enforcement and private enforcement may be balanced, while still leaving the performance of the specific balancing exercise to the national judge. By way of epilogue, Jürimäe notes how the Donau Chemie judgment35 of the Court of Justice pre-empted blanket national bans on disclosure and refused to allow a general exemption from disclosure for leniency documents. She concludes that only time will tell whether the Commission’s draft Damages Directive, which envisaged, inter alia, an exemption from disclosure for corporate statements has, indeed, struck the right balance. In April of 2014, the Parliament adopted a compromise text, and it is now clear that leniency materials in a strict sense (i.e., corporate statements) will be withdrawn from the Pfleiderer balancing regime, an element that will surely be criticized by certain commentators. However, the seemingly absolute protection for corporate statements in the context of damages actions (Article 6(6) of the Directive as adopted) is without prejudice to the possibility of invoking the Transparency Regulation in order to obtain such statements (Article 6(2) and Recital 19). Given the jurisprudence of the ECJ, which repeatedly diverged from the General Court’s approach,36 the possibility of claiming that a corporate statement is akin to an essential facility for the purpose of maintaining a damages Case T-437/08 CDC Hydrogene Peroxide v Commission [2011] ECR II-8251. Case C-360/09 Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161. See Philip Lowe and Mel Marquis, eds., European Competition Law Annual 2011: Integrating Public and Private Enforcement – Implications for Courts and Agencies, Hart Publishing, 2014. 35 See Case C-536/11 Bundeswettbewerbsbehörde v Donau Chemie AG and Others. 36 See most recently Case C-365/12 P Commission v EnBW Energie Baden-Württemberg AG, judgment of the ECJ of 27 March 2014. 32 33 34

Competition, Regulation and Public Policy Issues: An Overview 

xxxi  

suit, and that consequently there is an ‘overriding public interest’ (Article 4 of the Transparency Regulation) in disclosure by the Commission, remains open. However, we expect that possibility to be rather theoretical. It seems likely that the ECJ will confine the use of the ‘overriding interest’ concept to truly extraordinary cases, and this would be helpful from a number of angles including not least legal certainty. Rosa Perna’s chapter is called ‘The Role of Courts in Reconciling Competition, Non-Competition and Constitutional Imperatives: The Italian Experience’. Overall, this chapter revolves around the constitutional ascendance of the market as a ‘positive value’ in Italy, a development largely linked to the constitutional reform of 2001. Another implicit theme is the difficult role(s) of national judges in the EU context, which must be mindful not only of their respective national legal orders and juridical-cultural conditioning but also of the superimposed patterns and principles of EU ‘federal’ law, which seems to be perpetually chipping away at the margins of those national orders.37 Perna begins with the observation that the protection of free competition is part of the ‘genetic code’ of EU law. But is the same true of national law? In mixed economies such as Italy, one is tempted to view free competition as more of a mutation, rather than an established gene, a shift that might enhance a country’s chance of survival in a globally competitive environment – but only if the mutation itself is nurtured. In this regard, EU law can make a positive impact but it would be an error to suppose that the existence and application of competition law is enough to address relatively uncompetitive economies without additional efforts deployed at multiple levels of politics and society. Putting this perhaps intractable issue to one side and returning to the paper, Perna discusses a number of issues relevant in the Italian context, such as Article 41 of the Constitution (freedom of economic initiative – tempered in reality by the concept of ‘social purposes’, which appears in the very same Article), the constitutional recognition of the ‘competition principle’ (especially in Article 117(2)(e) of the Constitution) and Italy’s flirtations with national liberalization programs (e.g., in connection with the Decreto salva-Italia of 2011). Part II of the paper discusses the judicial review of administrative decisions (specifically those of the Autorità Garante della Concorrenza e del Mercato), and Part III comes to the difficult issue of reconciling competition, non-competition and constitutional values. With regard to the latter, the master-principle of the hierarchy of norms is clearly essential. An open question, which perhaps cannot be answered satisfactorily in the abstract, is what a judge must do when principles of equal rank collide. When this occurs, decisions may be made according to policy preferences not easily linked to purely legal motivations. When those preferences are revealed, practice and path dependence will likely take over and drive future cases. This may provide some context for the cases Perna discusses in which (distortive) regulation or taxes are challenged in particular 37 To cite only the most recent example as of this writing, see Case C-557/12 Kone AG and Others v ÖBB-Infrastruktur AG, judgment of the Court of Justice of 5 June 2014, not yet reported.

xxxii  Mel Marquis and Veljko Milutinovič sectors: pharmacies and other public health matters; mobile phone services; and ‘Sunday trading’ rules. A telling observation here is that, in some cases, conflicts have been decided without any resort to the principles of necessity and proportionality. Of course, where a conflict also concerns principles of EU law – ‘effectiveness’ being a classic example – additional obligations on national judges come into play, as seen in innumerable cases decided by the ECJ. The next chapter, by Douglas Ginsburg and Daniel Haar, is called ‘Resolving Conflicts between Competition and Other Values: The Roles of Courts and Other Institutions in the US and the EU’. In this chapter, Ginsburg and Haar discuss predominantly institutional aspects in the EU and U.S. that are relevant to resolving conflicts between competition and other values. The authors begin by asking the vital substantive question: is there only a conflict of means or also a conflict of ends between two laws? When there is a confict of means, the means may be reconciled with the same end by an institution with competence to apply both. Conflicts of ends, however, must be resolved by one of the ends giving way, at least partially, to the other. In terms of institutions that resolve conflicts, these may be courts, competition agencies or the legislature. The role of competition agencies is the most limited, since, in the authors’ understanding, they can usefully reach only as far as economic efficiency outcomes are sought. When such outcomes are not sought or are willfully avoided, competition authorities will be out of place. The role of the courts is freer and relatively straightforward when express derogations from competition principles are found in legislation. Ginsburg and Haar note, among other things, the difference of historical context between the EU and the U.S. In the U.S., unlike the EU, the basic antitrust law came into force very early, and indeed it took effect long before most other federal statutes were adopted. Consequently, the issue of lex posterior derogat legi priori in relation to the Sherman Act arose rather frequently. An interesting point that arises implicitly here already is that the operation of lex posterior in this context tends to debunk the claim seen occasionally according to which the Sherman Act has been elevated to quasi-constitutional status in the United States;38 this romantic conception of U.S. antitrust law is mercifully laid to rest by the authors, although doing so is not an aim of the chapter. With regard to resolving conflicts, the authors identify four judicial methods: (i) interpretation of an express statutory exemption to preclude application of the antitrust laws; (ii) interpretation of a statute to require a case-by-case balancing of competition and other values; (iii) interpretation of a conflict between antitrust law and a later law to preclude application of the antitrust law; and (iv) interpreting the vaguely worded antitrust laws narrowly so as to avoid a conflict with a later law. Most interesting are implied repeals of the Sherman Act by Congress: because the U.S. Supreme Court is generally averse to them, lower courts tended to limit the notion 38 See Mel Marquis, ‘Idea Merchants and Paradigm Peddlers’, 28(1) Pacific McGeorge Global Business and Development Law Journal (2015).

Competition, Regulation and Public Policy Issues: An Overview 

xxxiii  

of ‘conflict’ to situations of incompatible outcome – until Credit Suisse,39 which has significantly altered the landscape. In terms of the last option, limiting the scope of the antitrust rules, the authors discuss Trinko,40 where the Supreme Court, instead of finding a partial repeal of antitrust in the telecommunications laws, simply limited the reach of the antitrust laws. With regard to the EU, Ginsburg and Haar note the ‘constitutional’ (primary law) nature of the EU antitrust rules – which, again, distinguishes them from the rules contained in the Sherman Act. According to the authors the rigidity of the EU rules is somewhat offset by the existence of Article 101(3) TFEU but, overall, the ‘supremacy’ of the antitrust rules over other (secondary law) rules in the European system is far more pronounced than it is in the U.S. system. The EU has an additional peculiarity in the sense that, because Articles 101 and 102 TFEU are ‘constitutional’, the legislature itself may well be stuck with the interpretations of the Court of Justice.41 By contrast, since the Sherman/Clayton Acts are ultimately no more than ordinary statutes, the U.S. Supreme Court can always be overruled by Congress. Rather than supporting more judicial activism at the EU level, precisely due to such cementing constitutional consequences of a judgment, the authors suggest that a well-placed institution for anticipating conflicts between competition and other goals would be DG Competition, not least through its system of block exemptions.42 This would represent a perhaps questionable departure from the traditional understanding of block exemptions, and there may be limits to DG Comp’s capacities in this regard (although the Commission’s other DGs would surely be consulted if ever such an initiative were attempted). Be that as it may, the authors pass on to a final point, namely that a distinct advantage of a system largely based on judicial decisions that pass from first instance to the Supreme Court, as in the U.S., is that conflicts do not have to be anticipated: they are addressed based on a large amount of data that accrues in individual cases. Ian Forrester encapsulates many of the Workshop’s explicit and underlying themes in his chapter, ‘Competition Law and Public Policy Considerations’. Here he addresses the perennial and ceaselessly evolving issue of ‘how to reconcile common sense, public interest considerations and the ostensibly bleak prohibitory language’43 of the EU competition rules. His suggested approach is to challenge Credit Suisse Securities (USA) v. Billing, 551 U.S. 264 (2007). Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, 540 U.S. 398 (2004). 41 While the EU is the paradigmatic model of ‘constitutionally’ based competition law, it is doubtful that the European public ascribes constitutional gravity to it – in contrast for example to civil rights or the principle of democratic government. See page 342, footnote 12 (citing Gareth Davies, ‘Democracy and Legitimacy in the Shadow of Purposive Competence’, European Law Journal, DOI: 10.1111/ eulj.12079 (2013)). 42 For an earlier suggestion to use of block exemption regulations (as well as Article 10 decisions under Regulation 1/2003) as devices to mediate between public policy goals, see Holger Quellmalz, ‘Die Juztibilität des Art. 81(3) und die nicht-wettbewerblichen Ziele des EG-Vertrags‘, (2004/4) Wettbewerb in Recht und Praxis 461, 466-467. 43 Page 437. 39 40

xxxiv  Mel Marquis and Veljko Milutinovič the market integration bias of EU competition law and to look for solutions that are more in line with contemporary market theory and reality. Before discussing anything contemporary, however, Forrester conducts a fascinating analysis of more ancient English rules, including the restraint of trade doctrine. Already in late 1800s England, he finds a judicial balancing exercise between free competition and freedom of contract. Coming to more recent history, the discussion turns to the period in the 1970s and 1980s when the Commission was relatively lenient with regard to ‘crisis cartels’. Famous cases, or infamous for some, like Synthetic Fibres44 and Dutch Brick-makers45 are cited, and Forrester shows how ‘industrial policy’ has come to play a much more muted role in competition policy at EU level in recent years – a remarkable development given Europe’s economic malaise over the last half-decade. Both the Commission and the Court of Justice had a role to play in this respect, notably in the Irish Beef preliminary reference case46 and its fallout in national legal proceedings. Notwithstanding that, in a Union of increasingly complex goal structures, there may be a trend toward the rélativisation of the role of competition policy (and of the economic dimension of European integration more generally) where social objectives are at play,47 the particular paradigm of industrial self-help purportedly to alleviate actual or feared social dislocation has shifted rather dramatically. Forrester also analyses cases of price-setting by Member State authorities and recalls that, where there are no signs of an agreement prohibited by Article 101, Member States have substantial room for manoeuvre, subject to the Treaty’s free movement rules. Occasionally, borderline cases appear, such as Arduino48 and Cipolla,49 where Italy had essentially approved privately arranged tariffs – with the possibility of a ministerial override. (The Italian regime was later liberalized via legislative reforms.) Borderline cases of a somewhat different nature are also discussed, such as Wouters50 and Meca-Medina,51 where the Court shows Commission Decision of 4 July 1984 – Synthetic Fibres, 1984 OJ L207/17. Commission Decision of 29 April 1994 – Stichting Baksteen, 1994 OJ L131/15. 46 Case C-209/07 The Competition Authority v Beef Industry Development Society Ltd, Barry Brothers (Carrigmore) Meats [2008] ECR I-8637. 47 The apparent recasting of the Union’s essential character by the ECJ was originally seen in the free movement field (Viking and Laval, although the Court’s statement has been disparaged by some as disingenuous given the outcomes in those cases), but this move had wider implications, and the ‘not only an economic but also a social purpose’ formulation is spreading, predictably, to other sectors of jurisprudence such as state aid. This appears starkly in a 2009 judgment of the Court, where it is observed that, ‘since the Community has not only an economic but also a social purpose, the rights under the provisions of the Treaty on State aid and competition must be balanced, where appropriate, against the objectives pursued by social policy, which include […] proper social protection […]’. Case C-319/07 P 3F, formerly Specialarbejderforbundet i Danmark (SID) v Commission [2009] ECR I-5963, para 58 (emphasis added). Although the Court refers to the ‘rights’ under the competition rules, given the nature of the case (i.e., an action against the Commission for deciding not to object to a tax relief scheme), we presume that the same principle must also apply with regard to obligations under those provisions. 48 Case C-35/99 Manuele Arduino and Compagnia Assicuratrice RAS SpA [2002] ECR I-1529. 49 Joined Cases C-94/04 and C-202/04 Cipolla and Macrino [2006] ECR I-11421. 50 Case C-309/99 J. C. J. Wouters, J. W. Savelbergh and Price Waterhouse Belastingadviseurs BV v Algemene Raad van de Nederlandse Orde van Advocaten [2002] ECR I-1577. 51 Case C-519/04 P David Meca-Medina and Igor Majcen v Commission [2006] ECR I-6991. 44 45

Competition, Regulation and Public Policy Issues: An Overview 

xxxv  

deference to public interest considerations under certain conditions. Various arguments in support of public policy, including some of the less convincing ones (such as those raised by the Italian Government in the British Telecom case52) are analysed. The discussion also covers the Commission’s well-known CECED decision, where environmental considerations were explicitly taken into account under Article 101; and the ECJ’s Sot. Lélos judgment,53 an Article 102 case where the single market imperative was relevant in relation to the unilateral curbing of parallel trade in pharmaceuticals but where, nevertheless, the Court left the door open for a dominant firm to take proportionate steps – even if parallel imports are thereby squelched – to protect its commercial interests.54 The chapter concludes with the observation that, by and large, public policy considerations have been ‘decorative rather than substantive’ in EU competition law cases, as decided by the Commission and by the Court of Justice, with a broad exception for price-setting by public authorities, which has largely been tolerated under the competition rules. Forrester hopes for a re-evaluation of certain old policies, such as the primacy of market integration over ‘normal competition analysis’. Philip Lowe once again provides ‘Conclusions’ to the discussions held at the Workshop on which this book is based. As usual, Lowe deftly and succinctly captures a variety of points that were made. We do not propose to summarize his chapter – res ipsa.

Case 41/83 Italy v Commission [1985] ECR 873. Joined Cases C-468/06 to C-478/06 Sot. Lélos kai Sia EE v GlaxoSmithKline [2008] ECR I-7139. The breadth of this exception to the usual opprobrium the Court attaches to the partitioning of the Single Market is not altogether clear; the burden of proof appears to be on the dominant firm, and the cognizable scope of ‘commercial interests’ may have its limits. While the tenor of the judgment is not unequivocal, the sympathy for commercial interests (i.e., the interest in shutting down parallel trade) may be seen in paras 71 in fine and 76. In the former, the Court states that it is permissible for a dominant pharma company ‘to counter in a reasonable and proportionate way the threat to its own commercial interests potentially posed by the activities of an undertaking which wishes to be supplied in the first Member State with significant quantities of products that are essentially destined for parallel export’. In the latter, the Court states that such a company ‘must be in a position to protect its own commercial interests if it is confronted with orders that are out of the ordinary in terms of quantity. Such could be the case, in a given Member State, if certain wholesalers order [medicines from that producer] in quantities which are out of all proportion to those previously sold by the same wholesalers to meet the needs of the market in that Member State.’ 52 53 54

2  Competition, Regulation and Public Policies

Welcome Philip Lowe: First of all, thank you all for accepting the invitation to be here and for delivering the necessary visiting cards of the papers and interventions you’ve sent us. Secondly, many thanks to Mel for all the preparation in advance, and to those who are about to chair our sessions over the next 36 hours. Thirdly, a very sincere thanks to our sponsors for their indispensable support. In choosing themes to focus our discussion on this year, we have been trying to vary it to the extent that there may be more a judicial focus in some sessions but in others a little more economic or regulatory focus. For those who are disappointed that there isn’t more of a judicial element this year – although we do have courts represented, as we’ll find out tomorrow morning – don’t worry, next year we’ll come back to a heavier emphasis on legal debates. The reason why Mel and I focused on this bouquet of issues is not just the economic context, but the increasing interplay of competition policy and enforcement and the impacts of other public policies. You can look at it first of all in the narrow dimension of what may be admissible under competition law because of exceptional circumstances of low economic activity, for example structural over-capacity, as Ian has very eloquently pointed out in his paper. There we get into issues like crisis cartels or agreements to reduce capacity, but there may also be a concern to ensure that there’s effective market functioning in a somewhat more noble sense of promotion of industry-wide standards or technical standards or standards of professional conducts, or environmental and safety standards. Here again there’s the extent to which private agreements are condoned or endorsed by public authorities and the significance this has in determining whether they’re legally permissible. If you look at it in the wider dimension of the interplay between competition policy and other public policies, well there are plenty of areas in many jurisdictions these days where, if you press government ministers and parliamentarians they will say that competition policy in the present macro-economic context is of less significance than other concerns, for example the stability of the economic and financial system, the need for systemic risk to be covered, the possibility that that could be actually promoted by – in some people’s eyes – more concentration of market power rather than less, which is something which divides many people in a number of our countries. Or take the concern in some jurisdictions to prevent complete economic dependence on outside sources of goods and services. In my own area of energy at the moment we have some very pernicious clauses on foreign ownership of concerns. They are usually referred to as the ‘Gazprom’ clauses, and they prevent virtually any supplier of anything in the world upstream in energy from owning or partially owning any network or pipeline. This of course has led to huge unintended consequences; you get guys who are owners or partowners of a wind turbine or a wind farm in Argentina and they want to take a financial participation in a gas-pipe network in Portugal and the European

Welcome and Panel I 

3  

Commission says ‘No you can’t do that, you’re not fully un-bundled’. But that’s because of the Gazprom clause. Then of course there are restrictions which are pretty legitimate in terms of ensuring security of supply and ensuring a level playing field with third country operators. Now there’s a third dimension which is equally significant and more and more significant in network industries and network service industries, which is the extent to which competition policy is serving aims which are shared by other sectoral policies, and where the question is really what the best instrument is to solve the problem. And in consequence we have in most jurisdictions armies of regulators who are absolutely convinced that their action ex ante or even ex post is a more effective method of achieving the aim of making markets work than waiting for us, or them (whichever side you’re on), to finish a competition case. So the issue there is not necessarily one of different objectives, but of different degrees of effectiveness in achieving the result. So those are three dimensions of this interplay that appear to me to be very important, and they are reflected in each of the sessions we’re going to have today. In the last session we will be very interested to hear everyone’s comments on how courts have interpreted these different interplays and the different dimensions over the years, whether in terms of effectiveness, or trade-offs of policy objectives. So I’m going to hand over to Giorgio now for our first session, which is really quite a short one, but a global one.

PANEL I: Economic context, market malfunction and public action Giorgio Monti: Thank you, Philip. Allow me to welcome all of you to the European University Institute. I trust this will be another successful edition of the Workshop, and I won’t say any more than that, except that I’ll ask the speakers to keep to ten minutes so that we allow for a bit of time at the end for discussion during the roundtables. Kai-Uwe Kühn: I want to start off with the importance of the idea of market failure in competition policy. In antitrust and merger policy we’re often very consciously leaving out anything that has to do with market failures and asymmetric information because we say that’s not for this instrument. On the other hand, market failure plays a key role in state aid policy. If you’re thinking of competitive distortions in that context, they have little to do with market power; usually they relate to other kinds of other failures in markets. Also, the legitimization of state aid usually asks the question ‘where is the market failure that is supposed to be corrected?’ Not every other policy instrument or every other policy looks at these questions in terms of market failures, and the potential for policy conflict there is pretty large.

4  Competition, Regulation and Public Policies So I just wanted to briefly touch on three challenges that competition policy is facing in this interplay of different policies, especially in a time where there are broad policy concerns that people want to fix quickly. So the first question is: Does and should competition policy impinge on other policy fields that don’t think about market failure in quite the same way? I think we’ve had that whole discussion in merger policy versus industrial policy, and European or national champions. There, it seemed to be resolved to some extent although it always comes back, but the problem was that anyone who was focused on national champions had a hard time answering what exactly the market failure was that should lead to concentration. But we’re getting much more difficult discussions now, for example with patents in antitrust. You can think about the patent problems we have as hold-up problems where investments are not taking place because of potential market power ex post. You can think of that as an excessive pricing problem that you can deal with appropriately with competition policy, but it is interfering with patent policy and possibly better instruments in that area. The question to what extent one should intervene with competition policy there or not is an important one, and I’m going to get back to that in a minute. On the state aid side, I think this theme comes up when we’re talking about state aid modernization. The one way in which we’ve very often avoided conflicts is that although we’ve said we’re ‘more economic’ and so we have to address a market failure, if I look at past state aid cases the ‘market failure’ often might have been a lot of uncertainty in a market; that’s not exactly a market failure. Or that this product isn’t produced and I would like it to be; well that’s not exactly a market failure either. So, once you start asking ‘where exactly is the market failure?’, and when you try to incorporate that in state aid control – you might be in for a lot of intense discussions. For example, with R&D aid you might ask: why are you subsidizing helicopters and not fast computers, where’s the market failure in helicopters? And in infrastructure investments – in broadband it may be very legitimate to have goals for where you want to go with broadband, but the question from an economic point of view is: where is the market failure that dictates some higher rate of speed? Maybe the market is getting it right. Maybe the bigger business is not in the highest speed but in other things. The questions get much more difficult, and I think that one of the problems is that decision makers in a lot of policy areas are not necessarily thinking about these issues in terms of market failures. The second, related issue concerns the abuse of competition policy for policy purposes that have little to do with solving market failures. This just comes very often from political pressure; if you’re looking at food price increases or petrol price increases these are things that are driven by long trends, but a lot of pressure is put on authorities to intervene: prices are going up and people feel that must be caused by market power. There are a lot of examples of this. The big danger is that we then get regulation that actually does facilitate the exercise of market power. One of the examples relates to credit rating agencies, where it was said, mostly by politicians, that this was a competition problem: we have oligopolistic markets

Welcome and Panel I 

5  

and so on. If you look at this, there’s an asymmetric information problem; actually, the things that people got most upset about – namely, the ratings on sovereign debt – are not even monetized. So the question is – where is the oligopoly there? These are questions where I think it’s very important to do advocacy on the competition policy side in order to avoid unnecessary or even counter-productive interventions. The third issue that ties these together is: what are the appropriate instruments and regulatory structure to address issues of market power? I think that very often we get to the limits of competition policy instruments; that’s certainly true in the patent world, so I think the question is why we’re going in with a competition policy instrument when potentially the right answer is to do something about the duration of patent protection. The answer to that is that unfortunately some of the instruments that we need – and I mean legislation in order to intervene there – takes so much time that no one believes that it’s actually happening. Then one has to think about the third or fourth best instruments to address these types of problems. Then there’s a question of appropriate regulatory structures. Do enforcement agencies have the right incentives to bring regulatory initiatives into the policy world if those are necessary? And with regard to the proper integration of competition policy and regulatory policy, are we thinking about it in the right way in terms of the complementarity of the instruments to address market power questions? I’ll leave it with those questions, thank you very much for your attention. Juan Delgado: I’m also going to start from a broad policy perspective and ask: what is the objective of state intervention? Ideally, the objective of the intervention is to maximize social welfare, or any kind of welfare. So, paradoxically there are lots of market failures in the market economy and the aim of state intervention should be to correct such market failures through sectoral regulation, competition policies, or many other policies such as innovation, climate, etc. There are of course potential tradeoffs, but we need to keep in mind when analysing those tradeoffs that there is a single objective of social welfare. In finding the right point in the tradeoff we have to keep in mind that we are looking for the combination that maximizes that welfare to which we attach value. In this sense, restrictions of competition might be acceptable only if they are essential to realize some efficiency gains by promoting other policies. But policies are only instruments to get to an objective; they’re not objectives in themselves. We know that competition policy and environmental policies are instruments, but we often make the mistake of thinking that it is the policies that constitute the objectives. For example, if we remove barriers to cross-border competition, this will generally lead to lower prices and price convergence, and this implies greater welfare. But this has not always been the case; sometimes single market policies have been used as an objective and not as an instrument. For example, there are many cases where there has been an imposition of price homogeneity in some sectors, and this does not guarantee lower prices or higher prices. We have lots of examples –

6  Competition, Regulation and Public Policies car prices, iTunes, roaming charges, etc. – where authorities have imposed price homogeneity as a way to implement the single market. This is not the single market but price homogeneity is a consequence of single market policies. It is an objective, but not an instrument in itself. We have to also keep in mind that social welfare maximization strategies are not independent from the economic environment, and the combination of policies that maximize social welfare can vary in different economic environments. For example, guaranteeing stability might require more substantial state intervention under financial distress, and if the economic situation is healthy, active policies aiming at financial stability might be less important. So the optimal policy path should be flexible enough to adapt to changes in the economic environment; but let’s analyse what the role of competition policy is in this context of policies. We all know that competition – the standard textbook approach to competition policy – is the optimal allocation of resources, lower prices, incentives for innovation and growth, etc. And usually a dose of competition is essential in any policy recipe, and the financial crisis is not a good justification to put competition policy aside. We may want to vary the combination but we should not completely forget about competition policy because the way competition acts as a lubricant for the economy is valid both in good times and in bad times. So, as I said before, competition policy should only be relaxed if such relaxation is essential for the implementation of other policies that can make a greater contribution to social welfare. This means that, for example, crisis cartels are usually difficult to justify on an efficiency gains basis, and even a relaxation of the analysis of an abuse of a dominant position is usually difficult to justify under times of economic stress. But what kind of conflicts arise in times of crisis? The typical example is the banking industry; there may sometimes be a tradeoff in that industry between competition policy and financial stability. Banking is a strange animal from a competition point of view; usually in the banking industry there is a reasonable rate of market entry, but market exit is rare. Of course, governments do not like exits in the banking industry because they know what the consequences might be, and they don’t want to face the uncertainty that accompanies exit in that sector. So apart from the typical market failures of asymmetric information, we have a bigger market failure now, which is the failure to provide financial stability, a public good. During the financial crisis we have observed that there have been distortions of competition due to public intervention with the aim of protecting financial stability. Usually these kinds of distortions can technically be put in the state aid basket, with all the bank bailouts, etc. This amounted, of course, to massive public intervention to protect a public good, but there was a clear distortion of competition. There is also the idea that this intervention guarantees short-term financial stability, but not long-term business sustainability; sometimes a state intervention does not guarantee that banks will survive in a competitive market. There’s another kind of intervention or distortion of competition, which is seen in the context of merger policy. In many countries, like in Spain, there has been a lot of market concentration, but this is basically driven by regulators rather

Welcome and Panel I 

7  

than by the market. And these regulators have taken not the level of competition, but the short-term financial health of the industry as the only criterion for this intervention. So nothing guarantees that these mergers would have occurred in a perfectly working market. And even if dirigiste mergers can guarantee short-term financial stability, they do not necessarily guarantee the long-term viability of the business and healthy competition in the future. I will stop here but my point is that competition is an essential piece of the policy map and can act as a lubricant to implement other policies, so we cannot suspend competition policy in these circumstances. If we think about state aid, we have restricted access to capital so this can be incorporated into the private investor test and this is not really an exception to the rules. It may be true that tradeoffs are usually more difficult to analyse in times of financial distress; we have seen the case of the banking industry. It is difficult to compare the metrics of the restriction of competition with the contribution to financial stability. But that does not mean it is judicious to be more lenient as regards restrictions of competition. Lorenzo Coppi: I’ll focus on the background economic point that Philip was making in the beginning. In particular, and echoing Juana a bit I’d like to make a few remarks about whether in times of hardship competition authorities should be more lenient than in normal times. For those of you who, like me, didn’t go to primary or secondary school in the UK, you might be wondering who on earth is Florence Nightingale, the name on my slide. Florence Nightingale is the lady who has been credited with inventing modern nursing. She nursed and healed a lot of British soldiers back into combat at the time of the Crimean war. So the slide raises the question of whether competition authorities should be compassionate care-givers and nurse companies back to health during recessions, like Florence Nightingale, or should they take a more Darwinian approach and just let nature – in this case, economic nature – run its course? I’d like to address this by asking three sub-questions. The first one is: Is there any empirical evidence that competition authorities are more lenient in hard times? The second sub-question is: Should competition authorities be more lenient in hard times? And third, once we answer both of these, what are the practical implications for policy? Now, on the first question, is competition enforcement more lenient during the recessions? If one looks at the literature there a lot of examples that are given for why in times of crisis antitrust comes under attack and is sometimes set aside. The classical examples are all however in the first half of the 20th Century and they all come in times of very serious crises like wars, or in the case of the New Deal there was a complete wholesale change in economic policy. So, are these good examples and can we extrapolate from those? It is true that in those times antitrust policy was set aside but I would put to you that – as bad as the current recession is – it’s just a recession. It’s not like World War Two, so can we really think that that is going to happen here? Well, there are other examples that are more relevant. For instance, in Japan in the 1990s they tried to get rid of over-capacity through agreements within the stricken industries, as is common in Japan; and

8  Competition, Regulation and Public Policies in some of those industries this is actually credited with prolonging the crisis and contributing to the ‘lost decade’. But in reality there is some legal research which suggests that in times of crisis there will be more pressure from industry and therefore from politicians to relax competition policy. It is thus argued that competition policy will tend to be somewhat laxer in times of crisis. However, if we turn away from those examples and look at the actual empirical research, it is not clear that that’s the case. In fact, there is some empirical research that suggests that competition policy is counter-cyclical, meaning that when times are tight there is more enforcement. Now, that’s counter-intuitive and the evidence, I must say, is mixed. Posner, for instance, argues that it’s the other way around. But there is one strong result, I think, which is that the defence of small- and mediumsized enterprises is stronger in times of crisis. For example, there was a noticeable increase in Robinson-Patman cases during times of crisis. These are basically cases concerning small distributors being protected from the big industry bad guys. That is something that happens during recessions, but otherwise is there a link empirically? There the evidence is mixed, but there are some indications that it could be counter-cyclical. There are various explanations; some say there are more violations when economic activity is weaker, but it is difficult to really have an idea because, for instance, for cartels we don’t really know theoretically whether there are more cartels when times are good, or whether cartels break down more when times are good, or when times are bad. It is possible that businesses are weaker and consumers stronger during recessions, and therefore there may be more enforcement. In any case, it’s all very complicated and there isn’t a clear answer. But we can conclude that there is not an indication that authorities behave like Florence Nightingale. There is no compelling empirical indication that competition enforcement is more lenient when times are hard. So let me move to the second question. If competition authorities stay their course when times are hard, is that the right approach? Or should they be more ‘compassionate’? Well, first of all we have to decide whether recessions are bad. If you ask anybody – the man on the street, or politicians especially – they would say it’s like asking whether the earth is round. Recessions are bad, people lose their jobs, suicide rates go up and all these sorts of things, so recessions must be bad. There are also some quasi-economic arguments that say that recessions ‘destroy value’. You have a company which has been building cars for a long time, but then comes the recession, the company goes bust, all this value that has been created by putting skills together is dispersed, it goes to Asia and never comes back when times are good, and that’s value destruction. In fact, it’s unclear from an economic perspective whether recessions are bad. First of all, it’s unclear whether they’re just temporary blips that go away and don’t really affect long-term growth rates, in which case it’s a bad time but it doesn’t really have a long-lasting effect. But also, there is a sense in which they can actually help the economy because there may be periods of creative destruction. This is the time when inefficient firms are supposed to exit the market, and this is necessary for a better reallocation of resources from inefficient firms to the efficient firms that drive growth. There are

Welcome and Panel I 

9  

also some studies that suggest that the wide Europe-US productivity gap is also related to lower entry and exit rates. So, as bad as they may seem, we shouldn’t really fight against recessions. However, it isn’t clear that this process of creative destruction, of inefficient firms exiting the market, is actually stronger during recessions. In fact, there is research that says when times are good, firms enter and exit on a frequent basis. When times are bad, firms don’t exit that much. Why? It’s because firms seem to be able to get by during recessions by just firing the workforce, reducing wages, renegotiating credits, and so on. So actually the observed exit rate is not significantly higher during a recession. Now, what is the implication of this economic research? There is no need to really prevent exit in recession, because there isn’t that much more exit. It would be much better to focus on encouraging entry, because entry rates are reduced in times of recession. This is especially the case with regard to small- and medium-sized enterprises, which would often be a source of innovation but for the downturn. So, let me go to the possible recommendations – and this is where I go a little out on a limb. Let me start with something we all agree upon, I think, which is the implication for cartels. You sometimes hear a compassionate view towards crisis cartels. The argument is that crisis cartels may sometimes be necessary to manage over-capacity and to restructure the industry. Provided they do restructure the industry in an efficient way, maybe they are not such a bad thing. However, many of us take a more Darwinian approach to that: cartels reduce output, and so they deepen recessions – they are not particularly good recession-easers. They do provide incentives to avoid using capacity, but not necessarily to take that capacity out of the market. And as we said, exit rates are not high during recessions, so actually there is no need to have a cartel in order to save firms from going bust. And while we would need something to facilitate entry, cartels are clearly not entry-facilitators! So the bottom line here is that strong cartel enforcement is still needed during recessions. But what about state aid? State aid is trickier, because you could take the compassionate view that a lot state aid is needed. Why? Because it can be outputincreasing. After all, it is just another fiscal policy lever: if you give money to consumers so they can spend more and kick-start growth, you can also give money to companies to invest more and kick-start growth that way; what’s wrong with that? People also say that market failures are more prominent in recessions. Well, I’ve found no proof of that, but there is an intuitive argument behind that – indeed, many say that recessions are market failures. I’m not sure that they qualify as a market failure in technical terms. One potential additional argument for why state aid is needed in times of crisis is that bankruptcy law doesn’t work very well. We don’t have Chapter 11 bankruptcy in Europe, and so we don’t have a good way to restructure companies. We therefore end up liquidating companies much more than we ought to. So, rather than liquidating them we might as well keep them alive with state aid. However, once again the Darwinist approach would say that we don’t have a problem with exit because there is not too much exit during a recession. We shouldn’t help keep firms alive, we should instead focus

10  Competition, Regulation and Public Policies on encouraging entry. How can you do that with state aid? Well you could allow entry-inducing aid, for instance aid for SMEs or R&D, something that helps firms enter the market. This can be especially helpful if there is a market failure in access to capital or something like that. Exit-preventing aid, like rescue restructuring, which is the bread and butter of aid during a recession, is not a good area where policy could be made more lenient. Finally, let me say a word on merger control. This is probably where it gets a little more controversial, but mergers can be an orderly way to restructure and to have a healthy shake-out of inefficient firms from the industry. Especially if an industry is in equilibrium, merger policy can help to promote orderly exit. However, obviously we know that mergers may create market power. And at their worst, mergers may also be a hidden form of crisis cartels – they may be ‘defensive mergers’. So, my conclusions are very simple. First of all, economic distress and recession may cause a national regulator to relax enforcement. Regulators empirically are typically able to withstand the pressure, but some finessing of their policy should probably be considered. This finessing should not occur in relation to cartels or rescue and restructuring aid, which is where a lot of the pressure would probably be. But maybe where aid is entry-inducing, a more lenient policy is not a bad thing. And there is a question mark whether in selected cases – and for full disclosure, I don’t have any client that has such a case at the moment – a merger might actually lead to a positive shake-out in the industry concerned. Thank you. Monti: We now have about half an hour for a discussion. If you’re new to the Workshop, to contribute just lift your flag up and I will try to pick you out in the order in which you’ve lifted them. To give you some time to reflect, let me pick up on some of the points people have made. It strikes me that if someone makes the argument that a recession is creative destruction it’s perhaps economically correct but very difficult to sustain politically. The institutional design of a competition authority with more independence that is allowed to run arguments such as ‘we only base our reasoning on market failure’ may be economically effective but politically very dangerous. So we may be missing a dimension in this discussion, which is the risk of running too much of an economic argument and losing out on the – you might say populist, or you might say democratically legitimate – concerns of individuals. I see Joe Farrell has his flag tilted, would you like to start us off, Joe? Joseph Farrell: Yes, thanks, a couple of comments. First, on Lorenzo’s very nice Darwin/Nightingale personification: I wonder what people think of the biodiversity angle on that? Biodiversity is a policy concern that rests on a belief that biological eco-systems, perhaps particularly when manipulated by humans, are not very good at trading off current competition for long-term competition. In other words, the concern is that current competition may be too strong, particularly to the extent that life forms use humans to give themselves advantages over their rivals,

Welcome and Panel I  11   and that may be bad from the point of view of long-term biological competition. So perhaps we should intervene to protect the options for future competition or long-term competition at the cost of softening current competition. That, I think, underlies some of the proposals for softening ordinary competition policy in crisis times because it is essentially saying that this exit (which from a very broad Schumpeterian perspective might be a good thing) might actually, in an industry where entry is difficult, lead to exit in sort of the same way as predatory pricing might. So I think biodiversity concerns and predatory pricing concerns and crisis exit concerns are all somewhat the same. Second, I think some of the speakers said something I hope they didn’t mean, but could be misinterpreted. We do operate competition policy in a context with other micro-economic policies, and I think there’s been a lot of progress in making competition policy fact-specific and effects-based. But it would be a big mistake if we allowed those things to dissolve the boundaries between competition policy and other policies. Competition policy is something distinctive. If it loses its distinctiveness and its clarity, I think that would be a bad thing. Frédéric Louis: I have two questions for Lorenzo. First, when you say that there is no evidence that there is less exit during a recession, how do you define exit? Because certainly in Belgium, when we have a recession, the media says that there are far more bankruptcies. So for you, bankruptcy is not an exit? I would like to understand what you mean by saying that there is not more exit. Second, you say there isn’t more exit, therefore there shouldn’t be a more flexible application of competition policy. But it may be that the reason why there isn’t more exit is because there is a more flexible application of competition policy, but nobody really admits it. How do you factor that into your assessment? Giorgio Monti: Okay, perhaps given the specificity of the question, Lorenzo, do you want to answer that now? Coppi: Yes, thank you. I have not carried out the research myself but as to your first question, the research is based on U.S. data. It is a relevant question as to whether things in Europe are significantly different, but the research has verified several times that, as far as the US is concerned, the difference in exit rates is not statistically significant. Overall, there is a little bit more exit of small firms and not any difference in large firms. Europe may be significantly different but unfortunately we don’t have research addressing Europe specifically. The second point as to whether there may be an issue that we are not seeing the competition policy being more lenient and therefore we have less exit because of more lenient competition policy or more lenient enforcement, well that is the other part of the research, which again is unfortunately all US-based. But if this US-based research is right, ie, if it is correct that there is no clear story of more lenient application of competition law during recessions, then more lenient enforcement cannot be an explanation for the reduced rate. But again, Europe may be different.

12  Competition, Regulation and Public Policies Monti: Thanks. Maybe at this point I can ask the other panel members to comment before I take other questions. Kühn: Let me remark first on the creative destruction point and then on the role of competition policy. I think Lorenzo is right to say that probably the big factor in promoting growth, at least productivity growth, is the entry-andexit process and the sorting process that goes on. Recent research shows that in Europe generally there is less exit, and the firms that stay in the market are not growing as fast in the US. That is the source for the lagging productivity improvements in Europe relative to the US. Now, that is a matter for long-term policy, but the question is whether this is really something that is decisive in a recession. I would reformulate a little bit what Joe said, because I usually don’t like to go into concepts that I don’t understand like biodiversity. But certainly, in every recession option values of staying in the market are particularly high in economic terms. We also know that, if you have competition, that will eliminate option values from the consideration. So I think there is a good case to be made that there would be excess turnover – too much out and then back in afterwards – so that incentives for smoothing the process when you’re in a recession are there from a social point of view. So whether that really means that competition policy has that much of an impact on the entry and exit process is I think another question, and whether we really have to change the standards, that is another one. Even in cartel policy, where one might be concerned that you’re hitting them so hard that they have to leave the market because of your fine, it doesn’t seem to happen that often really, and there are mechanisms to alleviate that. Second, with regard to competition policy, I fully agree with Joe Farrell that generally in competition policy you want to be very clear that it’s about market power and nothing else but market power. That still gives you two boundaries that are complicated with other policies that are also addressing market power through regulation, and there are conflicts where there are interests other than just how to deal with market power, and you have to balance those. The bigger concern here is – and I think that’s the difference at least at the European level – is that once you are in the business of controlling state aid in industries, you cannot make those clean distinctions. In that context you practically cannot avoid that type of policy conflict Delgado: I just wanted to say that competition policy was one more instrument to increase welfare. I was not denying the role of competition policy in this policy map. In fact, the efficiency effects of competition policy are important in any context. What I was saying is that we have to find the right combination of policy instruments. In the 1990s in Europe we went through a process of deregulation in the energy sector driven by reducing market power. And now we have a new shock, which is climate change, and there are some obligations to use renewable energy sources. So we have a non-market intervention in a market that we have to accommodate in some sense. That is what I was trying

Welcome and Panel I 

13  

to suggest because it seems that we want to be able to combine different policy instruments that aim in the same direction and increase welfare, even if they do so in different ways. Alexander Italianer: I have a question for Juan because he was proposing financial stability and competition policy in the form of state aid, saying financial stability is a public good so if there is no financial stability it’s a market failure and that justifies state aid. But if you look at the financial crisis, I would say that the financial stability issue is more of a fallout of financial instability. The policy answer that is being given, at least in Europe but also in the US with the DoddFrank Act, is more regulation. My question is whether, with hindsight, if you assume that there is a certain complementarity between regulation and competition policy, what would competition policy enforcement need to have been in say the mid-2000s just before the crisis in order to avoid it? In other words, wasn’t the tradeoff at that point in time different from what it is now? Jon Stern: Thank you. I just wanted to put up a marker about trade policy, and I think a lot of the discussion – in particular the last two or three contributions – have brought up the question of the boundaries of competition policy and other policies at a time of high recession. I’m not a competition specialist, but I know that defining what is the relevant market is crucial in competition policy. In a recession a major company may fail, and then there is a diminution of supply for that market, so it is crucial to identify what is the relevant market. In turn this raises the quest of what is the relevant set of producers, and whether the set includes foreign producers, which may bring in trade policy concerns. This is important for defining the scope of competition policy, and that clearly relates to the whole question of state aid, where the pressure to grant and tolerate aid comes in part from the concern that a local producer will go down and people will be dependent on imports. And so I wonder whether in fact the question of the boundary has become more difficult and, if it has, what has been the response. Lowe: My remark comes back to a kind of Monty Python question about what has competition policy done for us. When you get into a recession, what is the fundamental problem? First of all, everyone has the prospect if not the reality of being less prosperous. Some people are unemployed, companies who under boom times look economically sustainable suddenly, when the water level goes down, discover that they are not fundamentally competitive in recessionary periods. The question is, what does competition policy do? Well, we preserve competition, we maintain competition, but we don’t necessarily have the instruments to promote it. We don’t have the capacity to say: right, here we want new entrants who are innovative and can really bring new technologies into this area and make us more competitive. It’s that proactive element which makes competition policy look less relevant for government ministers in a downturn, even though in a period of boom, where people see excessive pricing and the effects of market power they

14  Competition, Regulation and Public Policies may be very happy to see competition policy act to stop exploitation and stop rents being moved from consumers to producers. Then if you extend the boundaries of competition policy to state aid policy, you come to the question of the extent to which state aid policy should be oriented toward new entrants, to innovation and to research and development, or the extent to which state aid may actually help to preserve the competitiveness of those already on the market. And what happens, unfortunately, is that the research and development budgets get cut, at a private level and at a public level, and those are the ones which would promote new entrants and put pressure on incumbents. Then, in the financial sector we can ask ‘what would have been the correct policy in 2000?’ Well, the correct policy would have been to encourage the restructuring of banks which were basically in the long-term unsustainable as business models. Now, what could competition policy have done for us at that point? Well, perhaps it could have allowed mergers apt to eliminate inefficient banks. If you leave what is basically an inefficient sector in certain areas as it is, that may preserve bio-diversity in one sense, but it may lead you into recessionary conditions and politicians may say: this is all systemic, don’t touch it. I know that in Ireland, in Spain, and to a certain extent in Germany, the political pressure to avoid touching banks on the ground that you cannot let these banks exit because they are systemic has been a major cause of some of the sovereign debt problems in Europe. But coming back to the wider point, if you want a competition policy which is proactive then you have to engage in other policies such as research and innovation, reducing barriers to entry and stimulating new technologies. Mario Siragusa: What Philip was saying is of course very right: What are the areas in which competition authorities can be proactive during a recession? They are limited, but maybe at least in the European system we do have a tool that could be useful because the European Commission also has the power to intervene, applying not just the competition rules but also other rules of the Treaty to try to de-block situations which are blocked by national regulation or by limitations on market entry. In Europe we are, unfortunately, still plagued by a number of restrictions, so I think that in a time of recession probably competition authorities and the Commission should step up their roles as guardians of the Treaty and try to intervene in freeing certain economic sectors from limitations and restrictions that may contribute further to the recession. Monti: That comment also applies to national competition authorities because of the powers that have been conferred to them, particularly by the Fiammiferi judgment1 and perhaps the Italian competition authority’s approach is actually moving along those lines of freeing up markets. Which leads me to one comment – if I can abuse my role as Chair – that this makes me think about enforcement tools. If we think about antitrust law maintaining competition rather than 1 Case C-198/01 Consorzio Industrie Fiammiferi (CIF) v Autorità Garante della Concorrenza e del Mercato [2003] ECR I-8055.

Welcome and Panel I 

15  

promoting it, because in a sense the tools are prohibitions (‘you must not merge if it substantially lessens competition’, ‘you must not abuse a dominant position’), then some competition authorities have broader powers, for example to do market investigations that do not require an offence caught by a prohibition but require the identification of a market failure, and then there is a question of how to remedy it. So query whether those kinds of more sophisticated tools, although errorprone because of their breadth, might be given some consideration. A number of national competition laws are currently being reformed, and so it’s timely to consider whether some broader tools might be suitable in a time of recession. But I leave the panellists to conclude the session. Kühn: Philip brought up a lot of points also related to what Alexander said that I think I want to offer a couple of thoughts on. There was this idea that it was more important to stimulate entry than to prevent exit even in bad times, and I think that might actually be a problematic approach. The reason why that might be problematic is that it might not be a very good use of funding. If it’s very difficult to get people into markets because the markets have just collapsed, if you have more concentrated markets and you’re trying to get them to invest in R&D, you’re throwing away a lot of money and you’re still not getting the effects out. So on the other hand, as I said, if there’s something to this idea about option value, that you’re underperforming for a while but you know that you’re going to come out of this then it might actually be better to smooth that process. One example of this is in Germany where you’re subsidizing employment for a while, that worked really well in terms of the pick-up of employment in Germany, and that’s an example of how you deal with this option value type of problem and you’re not severing the firm-specific skills that have been accumulating. So one has to take a lot of care, and in terms of policy intervention, if you’re in a crisis, other measures that you might want to take in order to improve competition – any sell-offs or anything like this – are going to be very hard because you’re going to need to find a buyer and if we’re looking back in a couple of years at the remedies we’ve tried and the pro-competitive remedies over the years of crisis, they were probably much less successful than remedies in other times. That should not be surprising. But I think that’s something we’re going to learn from this crisis. In terms of the financial markets, one has to distinguish two effects that have made the crisis this severe. One is the typical instability of financial markets: if everybody stops lending to everybody else, that’s going to get you to another equilibrium, a bad one. And that gets augmented a lot by asymmetric information problems. That has brought about a particularly severe crisis, and getting out of it is a real problem. The problem that we’re facing is that you have to reform the structures that led to the asymmetric information problems, and that was basically a problem of financial regulation in the 1990s, it was a result of how competition policy was applied. At the same time, you don’t want to be too severe. We need to get the lending mechanism between the banks functioning again. We still have not achieved that. In the end, that is not really a competition policy issue. I think we

16  Competition, Regulation and Public Policies might have avoided some of these things with smaller banks because the implicit subsidy would have been lower, and I think one should be thinking about some kind of an instrument in which one has a systemic element of merger control that might not sit with the competition authority but with the financial supervisor. I think one really has to think about that, and of course that type of system exists in the US. But one has to look at it from that perspective and not just from the market power perspective. Just one more point on market investigations: I think market investigations are very important instruments, but there I want to come back to my point regarding the organizational structure and the context in which these inquiries take place. It’s always a question of what you can do with a market investigation. If you’re structured in such a way that your only way of having a real impact is running cases, then your market investigation has to generate cases. If you have a good instrument by which you do competition advocacy, you can use that too, and it may be equally effective if not more so. The Competition Commission in the UK can actually go farther and ask the question ‘Well, if we’re finding problems, what is the right instrument to address this?’ And very often when we’re looking at markets we’re actually seeing that the underlying problems stem from regulation and from the lack of entry of exit. We see that there’s a competition problem but we don’t have the right instrument to remedy it. And there I think we have to think about the bundling of instruments, that is to say, to use them in a way to maximize their complementarities. Delgado: When I was mentioning competition policy in my presentation, I referred to these two angles that Philip has been discussing: infringements and mergers on one side, but also a proactive promotion of competition. I think if we just focus on antitrust policy we are making a competition policy for incumbents but not for new entrants. And second, I did not mean to say that the financial crisis was a failure of competition policy, but that this public good of financial stability was not properly regulated. One of the reasons was that exit was not allowed in this market, and we didn’t have the Darwinist approach that Lorenzo was describing. But because of the financial crisis there have been state interventions that led to market distortions. Shovelling money into the banking sector could be justified from a financial stability point of view, but it is a market distortion, so this was my point. The long-term answer to the financial crisis is proper regulation, but initially the short-term answers imply competition distortions and we have to be aware that this is going to have a consequence on the future playing field of the industries concerned. Coppi: Let me go back to three main points very quickly because I don’t want to encroach on the next session. The first one is that I was intrigued by Joe’s point about biodiversity. Is bio-diversity something necessarily good per se? For instance, the dodo is extinct and it’s a pity, we would all like to have the dodo around. But the dodo couldn’t fly, couldn’t run, couldn’t do much. Should it really

Welcome and Panel I 

17  

still be around? Wouldn’t necessarily want to have dinosaurs around, I would guess. It’s an intriguing concept but I just don’t have an answer as to whether that kind of biodiversity is necessarily a good thing. On the boundaries of policies, I have always taken the view that, as Joe said, we need to focus on market power because competition policy is about market power. In particular, I’ve made that point several times where there is the interface between antitrust and IP policy, which we will discuss later today. I think that sometimes IP cases should be solved by fixing IP policy as opposed to using competition policy. However, where there is a failure of regulation and where some IP regulators are actually calling for competition policy to help them to sort out a problem that is seemingly untreatable for them, then I am starting to change my position a little bit and seeing how there may be a regulatory void. In that case, competition, blunt though it may be, is the only tool that we have. This takes me to Alexander’s point on the financial sector. Even there, I agree with the point that everybody has made on what competition could have done. The answer is: probably nothing. It was a regulatory failure. But that’s where competition policy in the form of state aid has tried to step in and fill the void of not having a European system for winding up failed financial institutions. So, sometimes I agree with you that it’s a blunt tool; you lose focus if you use that tool. But it might be the only tool you have.

Giorgio Monti*

Regulatory Holidays in Utilities Regulation and EU Competition Law: A Case Study on the Role of Efficiency Considerations in Economic Law 1. Introduction Stimulating investment is important in general, and in network industries in particular. We rely on market mechanisms or subsidies to achieve this, we do not (usually) require firms to invest. In many situations, competitive markets are said to provide a prize for innovation: sales move to those that innovate and these sales are the reward for the investment. This prize constitutes an incentive to invest when the profits from the investment are greater than the profits the firm would have made had it not made the investment.1 In cases where the investment results in a public good, we use intellectual property rights as a stimulus: the inventor has the exclusive right to enjoy the economic returns of its invention.2 Those who take this view normally conclude that the application of competition law does not need adjustment: application of a short run consumer welfare standard ensures both static and dynamic efficiencies because the source of innovation is the competitive market. In newly liberalised sectors like gas and electricity, the policy of the EU is to facilitate entry of new competitors in markets that had been vertically integrated and under the control of a state-supported firm for several decades. Here too the view is that increasing competition is said to stimulate both static and dynamic efficiency. On the other hand, by imposing obligations on the former monopolist (ie, the incumbent) to facilitate market access of competitors, that firm’s incentives to innovate are dented.3 In cases where the incumbent owns infrastructure which is essential for that market to function, removing the incumbent’s incentives completely is an unwise move. In this scenario a tradeoff is required: fewer short term gains in competition in exchange for more investment to achieve long * Professor of Competition Law, European University Institute, Florence. 1 Richard Gilbert, ‘Looking for Mr Schumpeter: Where Are We in the Competition-Innovation Debate?’, in Adam Jaffe, Josh Lerner and Scott Stern (eds), Innovation Policy and the Economy, Vol. 6, MIT Press, 2006. 2 It is beyond the scope of this chapter to explore the link between IP protection and dynamic efficiency. For discussion, see, eg, John Vickers, ‘Competition Policy and Property Rights’ 120 Economic Journal 375 (2010) and James Bessen and Eric Maskin, ‘Sequential Innovation, Patents and Imitation’, 40 RAND Journal of Economics 611 (2009). 3 Jan Bouckaert, Theon van Dijk and Frank Verboven, ‘Access Regulation, Competition, and Broadband Penetration: An International Study’, 43 Telecommunications Policy 661 (2010).

56  Competition, Regulation and Public Policies term benefits for everyone. In this chapter, I note that regulatory holidays are one plausible means of stimulating investment by the incumbent. A regulatory holiday entails the disapplication of certain regulatory (and antitrust) obligations on the party who carries out certain types of investment. It generates rewards that can stimulate investment.4 In this chapter I address two related questions – first how far regulatory holidays exist in EU law. Finding that there is no scope for such a measure, I turn to consider why this is so and find that for a number of legal and policy reasons the legal system does not allow one to trade off, in an explicit manner, short term losses in allocative efficiencies for greater gains in dynamic efficiencies. In an era where antitrust law is dominated by economic analysis dictating an effects-based approach, with an emphasis on case-by-case examination of markets, the unwillingness of the legal systems and of policymakers to engage in express balancing of costs and benefits of market conduct is puzzling and warrants explanation. The chapter is structured as follows. Section 2 contains an explanation of the notion of regulatory holiday. Section 3 is a review of the rejection of regulatory holidays in the field of electronic communications and section 4 is an account of their apparent acceptance in energy markets. However, the difference in approach between the two sectors is not as marked as it seems – the bottom line is that a pure regulatory holiday is not available under either regulatory framework. This leads to question whether under EU economic law there can ever be a tradeoff between dynamic and static efficiency, or between long term gains and short term losses. This is the question discussed in section 5. Here, I suggest that the approach in utilities duplicates that found in competition law and in this sphere we find a variety of reasons why tradeoffs are not made explicitly. The implications of this insight are explored in the final two sections.

2. Regulatory holidays Optimal conditions for innovation are unknown and vary from industry to industry and across time.5 Add to this the poor choices government tends to make when exercising industrial policy options (with exceptions of course) and you have a scenario where the right regulatory choice to stimulate innovation is difficult to make.6 A variety of mechanisms are used to stimulate innovation. 4 This is one policy tool among others, but the aim here is not to examine whether it is in each case the best policy tool. For discussion, see Rainer Nitsche and Lars Wiethaus, ‘Access Regulation and Investment in Next Generation Networks – A Ranking of Regulatory Regimes’, 29 International Journal of Industrial Organization 263 (2011). 5 The most comprehensive review of the literature is found in Gilbert, cited above note 1. 6 For successful examples of industrial policy in South Korea, see Ha-Joon Chang, 23 Things They Don’t Tell You About Capitalism, Bloomsbury Press, 2010, pp 125 et seq. For suggestions on how to best target state aid to stimulate innovation, see Philippe Aghion, Julian Boulanger and Elie Cohen, ‘Rethinking Industrial Policy’, Bruegel Policy Brief No 2011/04.

Giorgio Monti 

57  

In some contexts, opening markets to competition for instance by unbundling vertically integrated energy or telecoms markets, is said to invite new entrants eager to introduce new products and working methods.7 Furthermore, some have espoused a theory that in certain markets there is an investment ladder that can be designed, whereby one can facilitate new entry at the downstream levels of the market first, by affording the new entrant access to upstream facilities, and subsequently the new entrant will invest further upstream, reducing the firm’s dependence on the incumbent who, at the start was the sole, vertically integrated, actor on the market.8 For others, too much competition may be undesirable if one is concerned about dynamic efficiency.9 Government can stimulate investment through well-targeted state aid.10 However in times of austerity this option is not perhaps as widely available as in times of plenty. In regulated sectors, innovation is stimulated by giving the regulated firms appropriate rates of return. In the language of the British energy regulator (Ofgem), this is referred to as RIIO: Revenue set to deliver strong Incentives, Innovation and Outputs.11 This can be achieved by increasing the regulated rates set by the firm, but the problems associated with this option are economic (to determine by how much, and for how long, to increase the firm’s rates of return so as to ensure there is an incentive to innovate) and political (avoiding the risk of the public objects to the regulated firm’s extra profits).12 We might label this option an incomplete regulatory holiday. Usually regulation truncates the revenues that the firm could earn, while a partial regulatory holiday (like a relaxation of the rate of return) allows higher revenue to pay for the investment. A proper regulatory holiday instead, would excuse the investor from all obligations, thus removing any price controls or any duty to give access to the product or service that has been generated as a result of the investment. It is said that this regulatory approach removes the hold-up risks resulting from and uncertain or aggressive regulatory scheme.13 Vickers, ‘Competition Policy and Property Rights’, cited above note 2. This approach is normally associated with the work of Martin Cave, ‘Encouraging Infrastructure Competition via a Ladder of Investment in Telecommunications’ 30 Telecommunications Policy 223 (2006). 9 See generally Philippe Aghion et al, ‘Competition and Innovation – An Inverted U Relationship’, 120 Quarterly Journal of Economics 701 (2005). 10 This is, increasingly, the approach of the Commission. See Commission, State Aid Modernisation, COM(2012)209 final. 11 OFGEM, RIIO: A New Way to Regulate Energy Networks (October 2010) 12 See further Robert Baldwin, Martin Cave and Martin Lodge Understanding Regulation: Theory, Strategy and Practice (2nd edn, Oxford University Press, 2012) chapter 26, where the authors identify various means by which the correct revenue may be generated. 13 See generally Joshua Gans and Stephen King, ‘Access Holidays and the Timing of Infrastructure Investment’, 80 Economic Record 89 (2004). This is a fairly formal analysis; a more accessible paper with a critique is found in Aldo Spanjer, ‘Do Article 22 exemptions adequately stimulate investment in European gas markets?’, (2008) Zeitschrift für Energiewirtschaft 46. The implied suggestion in certain Opinions of Advocates General in refusal to deal cases appears to be that the right to refuse to deal is the reward for the innovation. See, eg, the Opinion of Advocate General Maduro in Case C-109/03 KPN Telecom BV v Onafhankelijke Post en Telecommunicatie Autoriteit (OPTA) [2004] ECR I-11273, para 39. 7 8

58  Competition, Regulation and Public Policies One might argue that a complete regulatory holiday is too blunt a tool. Instead the partial holiday whereby price regulation is relaxed is a more refined option which yields both investment incentives and competition. However the cost of finding the right access price, and reviewing for how long this higher price should be charged for (and the procedures for monitoring this in the absence of a sector specific regulator) may well outweigh the benefits of greater precision.14 A regulatory holiday should not be confused with the grant of an intellectual property right. The award of a patent is the creation of a property right and does not create immunity from other rules of law, on the contrary owners of intellectual property rights holders have obligations. Granted, the economic considerations are similar: the patent is a reward to stimulate investment, but the tradeoffs made by the rules of intellectual property between granting rights to stimulate investment and limiting those rights to prevent inefficient exclusion are outside the scope of this chapter.15

3. Electronic communications In this market, the question might be put whether an access holiday might be considered as a proper strategy when a firm makes a large investment in New Generation Access technology. The Commission’s relevant soft law notices appear to exclude this possibility. The judgment in Commission v Germany also suggests that regulatory holidays are an inappropriate tool in this sector.16 The judgment was a challenge against the German legislation transposing the EU’s regulatory framework and exempting new markets from the regulatory framework. The Commission alleged that the directives did not afford a Member State this option because, inter alia, there was no hierarchy in the regulatory objectives stipulated therein (eg ensuring consumer welfare, avoiding restrictions of competition, encouraging efficient investment and ensuring access to universal services).17 The Court of Justice largely agreed with the Commission’s allegations, ruling that the ranking of regulatory objectives was a matter for the national regulatory authority, which is to be independent from state control, but dependent 14 For discussion of the relative merits of these two approaches, see Gans and King, ‘Access Holidays and the Timing of Infrastructure Investment’, cited previous footnote. 15 See, generally, Ronald Cass and Keith Hylton, Laws of Creation – Property Rights in The World of Ideas (Harvard University Press, 2013). 16 Case C-424/07 Commission v Germany [2009] ECR I-11431 17 Directive 2002/21/EC on a common regulatory framework for electronic communications networks and services (Framework Directive), 2002 OJ L108/33, Article 8. The Framework Directive was modified by Directive 2009/140/EC amending Directives 2002/21/EC on a common regulatory framework for electronic communications networks and services, 2002/19/EC on access to, and interconnection of, electronic communications networks and associated facilities, and 2002/20/EC on the authorisation of electronic communications networks and services, 2009 OJ L337/37. The amended Directive retains a number of multiple objectives in Article 8.

Giorgio Monti 

59  

on the Commission’s policy line.18 While this might appear to leave some wriggle room for the regulator to impose a regulatory holiday, this is not so in practice: (i) the regulator must first identify whether a firm or firms hold significant market power, and if this is answered in the affirmative, then it must regulate the market to mitigate the adverse effects of market power, so some regulation is inevitable upon proof of market power; (ii) if the Commission’s guiding hand governs, then that institution’s reluctance towards allowing regulatory holidays would limit the regulators’ option. This was precisely why the German legislator had tried to create space for regulatory holidays by legislative fiat: the regulator would be unable to exercise this policy choice under the legal framework. Thus, a plausible criticism of the judgment and of the Advocate General’s Opinion is that by relying simply on constitutional niceties, they ignored the reality of the institutional makeup: true regulators are supposed to be independent of their governments, and they are best placed to articulate the appropriate policy for the relevant market that they have the competence to review. But if this seemingly legitimate institutional makeup is then capped by the Commission restricting policy options, then the regulator’s autonomy to balance the various goals is undermined, so the effectiveness of the policy takes second place to the harmonisation of national regulatory policies by the Commission. One might even respond to the Court’s legalistic argument with another: that the state must reserve the right to make tough political calls on matters of general interest and cannot be said to have delegated too much policy discretion on the regulator.19 The emphasis on uniformity is at odds with the views of some that consider the EU system as being particularly well placed to afford possibilities of experimentalist governance. That is to say, informed divergence by national regulators or competition authorities, with the option of sharing experiences with different regulatory models.20 This approach seems to be lacking in the sector of electronic communications where one Member State’s experiment with a specific form of regulation is curtailed by the supremacy of EU law. At the same time, the Commission is aware that the large scale investments are only likely to be carried out by large firms, possibly holding market power in other electronic communications markets, so what alternatives are there to a holiday are provided? The preferred method is the award of a risk premium.21 It means that the developer of a new facility must give access to third parties but that he may be allowed to charge a higher rate to recoup the costs of investment. However, the Commission has had to refine the original notice to provide greater clarity and Commission v Germany, cited above note 16, at paras 53–61 and 91–93. See Marek Szydlo, ‘The Promotion of Investments in New Markets in Electronic Communications and the Role of National Regulatory Authorities after Commission v Germany’, 60 International and Comparative Law Quarterly 533 (2011). 20 On this approach, see especially Charles Sabel and Jonathan Zeitlin, Experimentalist Governance in the European Union (Oxford University Press, 2010). 21 See Commission Recommendation of 20 September 2010 on regulated access to Next Generation Access Networks (NGA Recommendation), 2010 OJ L251/35, para 25; and Case UK/2010/1064-1065 SG-Greffe (2010) D/7658 of July 1, 2010. 18 19

60  Competition, Regulation and Public Policies better incentives.22 Moreover, the approach of a regulator would be to allow the firm to set higher access prices for an undefined period: regular reviews would then determine when the relevant costs were recouped and then the risk premium can no longer be charged. This requires a lot of work form the regulator, and query whether this is really an incentive scheme: would firms not wish for a reward that went beyond the cost of doing the work? This ultimately comes down to whether we prefer over-enforcement of the regulator’s obligation (the possible effect of the Commission’s position) or under enforcement (the stance taken by those that support a regulatory holiday). A second option is co-investment: a scenario where several firms pool their resources to develop the new infrastructure. If one has substantial market power, then regulation is avoided if it has enough co-investors that the regulator may conclude that there is no risk to competition.23 This scheme has proven popular although in some instances concerns over market power have been aired by regulators or addressed by the parties themselves. So there is no regulatory holiday here either unless the investment group is so large that extra rents from market power cannot be earned.24

4. Energy The regulatory obligation in energy sector that we are mostly concerned with here is third party access to transmission/distribution infrastructures. Access rules for competitors are found in the relevant legislation and may also be grounded on competition law. These stimulate competition but may undermine the incentive to invest, so the Commission has designed a policy to try and balance competition and investment by providing for exemptions from the duty to provide third party access in cases where new infrastructure is constructed. This is a form of regulatory holiday, but as noted below, while the application of this exemption has been carried out with increased sophistication, one cannot call this procedure a regulatory holiday. The procedures for exemption are similar in gas and electricity: first the National Regulatory Authority (‘NRA’) reviews the request for exemption. If its opinion is positive decision the request is then reviewed by the Commission, and three options are open to it: consent to the decision, require amendments, or require the exemption to be withdrawn. The criteria for exemption are broadly similar 22 Commission Recommendation on consistent non-discrimination obligations and costing methodologies to promote competition and enhance the broadband investment environment, C(2013) 5761 final. 23 NGA Recommendation, cited above note 21, para 28. 24 For a discussion of the challenges of these two types of regulatory modality, see Marc Bourreau, Carlo Cambini and Steffen Hoernig, ‘Ex Ante Regulation and Co-investment in the Transition to Next Generation Access’, 36 Telecommunications Policy 399 (2012).

Giorgio Monti 

61  

in electricity and gas.25 There is a consensus that the Commission has tightened up the scrutiny of these exemptions, and a concern that the regime is legally uncertain.26 However the focus of inquiry for present purposes is to consider what form of regulatory holiday is afforded. In outline, exemptions from third party access obligations may be granted if the following, cumulative, conditions are met:27 • • • • •

The investment must enhance competition The level of risk attached to the investment is such that the investment would not take place unless an exemption was granted; The facility must be owned by a natural or legal person which is separate at least in terms of its legal form from the system operators in whose systems that infrastructure will be built; Charges are levied on users of that infrastructure; The exemption is not detrimental to competition or the effective functioning of the internal electricity or gas market, or the efficient functioning of the regulated system to which the infrastructure is connected.

It is said that there is a reluctance to award exemptions when a dominant player is the would-be beneficiary because this would indicate that the competitive balance is negative.28 However, the Commission’s more recent communication suggests a willingness to exempt on the grounds that a dominant generator cannot exploit all of the new investment.29 For present purposes the key question is how the first and the final condition are managed: investment must enhance competition but the exemption must not be detrimental to competition. Some general remarks are in order first. The Commission takes the view that the first condition requires a balancing of the pro- and anticompetitive effects of the investment, and that this analysis can be approached by analogy with antitrust law – indeed in the more recent cases the decisions make explicit reference to antitrust case law. The final condition instead looks at other negative effects that arise from the exemption: the effect that the exemption may have on the project under scrutiny or on other projects. It should be noted also that of all the conditions, the Commission has expressed the vast majority of its concerns to the impact of the investment on 25 One difference is that, in gas, an exemption requires proof that the infrastructure contributes to security of supply. 26 It was in response to this uncertainty that the Commission published its Staff working document on Article 22 of Directive 2003/55/EC concerning common rules for the internal market in natural gas and Article 7 of Regulation (EC) No 1228/2003 on conditions for access to the network for crossborder exchanges in Electricity, SEC(2009)642 final (‘Exemption working document’). 27 Regulation 714/2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003, 2009 OJ L211/15, Article 17; Directive 2009/73 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC, 2009 OJ L211/94, Article 36. 28 Adrien de Hautecloque and Vincent Rious, ‘Reconsidering the European Regulation of Merchant Transmission Investment in Light of the Third Energy Package: The Role of Dominant Generators’, 39 Energy Policy 7068 (2011). In fact, much of this paper seems to pre-date the working document cited above, which addresses some of the concerns identified by the authors. 29 Exemption working document, cited above note 26, para 34.

62  Competition, Regulation and Public Policies competition. That is to say, it was uncontroversial that the risk of the investment was such that an exemption was necessary.

4.1 Electricity: merchant interconnectors There are four decisions in this market, and in three of these the Commission has intervened because of concerns that the exemption would risk harming competition.30 However, some of the reasoning is a little difficult to discern and is a bit reminiscent of Article 101 TFEU exemption decisions where the same facts led the Commission to say that agreements restricted and enhanced competition at the same time.31 This is exemplified by the approach in BritNed. It is a project to build the first interconnector between the UK and The Netherlands, and in the first part of its review the Commission noted the pro-competitive impact of this new facility, but then it moved on to observe that the parties might well have constructed a facility with suboptimal capacity which would allow it to charge higher prices than if there were more capacity. As a result the 25 year exemption is subjected to a review after 10 years when the rate of return on the project is evaluated. If this is one per cent higher than that which had been estimated, then Britned has two choices: (i) increase interconnector capacity; or (ii) accept that the profits exceeding the estimated rate of return are capped and used to finance the regulated asset base in the UK and the Netherlands.32 However, is the analysis here weakened by what one might call the O2 fallacy? It will be recalled that this was a roaming agreement to facilitate a new entrant on the German telecoms market, which the Commission had condemned under Article 101(1) because roaming, as opposed to the new entrant devising its own facilities, would allow the parties to coordinate prices. The General Court rightly noted that absent the roaming agreement, new entry would not have occurred, so the agreement did not reduce competition, but enhanced it.33 Likewise in BritNed, absent the investment, there would have been no interconnector between the two markets, so while the firm would likely under-invest to reap higher returns, is the market not better off with a sub optimal investment than with no investment at all? For surely, one of the risks of not granting a regulatory holiday is that future would be investors are put off making other investments. 30 These decisions may be found here: http://ec.europa.eu/energy/infrastructure/exemptions/ exemptions_en.htm. 31 The exemption granted by the Commission in the TPS decision is illustrative. In this case the Commission began its analysis under Article 101(1) and found that certain agreements in a joint venture designed to introduce a new player on the French pay-TV market would restrict competition. However, under its Article 101(3) assessment in the same case and on the same facts it then found that those restrictions ‘far from eliminating competition, the TPS agreements are pro-competitive.’ TPS, 1999 OJ L90/6, para 135. This analytical stance was approved by the then-Court of First Instance. See Case T-112/99 Métropole télévision (M6), Suez-Lyonnaise des eaux, France Télécom and Télévision française 1 SA (TF1) v Commission [2001] ECR II-2459. 32 BritNED (UK/NL) – CAB D(2007) 1258, http://ec.europa.eu/energy/infrastructure/exemptions/ doc/doc/electricity/2007_britned_decision_en.pdf. 33 Case T-328/03 O2 (Germany) GmbH & Co. OHG v Commission [2006] ECR II-1231.

Giorgio Monti 

63  

In subsequent cases the Commission has also been cautious. For example, in Arnoldstein-Tarvisio, a project to increase interconnection capacity between Austria and Italy. The NRAs had agreed an exemption for 16 years, subject to an annual review so that if the investment costs were recouped earlier, the regulatory holiday would be cut short. This approach probably learns the lessons of the BritNed case, and avoids the penalty of additional regulation that was imposed there. Nevertheless, the Commission considered that an exemption of 50 per cent of the facilities from third party access was not warranted because it was not clear how this would contribute to the investors recouping their costs, since auctioning such capacity would be sufficient to manage the risk inherent in the investment. This effectively removes the benefit of the exemption, which was still formally granted. Moreover, the exemption expires in five years, should the project not start by then. This seems to be designed to avoid a scenario where an investment is announced to deter others from entering the same market.34 These decisions address some of the risks that can arise from parties applying for regulatory holidays: for example that a new investor under-invests to maximise returns, or a new investor announces an investment to deter competition. From this perspective, the Commission’s oversight is quite astute. However, from another perspective the approach taken to these exemptions reveals certain problematic features. First, in neither case is the full regulatory holiday granted, because the risk that competition (understood in terms of allocative or productive efficiency) may be suboptimal even if the investment itself managed to connect two separate geographical markets. A pure regulatory holiday, where dynamic efficiency considerations trump competition concerns, is not to be found. Second, in the electricity markets, the investors are mostly transmission system operators, who already have statutory duties to carry out similar investment, so the regulatory framework appears to merely open a second means for TSOs to invest, designed to address high risk scenarios where the sunk costs are high and long term contracts with suppliers are perceived to be the best way to hedge against those risks.35 Finally, given the need for this kind of investment it is legitimate to suggest that a more relaxed attitude to exemptions would be beneficial.36

4.2 Gas infrastructure At the time of writing, thirteen exemptions have been granted for the construction of liquid natural gas (LNG) regasification terminals.37 The exemptions allow the Arnoldstein/Tarvisio (AT/IT) – SG D(2010)16980 and Ares (2011) 42548. See Exemption working document, cited above note 26, paras 40–43. Michael Cuomo and Jean-Michel Glachant, ‘EU Electricity Interconnector Policy: Shedding Some Light on the European Commission’s Approach to Exemptions’, Florence School of Regulation Policy Brief No 2012/06 (June 2012). 37 These decisions may be found here: http://ec.europa.eu/energy/infrastructure/exemptions/ exemptions_en.htm. 34 35 36

64  Competition, Regulation and Public Policies owner to reserve some of the capacity for a period of between ten and twenty years, and to allocate this to interested customers. These long term arrangements give the builders the security they require to sink significant costs. However, it is not clear whether these exemptions may really be called regulatory holidays. First, it appears that all of the parties seeking exemption did not hold a position of significant market power, and that the facilities they were building would normally accommodate new entry. For example, if we look at the four exemptions for the UK market, it appears that after the latest round of construction there would be between 11 and 13 new entrants, so in fact the exemptions serve to increase competition, with little anticompetitive risk, or any other risk to the liberalisation project.38 Similarly, if we consider the four exemptions granted for projects in Italy and the three granted for projects in The Netherlands, it transpires that all would serve to challenge the incumbent in the wholesale market (ENI and Gas Terra, respectively) and so again, the exemption stimulates competition by ensuring new entry.39 Therefore, these are not instances where we suppress competition for an improved development of the relevant markets – no real balancing is carried out.40 Second, the exemption in some of the decisions is conditional on the parties taking steps to avoid the risk of anticompetitive effects. For example some exemptions forbid the dominant gas supplier from holding more than a given share of the new capacity, so as to ensure that there is a credible challenge to the incumbent.41 But there are even cases where such anti-hoarding provisions are imposed even when the relevant market appears competitive so that the risks of anticompetitive effects appear low.42 In many cases the exemption expires if construction does not start after a given time period, which is designed to prevent the hoarding of an exemption, which would serve to raise entry barriers to new investment. Surprisingly this condition is found even in cases where the investors have low market power so that it is not clear what the commercial incentive is to act strategically to exclude other LNG terminals.43 The Commission’s decisions are quite brief, and it would be helpful if it were made more apparent how real the competition risks are to warrant these conditions. There have also been thirteen exemptions relating to the construction of six pipelines across Member States. Seven decisions address the ill-fated Nabucco pipeline.44 As with LNG storage decisions, all of the pipelines are judged to improve See, for example, the discussion in National Grid Grain LNG (UK) – C(2013) 3443. See, for example, LNG Porto Empedocle (IT) – C (2012) 3123 and LNG Eemshaven (NL) – C (2009) 4006. 40 This is contrary to what the Commission and most commentators appear to assume should be happening. See, eg, Exemption working document, cite above note 26, para 11; Kim Talus, EU Energy Law and Policy: A Critical Account (Oxford University Press, 2013) p 97. 41 See, eg, LNG Eemshaven (NL) – C (2009) 4006. 42 National Grid Grain LNG (UK) – C(2013) 3443. 43 See, eg, LNG Livorno (IT) – C (2009) 10172. 44 At the time of writing it appears that this project will not go forward, and that the TAP pipeline will be the new route for gas imported from Azerbaijan. See ‘EU-backed Nabucco Project ‘Over’ after Rival Pipeline Wins Azeri Gas Bid’, Euractiv (27 June 2013). For background on the many facets of this pipeline, see Aleksei Tarasov, ‘The Making of Empires: Russia’s Gas-exporting Pipelines v Nabucco’, 4 Journal of World Energy Law and Business 77 (2011). 38 39

Giorgio Monti 

65  

competition because they will likely increase gas flows from parties other than those that hold significant market power in the Member States concerned. Indeed, the most common concern of the Commission is that the newly built capacity should not go to the dominant players in such a way that they expand their market share or foreclose competition, so capacity caps are imposed to avoid this risk.45 Furthermore, the Commission insists that capacity increases over the time of the project are made more rapidly than the parties and the NRAs had proposed as a way of stimulating competition further. Thus again, most of these decisions do not really provide regulatory holidays but modify the regulatory framework in such a way as to maximise the gains of the new investment – no balancing of positive and negative effects is carried out. To this there is one exception in the decision relating to the Gazelle pipeline.46 This project would connect, via the Czech Republic, two areas of Germany. The Commission foresaw a risk that this new pipeline might lead to other pipelines being under-utilised with resulting higher costs for those using that other pipeline, however the Commission noted that the tariffs for this pipeline were regulated for a long period of time and that the freed up capacity might be taken by other new entrants. Thus, any possible price increase would be small, while the benefits resulting from the exemption would be significant.47 This is as close as one gets in the decisions to a balance between the costs and benefits. Finally, it is worth noting the Commission’s position in its working paper offering guidance to national regulators. Here the Commission’s view is that if the infrastructure has the qualities of a natural monopoly then this weakens the case for exemption.48 However, this is precisely when the balance would kick in: allowing an investor to secure a monopoly would be the incentive to construct the relevant infrastructure and one would then be called upon to balance the increase in investment with the diminution of allocative efficiency, but the steer from the Commission is not to exempt in these circumstances. In sum then, as for electricity markets, the exemption does not serve as a regulatory holiday: rather it removes a regulatory obligation in those cases where that obligation would be counterproductive to the aims of the energy directives. This is not to be scoffed at, because a flexible application of the regulatory framework that accelerates the achievements of the Union’s policy is a wise move, and the Commission’s increased sophistication in the decisions illustrates a careful assessment of the competitive risks.49 The concern for the purposes of this chapter is that the exemption procedure is not about balancing risks to competition and increases in investment. 45 This is found in all the Nabucco decisions (eg Nabucco – HU – C (2009) 3034; SG Greffe D(2009) 2298); and in most other decisions (see, eg, Trans Adriatic Pipeline – C(2013) 2949). 46 Gazelle (CZ/DE) – C (2011) 3424. 47 See Exemption working document, cited above note 26, paras 67–69. 48 Ibid., page 8, Box 2. 49 Thus, while the first few decisions are quite terse, the Commission issued the Exemption Working Paper in 2009, and some of the more recent decisions structure the competition analysis closely, looking for example at the incentives and capacity of the project participants to foreclose the market.

66  Competition, Regulation and Public Policies

5. Law’s problem with efficiency as a defence The aim here is not to advocate that the Commission should be more willing to grant regulatory holidays, nor to argue that the strict approach in the energy exemption procedure is in line with the legislative norm.50 Rather, the focus is on how efficiency arguments are managed by looking at the way regulatory holidays are discussed and applied. Regulatory holidays are a very extreme form of the efficiency defence: support dynamic efficiency for long term gains and sacrifice competition/allocative efficiency in the short term in exchange for increased consumer welfare in the long term. They remind one of the stark Williamson tradeoff: allow a merger to monopoly when the productive efficiencies of the merger outweigh its allocative inefficiencies.51 In the utilities sector, the argument for regulatory holidays is relatively stronger than in other markets. However, as we have suggested above, no application of a proper regulatory holiday may be found: the exemptions in energy markets merely correct the blunt tools of the regulatory framework, which is a different kind of regulatory technique. The evidence in sections 3 and 4 above suggests that the regulators are unwilling to make this kind of tradeoff. Having identified this reticence the question is why tradeoffs are so difficult to find, and to address this question we can shift our focus away from regulatory holidays specifically, to the issue of whether efficiencies may be a defence. This is justified by the increased alignment of regulation and antitrust in the regulation of energy and electronic communications markets.

5.1 Aligning regulation with antitrust A number of the practices engaged in by energy firms considered in section 4 would also likely be considered under the antitrust rules. Given that the antitrust rules are not displaced by the application of sector-specific regulation, this means that the grant of a regulatory holiday under the energy legislation could risk being frustrated by the application of antitrust law to compel the parties to deal. The means by which this is avoided is by aligning the conditions for granting an access holiday with the conditions for exemption under Article 101(3) TFEU, which are also, as far as the case law has developed on this point, the same conditions that apply to establish that conduct is not in breach of Article 102.52 In parsing the 50 The latter view is taken in Tjarda van der Vijver, ‘Third Party Access Exemption Policy in the EU Gas and Electricity Sectors: Finding the Right Balance between Competition and Investments’, in Martha Roggenkamp et al (eds), Energy Networks and the Law (Oxford University Press, 2012), on the basis that exemptions should be construed strictly. This is a fair point, but it fails to explain why the precise form of strictness we see in the decisions is appropriate. 51 Oliver Williamson, ‘Economies as an Antitrust Defense: The Welfare Tradeoffs’, 58 American Economic Review 18 (1968). 52 The efficiency defence in Article 102 is considered in particular in Case C-95/04 P British Airways plc v Commission [2007] ECR I-2331; Case C-209/10 Post Danmark A/S v Konkurrencerådet [2012] ECR I-000.

Giorgio Monti 

67  

exemption decisions it is evident that as the Commission’s analytical approach becomes more detailed, it draws increasingly upon antitrust case law and methods. From this perspective, two observations arise. First, it made sense for the legislator to avoid designing a regulatory holiday under the energy legislative framework which would be frustrated by the application of competition law. Otherwise the beneficiaries of the exemption would fear an antitrust challenge by a disappointed third party and this would wreck the finality necessary for the investment to commence. Second, this means that to answer the question of why regulatory holidays are not tolerated, we can to look to antitrust law for an answer, given that there too, the efficiency defence has been explored at length.

5.2 The efficiency defence in competition law The application of an explicit efficiency defence is a rarity in antitrust law. That is to say, a scenario where an agreement restrictive of competition is saved because the efficiencies that result are greater than the inefficiencies resulting from the restriction of competition is hard to find. There are four interlinked reasons for this, which also serve to explain the reluctance to tolerate regulatory holidays.

5.2.1 Procedural limits A first remark relates to procedural considerations, which are specific to antitrust.53 Under the old procedural rules the Commission had a monopoly in the grant of an exemption under Article 101(3). While it was probably possible for national competition authorities (NCAs) to decide that a practice was objectively justified and thus did not breach Article 102, under Regulation 1 (which was supposed to facilitate decentralised enforcement) NCAs lose even the power to declare that Article 102 is not infringed.54 It means that NCAs cannot declare, by a decision, that an agreement is justified because of efficiencies. Moreover, the Commission’s power to make such declarations is restricted to situations where the criteria in Article 10 of Regulation 1 are met, that the public interest of the EU requires such a decision.55 Since 2004 no such decisions have been reached. This leaves the national courts as the sole players who may declare agreements exempt if efficiencies are demonstrated. From the perspective of a would-be investor this framework is problematic because no reassurance is available from the authorities, save in the form of a commitment decision, whereby both the Commission and the NCA may authorise 53 In merger cases and in exemptions in energy markets there is a notification and exemption procedure that ameliorates the concerns expressed in this section. 54 Case C-375/09 Prezes Urzędu Ochrony Konkurencji i Konsumentów v Tele2 Polska sp. z o.o., devenue Netia SA [2011] ECR I-3055 55 Regulation 1/2003 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, 2003 OJ L1/1.

68  Competition, Regulation and Public Policies an agreement that yields efficiencies. The disadvantage of this approach however is that the commitment decisions do not, so far, articulate clearly the theory of harm, nor the reasoning for accepting the efficiency defence, and furthermore there appears to be only one such decision where it appears that an analysis under Article 101(3) was in fact carried out.56 This makes the law less certain for other investors looking for guidance. As we saw above, this problem is ameliorated in the energy exemption procedure where an exemption is granted. There is a second procedural disadvantage in antitrust cases: under Regulation 17, when the Commission explored the application of Article 101(3) it would itself look for evidence to support the possibility of this provision applying. This was made clear in both the seminal Consten & Grundig judgment and in one of the last cases under Regulation 17, GlaxoSmithKline.57 In GSK the Court recalled that under the exemption procedure, it was for the Commission, on the basis of evidence and arguments in its possession to determine if there were advantages such that an exemption may be granted, and that it was for the Commission to convince itself of this fact.58 Procedurally this was awkward for well-known reasons: the Commission acted with three hats simultaneously: first as prosecutor in trying to find evidence that Article 101(1) is infringed, then as advocate for the defence in trying to show that there may be countervailing benefits, and finally as adjudicator to determine whether the arguments of the defence are convincing. However, one advantage of this for the parties subjected to an investigation is that their sole burden was to give the Commission what evidence they could and indeed they can expect the Commission to also look for evidence that Article 101(3) is applicable, as a matter of good administration.59 Contrarily now, in front of the national courts, the defendant bears the burden of showing that the efficiency defence is applicable on the facts. In sum, procedurally the channels through which an efficiency defence may be pleaded in antitrust cases are fewer, and the burden on the defendant is higher when compared with the situation under Regulation 17.60

5.2.2 Evidentiary limits Moreover, there are high evidentiary hurdles where many claims have failed under the merger Regulation. The main difficulty here is that substantiating a 56 However it has also been suggested that the possibility of using commitment decisions as a functional substitute for exemption decisions could be more widespread, see Heike Schweitzer, ‘Commitment Decisions under Article 9 of Regulation 1/2003: The Developing EC Practice and Case Law’, in Claus-Dieter Ehlermann and Mel Marquis (eds), European Competition Law Annual 2008: Antitrust Settlements (Hart Publishing, 2010) pp 547 et seq. 57 Joined Cases C‑501/06 P, C‑513/06 P, C‑515/06 P and C‑519/06 P GlaxoSmithKline Services and Others v Commission and Others [2009] ECR I‑9291. 58 Ibid., paras 93–94. 59 Joined Cases 56/64 and 58/64 Consten and Grundig v Commission [1966] ECR 299, 347. 60 This is also supported, using a different analytical framework, by Christian Growitsch, Nicole Nulsch and Margarethe Rammerstorfer, ‘Preventing Innovative Co-operations: The Legal Exemption’s Unintended Side Effect’, 33 European Journal of Law and Economics 1 (2012).

Giorgio Monti 

69  

future benefit is a tall order, and the more high-risk the venture, the more difficult it will be to demonstrate that the efficiencies are likely to materialise.61 This does not mean that efficiencies can never be shown, as the Tom Tom merger decision illustrates.62 This was a vertical merger between Tele Atlas, a supplier of digital map databases and Tom Tom, a manufacturer of portable navigation devices and software for such devices. The Commission was convinced that the merger would eliminate the risk of a double mark-up and so lead immediately to a small decline in prices.63 However, in the same decision the Commission also showed its reluctance to accept more speculative efficiency claims. The parties argued that the merger would allow Atlas to benefit from the map corrections that the users of Tom Tom’s devices sent in so that then better maps would be deployed more quickly. The Commission did not accept these claims because it believed that one of the studies overestimated cost savings and another study misunderstood how the merged entity would react.64 However, parties that make claims of efficiencies that will materialise in the future face a more difficult challenge. As the Commission puts it, ‘in general, the longer the start of the efficiencies is projected into the future, the less probability the Commission may be able to assign to the efficiencies actually being brought about.’65 This biases the decision-maker in favour of agreements that yield shortterm productive efficiency gains in terms of cost savings, and against longer term investments. This should not be read as a criticism of the system; on the contrary it is arguable that the domain of competition law is the short run: just as the forecasts for anticompetitive risks tend to be about short-run effects of a given practice, so are the forecasts of a more rosy future. The key point to bear in mind is that this structural setup makes it very difficult to succeed in arguing dynamic efficiency gains.

5.2.3 Legal limits Furthermore, there are some limits to the way the efficiency claim is handled which have particular salience when one pleads dynamic efficiencies, even if the evidentiary hurdle is overcome. In the context of Articles 101 and 102, once the efficiencies have been demonstrated then it must be shown that consumers obtain a fair share. According to the Commission, the consumers that benefit should be those who also suffer from the negative impact of the practice in question.66 61 These and other related difficulties are considered in Lars-Hendrik Röller, Johan Stennek and Frank Verboven, ‘Efficiency Gains from Mergers’, Discussion papers WZB, Wissenschaftszentrum Berlin für Sozialforschung, Forschungsschwerpunkt Marktprozeß und Unternehmensentwicklung, No. FS IV 00-09 (2000). 62 Commission Decision of 14 May 2008 in Case COMP/M.4854 – Tom Tom/Tele Atlas. 63 Ibid, paras 238–243. 64 Ibid, paras 244–248. 65 Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, 2004 OJ C31/5, para 86. 66 Guidelines on the application of Article 81(3) of the Treaty, 2004 OJ C101/97, paras 43, 85 and 86. For a more expansive reading of the case law discussed in this chapter, see Damien Geradin, Anne

70  Competition, Regulation and Public Policies This restrictive interpretation means that one cannot make an argument along the following lines: higher prices today will result in superior medicines in the future, thus yielding benefits to future generations that are larger than the losses imposed on today’s consumers. This restrictive approach has the support of the European Courts, which has yielded only little on this point. In Asnef-Equifax the ECJ held that not all the consumers in the class must receive the benefits,67 and in Mastercard the General Court has taken the view that benefits to other parties may well be taken into consideration (in that case, users of credit cards) but that some of the benefits must also be passed on to those harmed by the restrictive practice (in that case, merchants).68 The Commission will also likely entertain a scenario where the anticompetitive effects occur in one market, the efficiencies in a second market, and where the consumers for both are the same and consume the two goods in similar proportions. Then here too the consumer may be said to receive a fair share of the efficiencies. However this scenario has only limited application. In economic terms, then, the efficiency defence requires one to show a Pareto improvement in welfare. This makes it impossible to trade off short and long term efficiencies, or gains to one group over gains to another. Similar consumer benefit analysis applies in abuse of dominance cases. At first blush it might appear that matters are different under Article 102, at least insofar as refusals to supply are concerned, because the Commission has indicated, in its Guidance on exclusionary abuse, that it would consider claims that the refusal is justified either because this allows the recoupment of prior investment or because a duty to deal will reduce future incentives to innovate.69 However this statement should be read subject to the general principles that the Commission articulated earlier in the same document, whereby consumer welfare in the affected markets is assessed, thereby making it harder to consider dynamic efficiencies that are projected into the future.70 It would be unusual if different principles applied given the shared aim of these two provisions. Layne-Farrar and Nicolas Petit, EU Competition Law and Economics (Oxford University Press, 2012) §§ 3.238–3.243. Case C-238/05 Asnef-Equifax and Administración del Estado [2006] ECR I-11125. Case T-111/08, MasterCard, Inc and MasterCard Europe v Commission [2012] ECR II-000, para 228; on appeal: Case 382/12 P. The court cited Case T-86/95 Compagnie Générale Maritime and Others v Commission [2002] ECR II-1011. However, in the latter case the then-CFI appeared not to require a specific pass-on to those harmed. Similarly, the CFI did not seem to require a pass-on to the consumers harmed in GlaxoSmithKline (see Case T-168/01 GlaxoSmithKline Services v Commission [2006] ECR II-2969, para 284), but this judgment can probably safely be disregarded given that its interpretation of Article 101 was quashed by the ECJ on appeal. As of this writing the MasterCard case is on appeal to the ECJ. In an Opinion of 30 January 2014, Advocate General Mengozzi approved of the approach taken by the General Court. See Opinion of the Advocate General in Case C-382/12 P MasterCard and Others v Commission, not yet decided. 69 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 (‘Guidance Paper’), para 89. For discussion, see Renato Nazzini, The Foundations of European Union Competition Law (Oxford University Press, 2012) pp 313–317. For a more circumspect assessment, see Pierre Larouche, ‘The European Microsoft Case at The Crossroads of Competition Policy and Innovation’, 75 Antitrust Law Journal 601, 614–616 (2008). 70 Guidance Paper, cited previous footnote, para 30. 67 68

Giorgio Monti 

71  

There is nothing to suggest that the consumer pass-on test is not the same in merger cases. But even more problematic is the fact that in merger cases, where there is more discussion of the efficiency defence, the precise status of efficiency claims is unclear. In Tom Tom/Tele Atlas, as we saw above, the parties succeeded in showing modest cost savings, but after assessing the efficiencies, the Commission concluded that this exercise had anyway been in vain ‘given the proposed transaction’s lack of anticompetitive effect irrespective of efficiencies.’71 It would appear that the only good reason for the Commission to illustrate a successful application of the efficiency defence was to announce its intention to take efficiencies into consideration, but one is still missing a decision where the efficiencies serve as a proper defence. In other decisions the Commission appears to suggest that efficiencies can weigh in as evidence that the merger is not harmful, for example in Korsnäs/Assidomän Cartonboard the Commission found that for several reasons the merger posed no risks and also found that there were some efficiencies. It concluded that these efficiencies ‘strengthen the conclusion that the proposed transaction will not significantly impede effective competition as a result of non-coordinated effects.’72 With respect, it is hard to make sense of this assertion. The legal basis for blocking a merger is that it should substantially impede effective competition. Therefore, if we find that the merger is not likely to cause any such harm, it is not clear why the benefits that the merger yields in terms of efficiencies are relevant to support the assertion that the merger is not harmful. A final legal limit pertains to regulatory holidays authorised by NRAs. Even with the procedure whereby these proposals are vetted by the Commission (as is the case for energy exemptions discussed above), it remains the case that compliance with sector-specific regulation does not justify an infringement of competition law.73 Therefore, a sector-specific regulatory holiday is unworkable. Granted, it will be argued that the Commission cannot grant an exemption with one hand only to remove it with the other, but this would be to ignore the risk that a national competition authority that had not been involved in the exemption request has no legal limits to prosecute, and it ignores the risk that private parties (and here it is clear that competitors of the beneficiary will have an interest) may use antitrust law to secure what the regulatory framework denied them.74

5.2.4 Ideological limits Above we have sketched some of the formal difficulties that make efficiencies hard to plead, but there are two more fundamental reasons why efficiency claims are problematic. The first is that the reason for prohibiting certain practices does Tom Tom/Tele Atlas, cited above note 62, para 250. Commission Decision of 12 May 2006 in Case COMP/M.4057 – Korsnäs/Assidomän Cartonboard, para 64. 73 Case C-280/08 P Deutsche Telekom AG v Commission [2010] ECR I-9555. 74 This point is elaborated well in Pierre Larouche, ‘Europe and Investment Infrastructure with Emphasis on Electronic Communications’, TILEC Discussion Paper DP 2007/031. 71 72

72  Competition, Regulation and Public Policies not rest on concerns that the conduct is inefficient, but it is based on conduct that is ‘anticompetitive’. This word is slippery and ill-defined.75 In spite of increased emphasis on developing theories of harm, it remains possible to condemn agreements or abuses of a dominant position simply on the grounds that they restrict the competitive process.76 According to Posner, the reason for this is that finding a restriction of competition is a sufficiently good proxy to ensure that antitrust enforcement yields efficiency.77 In the EU one might even go further and intimate that the protection of the competitive process is itself a value which is protected by the law, and while this likely leads to better economic outcomes, this is not a necessary condition for enforcing the law. Of course this has important implications for the efficiency defence in antitrust cases. To put it bluntly it places decision-makers in an impossible position because they are required to balance a restriction of competition on the one hand, with efficiencies on the other, and the metric by which one compares the two is not provided. More generally, it should be recalled that the basic hierarchy is that the competitive process comes first, and efficiencies only count provided the arrangements in question do not damage the competitive process too much. Therefore, whether one applies the efficiency defence under Articles 101 and 102 or under the Merger Regulation, one must show that there remains some competition.78 Thus for example the fourth condition in Article 101(3) is that competition is not eliminated. The Commission explains this requirement thus: ‘[u]ltimately the protection of rivalry and the competitive process is given priority over potentially pro-competitive efficiency gains which could result from restrictive agreements.’79 In applying this test when considering the efficiency defence in Article 101(3) one considers how far the parties to an agreement face actual or potential competition.80 It is less clear how this requirement may apply under Article 102 since the dominant firm is already in a position to withstand competitors. The primacy of competition over efficiency is also found in the energy sector, where even when investments are found to promote competition in the long term, by creating a more unified market, the short term interests of consumers are kept in mind and the interests of present and future investors take second place. In sum, to coin an aphorism, if for the 75 For a clever method to surmount this, see Aaron Edlin and Joseph Farrell, ‘Freedom to Trade and the Competitive Process’, NBER Working Paper No. 16818 (2011). 76 Or indeed, competition as such, as the ECJ said in GlaxoSmithKline, cited above note 57, para 63. 77 Richard Posner, Antitrust Law (2nd edn, Chicago University Press, 2001) at 29 (‘Efficiency is the ultimate goal of antitrust, but competition a mediate goal that will often be close enough to the ultimate goal to allow the courts to look no further.’). 78 See, eg, the Horizontal Merger Guidelines (cited above note 65), which claim that ‘it is highly unlikely that a merger leading to a market position approaching that of a monopoly, or leading to a similar level of market power, can be declared compatible with the common market on the ground that efficiency gains would be sufficient to counteract its potential anti-competitive effects.’ For Article 102, see the way the efficiency defence is set out in British Airways, cited above note 52, para 86, confirmed in Post Danmark, cited above note 52, para 41. 79 Guidelines on Article 81(3), cited above note 66, para 105. 80 Ibid, paras 107–116.

Giorgio Monti 

73  

Chicago School efficiency is justice,81 for the EU antitrust lawyer, efficiency without competition cannot be tolerated. There is also a second ideological matter that is related: using total welfare as a tool for market regulation is controversial. In Canada, when the competition authority and the courts sought to apply a total welfare standard to authorise a merger to monopoly there were calls for the law to be reformed.82 A study of the New Zealand regulatory experience in telecoms regulation suggests that efficiency based regulation is only successful insofar as politicians are willing to support it, but that it is not easy to secure a stable commitment to it. Perhaps worryingly the New Zealand example suggests that politicians favour static competition over efficiency because its benefits are more politically visible.83 It is apparent from the discussion above that there is no evidence of a commitment to economic efficiency above all else in the EU’s regulatory and competition law framework, rather the political commitment is to the competitive process, however this is not premised on short term political calculus, but on a long-term vision of the role of competitive markets.84

6. Finding a space for efficiencies It appears paradoxical that more than a decade since the more economicsbased approach has been inaugurated competition law should still struggle with efficiency claims.85 The way efficiencies have been integrated has instead been more subtle, but even so one may still wonder if more courageous approaches should not be tried.

6.1 Implicit efficiency defences The way antitrust law has tackled the problem of efficiency is to develop systems whereby the prohibitions do not apply when it can be safely assumed that the risk of harm is usually lower than the possibility of gain. Perhaps the clearest illustration of this is found in the use of Block Exemption Regulations. These provide that parties with low levels of market power and who do not include particular clauses 81 Eleanor Fox and Lawrence Sullivan. ‘Antitrust – Retrospective and Prospective: Where Are We Coming From? Where Are We Going?’, 62 New York University Law Review 936, 958 (1987). 82 See Giorgio Monti, ‘Merger Defences’, in Giuliano Amato and Claus-Dieter Ehlermann (eds), EC Competition Law: A Critical Assessment (Hart Publishing, 2006) pp 519 et seq., at pp 521–522. 83 Bronwyn Howell, ‘Politics and the Pursuit of Telecommunications Sector Efficiency in New Zealand’, 6 Journal of Competition Law and Economics 253 (2011). 84 Roger Zäch, ‘Freedom to Compete and the More Economic Approach – Limits Imposed by Law’, 40 IIC International Review of Intellectual Property and Competition Law 623 (2009). 85 Jacques Bourgeois and Denis Waelbroeck, Ten Years of Effects-based Approach in EU Competition Law: State of Play and Perspectives (Brussels, Bruylant, 2012).

74  Competition, Regulation and Public Policies in their agreement are automatically exempted. The rationale here is that while there is a risk of some harm to competition from some of these block exempted agreements, overall society gains from block exempting all. Similarly in merger cases, a market power screen operates to identify whether there is a prima facie risk of anticompetitive effects, and cases are closed quickly if there is no risk signalled after a preliminary consideration of certain factors. The rationale here too is that mergers below the risk thresholds are efficient.86 In other words, the system tolerates Type 2 errors in scenarios where the parties have low market power, because ‘letting the guilty go free in antitrust is generally a self-correcting problem’.87 Another mechanism to reach a similar result is to raise the standard of proof for the plaintiffs in such a way that they will consider carefully whether or not to sue and will thus allow efficient practices to go unchallenged. This policy is prevalent in the United States because of the perceived risk of private litigants using antitrust laws to secure lucrative settlements, but it is also found in the Commission’s commitment to use a more economics-based approach before challenging mergers and dominant firms. One of the effects of the revision of the merger Regulation has been that the Commission uses more indicators to test for anticompetitive risks.88 And the expected effect of the Guidance Paper on Article 102 is precisely to make it more difficult for the Commission to bring a case by requiring more economic evidence than at present.89 A third mechanism to handle efficiencies is to use them in such a way as to nullify the anticompetitive risk. This is particularly useful in merger cases, where a merger might serve to disrupt an anticompetitive equilibrium by creating a firm with the capacity to act as a maverick and disrupt the existing state of affairs. Obviously in this case we do not speak of an efficiency defence; rather efficiencies are evidence that the merger is likely to increase competition between the merged entity and others.90

6.2 Bolder approaches A number of scholars have argued in favour of a more robust acknowledgment that innovation and dynamic efficiency in antitrust law.91 It is beyond the scope of this chapter to review all of the suggestions. In particular, it is unnecessary to 86 Directorate-General for Economic and Financial Affairs, ‘The Efficiency Defence and the European System of Merger Control’ (2001) European Economy No.5 (2002). 87 Fred McChesney, ‘Talkin’ Bout My Antitrust Generation: Competition for and in the Field of Competition Law’, 52 Emory Law Journal 1401, 1412 (2003). 88 Nicholas Levy, ‘The EU’s SIEC Test Five Years On: Has it Made a Difference?’, 6 European Competition Journal 211 (2010). 89 Of course, such self-control only works if there are no alternative avenues, and what we have witnessed in the context of Article 102 is an increased use of commitment decisions, with imprecise theories of harm. 90 Two examples are discussed in Monti, ‘Merger Defences’, cited above note 82, at 525–527. 91 For my review of the literature a decade ago, see Giorgio Monti, ‘Article 82 EC and New Economy Markets’ in Cosmo Grahakm and Fiona Smith (eds), Competition, Regulation and the New Economy (Hart

Giorgio Monti 

75  

consider the debate between those who urge that antitrust enforcement should be sharpened to address specific antitrust harm that may occur in innovation, for example by suggesting that agencies focus more on exclusionary conduct,92 and those who are more cautious about Type 1 errors from this kind of enforcement strategy.93 At any rate, as we have seen in the context of energy exemptions, the Commission is well aware that sham innovative steps can serve exclusionary purposes, which explains why exemptions in energy projects expire if the parties do not start the project after a given timeframe.94 For present purposes, the focus is on proposals that explore to what extent conduct suppressing competition might be outweighed by gains in dynamic efficiency. One line of argument is that the thresholds-based approach to identifying anticompetitive risk is unhelpful because it is only based on static considerations. Accordingly defining markets and assigning market shares are not helpful when markets are dynamic: it is not easy for example to apply the hypothetical monopolist test in high-tech industries.95 One suggestion is to focus on the capabilities of firms, which would measure their strength on the market in a dynamic way.96 Capability refers for example to the capacity of a firm to innovate. The principal proponent of this view, David Teece, gives as an example a focus not on Honda’s market share in its products (cars, lawnmowers or electric generators) but perhaps on Honda’s capacity to develop engines.97 This approach suggests looking more closely at the processes within the firm as a proxy for gauging its capacity to compete. Mergers of two firms with the highest capabilities for innovation in the same sector would then raise competition concerns, irrespective of market shares. This approach continues the implicit efficiency defence we noted earlier, but refines it to take into account criteria that are more pertinent to dynamic efficiency considerations. An approach which tries to balance short run and long run incentives has been sketched by Professor Wu: a competition authority should give a firm that has Publishing, 2004) pp 17 et seq., at 18–24. See more recently Carl Shapiro, ‘Competition and Innovation: Did Arrow Hit the Bull’s Eye?’, in Joshua Lerner and Scott Stern (eds), The Rate and Direction of Inventive Activity Revisited (University of Chicago Press, 2012); Howard Shelanski, ‘Innovation, Information and Competition Policy for the Internet’ 161 University of Pennsylvania Law Review 1663 (2013). 92 Tim Wu, ‘Taking Innovation Seriously: Antitrust Enforcement if Innovation Mattered Most’, 78 Antitrust Law Journal 313 (2012). 93 Joshua Wright, ‘Antitrust, Multidimensional Competition and Innovation’ in Geoffrey Manne and Joshua Wright (eds), Competition Policy and Patent Law Under Uncertainty (Cambridge University Press, 2011). 94 Moreover, the debate between Wu and Wright has less resonance in the EU, where the Commission and the Courts are both committed to a stance that promotes and protects the innovation potential of new entrants. See, eg, Case T-201/04 Microsoft [2007] ECR II-3601, where the Commission condemned exclusion which harmed new entrants; and Case C-52/09 Konkurrensverket v TeliaSonera Sverige AB [2011] ECR I-527, where the doctrine of margin squeeze is designed to facilitate new entry. 95 Michael Katz and Howard Shelanski, ‘Mergers and Innovation’ 74 Antitrust Law Journal 1 (2007). 96 This approach builds on that which identified ‘innovation markets’. See Richard Gilbert and Steven Sunshine, ‘Incorporating Dynamic Efficiency Concerns in Merger Analysis: The Use of Innovation Markets’, 63 Antitrust Law Journal 569 (1995). 97 David Teece, ‘Favouring Dynamic over Static Competition – Implications for Antitrust Analysis and Policy’, in Manne and Wright (eds), Competition Policy and Patent Law Under Uncertainty, cited above note 93, pp 224 et seq.

76  Competition, Regulation and Public Policies acquired a monopoly some breathings pace to make profit before assailing it with an antitrust lawsuit. The argument is based on analogy with patent law, which gives the inventor a temporary respite from competition. In his view this can create a credible commitment for firms to invest because they would be safe in the knowledge that, having achieved market power as a result of their investment, they can exclude rivals without fearing antitrust liability.98 This suggestion is precisely in line with the regulatory holiday suggested in this chapter. The advantage of Wu’s approach is that rather than requiring a change in the legal standard, the suggestion is that agencies declare a prosecutorial intention to leave innovators alone up to the point when the monopolist can be assumed to have earned back its investment. Of course this is also the proposal’s weakness for private enforcement cannot be stopped by agencies. What is also worth noting in this proposal is that it eschews balancing between short term harm and long term gains, and so fits within the implicit balancing tests noted in section 5 above. It appears too risky to suggest that an agency balance static and dynamic efficiencies in a direct manner.

7. Conclusion We began by the story of how regulatory holidays do not fit easily into the EU’s regulatory framework in industries that are being liberalised. In spite of good reasons for facilitating the construction of new infrastructure to resolve many of the EU’s energy concerns, the Commission has remained concerned of possible short run anticompetitive effects. We have then traced this reluctance to a core area of the EU, competition law. The procedural, legal and practical hurdles we have identified that make the application of a balancing exercise whereby dynamic gains are traded off against static losses stem from a basic constitutional setup in favour of competition as a process. The real question then is whether the European Union should to rethink the so-called economic constitution so that more space is afforded to investment stimuli and total welfare considerations. Perhaps this will occur in view of the new energy chapter in the TFEU (Article 194), where it is said that Union policy should aim to, inter alia, ensure the functioning of the energy market and promote the interconnection of energy networks. There is no mention that these markets should be competitive, perhaps then the priority is tilted towards goals other than competition. That said, it remains to be seen how the legislators use this new legal basis. The lesson from competition law is that the most convenient way forward is to design rules of thumb that exclude the application of competition considerations. Bizarrely, then, efficiency claims are most easily accepted when concealed from view. 98

Wu, cited above note 92, at 326–328.

Anthony Pygram*

Competition, Regulation and other Public Policies in the UK Energy Sector

Introduction This chapter offers some reflections on how competition can benefit consumers in a regulated sector and how a regulator can promote competition.1 It also looks at the balance that is struck when addressing an issue using competition law tools and sectoral regulation tools, and how some public policies cannot succeed by relying on competition alone.

The duties of the regulator The Office of Gas and Electricity Markets (Ofgem) is a creature of statute. Parliament sets out the duties. The Authority governing Ofgem, the Gas and Electricity Markets Authority (GEMA), has as its principal objective: to protect the interests of existing and future consumers in relation to gas conveyed through pipes and electricity conveyed by distribution or transmission systems.  The interests of such consumers are their interests taken as a whole, including their interests in the reduction of greenhouse gases and in the security of the supply of gas and electricity to them, and their interests in the fulfilment by the Authority of certain objectives specified in the gas and electricity directives. The Authority is generally required to carry out its functions in the manner it considers is best calculated to further the principal objective, wherever appropriate by promoting effective competition between persons engaged in, or commercial activities connected with, • the shipping, transportation or supply of gas conveyed through pipes • the generation, transmission, distribution or supply of electricity • the provision or use of electricity interconnectors * Partner, Enforcement and Competition Policy, Ofgem. The views expressed are those of the author and may not reflect those of Ofgem. 1 The chapter has been updated to reflect changes between July 2012 and October 2013, including the passage of the Enterprise and Regulatory Reform Act 2013 and the publication of consultations on related secondary legislation and guidance.

78  Competition, Regulation and Public Policies in the UK Energy Sector However, before exercising functions for the purpose of promoting competition, the Authority has a positive duty to consider the extent to which the interests of consumers would be protected by the promotion of competition and whether alternative ways of exercising its functions would better protect the interests of consumers. Promoting competition is the starting point. But Ofgem is given a clear steer that methods other than the promotion of effective competition in gas and electricity markets should be used to look after the interests of present and future consumers where they would better protect those interests. And that is surely right – as an expert regulator we should be looking to achieve results as best we can – using the best tool at our disposal for each scenario as best we can to achieve lasting benefits for consumers.

Ofgem is a concurrent regulator Ofgem is a concurrent regulator – which is unusual in Europe. That means that Ofgem is the first port of call for the investigation of alleged breaches of Articles 101 and 102, and their UK equivalents Chapters 1 and 2 of the Competition Act 1998 (CA98) in respect of the energy sector in Great Britain. There are clearly some benefits associated with Ofgem being able to carry out these investigations. First, Ofgem is an expert regulator, and so it should be better placed to get to grips quickly with the issues raised in the particular markets under investigation. It also allows Ofgem directly to take action where there are problems in a market. That might include deciding that in certain circumstances it is more appropriate to use its regulatory tools rather than its competition law tools. However, a concurrent regulator can be open to criticism for its choices as to which tool to use or, having decided to do so, how it uses that tool;2 there can be perceived tensions about use of different regulatory tools. A concurrent regulator may consider it appropriate to take the quickest path to resolving something for the more immediate benefit of consumers but at the expense of pursuing a case to completion – whereas a non-industry-specific regulator might see the same case as setting an important precedent across industries. So a concurrent sectoral regulator could face suggestions that it has its eye on its aims more than the aims of the whole competition system. That said, where the Authority is exercising its competition law functions (i.e. CA98, but not Enterprise Act market investigation references) its Principal Objective does not apply and therefore it is acting purely as a competition authority. 2 See ‘Growth, competition and the competition regime: Government response to consultation’ (March 2012), section 8. Available at https://www.gov.uk/government/uploads/system/uploads/ attachment_data/file/31879/12-512-growth-and-competition-regime-government-response.pdf.

Anthony Pygram 

79  

Ofgem has addressed these tensions by working closely with the Office of Fair Trading. Going forwards, changes introduced by the Enterprise and Regulatory Reform Act 2013 and related legislation require Ofgem and other concurrent regulators in the UK to work closely with the OFT’s successor the Competition and Markets Authority (CMA), sharing information and expertise. The Act also recognises the tensions which can arise between the regulator’s competition law and regulatory roles, by giving the CMA the power to decide it is to investigate a particular case rather than the regulator or take over an open investigation from the regulator where the CMA is satisfied that its doing do so would further the promotion of competition, within any market or markets in the United Kingdom, for the benefit of consumers3, and for the Government to remove concurrent powers from a sectoral regulator where it considers it appropriate to do so for the purposes of promoting competition, within a market or markets in the United Kingdom, for the benefit of consumers.4 It is also worth noting that some things are easier to enforce than others. For example, a case alleging abuse of a dominant position might be easier in a market with only one player – such as a market involving a single system operation – than one with six large players and a few smaller ones – such as the retail markets for gas and electricity in the UK. This may be another factor a concurrent regulator might take into account when deciding which tool to use.

Not all problems lend themselves to Chapter I/Chapter II solutions Of course not all imperfections in a market are a direct result of anti-competitive behaviour which could be prosecuted under articles 101 or 102. In the UK, despite over two decades of privatisation, it is clear that retail markets are not functioning effectively. There are a plethora of different tariffs consumers can sign up to, yet around ¾ of customers are on their suppliers’ standard tariffs, which will be one of the supplier’s most expensive tariffs on the market, and indicates that the consumer is unlikely to be actively engaged in searching for the best deal, despite energy prices being a significant part of consumers’ outgoings. This is probably because of a combination of factors, and so a combination of responses may be appropriate. One part of the solution is to take action to incentivise suppliers to do what they should do as a matter of course – for example 3 See draft Competition Act 1998 (Concurrency) Regulations 2014, Regulations 5 and 8. The Government is consulting on these draft Regulations as part of its package of legislation relating to the creation of the new CMA. See the consultation document at https://www.gov.uk/government/uploads/ system/uploads/attachment_data/file/245994/Competition-regime-draft-secondary-legislation-parttwo.pdf. 4 Enterprise and Regulatory Reform Act 2013, section 52. Before exercising such a power, the government is required to comply with certain procedural requirements including consultation with the regulator in respect of whom it proposes to remove concurrent powers. See section 53 of the Act.

80  Competition, Regulation and Public Policies in the UK Energy Sector Ofgem is investigating 2 of the big 6 energy suppliers for possible mis-selling of energy contracts, and has concluded investigations into three other suppliers, in one case resulting in a fine of £10.5 million and in the other two cases redress, to agreed customers and schemes, and compensation packages totalling £13 million have been agreed. Another part of the approach is to look across the market. In 2010 Ofgem started its Retail Market Review, identifying some concerns about a lack of effective competition in retail markets. In June 2013 Ofgem announced detailed rule changes that will deliver a simpler, clearer and fairer energy market.5 It is possible for Ofgem to conduct a Retail Market Review with teeth because it has the option of making a Market Investigation Reference to the UK’s Competition Commission (CC) if it suspects there are features of the market which prevent restrict or distort competition. Such investigations can themselves be very powerful, and can lead to significant changes in market structure being required – the CC’s investigation into BAA’s ownership of airports, including the three large London airports, resulting in BAA having sell Gatwick and Stansted Airports, as well as one of Glasgow and Edinburgh Airport. Similarly, the investigation into payment protection insurance led to a prohibition on selling the insurance at the same time as the underlying credit, to open up the possibility for competition to provide the insurance. Some issues may not lend themselves to competitive solutions without a nudge. Protecting the interests of consumers in relation to the security of supply of gas and electricity to consumers is part of Ofgem’s Principal Objective. Keeping the lights on is important for consumers, suppliers, and society more generally, but the impact on those groups, or even within these groups, of a security of supply problem is not necessarily the same. So it is important that we incentivize suppliers to meet security of supply concerns at a level that meets the needs of society in both the short and longer term. Regulation in its broadest sense can play an important role in this. For example, Ofgem is in the final stages of a Significant Code Review for gas, which seeks to strengthen the price signals that shippers face, in order to encourage them to procure sufficient gas should a gas emergency arise. This is different from replacing competition with price regulation, but gives suppliers the incentive to look at solutions they might not otherwise look at – whether it is by entering into interruptible contracts (most likely with large I&C customers) additional storage, diversifying the routes used to procure gas, or a combination of methods. Similarly, reducing greenhouse gases is an important issue, but one which is costly, and without being nudged in a certain direction the incentives on suppliers might otherwise be to keep to cheaper ways of supplying gas and electricity. The Government sets strict environmental targets. The regulator has a role to play in this – for example by administering schemes where suppliers carry out measures to reduce carbon emissions, which the Government sets up to incentivise suppliers 5

See https://www.ofgem.gov.uk/retail-market-review.

Anthony Pygram 

81  

to meet those targets. If necessary Ofgem can enforce against companies where the scheme rules are not followed or the required carbon savings are not achieved.

Conclusion The role of Ofgem as the gas and electricity regulator in the UK is to look after consumers. Competition is a key means to this end. In many markets effective competition can look after the needs of consumers, bringing a good mix of price, quality, range and service. However, in many sectors this is not enough – in the UK, markets as diverse as supermarkets, insurance products, airports and local bus services have needed and benefitted from intervention by competition authorities. Competition can bring benefits for consumers in regulated sectors – in the UK the retail supply of gas and electricity is open to competition and Ofgem is working to maximise the benefits from competition for consumers. Sometimes competition alone cannot meet the competing demands placed on a sector. In the generation and supply of gas and electricity there are competing demands which cannot all be obviously met by unfettered competition alone. Sometimes regulation aimed at ‘oiling the wheels’ may be necessary, as the Retail Market Review is aiming to do, and sometimes yet more interventionist measures may be needed to ensure that the incentives on suppliers align with those of society as a whole

Geert Goeteyn*

Regulation and Competition in the Aviation Sector: Focus on Airline Content and Global Distribution Systems

Update (3 March 2014) Since the Workshop of July 2012, there have been a number of developments in the aviation sector, including most recently the merger between American Airlines and US Airways. In addition, American Airlines settled its antitrust action against Sabre on 31 October 2012 and its actions against Travelport and Orbitz (which is partly owned by Travelport) on 13 March 2013 and 1 April 2013 respectively. The US Airways/Sabre litigation continues. However, the issues and questions raised in this chapter have not been resolved in the intervening time.

Introduction and background (i) Competition Policy & Enforcement in Context: A Decade in Review A little over a decade ago, in March 2002, Joos Stragier, the then-Head of the Transport Unit of DG Competition, addressed the Annual Conference of the Annual Conference of the Guild of European Business Travel Agents on EC Competition Policy in the Aviation Sector: State of Play and Outlook.1 Referring to the gradual liberalisation of air transport over the preceding decade, Stragier stated that ‘[c]ompetition policy can usefully contribute to the creation of a healthy environment for the development of an efficient and competitive air transport industry’. Stragier stressed, however, that ‘[c]ompetition enforcement must secure that airlines’ responses to the new competitive environment do not undermine the benefits, which liberalised air transport markets can provide to travellers’. (emphasis added) Against this background (and taking account of the recent market developments), Stragier indicated that the Commission’s enforcement priorities in the air transport field concerned three main areas: (i) alliances, mergers and other forms of co* Partner, Shearman & Sterling. This outline was prepared together with Collette Rawnsley. 1 EC COMPETITION POLICY IN THE AVIATION SECTOR: STATE OF PLAY AND OUTLOOK, Annual Conference of the Guild of European Business Travel Agents (GEBTA), Lisbon, 22 March 2002, http://ec.europa.eu/competition/speeches/text/sp2002_009_en.pdf.

84  Competition, Regulation and Public Policies operation (eg code shares) between airlines; (ii) hard-core restrictive agreements between airlines and abusive behaviour; and (iii) industry-wide horizontal cooperation, in particular in the IATA framework. Stragier reiterated the position later that year (2002), stating that: In the aviation sector, there are three main competition enforcement areas, which deserve […] our attention. The first obviously concern mergers, alliances and other type of horizontal co-operation agreements between air carriers. The second area relates to unilateral abusive behaviour, in particular foreclosure practices, such as loyalty incentive schemes for travel agents and predatory behaviour. Finally, certain traditionally accepted industry-wide restrictive practices and agreements, mainly in the context of IATA need to be reviewed.2

Stragier (rightly) believed that ‘airline alliances and mergers [would] continue to play an important role within the ongoing industry restructuring process’. (emphasis added) While ‘[i]t might be expected that in the short run industry evolution [would] follow largely the same pattern as in the past decade, with a strong preference for increasingly integrated alliances’, Stagier recognised that in the medium and long term ‘full-scale mergers between airlines would not any longer be hampered by regulatory restrictions’. (emphasis added) Similarly, Stagier acknowledged that the modernisation of anti-trust procedures3 were an ‘important development’ that would ‘have a significant impact on competition policy and enforcement’. In the decade since Stragier’s speeches, there have been relatively few (official) DG Competition speeches dedicated to air transport.4 Although limited in number, these speeches reveal a trend in competition enforcement. In 2003, the then-Commissioner, Professor Mario Monti, noted that ‘[o]ver the last few years aviation has become an important area for competition policy and enforcement. Various problems have passed the competition review: from alliances and mergers to abusive practices, cartel behaviour and industry wide co-operation’.5 Noting that aviation was shifting ‘to a progressively liberalised regulatory framework’, the Commission observed that: Efforts of liberalisation go hand-in-hand with the industry’s plea for consolidation. The Air France KLM move is just the most recent example of how important it is for carriers to make concerted efforts and create synergies which will allow them to survive in a global economy. But before the word ‘merger’ was even mentioned in the industry, carriers found other inventive ways to co-operate: code-sharing, joint services and 2 Emphasis added. See OUTLOOK OF EUROPEAN COMMISSION’S COMPETITION POLICY AND ENFORCEMENT PRIORITIES IN AIR TRANSPORT, EALA, Conference of 22 November 2002, Stockholm, http://ec.europa.eu/competition/speeches/text/sp2002_038_en.pdf. 3 Pursuant to Council Regulation 1/2003 on the implementation of the rules on competition laid down in Articles [101] and [102] of the Treaty, 2003 OJ L1/1. 4 According to the DG Competition website there have been three such speeches, one of which was Commissioner Joaquín Almunia’s announcement of the prohibition of the proposed merger of Olympic Air and Aegean Airlines. See http://ec.europa.eu/competition/speeches/index_theme_28.html. 5 Mario Monti, ‘Recent Developments in European Air Transport Law and Policy’, Brussels, SPEECH/03/522 of 6 November 2003.

Geert Goeteyn 

85  

fully-fledged alliances. The Commission takes a generally positive approach to alliances because in most cases they generate several benefits. However, it is my role to ensure that consolidation is not done at the expense of competition and consumer welfare.

Addressing the ‘role of competition policy in these times of change’, Commissioner Monti referred ‘schematically to the current main issues of competition enforcement in the air transport field, as regards alliances & mergers, hard-core restrictions and industry-wide agreements’. Predictably, three years later, in a speech entitled ‘Competition in the Aviation Sector: the European Commission’s approach’,6 Commissioner Neelie Kroes focused on the Commission’s ‘recent enforcement practice in relation to airline mergers and alliances’.7 (emphasis added) A brief review of the speeches above suggests that, despite significant changes in the aviation sector, enforcement priorities have remained almost unchanged for a decade.8 To a certain extent the focus on airline mergers and alliances is understandable – there continues to be considerable consolidation in the sector, a trend that looks set to continue as a result of the ongoing global economic crisis. Indeed, the intervening period has seen significant consolidation in the aviation sector both in terms of mergers9 and participation in alliances.10 Leiden, SPEECH/06/247 of 24 April 2006. But she recognised that ‘[d]ynamic and entrepreneurial airlines [we]re shaking up the industry and bringing prices down’ and that ‘[c]itizens can choose between more airlines and more routes when travelling within Europe’. 8 There have been a limited number of cartel investigations in the sector during the last 10 years; notably in respect of (i) Air Cargo (see Case T-9/11 Air Canada v Commission (2011 OJ C72/25), Case T-28/11 Koninklijke Luchtvaart Maatschappij (2011 OJ C72/30), Case T-38/11 Cathay Pacific Airways v Commission (2011 OJ C72/32), Case T-36/11 Japan Airlines v Commission (2011 OJ C80/25), Case T-39/11 Cargolux Airlines v Commission (2011 OJ C80/26), Case T-40/11 Lan Airlines and Lan Cargo v Commission (2011 OJ C80/27), Case T-46/11 Deutsche Lufthansa and others v Commission (2011 OJ C80/31), Case T-48/11 British Airways v Commission (2011 OJ C80/32), Case T-62/11 Air France-KLM v Commission, Case T-63/11 Air France v Commission and Case T-67/11 Martinair Holland v Commission (2011 OJ C95/8)); (ii) Freight Forwarding (see Commission Decision of 28 March 2012 in Case COMP/39462 – Freight Forwarding); and (iii) the abortive investigation into international airline passenger services (see Administrative Closure notice for Case COMP/39.419 – International airline passenger services). Interestingly, there have been no specific abuse of dominance cases (ie, outside consideration of alliances) under Article 102 TFEU since Commission Decision of 13 December 1999 in Case IV/24.780 – Virgin/British Airways, 2000 OJ L30/1. 9 See Case COMP/M.2672 SAS/Spanair, 2002 OJ C93/7; Case COMP/M.3280 Air France/KLM, 2004 OJ C60/5; Case COMP/M.3770 Lufthansa/Swiss, 2005 OJ C204/3; Case COMP/M.3940 Lufthansa/Eurowings, 2006 OJ C18/22; Case COMP/M.4600 TUI/First Choice, 2007 OJ C137/6; Case COMP/M.4439 Ryanair/Aer Lingus, 2008 OJ C47/9; Case COMP/M.5181 Delta Airlines/ Northwest Airlines, 2008 OJ C281/3; Case COMP/M.5335 Lufthansa/SN Airholding (Brussels Airlines), 2009 OJ C295/11; Case COMP/M.5364 Iberia/Vueling/Clickair, 2009 OJ C72/23; Case COMP/M.5403 Lufthansa/bmi, 2009 OJ C158/1; Case COMP/M.5141 KLM/Martinair, 2009 OJ C51/4; Case COMP/M.5440 Lufthansa/Austrian Airlines, 2010 OJ C16/11; Commission Decision of 14 July 2010 in Case COMP/M.5747 – BA/Iberia; Commission Decision of 27 July 2010 in Case COMP/M.5889 – United Airlines/Continental Airlines; Commission Decision of 26 January 2011 in Case COMP/M.5830 – Olympic/Aegean; Commission Decision of 30 March 2012 in Case COMP/M.6447 IAG/bmi. 10 Most recently the Commission’s has investigated the Air Canada/Continental/Lufthansa/United Airlines, British Airways/American Airlines/Iberia and Air France-KLM/Alitalia/Delta alliances. 6 7

86  Competition, Regulation and Public Policies This chapter raises the question whether the current competition policy and enforcement, which is almost exclusively aimed at airlines, should be broadened to reflect the fact that the airlines themselves are looking at the bigger picture (beyond consolidation) and that there are wider issues to be addressed in the sector. Recent developments have underlined a shift in market dynamics, in particular vis-à-vis the relationship and interaction between the various market participants (notably those involved in travel distribution, in what is generally recognised as a two-sided market).11 In this regard, there is a real risk that, if it continues to plough the same, familiar, furrow, competition policy and enforcement in the EU will be off the pace – addressing historic issues and storing up problems for the future. Against this background, this chapter will consider whether the current Code of Conduct,12 which was initially intended to deal with travel distribution by way of computerised reservation systems (CRSs), is fit for purpose. Or whether, in a dynamic and fast-moving market, the time has (finally) come for outdated ex ante regulation to give way to active competition law enforcement which is able to take account of the features of the market today, rather than a historic snapshot of what used to be. The latter approach would be consistent with that in the US, where the issues (and the market participants) are invariably the same. At the moment, however, the more pressing concern is that an enforcement lacuna exists, which is enabling market participants to engage in (and continue) conduct (which could arguably be characterised as industry-wide restrictive practices and agreements) that risks damaging the sector and, ultimately, European passengers (as end-consumers).

(ii) The CRS Code of Conduct: a blunt & unsophisticated tool approaching extinction? CRSs were originally part of the internal reservation systems of carriers.13 They were developed by airlines to provide a single point interface for subscribers (ie, travel agents) in the wake of industry deregulation in the US However, airline ownership and control provided the incentive and ability to engage in practices (eg, so-called display bias), to favour the content of controlling airline(s). 11 In Case COMP/M.4523 – Travelport/Worldspan the Commission determined that Global Distribution Systems (GDSs) operate in a two-sided market connecting two distinct categories of customers. In the upstream market, multiple travel service providers (eg, airlines participating in the GDS) offer the GDSs access to Airline Content for aggregation and distribution to travel agencies subscribing to the GDSs’ data processing system, so that travel agents in the downstream market can search and compare Airline Content and process reservations and ticket sales for involved carriers using the GDS as their technology platform to interface with airline host reservation systems. 12 Regulation 80/2009 of the European Parliament and of the Council of 14 January 2009 on a Code of Conduct for computerised reservations systems and repealing Council Regulation 2299/89, 2009 OJ L35/47. 13 1964 saw the launch of the first CRS – Semi-Automated Business Research Environment – or SABRE, a collaborative effort between American Airlines and IBM. Other airline-sponsored/created CRSs followed including Amadeus (group of European airlines), Galileo (group of European airlines), Apollo (United Airlines), PARS (TWA) DATAS II (Delta).

Geert Goeteyn 

87  

July 1989 saw the adoption of Regulation 2299/89,14 which established a ‘code of conduct’ for CRSs. Regulation 2299/89 was designed to ensure nondiscriminatory access by all carriers to the CRS Systems. Nonetheless, in 2007, the second public consultation on Regulation 2299/89 (as amended by Regulations 3089/93 and 323/99), indicated that further, more significant revision was required. The consultation identified a number of serious issues, notably that the Regulation was ‘increasingly ill-adapted to the changed market conditions’ and that the ‘non-discrimination requirement for booking fees stifles price competition, and the prohibition for airlines to differentiate content between CRSs significantly restricts their negotiating freedom. The ensuing lack of competition leads to higher CRS booking fees and creates a system of economic rents in favour of CRSs and travel agents, at the expense of airlines and their passengers.’15 In addition, the consultation highlighted the need for regulation to adapt to, and take account of, technological and market developments, in particular the fact that airline ownership/control of was almost extinct (with Amadeus being the notable exception). The current CRS Code of Conduct was intended to address the concerns identified in the consultation process and also simplified the legislative framework by giving more flexibility to system vendors (ie CRSs) and air carriers to negotiate booking fees and fare content – allowing them to adapt in a flexible way to the needs and requests of travel agents and consumers and to distribute more efficiently their transport products. Despite the acknowledged change in market conditions, certain provisions to prevent bias granting undue advantages to so-called ‘parent carriers’ and to ensure the supply of neutral information to consumers were retained. Article 18(3) of the current CRS Code of Conduct provides that ‘by 29 March 2013, the Commission shall draw up a report on the application of this Regulation which shall assess the need to maintain, amend or repeal this Regulation’. (emphasis added) To this end, in summer 2011, the Commission (DG MOVE) undertook a stakeholder consultation.16 Industry commentators noted that ‘[a] lthough examination of the Code is a standard process (other similar reviews saw amendments made to the original 1989 Act in 1993 and 1999), many believe so widespread is the shifting landscape of travel distribution that the review could see some significant changes next year’.17 One such shift, has been the change to the distribution model around merchandising, so-called direct-connect and mobile. In this regard it has been suggested that: 14 Council Regulation 2299/89 on a code of conduct for computerised reservation systems, 1989 OJ L220/1, amended by Council Regulation 3089/93, 1993 OJ L278/1. 15 Commission MEMO/07/463 of 15 November 2007. 16 See Kevin May (tnooz), ‘European Commission Begins Review of GDS Code, Realpolitik Expected’, 15 July 2011, http://www.tnooz.com/2011/07/15/news/european-commission-beginsreview-of-gds-code-realpolitik-expected/. At the time of writing, neither the responses to the consultation, nor the Commission’s conclusions have been made public. 17 See ibid.

88  Competition, Regulation and Public Policies European officials are apparently aware of and keen to see these changes in the marketplace also reflected in the Code, although some suggest this is actually also double-edged sword for the GDSs. On the one hand the changes to the distribution of air fares and associated products should be regulated, but equally including these new processes in the Code then shows acceptance in the wider industry that, at least in terms of Direct-Connect, the model is evolving – a change that some of the GDSs are fighting against in the courts in the US.18

Given the important changes to the market for travel distribution, an obvious question is whether the time has come to follow the US example and deregulate the market.

The travel distribution landscape: widespread shifting The current CRS Code of Conduct and DG Competition’s enforcement and/or policy objectives do not focus on, or even take account of the significant changes in this market. When CRSs were first developed, local travel agents habitually acted as intermediary between the airline and the consumer, making bookings using a CRS. Online computing systems were nascent technology and, as such, wider distribution of and access to CRS was both impractical and prohibitively expensive. This resulted in airline led initiatives to create and roll out CRSs. Now, the internet is almost ubiquitous19 and, consequently, access to online information has become much more widespread and significantly cheaper. The internet booking enables most consumers to purchase airline tickets without the need to visit a local travel agent. In fact, a number of companies (eg Travelocity, Expedia and Orbitz) specifically cater to consumer demand for online travel booking facilities, allowing consumers to search for available flights, schedules and fares, as well as information on related products/services such as rental car and hotel accommodation. The revolutionary effect of the Internet in this sector should not be underestimated. Today, the CRSs have evolved into four large GDSs, which are controlled by just three suppliers: (i) Galileo/Apollo and Worldspan (both owned by Travelport); (ii) Sabre; and (iii) Amadeus. Airline ownership or control of GDSs has given way to a new form of vertical integration (ie, vertical integration is now prevalent between GDSs and (online) travel agents). In this regard, Travelport has ownership interests in brands such as Orbitz, ebookers and CheapTickets; Sabre owns Travelocity, lastminute.com and others; and, until relatively recently, Amadeus held a controlling share in Opodo. Ibid. The United Nations has recently ruled that Internet access is a basic human right that should be guaranteed and protected by states. See http://www.regeringen.se/content/1/c6/19/64/51/6999c512. pdf. 18 19

Geert Goeteyn 

89  

Moreover, deregulation of the previously fragmented aviation sector in the EU and the success of low cost carriers (ie, the ‘[d]ynamic and entrepreneurial airlines’ that Commissioner Kroes stated ‘[w]ere shaking up the industry and bringing prices down’) have not only increased competition on certain routes, but have encouraged the commercialisation and ‘unbundling’ of airline products – the customer no longer simply buys a seat on an airplane, but wants to have greater control over the entire experience (eg from booking, to check-in, seat selection to baggage allowances).20 In recent times, the ‘traditional full service’ airlines have sought to render their distribution systems more efficient. Following the example of the low cost carrier, they have sought to differentiate their service and product offerings. They also have promoted bookings through their own websites and tried to leverage their freedom resulting from the last revision of the CRS Code of Conduct to offer differentiated content agreements to the different GDSs in order to obtain better contract provisions. More recently, there have been great efforts by a number of carriers to roll out airline Direct Connects allowing agents to book directly into the reservation systems of the carriers, rather than having to making bookings though a GDS, leading to creation of content aggregators who can consolidate numerous Direct Connects. Nonetheless, despite these developments, the Commission’s analysis in the Travelport/Worldspan merger decision as regards the airlines’ countervailing buyer power21 does not apply today, or at least not as fully as expected at the time. The GDSs continue to have an iron grip on the airline content distribution chain.22 This is a stark contrast to the picture of the market painted in the Notification of the Travelport/Worldspan merger.23 The Commission Decision (para 24 and footnote 17) states that ‘[t]he notifying party states that several companies have recently developed alternative technological platforms providing travel content and booking capabilities. Travelport mentions the companies G2 Switchworks, ITA Software, Farelogix, Hitch-Hiker, Ypsilon and Dolphin’. This begs the question whether continued regulation is necessary, appropriate and/or desirable in the current landscape, or whether practices in the industry would be best examined on the basis of the EU competition rules enshrined in Article 101 and 102 TFEU. Competition Enforcement Over Ex Ante Regulation: Should the EU Follow the US Example? 20 This is no longer the preserve of the leisure traveller. In the current economic climate, businesses are also demanding more for their money. 21 For example, para 96 of the Decision indicates that travel service providers, such as the airlines, are capable of forcing GDS providers to lower their prices in exchange for full content, or alternatively, in order to avoid surcharges being applied on the [Travel Agents] they have under contract. Indeed, para 18 of the Decision envisages that an airline can ‘withhold specific content from one GDS and not from another’ to introduce an element of differentiation. Moreover, para 98 of the Decision states that ‘[e]ven in a situation with only three GDS providers none of them will be able to increase prices because [Travel service Providers] will maintain sufficient bargaining power’. 22 According to the European Commission, by 2007 whilst approximately 40% of all airline tickets in the EU were booked through ‘alternative’ distribution channels 60% were booked through the GDSs. 23 Case COMP/M.4523 Travelport/Worldspan.

90  Competition, Regulation and Public Policies Absent specific regulation, in the US there have been a number of contemporaneous antitrust actions involving carriers and GDS suppliers (ie, American Airlines v. Travelport;24 American Airlines v. Sabre;25 and US Airways v. Sabre26). All actions involve allegations of monopolization (abuse of dominance) and anticompetitive agreements. The actions broadly coincide with the scheduled expiry of the GDS-carrier agreements. As such, it is arguable that these are just commercial disputes playing out in US courtrooms. However, the fact that the US Department of Transportation (DoT) is also investigating the GDSs’ conduct suggests that this is more than mere commercial posturing.27 This suggests that the US cases touch upon more profound questions about how airlines sell their content to the traveling public and the GDSs response to the competitive threat posed by alternative distribution methods. In this regard, it is notable that pleadings in the US proceedings point to the existence of Most Favoured Nation (MFN) clauses in agreements between carriers and the GDSs, as well as ‘full content’ or ‘content parity’ provisions. Such agreements may have appreciable market foreclosure effects as they limit a carrier’s’ ability to encourage the use of one GDS over another or the use of alternative providers of airline booking services other than GDSs. In addition, according to filings with the United States Securities and Exchange Commission, Orbitz has an agreement with its controlling GDS, Travelport, whereby Travelport effectively pays Orbitz not to establish a direct connection link any airline.28 As noted above, the issues and the market participants are, essentially, the same at either side of the Atlantic. In such circumstances, is it appropriate (or, indeed, desirable) to have a divergence in approach, particularly in circumstances where regulation in the EU has shown itself as being slow to adapt to significant market changes, including an (apparent) shift in the balance of power from carriers to the GDSs they once controlled? The competition enforcement toolkit appears to fully loaded (and the more appropriate mechanism) to deal with issues of foreclosure; 24 Which follows American’s unsuccessful attempts to establish Direct Connect link with Orbitz, an online travel agent in which Travelport holds a large ownership interest. The conduct at issue in this much publicised dispute involves termination and reinstatement by American of Orbitz ticketing authority, increased GDS booking fees, fees charged by American to travel agents and alleged display bias. 25 Which involves alleged fair bias and increased booking fees. 26 Which follows a distribution agreement struck between the parties in February 2011, which US Airways claims is an anticompetitive arrangement in which it was forced to participate by Sabre. 27 In February 2011, the DoT warned that it had received reports that unnamed GDSs and online travel agents appear to be engaging in display bias which amounts to ‘unfair and deceptive practices’. In March 2011, the DoT outlined what it expects from GDSs and online travel agents to avoid their displays being considered ‘unfair and deceptive practices’. In May 2011, Antitrust Division of the US Department of Justice confirmed sending Civil Investigative Demands (‘CIDs’) to various participants in the industry including carriers and the GDSs, seeking information about the GDS industry. In April 2012, the DOJ broadened its investigation of the GDS industry to include travel agencies. A large U.K. agency with extensive operations in the US indicated reportedly received a CID for ‘an enormous amount of information’ about its relationships with GDS companies (see http://www. travelmarketreport.com/technology?articleID=7155&LP=1). 28 See the Orbitz Form 10K filed with the United States Securities and Exchange Commission in March 2010 and the Orbitz Form 8K filed on 4 February 2011.

Geert Goeteyn 

91  

access; interoperability; innovation; and technological advancement driven by and in response to consumer demand for choice. However, this would require a shift in focus (and policy) by DG Competition. After a decade of concentrating of airline conduct, has the time come to focus more widely on all players involved in the provision of airlines services, including global distribution systems, whose conduct so far may (arguably) have benefited from a gap between regulation and competition?

Jon Stern*

A Market Design for Water Trading in England and Wales: The Role of Water Rights1 Introduction Around the world, the water industry tends to be organised as a large number of small vertically integrated companies. For instance, in Germany, there are over 600 water companies and in Japan over 15,000. Sometimes, as in Germany, water and waste-water (sewerage) services are separate; sometimes they are combined (as in the Netherlands and most of Britain). Countries with combined water and sewerage, like Britain, tend to have larger and fewer, but still vertically integrated companies. Competition in the water supply market is unusual and, in sewerage, it is virtually non-existent other than for sludge disposal. In water supply, the UK is a leader in the development of retail competition. The Scottish Government introduced an open retail market for all non-household customers in 2008 whereby Scottish Water sales to non-household customers are carried out by a separate subsidiary company – Business Stream. This has led to a contestable retail market but, apparently, with only a low level of switching until 2012–2013. A similar retail competition model is planned for England from 2017 but retaining a considerable amount of vertical integration. Upstream competition does not exist in Scotland because it has a single incumbent water company. It barely exists in England, but there are proposals to encourage its expansion – albeit slowly. Upstream water trade (including trade in water licences) is important in promoting competition in water supply – particularly competition in bulk water supply. In general, trade in water (mainly water licenses or water use rights) almost only exists in countries with regular chronic and acute water shortages, eg Australia, California and Chile, but England (at least South and Eastern England) seems to be joining the water shortage group of countries. In Australia, there is upstream competition via a single buyer in Melbourne, Victoria and Queensland. However, trade in bulk water can exist without upstream competition. For instance, there is some bulk raw water trading between * Centre for Competition and Regulatory Policy, City University, London. 1 This chapter is an expansion of the arguments in the briefing note on water rights and water trading in the UK written for the FLJS (Foundation for Law, Justice and Society). That note has been extended to focus more on water trade and competition. I am grateful to the FLJS for permission to use this material in this context. I would like to acknowledge helpful comments from Bettina Lange, Phil Dines and Mel Marquis on earlier versions. I remain solely responsible for the content of the text and the views expressed in it.

94  Competition, Regulation and Public Policies the English water companies, mainly on long-term contracts agreed 50–100 years ago. There is also some limited trade in treated water, primarily in South East and Eastern England. There is considerable water trade in Germany but without water competition at any level. Although upstream water trade can exist without wholesale or retail competition, the development of competition in water supply, particularly upstream competition, requires the expansion of water trade opportunities and this, in turn, requires the clear specification of water rights. Hence, the discussion of upstream water competition inevitably involves a discussion of water trading possibilities and this, in turn, requires a discussion of property rights in raw and treated water.

Property Rights and Water Other than moral philosophers, the people who take ‘rights’ most seriously are lawyers and economists. For legal purposes, rights must be defined and enforceable, while for economists, clear, secure, and enforceable property rights are crucial for the effective operation of markets. In particular, they are essential for the viability of contracts, especially long-term contracts. Such contracts are particularly important for infrastructure and resource industries — of which water is a classic example. Hence, establishing stable water rights is crucial for enabling an effective contractual basis for water trading and competition. Contracts are at the heart of economic activity. Most commercial trading is carried out under contracts and spot market transactions make up only a small proportion of sales, especially of business-to-business sales. For instance, the share of spot markets in coal and natural gas trading has always been very low (around 10 per cent or so) and still remains so. Even for crude oil, where the share of spot trades has been increasing steadily in recent decades, the share of contractbased trade was still over 50 per cent in 2007.2 To sustain the ten-year, twentyyear, or longer contracts that one observes in infrastructure industries requires secure property rights for sellers and buyers so that the long-lived associated investments can be securely financed at an affordable cost of capital. This has been a major theme in the New Institutional Economics, which has explored in detail the requirements for efficient contracts in commercial relationships and the associated sunk investments.

2 See Elitza Mileva, and Nikolaus Siegfried, ‘Oil Market Structure, Network Effects and the Choice of Currency for Oil Invoicing’, European Central Bank Occasional Paper No 77 (2007).

Jon Stern 

95  

Water trading and rights One important point to note in the context of water is that secure property rights do not necessarily require individual ownership. As Elinor Ostrom has emphasized, collective ownership of a resource can work very effectively. An example of this is in New Zealand fishery rights, where fishermen collectively own shares in the main offshore fishery resource pools, shares that can be traded between themselves or with others. For water, ‘rights’ issues occur in various guises. The most obvious concern retail consumers, particularly households and farmers, where major questions arise on whether and how far access to water is (or can be) maintained securely, how the quality of water supplied can be maintained, its terms of sale, and so on. A particular issue for households is whether they are liable to be cut off for not paying their water bills. Whilst these are one important consideration, the focus of this paper will be the role of property rights in water trading. Water trade primarily relates to wholesale water markets in which water companies and others buy and sell bulk water. Such trades are of two main types: (i) trade in abstraction licences (typically within a catchment area or river basin); and (ii) physical trade in bulk water, mainly raw water but also some treated water. But what property rights underpin this trade? In the first instance, abstraction rights are granted by abstraction licences, which are issued by government agencies — in Britain, by the Environment Agency (including the Scottish and Welsh environment agencies). Abstraction rights, owned by water companies and others, provide the foundation for wholesale and retail sales of water which are regulated by Ofwat and Scottish Water Commission supply licences. Abstraction Licence Characteristics in England and Wales In England and Wales, Environment Agency (EA) abstraction licences are required for any agent wishing to take more than 20 cubic metres of water per day from a river, bore-hole, canal, etc. Licences are only granted where additional abstraction would not be environmentally damaging and, since 2003, they have been time limited. Abstraction licences set annual authorised volumes (in cubic metres). Abstractions must be metered with regular reporting of abstracted volumes to the EA. The licence may contain a ‘hands off’ flow (or hands off level) condition which requires the owner to reduce abstraction as specified when river flows fall below a certain level e.g. in drought conditions. Abstraction licence holders have to pay licence fees but these are set so as to cover the costs of the EA abstraction licensing system. Under current law, the EA is not allowed to use these charges to provide economic incentives. This means that abstraction licence charges are relatively low and can be higher in water surplus areas than in water scarce areas.

96  Competition, Regulation and Public Policies The most interesting question around property rights in water concerns ownership of (a) the ‘original’ water (eg, rivers, underground aquifers, etc.); and (b) the extracted water to which the abstraction licences refer — and how these two relate to one another. Traditionally, water has been treated as a ‘commons’ good like air or fisheries in international waters. However, this breaks down when there is excess demand for the resource, as is well known from the literature on the ‘tragedy of the commons’. For water supply, the issue arises when there is insufficient water to meet both human demand (at the ruling set of prices) and to sustain the environment. As will be discussed further below, this problem has arisen to a significant degree in Australia, California, and other drought-prone jurisdictions. It has already become an important issue in the UK, particularly in the South and South East England, and the problems are expected to increase significantly in coming years, particularly towards 2050. In most legal systems, including English law, the ownership of the water to which the abstraction licences apply depends on whether it is river water or underground water. For flowing water (eg, rivers), nobody owns the water in them — it is a commons resource. Instead, riparian rights (reasonable use rights) apply. These are typically associated with land ownership on the river bank and were developed as a legal framework for fishing rights management, water use on the land bordering the river, and similar. At the opposite end of the spectrum, where there is an underground water source on a single property, the water is owned by the landowner. (Note that this is different from hydrocarbon resources like gas and oil which, in the UK, are owned by the state.) Abstraction rights are grafted onto these more fundamental water ownership rights, for example, by ‘grandfathering’.

Climate change and water scarcity Prospective water scarcity is becoming a significant problem in England and Wales. In 2008, the Environment Agency reported that out of 119 water catchments in England, 18 per cent of management units were over-licensed and 15 per cent were over-abstracted. Most of these were in South East England with a few in the South and the Midlands. More recently, the Environment Agency’s 2011 projections report that, not only are current levels of abstraction harming some nature conservation sites and the ecological health of catchments, but that the number and distribution of catchments facing unmet demand could increase significantly by 2050. As a result of climate change, population movements, and other factors, catchments across Wales, South West England and Northern England could face significant unmet demand as well as South East and Eastern England. Perhaps most importantly, the outlook is highly uncertain, with a wide range of possible water demand–supply scenarios both nationally and across the regions.

Jon Stern 

97  

The combination of lower supply, regional changes in demand, and major uncertainties bring out the very sizable potential benefits of increasing water trade. In England and Wales, trade in bulk water is currently much more important than trade in licences (unlike Australia and the Western US states). Traded bulk water accounts for around 4 to 5 per cent of delivered water in England and Wales, but rather more – around 8 per cent – in the South East. Those figures have been stable for many years. Of this traded water, around three-quarters is in raw water (eg, water transported from the Elan Valley in Wales into Birmingham). However, in South East England, almost two-thirds of the traded water is in treated water. For England and Wales, increased trade in bulk water as well as in water licences will be important in responding to the challenges and uncertainties of climate change. This, of course, raises many issues (like interconnection) that go beyond water rights, but the discussion that follows will focus on the relationship between higher trade and better defined property rights in water.3 However, the basic point remains. To reconcile the demand of water for human use and for the environment with available supplies is likely to require a considerable increase in water trade within and between regions — the more so the greater the degree of weather variability and the greater the uncertainties. Where there is significant excess demand, some form of rationing is required. In principle, that can be done either by quantity rationing or by pricing, for example, in a market framework or by some combination of quantity and price regulation. For instance, in the water context, price signals can be used to handle water scarcity for most periods unless and until severe drought conditions arise which require quantity controls to allocate between essential uses. For England and Wales (as in Australia and elsewhere), all the recent reform proposals have relied primarily on price-based methods of handling growing water availability problems — not least to encourage trade and water resilience. The main tool advocated from the Cave Review onwards and reflected in the 2011 Defra (Department for Environment, Food and Rural Affairs) Water White Paper is the development of scarcity-based abstraction prices. However, to make these practical would require major reforms to the abstraction licensing system so as to create well-defined and secure water property rights.

Scarcity-based water abstraction pricing There are a number of possibilities for how scarcity-based abstraction prices might be introduced. One option is to charge abstraction prices in areas of relative water shortage; another option (the main competitor to a simple pricing option) is 3 For a fuller discussion of the wider issues, see Jon Stern, ‘Developing Upstream Competition in the England and Wales Water Supply Industry: A New Approach’, CCRP Working Paper, March 2010; Jon Stern and Jonathan Mirrlees-Black, ‘A Framework for Valuing Water in England and Wales from 2015 Onwards’, CCRP Working Paper, October 2011.

98  Competition, Regulation and Public Policies to redefine abstraction licences in terms of percentage shares of available water and to allow trade in those shares. The first option is problematic for a number of reasons, not least because the prices would have to be set high enough to ensure water sustainability (ie, sufficient water to meet environmental needs such as minimum river flow requirements). In the absence of more interconnection and increased efficiency of water use, these initial prices could be very high — particularly wholesale and retail prices in water-scarce areas — unless there were a long (ten-year or more) transition. Besides these difficulties concerning how the initial prices are set and revised, there would also need to be careful design of (quantity-based) step-in rights at times of serious drought. The simple price option also presents serious challenges in creating secure property rights which arises from the current large overhang of existing and potential excess abstraction rights. This overhang means that if current licenceholders attempted to use a significantly larger proportion of their existing abstraction rights, the result would be a considerable increase in over-abstraction and environmental damage to rivers. In the UK, buy-back schemes have been introduced to address this problem but they are difficult and very slow moving. The second (abstraction shares) option does not have these problems as existing abstraction licences can remain in place in perpetuity, but with redefined rights concerning the quantity of water and the circumstances under which water can be abstracted. A good example of the second option is the proposal by Professor Mike Young to convert water abstraction rights from physical levels (eg, megalitres) into percentage shares of available water. These shares would apply each year to an announced level of available water in the appropriate river basin or other defined area, as declared by the relevant authority (eg, the Environment Agency). These shares could be bought and sold (within competition policy limits) and also traded short- and long-term, as well as providing the basis for water use by abstraction rights owners (eg, farmers). This second option is attractive for various reasons. Firstly, the economic logic behind it is strong — it is a well-targeted approach to the problems of varying degrees of excess demand and major uncertainty. The (annual or semi-annual) announced water availabilities would be set in each area at a level sufficiently low to ensure environmental needs. Household and industrial needs would be met either by available local abstraction and/or by trade, with water companies and others having strong, direct incentives to trade water (either via licence transactions or physical water). In consequence, the resulting water prices, which include the impact of environmental obligations, emerge from observed market transactions. In consequence, they reflect regional and local supply and demand pressures rather than having to be set centrally. Secondly, options of this type have been used in practice and found to be successful in Australia as in the southern connected River Murray system. Interestingly, it is also very similar in design to the New Zealand fishery shares

Jon Stern 

99  

example mentioned above, which has also worked very well. These fisheries have been in secular decline but with considerable (unforeseeable) year-to-year variability. Defining perpetual shares in available fisheries has provided the basis for this system and has helped stabilize and renew fish stock availability.4 Finally, this option would (if adopted) allow the continuation of existing abstraction licences, and it would not require buy-back of excess allocations as a precondition. Hence, it provides scarcity-based prices for existing water as a basis for trade both in licences and in bulk water, while also providing secure property rights. Available water volumes may change year-to-year, but this is known in advance — as should be the rules by which volumes are set. The underlying licences and the water shares within them do not need any time cut-off or subsequent imposed reallocation. In this way, the redefinition of water rights from physical volumes to percentage shares should allows ‘grandfathering’ of current rights without the need to renegotiate or terminate existing property rights. This should provide a strong set of incentives for efficient water trade within secure markets. Creation of such property rights commodifies water ownership and use. Other than the marginal uses, such as river fishery and similar, the water commons is converted into a secure property right which can support (and promote) trade. This follows the example of forests, grazing land, and many other cases. Even policymaking concerning the air we breathe — the ultimate commons good — is moving in this direction, with taxes or tradable permits being applied to emissions of sulphur dioxide, nitrous oxide, and now to carbon gas emissions.

Market-based versus quantity-based water rationing Why is a market-based property rights option for water in general preferable to quantity-based rationing? The most persuasive answer is that any quantity-based allocation system would require massive amounts of information for the initial allocation of abstraction rights. There would almost certainly be major arguments on their initial distribution – ie who should receive what volume of rights and why. (Note that this is also true of the pure abstraction price option discussed above, which is another reason why the ‘tradable shares’ option is superior.) In other circumstances, (eg airport slots), ‘grandfathering’ of rights is the effective starting point, but this may not be appropriate for water rights where the future supply of water is becoming increasingly variable. Indeed, planning solutions become increasingly less attractive as variability increases — particularly unpredictable variability which, for water availability, 4 See Ragnar Arnason, ‘A Review of International Experience with ITQs’, CEMARE Report No 58 (2002).

100  Competition, Regulation and Public Policies generates major uncertainty regarding future supplies. To this must be added the incentives and likelihood of corruption, from black markets through to high-level bribery. Of course, plan allocations can be combined with ‘white markets’, but, in that case, we have effectively reverted to a market- and price-based model. Market sceptics may express concern that essential users would be priced out of the market the more that wholesale water prices reflect scarcity conditions, yet a market-based approach offers a number of benefits over a quantity-based allocation system. Firstly, scarcity-based wholesale (and retail) water prices provide strong incentives for the efficient use of water — and more importantly for the development of efficient water-using technologies, as has been the case with the development of high-efficiency irrigation equipment industry in Australia. Furthermore, scarcity-based water prices help reduce and/or reallocate high water use activities away from water-scarce areas. Neither of these outcomes can readily be achieved in a quantity-based allocation system. In addition, in a market-based system, reserved allocations can be set aside for truly essential uses. This means that, at least outside emergency drought circumstances, the groups with identified priority needs have a guarantee of supplied water, irrespective of variations in the availability of raw water (eg from rainfall variability). Finally, a quantity-based allocation system will require periodic reallocations. The consequence that water ownership property rights would become much more insecure, with all the associated adverse implications for trade, investment, and efficiency growth in the water industry and among water users. For all of these reasons, market-based abstraction allocations based on percentage shares of available water are strongly preferable both to quantity-based systems as well as to market systems operating via pre-fixed scarcity-based abstraction prices for physical volumes of water.

Policy proposals Because of these advantages, the ‘water shares’ approach to water pricing and trade, as advocated by Mike Young, myself, and others5has become arguably the leading option for abstraction reform, with secure property rights to foster trade and innovation through a form of scarcity pricing. The proposal is primarily intended to encourage local/regional abstraction licence trading rather than bulk water trade, but the prices set in the water licence market will provide marker prices for physical inter-area trade in bulk water. This approach also provides incentives for regional water network planning and organization, regional wholesale markets, and (most importantly) interconnector investment. 5 See Mike Young, ‘Towards a Generic Framework for the Abstraction and Utilisation of Water in England and Wales’, UCL Environment Institute Report, 2012. See also Stern and Mirrlees-Black, ‘A Framework for Valuing Water’, cited above note 3, Annex A.

Jon Stern 

101  

In conclusion, it is clear that a redefinition of water ‘rights’ related to water scarcity is crucial for an effective policy towards water in England and Wales, particularly given the uncertainty caused by climate change. In this paper, the key underlying issues have been discussed and a specific approach has been put forward. Abstraction licence reform of the tradable water shares type proposed effectively severs the link between water ownership and water use. Riparian rights can continue for fishing, cultivating water plants, and similar uses, but the link to irrigation and commercial uses of water is broken and the scope of water commons is thereby greatly reduced. In consequence, an abstraction mechanism of this type can make a major contribution to commodifying the use of water while ensuring sufficient water for ‘the environment’. By providing secure property rights and ensuring that trade reflects secure water property rights and water trading are not just reconciled but can reinforce one another. Establishing such a system will, in turn be crucial if there is to be effective development of water competition, particularly upstream competition.

Rafael Allendesalazar*

Does Merger to Monopoly Become Efficient When it Refers to Sector Regulators and the Competition Authority?1

In February 2012, shortly after taking office, the new Spanish Government announced that it would merge (almost) all the sector regulators in Spain, together with the National Competition Authority, within a single body, the so-called National Commission for Markets and Competition (CNMC, ‘Comisión Nacional de los Mercados y de la Competencia). As of this writing, an initial draft Act creating the CNMC has been unofficially circulating on the internet and has been submitted to the affected bodies for their reports. The national regulatory authorities (NRA) that are due to merge into the new CNMC are: – the National Competition Authority (CNC); – the energy regulator (CNE); – the telecoms regulator (CMT); – the authority for audiovisual media services; – the postal regulator; – the railway regulator; – the airports regulator. The draft also provides for the disappearance of the gambling regulator. Most of these regulators had been created recently and had not effectively initiated their activities (media services, railways, airports and gambling). Thus, in practice the merger will mainly affect three high-profile regulators – the highly visible and often controversial CNC, CNE and CMT – and the less visible postal regulator. The Central Bank, the Stock Exchange and the nuclear energy regulator will remain independent. Others, such as the NRA for tobacco markets, seem to have been forgotten. It is also worth noting that, as a result of Spain’s quasi-federal structure, it is unclear to what extent the autonomous regions will decide to follow suit; if they do not, the future CNMC could continue to coexist with regional sector-specific regulators and Competition Authorities. * Partner at Martínez Lage, Allendesalazar & Brokelmann. 1 At the time of the Workshop in July 2012, the Spanish Government had only just announced its intention to merge the regulators and the competition authority. This chapter therefore discusses the project at its early stage. However, a postscript to the chapter describes the main features of the ‘Superregulator’ that was later established under Spanish law.

104  Competition, Regulation and Public Policies One main feature of the proposal is that all the functions and competences which are presently exercised by the regulators and which are not explicitly attributed to the CNMC will revert to the corresponding ministries. The project has been widely criticized. As a matter of fact, it has gained very little explicit support. Let us see whether these criticisms are well founded.

1. The goals of the proposal are unclear; lack of transparency in the procedure prior to the adoption of the draft Although many companies and practitioners would agree that some fine-tuning in the present functioning of the CNC and of the sectorial NRA could improve their processes and results, there is a practical unanimity in considering that the Government has acted with haste and lack of transparency when adopting its proposal. When the current Competition Act was adopted in 2007, the Competition Authority previously published a White Paper explaining the reasons for a change in the prior Act. This White Paper and the following initial draft of the Act were both submitted to an extensive process of public consultation that resulted in the amendment of the draft even before being sent to the Parliament. Thus, when the new Competition Act was finally approved in 2007, it received the unanimous support of all major political and social forces. By contrast, the proposed legislation creating the CNMC is due to be adopted by a Parliament having an absolute majority of the Government’s party, and without any previous public debate as to its appropriateness and reasonability. Moreover, the Government has provided almost no explanation of the motives behind the crucial changes that it promotes. The Government argued that the merger would favour the achievement of three goals: – To avoid conflicting decisions and increase legal certainty When several bodies (NRA, NCA, European Commission and national courts) can scrutinize the way undertakings compete in the market, and apply different sets of legislation, it is almost inevitable that the decisions of these bodies can differ in specific cases. Thus, the risk of conflicting decisions and the possibility of forum shopping should be considered as intrinsic to a decentralized application of such legislations. Furthermore, such discrepancies may prevent a petrification in the application of the law, and will sometimes facilitate a desirable evolution in the criteria with which such legislations are being applied. This being said, stakeholders unanimously agree that the avoidance of conflicting decisions and the increase of legal certainty are goals that should be generally pursued. Of course, merging all the bodies that apply such legislation reduces the risk of conflicting decisions. But those goals can also be attained through less intrusive measures, such as procedural reforms favouring a greater interaction

Rafael Allendesalazar 

105  

between NRAs and the NCA during the process leading to the adoption of ex ante sector regulations and ex post antitrust decisions. In any case, the creation of a single authority at a national level will not necessarily guarantee a consistent application of multiple regimes, since differing decisions can reflect a conflict between the NRA and DG COMP, as illustrated by the Deutsche Telekom and Telefónica cases. – To allocate scarce budgetary resources more efficiently Merging the NRAs could allow for the reduction of redundant personnel and facilities, while the remaining administrative staff would benefit from sharing their insights and experience across industries. The Government has estimated that the merger could save around … 4 million euros. This amount is so minute that it would seem to be a mere slip of the tongue. Furthermore, we should not only look at the savings in the operating costs of the merged NRAs and the NCA; we should compare this reduction in costs with the loss of output and positive externalities that will result from an inevitable — if only temporary — reduction in the enforcement of competition and regulatory laws while the future CNMC adjusts its internal procedures and becomes fully operative. Although we have no figures to carry out this comparison, it seems safe to guess that the cost of even a short period of under-enforcement in the telecoms or energy markets (not to mention a diminution in the enforcement of competition law across all sectors of the Spanish economy) would easily outweigh the estimated savings in costs. Furthermore, we could also assume that this merger will not be an exception to the golden rule that applies to most mergers of private companies, ie that efficiencies tend to be overestimated while unforeseen inefficiencies slash the expected benefits of the merger. All in all, the goal of the reform should be to improve the intervention by NRAs/NCA in the market to maximize consumer/total welfare. Budgetary savings should remain a criterion in opting for one or another model, but it should not as such be the goal of the reform. – To follow suit with an (alleged) tendency in other countries It has been pointed out that this latter reason, the alignment with other countries, seems a manifest overstatement. Among the main European countries, only Germany and The Netherlands might follow a similar institutional scheme. And it must be noted that in Germany, while there is a multi-sectoral NRA (the Bundesnetzagentur), the Bundeskartellamt remains completely independent. With regard to The Netherlands, the NMa is already both the competition authority and the NRA for energy and transport, and it is in the process of merging with the telecoms and postal regulators (OPTA) and the Agency for the protection of consumers; but the transition from the previous model to the single Authority has been extremely cautious since it began in 2011. Unfortunately, it does not seem that the Spanish Government will follow the prudent Dutch approach, as it is foreseeing an almost immediate transformation of the institutional model.

106  Competition, Regulation and Public Policies Furthermore, since the CNMC will be a ‘première’ within our neighbouring countries, the Spanish Government will not be able to analyse how similar institutional schemes have worked in practice, and to replicate their strong points and elude their weaknesses. In view of the lack of transparency in the adoption of the Government’s proposal and the absence of a consistent explanation to such a drastic revolution in the regulatory institutional framework, some critics have mentioned two spurious alternative explanations. First, they have pointed out that the reform seems to have been triggered by a report recently prepared by PWC on behalf of Telefónica; and since the latter company has been one of the most prominent and unfortunate ‘clients’ of both the CMT and of the antitrust authorities, they have cast doubts on whether the report’s real objective is to improve the application of sectoral and antitrust regulations. It has also been indicated that the Popular Party now in Government strongly criticized some of the appointments made by the previous Government to the NRA and the CNC, and that the reform is in fact a ‘Hungarian way’2 of replacing those appointees.

2. The CNMC is likely to be less independent The draft law that would create the CNMC enunciates, from among the functions that are presently exercised by each of the sectorial NRA, which of these functions will remain within the CNMC and which will revert to the corresponding ministries. Furthermore, it establishes the principle that all the functions not explicitly attributed to the CNMC will revert to the Administration. Finally, the draft law also provides that certain financial resources that were being levied by the sectoral NRAs have also been transferred to the corresponding ministries. The sectoral NRAs, and in particular the CMT and the CNE, have published very detailed reports explaining why these changes will undermine the effectiveness of the future watchdog and its independence with regard to the Government and the Administration. Furthermore, they have indicated that many such changes would be contrary to the European Directives that require the independence of the regulator for electronic communications networks and services (Directives 2002/21/EC (framework Directive); 2002/19/EC; 2002/20/EC; 2002/22/EC; and 2002/58/EC) and contrary to the Directives concerning the common rules for the internal market in electricity (Directive 2009/72/EC) and in natural gas (Directive 2009/73/EC). In this respect, suffice it to say that the Chairman of the BEREC (the Body of European Regulators for Electronic Communications) recently expressed his 2 The European Commission has initiated accelerated infringement proceedings against Hungary for measures considered to endanger the independence of the Central Bank, the Data Protection Authority and the judiciary (see Press release IP/12/24 of 17 January 2012).

Rafael Allendesalazar 

107  

concern regarding the independence of the future Spanish NRA. More tellingly, in its Assessment of the 2012 national reform program and stability program for Spain, the European Commission expressly stated that ‘the current draft Law that creates the CNMC does not guarantee that it will carry out its regulatory activity in an effective and independent way’.3

3. The CNMC is bound to be less effective, at least for a certain time The draft law creating the CNMC says very little about its practical functioning. It only regulates the existence of a decisional board, the ‘Council’, with nine members, appointed by the Government, for a non-renewable six-year period, that will elect amongst its members a President and a Vice-president, both for a non-renewable 3-year period. (Nothing, however, prevents a Putin/Medvedev type of situation.) It has been indicated that this institutional design is incompatible with having, permanently on the board of the CNMC, a set of experts with sufficient knowledge about competition law and about the specificities of all the regulated sectors to decide on these complex issues. Furthermore, in contrast with what is happening in The Netherlands, where the transition to a single regulator will last for almost two years, the draft Spanish law considers that the merger is due to take place immediately after the approval of the law. This completely overlooks the complexities of such a process. It is thus feared that the merger will inevitably be followed by a period of reduced and inefficient enforcement both of regulatory legislation and competition law in Spain, coinciding precisely with a critical period where the Spanish economy desperately needs all the impetus of the CNC and the NRAs to restore the competitiveness of its economy.

Outlining a possible alternative reform 1. Reducing the number of NRAs In Spain we have seen in recent years a proliferation of NRAs. Of the seven NRAs due to merge in the draft law, four were created in the last couple of years and have not yet started to work effectively (specifically: the NRAs for audio-visual media services, railways, airports and gambling). Furthermore, several regions have 3

COM(2012) 310 final, http://ec.europa.eu/europe2020/pdf/nd/swd2012_spain_es.pdf.

108  Competition, Regulation and Public Policies created their own regulatory and competition authorities. Against this background, a sound rule would be to avoid the proliferation of agencies that can have a variety of negative effects: the dissipation of expertise; the loss of economies that a single entity would have when performing related tasks; the greater need for and costs of coordination; and additional complexities for the undertakings concerned. Thus, it is certainly worth revisiting the soundness of this proliferation of entities. This re-examination should essentially ask three questions: – Given the specificities of each of these sectors, does their proper functioning require the existence of an independent NRA or can it be effectively managed through a combination of classical governmental regulation and competition enforcement? – Do these markets have a regional dimension and present all the characteristics of a distinct market within the national market? If not – and we are inclined to believe that the response would be negative – the existence of regional regulatory and competition authorities is unjustified and inefficient. – For those sectors where the existence of an NRA seems justified, should the regulator be an industry-specific, sector-wide or multi-industry agency? In this respect, commentators have identified that multi-industry regulators have specific advantages: sharing resources; facilitating learning across industries; reducing the risk of industry and political capture; reducing the risk of economic distortion among competing industries; and dealing with blurring industry boundaries. Thus, in sectors where it is possible to identify sufficient commonalities in the market structures and conditions, and where the regulator would have similar goals and means, creating a multi-sector regulator or merging the existing ones may be efficiency-enhancing. By contrast, where a sector has very specific features and the regulator is required to perform functions unique to that sector, an industryor sector-specific regulator will be preferable. (For example, assuming that a regulator is required for the media service sector and that its primary objectives include insuring the plurality of the media and their respect for privacy, what would be the efficiency gains in merging this regulator with, for instance, the transport regulator?)

2. Keeping the NRAs and the NCA separate Although they share a common goal, regulators and competition authorities are essentially different animals. They usually have a different legislative mandate, and their perspective regarding competition may also differ: – an NCA seeks (or should seek) to protect the competitive process, whereas sector regulators may need to temporarily protect nascent inefficient competitors.

Rafael Allendesalazar 

109  

– NRAs tell the agents what to do in the market, whereas an NCA tells them what they should not do. It is within the basic DNA of any NRA to adopt ex ante technical and economic regulations for a specific sector, whereas an NCA must be equipped to respond swiftly ex post to competition restrictions affecting all industries, but it generally does not need to adopt regulations. – Technical and economic regulations by NRAs generally address long-term structural issues, whereas enforcement by an NCA is usually short-term and behavioural. – NRAs require ongoing monitoring capabilities and sector-specific expertise, whereas an NCA will normally only intervene where a competition infringement is detected, and when it does intervene it should be able to rapidly acquire the necessary knowledge of the market, irrespective of the sector affected. – an NRA’s main activity is to regulate, and enforcement of this regulation could be accessory; it could even be done by other administrative bodies and courts (although this may not be very efficient). An NCA essentially focuses on enforcement of competition law, as shown by the level of fines imposed, and due process is essential in this context for effective enforcement. The resources (which are by definition scarce) available to the NRAs and the NCA, and their procedural framework, should take account of such differences. We therefore believe that the Spanish NCA and the NRAs should remain separate but complementary agencies. For this complementarity to be fully achieved, and for both types of agencies to properly perform their respective goals, cooperation between the NCA and the NRAs is indispensable. The reform should seek to favour greater interaction between the NRAs and the NCA during the process leading to the adoption of ex ante sector regulations and ex post antitrust decisions. The NRAs should be required to intervene in cases where the NCA intends to apply competition law to an undertaking subject to sectoral regulation, and vice versa, the NCA should intervene when an NRA intends to adopt a regulation or a specific decision having a direct impact on the competitive conditions within a regulated sector. In cases of merger control, a system of double vetoes could also ensure that the goals of both sets of legislation are taken into account. Finally, it could also be possible to reserve a number of seats within the decisional body of the NCA for members of the NRA’s decisional body to decide on cases involving regulated industries. And again, vice versa, the NCA could be involved when an NRA adopts regulations or individual decisions having a significant impact on the competitive structure or process in the relevant sector.

110  Competition, Regulation and Public Policies

3. Handling changes with caution Commentators have pointed to the fact that creating ex novo a multi-sectoral regulator is usually easier than merging pre-existing agencies. Thus, any decision to merge should provide for a smooth transition to avoid the welfare cost of a regulatory lag.

POSTSCRIPT As a result of the numerous criticisms and comments to the initial draft, the Government finally decided to introduce some important modifications in order to try to tackle these reproaches, particularly those expressed by the European Commission. Act 3/2013 creating the National Markets and Competition Commission was then approved by the Parliament on 4 June 2013. Only four months later, in October 2013, the new CNMC — often referred to as the ‘Super-regulator’ — was formally incorporated, and the former regulators and the National Competition Authority were extinguished. The main amendment introduced in the Act relates to the division of functions within the CNMC, which has resulted in a modification of its structure. The CNMC retains as its general function the preservation and promotion of competition in all sectors, together with its specific supervisory and monitoring functions across six markets or sectors. But instead of acting as a fully integrated authority, the Council of the CNMC, which will operate as the decision-making body, has been divided in two chambers: the Competition Chamber is devoted solely to competition enforcement, while the Regulatory Chamber is devoted to regulatory files. Each Chamber may issue an opinion on the files decided by the other. Both Chambers have five different members, who are appointed by the Government but must be examined by the Parliament before taking office. The President of the Council and of the CNMC is the President of Competition Chamber, while the Regulatory Chamber is chaired by the Vice-president of the Council and of the CNMC. The members of Council are appointed for a six-year period, and cannot be re-elected. It is foreseen in the Act that the members can rotate from one Chamber to the other. If this rotation finally takes place, it could facilitate the unification of the criteria of the Chambers. However, it may also prevent the members of the Chambers from going through their entire learning curve before changing to the other Chamber. The CNMC has four Directorates of Investigation (Direcciones de Instrucción): the Competition Directorate, the Telecommunications and Audiovisual Sector Directorate, the Energy Directorate, and the Transport and Postal Sectors

Rafael Allendesalazar  111   Directorate. Following an amendment introduced in the final Act, the Directors are appointed by the Council of the CNMC, and not by the Government. Resolutions can be adopted by either Chamber or by the Plenary. The Plenary, in which the President has a casting vote, will assume the role of unifying criteria and will be competent to deal with cases in which a divergence of interpretation arises between the Chambers, and in matters with a special impact on the competitive functioning of markets or on activities subject to supervision (the latter being subject to a qualified majority, and following a proposal of the President or of three members). The adoption of resolutions by the Plenary has already taken place in several cases. The CNMC, in its Resolution of 21 January 2014, S/0373/11 Correos 2, fined the Spanish Postal Service incumbent (Sociedad Estatal Correos y Telégrafos, S.A.) for an abuse of its dominant position, and due to its special impact on the competitive functioning of the affected market this Resolution was adopted by the Plenary. With regard to concentrations, the Plenary has also recently authorized, in ‘phase I’, concentration C/0527/13 Abertis/Activos Telefónica/Activos Yoigo in the telecoms sector. A second important amendment introduced in the Act refers to the allocation of powers previously attributed to the NRAs, some of which have now been returned to the Administration. Whereas the draft Act had foreseen that all functions not explicitly attributed to the CNMC would revert to the Administration, it was finally established that all the functions not explicitly attributed to the competent ministries by the Act are to be exercised by the CNMC. The Act mainly establishes the new institutional framework, and does not modify the legislative content in the areas of competition and regulatory law. In particular, the Act attributes all of the functions previously entrusted to the CNC to the CNMC. In addition, there has been a realignment of the functions traditionally implemented by the NRAs, between the CNMC and the relevant ministries. According to the Explanatory Memorandum of the Act, the ministries will assume all administrative functions for which a special independence is not required, as well as tasks of limited utility in achieving the objectives of the CNMC. With regard to telecoms, the Act attributes all functions of the former telecommunications regulator to the CNMC, with the exception of the collection of certain taxes, which will be attributed to the Ministry of Industry, Energy and Tourism. However, in the audio-visual field the Act confers on the same Ministry certain functions such as collecting the statements of initiation of activity or carrying the Registry of providers of audio-visual services. Other functions in this field are also attributed to the Ministry of Presidency. In energy matters, the Act also transfers to the Ministry the functions of inspection, initiation and conduct of certain enforcement proceedings and responding to claims made by consumers amongst others. The Ministry also retains the power to examine the taking up of stakes in the energy industry, without prejudice to the CNMC’s scrutiny of concentrations. The CNMC has expressed its concerns that it is now, among the various European telecoms regulators, the one with the least functions.

112  Competition, Regulation and Public Policies In addition, and beyond the scope of the creation of the CNMC, the new regulation of the General Telecommunications Act enables the CNMC, if it reaches the conclusion that competition concerns or market failures persist despite other measures applying to wholesale markets of access products, to require vertically integrated operators with significant market power (as an exceptional measure) to transfer activities related to the wholesale supply of access products to an independently operating business unit, which will provide access products and services to all enterprises on non-discriminatory terms. It should be noted that the CNMC has retained a function which is highly symbolic and which can facilitate the eradication of public restrictions of competition: the CNMC has locus standi to appeal administrative decisions and regulations that may impair competition. In fact, the former NCA used this function in a couple of cases. Some months after the approval of the Act, the CNMC was also granted legal standing to appeal against administrative decisions and regulations that could jeopardize the unity of the Spanish market.4 This has been a controversial issue, as some regional governments considered that this could jeopardize their competence as regards local affairs. However, the grant of standing has been welcomed by the business community as a means to eliminate barriers to a single Spanish market due to the proliferation of regional market regulations. To conclude, the creation of the new CNMC has been a very controversial decision that initially attracted the criticism of most commentators and practitioners based, not only on the difficulties of a model based on a unified authority, but also on the impetuous transition from one model to the other and the hasty extinction of the former NRAs. Furthermore, some of the appointments and some of the initial measures adopted by the CNMC have generated intense public polemics. This said, we should now all work for the success of the new institutional structure. This will surely require, on behalf of the CNMC, extracting the synergies derived from the unification while remembering that competition and regulation pursue the same final goal — maximizing consumer welfare — albeit through different means. The CNMC should also be open to listen to the opinion of the ‘operators’ that have been applying, actively or passively, the competition and sectorial rules related to the markets where the CNMC will act. And these ‘operators’, many of whom have been critical of the creation of the CNMC, should now look ahead and make constructive proposals to improve the functioning of the Super-regulator. Indeed, we must all bear in mind that the quality of the new institution and its ability to ensure healthy competition in all the markets, particularly in those sectors subject to specific regulation, will be the main pillars on which to build the recovery of competitiveness and the growth that the Spanish economy so badly needs.

4

Act 20/2013 on market unity.

Suzanne Kingston*

Competition and Environmental Protection: A Case of Ne’er the Twain Shall Meet?

1. Introduction The interaction between the EU’s competition and environmental protection policies is a topic largely neglected in policy documents as well as in the legal and economic literature to date.1 This is unsurprising for at least two reasons. First, the degree of overlap between the two policy areas may not be immediately obvious, at least without a familiarity with both policies. Second, recent years have seen the emergence in certain quarters of a perception that EU competition policy must in principle be insulated from non-competition or public policy concerns. This perception is often closely linked to the now-orthodox idea that the goal (or at least an important goal) of EU competition law is the enhancement of consumer welfare.2 Accepting this idea necessarily implies, for some, that environmental goals are extraneous to competition policy, and should be left to be implemented by the legislator via environmental legislation.3 By this view, competition policy * Suzanne Kingston is a barrister practising at the Dublin bar and a senior lecturer in law at University College Dublin. This piece updates, and draws on, some of the arguments in Kingston, Greening EU Competition Law and Policy (Cambridge University Press, 2012), to which the reader is referred for fuller detail. The useful comments of the participants at the Competition Workshop of 13–14 July 2012 are gratefully acknowledged. 1 For some notable exceptions see F O W Vogelaar, ‘Towards an Improved Integration of EC Environmental Policy and EC Competition Policy: An Interim Report’, in Barry Hawk (ed), International Antitrust & Policy: Fordham Corporate Law 1994, Juris Publishing, 1996, pp 529 et seq; Damien Geradin, ‘EC Competition Law and Environmental Protection’, 2 Yearbook of European Environmental Law 117 (2002); Vincenzo Baccaro, ‘Collecte et recyclage des batteries usagées en Italie: droits exclusives, exigences environnementales et droit de la concurrence’, (2001/1) Competition Policy Newsletter 39; Hans Vedder, Competition Law and Environmental Protection in Europe: Towards Sustainability? (Europa Law Publishing, 2003); Giorgio Monti, EC Competition Law (Cambridge University Press, 2007) pp 91–94; Kingston, Greening EU Competition Law and Policy, cited above. 2 See, eg, Case T-168/01 GlaxoSmithKline Services Limited v Commission [2006] ECR II-2969, para 118 (protection of consumer welfare as the objective of Article 101 TFEU); but see the divergent statements of the ECJ on appeal in Joined Cases C-501/06 etc. [2009] ECR I-9291 (protection of consumer welfare and of market structure). 3 See, eg, the approach of Giuliano Amato in Antitrust and the Bounds of Power (Hart Publishing, 1997) p 117; Okeoghene Odudu, The Boundaries of EC Competition Law: The Scope of Article 81 (Oxford University Press, 2006) Chapter 6 and p 7; Hans Maks, ‘The “New” Horizontal Agreements Approach in the EU: An “Economic” Assessment’. 37 Intereconomics 28, 28 and 32–33 (2002); Gerwin Van Gerven, ‘The Application of Article 81 in the New Europe’, in Barry Hawk (ed), International Antitrust Law & Policy: Fordham Corporate Law 2003 (Juris Publishing, 2004) pp 415 et seq, at 429–431.

114  Competition, Regulation and Public Policies should focus on maximising economic efficiency: its effectiveness as a policy area will be blunted if environmental goals are added to the mix. This paper seeks to challenge this assumption. Section 2 highlights the increasing interrelation between competition and environmental policies due, inter alia, to the current environmental policy trend of using market-based instruments to achieve environmental protection goals. Section 3 puts forward four arguments about what the role of environmental protection in EU competition policy should be. Section 4 considers the present attitude to environmental protection in EU competition law, taking a variety of examples from antitrust and State aid law. Section 5 offers some brief conclusions on the institutional and substantive implications of this approach.

2. The modernisation of environmental protection policy: using the market to achieve environmental protection goals Traditionally, environmental policy has favoured direct, command-and-control regulatory instruments, with rules set down by the State (legislators or regulators) and enforced by the State (prosecutors, regulators or courts). As with many other policy areas, however, the (Thatcherite/Reaganite) shift towards neoliberal ideology that took place in the 1980s brought with it a deep-seated belief in market values, not only in economic life but also in social and environmental matters classically viewed as falling within the exclusive purview of the State. In the environmental field, it led to a fundamental reassessment of the appropriate regulatory tools in this area, and to widespread efforts in many jurisdictions to harness market forces for achieving environmental protection goals. This resulted in the embrace, in a large number of jurisdictions worldwide, of market-based environmental regulatory instruments, broadly defined as instruments where market mechanisms are employed to provide incentives to guide behaviour towards an environmentally favourable outcome.4 Such instruments range from environmental taxes, to environmentally-motivated State subsidies, to the creation of markets in the right to pollute (tradable permit systems), to encouraging market actors to behave in a greener manner voluntarily (eg, by adopting voluntary environmental agreements).5 Thus, for instance, the State may hold off on adopting strict(er) environmental regulatory standards in circumstances where an effective voluntary environmental agreement exists within an industry.6 4 See generally, Bruce Ackerman and Richard Stewart, ‘Reforming Environmental Law’, 37 Stanford Law Review 1333 (1985); Richard Stewart, ‘The Importance of Law and Economics for European Environmental Law’, 2 Yearbook of European Environmental Law 1 (2002). 5 See, for instance, the voluntary eco-design industry agreement for set-top boxes likely shortly to be approved by the Commission: http://env-ngo.eup-network.de/fileadmin/user_upload/Produktgruppen/ Lots/Working_Documents/Explanatory_doc_07_05_2012.pdf. 6 See, for instance, the EU’s initial reliance on voluntary environmental agreements to set carbon emission standards for passenger vehicles within the EU; in particular, the 1998 agreement between

Suzanne Kingston 

115  

While the EU was somewhat slower to embrace market-based environmental instruments than the US, it has certainly now done so, albeit with its own distinctive take on the appropriate balance between regulation and market forces in achieving environmental goals.7 In line with this, the Commission has stated that ‘market-based instruments and fiscal policies in general will play a decisive role in delivering the EU’s policy objectives’8 as a means of achieving greater levels of environmental protection (by reaching the parts that direct regulation cannot reach) in the most cost-effective way for society. This win-win message is evident in a multitude of high-level EU policy documents in recent years, not least the EU’s flagship Europe 2020 strategy, one of the key messages of which is the need for sustainable growth as a means of increasing the EU’s global competitiveness. In this sense, it is undoubtedly the case that the EU’s environmental policy has been profoundly affected by the EU’s economic policies and, in particular, the EU’s efforts to define the balance that should be struck between the market and the State in its distinctive ‘social market economy’ (Article 3(3) TEU). Such efforts lie at the heart of any consideration of how EU competition policy should approach environmental protection goals.

3. Should environmental goals have any relevance in competition policy? 3.1 Welfare, efficiency and political choices Naturally, any consideration of the role of environmental protection in EU competition policy depends on one’s view of the goals of that policy, a debate that has raged long and hard for decades.9 Without attempting to rehearse that the European Commission and the European Automobile Manufacturers Association (ACEA). See also the related agreements between the Commission and the Japanese and Korean Automobile Manufacturers Associations. These agreements did not transpire to be effective from an environmental policy perspective, and were subsequently replaced by EU-level centralised regulation, Regulation (EC) No 443/2009 of the European Parliament and of the Council of 23 April 2009 setting emission performance standards for new passenger cars as part of the Community’s integrated approach to reduce CO 2 emissions from light-duty vehicles, 2009 OJ L140/1. 7 The centrality of market-based instruments to modern EU environmental policy became evident as far back as 1992, when the Community’s Fifth Environmental Action Programme (1992–2002) identified market-based (or, in its terminology, economic) instruments as central to its regulatory strategy over the next 10 years. This emphasis was continued in the Sixth Environmental Action Programme (2002–2012). 8 2007 Green Paper on Market-Based Instruments, COM(2007)140. 9 See, eg, David Gerber, Law and Competition in Twentieth Century Europe: Protecting Prometheus (Clarendon Press, 1998) Chapter XI; Rein Wesseling, The Modernisation of EC Antitrust Law (Hart Publishing, 2000) p 96; Massimo Motta, Competition Policy: Theory and Practice (Cambridge University Press, 2004) Chapter 1; Laura Parret, ‘Shouldn’t We Know What We Are Protecting? Yes We Should! A Plea for a Solid and Comprehensive Debate about the Objectives of EU Competition Law and Policy’, 6 European Competition Journal 339 (2010).

116  Competition, Regulation and Public Policies debate, it is vital to unpack the premises underlying the common assumption that environmental protection, as a non-economic public policy consideration, is irrelevant to competition policy. Certainly, if one were to embrace what we might call a Borkian Chicago School model of competition theory, one must readily conclude that there is no role whatsoever for environmental factors in competition policy. An inherent part of this model is that all goals other than consumer welfare (in the sense of allocative and productive10 efficiency) must be excluded from competition policy. Allocative efficiency in this sense is at its greatest when society’s economic resources are allocated so that consumers can satisfy their wants as fully as technological constraints permit.11 Fundamentally, however, this approach is embedded in deeply political choices rooted in a specific (deregulatory and highly conservative) neo-liberal vision of the function and independence of the market. Against this background, it is evident that the greatest of care must be taken before adopting this approach in the EU context. It is worth recalling that part of the reason why the EU never embraced the Chicago ‘revolution’ in the way US antitrust policy did was because the EU’s competition policy was always, as inspired by ordoliberalism’s Ordnungspolitik, conceived of as just one of the parts of the EU’s economic constitution.12 Far from being an end in itself, competition policy was a component part of a larger system of economic constitutional law set out in the Treaty of Rome. In this sense, the Member States’ best efforts to define the particular economic model strived for by the EU are at present represented in Article 3(3) TEU, the post-Lisbon version of which reads: The Union shall establish an internal market. It shall work for the sustainable development of Europe based on balanced economic growth and price stability, a highly competitive social market economy, aiming at full employment and social progress, and a high level of protection and improvement of the quality of the environment.

Clearly, the meaning of many of these goals is unclear and admits of no precise definition. What is undeniable, however, is that the Member States’ vision is of an economic model intertwined with certain fundamental values of European society, namely social progress and environmental protection. The vision of a social market economy bears obvious resemblance to the continental idea of a soziale Marktwirtschaft so central to the economic models of many Member States.13 Whilst the conceptions of a social market economy are many, and vary significantly by State, a defining feature of this economic model is the idea that The inclusion of productive efficiency is of course also itself contested. See, eg Robert Bork, The Antitrust Paradox, Basic Books, 1978. 12 See, eg, David Gerber, ‘Constitutionalising the Economy: German Neo-liberalism, Competition Law and the ‘New” Europe’, 42 American Journal of Comparative Law 25 (1994); Hans Grosskettler, ‘On Designing an Institutional Infrastructure for Economies: The Freiburg Legacy after 50 Years’, 21 Journal of Economic Studies 4 (1994); Christian Joerges, ‘A New Alliance of De-legalisation and Legal Formalism? Reflections on Responses to the Social Deficit of the European Integration Project’, 19 Law and Critique 235, 237 – 241 (2008). 13 See, eg, Joerges, ‘A New Alliance’, cited previous footnote. 10 11

Suzanne Kingston 

117  

economic progress, and the increase in the aggregate wealth of society, is not an end in itself.14 Those arguing that competition and public policy considerations must be kept separate may respond that the vision of a social market economy is perfectly compatible with their view. By this view, it is not the case that public policy considerations should play no role in the EU’s economic policies, but that they may only play a role in the hands of the legislator, and not in the case of enforcement by (non-legislative) competition enforcers such as the Commission, NCAs and the national courts. This is so because decisions involving public policy are by their nature ‘intensely political’, and for that reason should be left to the legislature alone as a democratically legitimated body, and because such decisions are inherently subjective in nature, so allowing them to play a role in competition policy increases uncertainty and reduces predictability in its enforcement.15 These points are reasonable, applied to public policy considerations as a general category and without regard to the specific context of EU competition policy. Yet, when considering environmental factors in EU competition policy, there are at least four strong countervailing arguments that no strict separation can or should be made between EU environmental and competition policies.

3.2 Arguments for an integrated approach First, from a legal perspective, the EU Treaties explicitly demand joined-up, noncompartmentalised decision-making in any field where environmental protection requirements may be relevant. The source for this approach is ultimately Article 3(3) TEU, as cited above, but it is further specified in the Article 11 TFEU integration obligation, requiring that Environmental protection requirements must be integrated into the definition and implementation of the Union’s policies and activities, in particular with a view to promoting sustainable development.

The environmental integration obligation is a key priority for the EU at the highest level, as emphasised most recently in the European Council’s June 2012 statement on the Seventh Environmental Action Programme.16 Article 11 TFEU contains no exclusion for competition policy. However, its effect clearly depends upon its legal status (a mere policy directive, or a condition of validity?) as well as upon the meaning of the requirement to ‘integrate’ environmental protection requirements. 14 This is so even if one disregards Lisbon’s demotion of the reference to ‘free and undistorted competition’ to a Protocol as more symbolic than substantial in effect. 15 See Bork, The Antitrust Paradox, cited above note 11, at p 25. 16 Despite the European Council’s grand 1998 launch of the Cardiff Process aimed at achieving environmental integration across all policy areas, achieving practical results from this Process remains a significant challenge for the EU.

118  Competition, Regulation and Public Policies On the former point, the EU courts have made clear that Article 11 TFEU carries legal force going beyond the merely programmatic, including as a tool for interpreting Treaty and secondary EU law,17 a limitation on the scope and exercise of the EU’s competence,18 and a limitation on and justification for Member States’ actions within the scope of EU law.19 In contrast, it probably does not (yet) constitute a self-standing ground for annulling EU secondary law.20 The meaning of the requirement to ‘integrate’ environmental protection requirements is perhaps an even thornier issue. What is clear is that it does not suffice for decision-makers simply to address their mind to environmental protection requirements at some stage in the decision-making process (procedural integration). Instead, decision-makers should demonstrate substantive integration by, for instance, interpreting EU laws in so far as possible in conformity with the EU’s environmental objectives.21 In this regard, the strength of the environmental integration obligation stands in contrast to the other cross-cutting integration clauses inserted into the TFEU by Lisbon (for instance, in relation to social protection, discrimination and consumer protection), which are all phrased in weaker terms. Second, an integrated approach to environmental protection requirements has been followed for many years now in the EU’s internal market rules applicable to States, as opposed to undertakings. The EU Courts’ jurisprudence integrating environmental concerns in interpreting the free movement provisions as well as public procurement legislation is well documented.22 Yet, as set out above, the traditional boundary between State and private functions in environmental regulation has disappeared due to the heavy reliance on market-based instruments in contemporary environmental policy. As the public/private divide in environmental regulation collapses, it becomes very difficult to maintain any logical justification for integrating environmental protection within the internal market’s rules governing State action (free movement), but not within those governing undertakings (competition policy).23 Third, an integrated approach follows from the basic principle that decisionmaking should be done in a coherent manner, an obligation set out in Article 7 TFEU and long embraced by the Commission as one of its five principles of See, eg, Case T-233/04 Netherlands v Commission [2008] ECR II-591, discussed further below. See, eg, Case C-300/89 Commission v Council (Titanium Dioxide) [1991] ECR I-2867. See, eg, Case C-379/98 PreussenElektra AG v Schhleswag AG, in the presence of Windpark Reußenköge III GmbH and Land Schleswig-Holstein [2001] ECR I-2099. 20 But see Case T-229/04 Sweden v Commission [2007] ECR II-2437. 21 See, eg, André Nollkaemper, ‘Three Conceptions of the Integration Principle in International Environmental Law’, in Andrea Lenschow, Environmental Policy Integration: Greening Sectoral Policies in Europe, Earthscan, 2002; Martin Wasmeier, ‘The Integration of Environmental Protection as a General Rule for Interpreting Community Law’, 38 Common Market Law Review 159 (2001); Nele Dhondt, Integration of Environmental Protection into Other EC Policies, Europa, 2003; Giorgio Monti, ‘Article 81 EC and Public Policy’, 39 Common Market Law Review 1057, 1078 (2002). 22 See, eg, Case C-389/96 Aher-Waggon GmbH v Germany [1998] ECR I-4473; Case C-297/05 Commission v Netherlands [2007] ECR I-476; Case C-2/90 Commission v Belgium [1992] ECR I-4421. 23 See further Kingston, Greening EU Competition Law and Policy, cited above, Chapter 3. 17 18 19

Suzanne Kingston 

119  

good governance within the EU.24 While, again, this approach is a familiar one in EU and national courts, there is a body of resistance to this notion as applied to DG Competition within the Commission. By this view, DG Competition should be seen to occupy a special status within the Commission, insulated from other Commission directorates-general and their policy concerns and, as a result, needing only a weak form of democratic control.25 This argument, which goes to the heart of the present discussion, will be returned to below. Fourth, even if one accepts maximising consumer welfare as the sole goal of EU competition policy, it is by no means evident that this necessarily means excluding consideration of environmental factors. Drawing on the rapidly evolving body of environmental economics aimed at placing a monetary value on environmental benefits and harms, an assessment of these benefits and/or harms resulting from a transaction could be incorporated in estimating post-transaction consumer surplus.26 This is quite apart from the dynamic efficiency gains that many environmentally friendly products may benefit from, by representing innovation and market creation.27 While environmental valuation techniques may make estimating consumer surplus more complex, such complexity already exists in numerous fields of EU competition policy (such as, for instance, attempts to value dynamic efficiencies) and can hardly in itself constitute sufficient reason to ignore environmental integration obligations.28 Taken together, it is reasonable to conclude that the practical implications for competition enforcers of these arguments are at least threefold. – Where there is no scope at all for interpreting the Treaty provisions in a manner favouring environmental protection, the environmental integration obligation is not relevant. (This means, for instance, that Article 101(1) TFEU only applies where there is a relevant competitive restriction – and not, for instance, simply environmental damage). – Where there is scope for interpreting the Treaty provisions in an environmentally favourable manner and there is no conflict with competition goals, these provisions must always be interpreted in that way. – Where there is scope for interpreting the Treaty provisions in an environmentally favourable manner and there is a conflict with competition goals, a proportionality assessment applies such that the measure least See, eg, Commission, White Paper on European Governance, COM(2001)428. See generally Claus-Dieter Ehlermann, ‘Reflections on a European Cartel Office’, 32 Common Market Law Review 471 (1995); and, eg, the proposal of Gordon Brown for an independent EU competition agency, reported in Financial Times (20 April 2005). 26 For this argument, see further Kingston, Greening EU Competition Law and Policy, cited above, Chapter 5. On environmental economics and environmental valuation, see, eg, Tietenberg and Lewis, Environmental and Natural Resource Economics (8th ed, Pearson, 2009); TEEP, The Economics and Ecosystems and Biodiversity (Earthscan, 2010). 27 See, eg, Ecorys, 2009 Study on the Competitiveness of the EU’s Eco-industries, available on DG Enterprise’s website. 28 On the inherent difficulties in applying economic tests in competition law see, eg, Franklin Fisher, ‘Economic Analysis and “Bright Line” Tests’, 4 Journal of Competition Law and Economics 129 (2007). 24 25

120  Competition, Regulation and Public Policies restrictive of competition that achieves the same environmental objective should be preferred.

4. Some examples of environmental factors in EU competition policy Clearly, the increase in market-based instruments in environmental policy means that competition enforcers will increasingly have to deal with cases raising environmental protection issues. The interface between environmental and competition policy has already arisen in multiple cases demonstrating that, first, in some cases environmental protection/policy factors are indeed given weight by competition enforcers but, second, this is not done in a consistent manner, as illustrated by the examples below.

4.1 Environmental protection in Arts 101 and 102 TFEU Concept of an undertaking An obvious first issue in the environmental policy/substantive EU antitrust law interface is the relevance of environmental policy to the concept of an undertaking. This was famously the issue in Calì & Figli, where anti-pollution surveillance in relation to loading and unloading of acetone products was held not to qualify as an economic activity on grounds that this constituted a task in the public interest forming part of one of the ‘essential functions of the state’ in protecting the maritime environment.29 The ECJ held that: Such surveillance is connected by its nature, its aim and the rules to which it is subject with the exercise of powers relating to the protection of the environment which are typically those of a public authority. It is not of an economic nature justifying the application of the Treaty rules on competition.30

As a result, Article 102 TFEU did not apply. Subsequent cases such as Ambulanz Glöckner (on emergency and ambulance services) have set out a test of whether the activities at issue have ‘always been, and are [...] necessarily carried on by’ bodies other than private operators.31 Applying this test, therefore, raises issues going to the heart of the public/ private divide and, ultimately, each Member State’s view of the proper limit to 29 Case C-343/95 Diego Calì & Figli Srl v Servizi ecologici porto di Genova SpA (SEPG) [1997] ECR I-1547. 30 Ibid, para 23. 31 Case C-475/99 Firma Ambulanz Glöckner v Landkreis Südwestpfalz [2001] ECR I-8089.

Suzanne Kingston 

121  

deregulation. The Irish domestic waste collection system is a case in point, as Ireland is one of the very few States within Europe that has chosen to open its private waste collection sector to open competition.32 Does this mean that waste collection operators constitute undertakings in this jurisdiction? In Panda Waste, the Irish High Court was asked to decide this issue in circumstances where Dublin’s local authorities had decided to change the applicable waste regulation to move from multiple waste collection operators to a single operator.33 In that case, the Court’s response was that local authorities constituted undertakings, meaning that the local authorities’ regulatory decisions were subject to review for compatibility with (national) competition rules. Central to the Court’s reasoning was that private operators could carry on this business under market conditions. In other jurisdictions, however, domestic waste collection might well be viewed as falling outside the scope of competition law, as an activity aimed at meeting an ‘essential public interest’ which is ‘provided for the benefit of the whole of the community’.34 Greater use of markets as tools to achieve environmental policy goals will mean, however, that more environmental services will fall within the definition of an undertaking.35

Market definition Market definition constitutes a further area where the impact of (different trends in) environmental regulation is closely felt. As consumers begin to differentiate between products on the basis of their environmental performance, competition enforcers are being asked to adjudicate on substitutability issues. In reality, however, preferences we may broadly class as ‘environmental’ in nature may differ greatly in nature, ranging from private health benefits (organic foods) to intangible personal benefits (avoiding guilt from purchasing a polluting product) to pure altruistic reasons.36 The behaviour of marginal consumers will be vital for market definition in most37 of these cases. More generally, as consumers become increasingly environmentally conscious, narrower market definitions for greener products and services will gradually emerge, entailing the risk of increased market power. 32 Along with Poland and Kosovo. See further Philip Andrews and Paul Gorecki, ‘The Panda Waste case: Conflicts between National Competition Law and Regulation’, 6 European Competition Journal 543 (2010). 33 Panda Waste Services v Dublin City Council and Others [2009] IEHC 589 (on appeal to the Irish Supreme Court). 34 See the Opinion of AG Cosmas in Calì & Figli, cited above note 29. 35 A good example is that of nature conservation which, although ostensibly a classic instance of pure public interest action, is itself being ‘economised’ by the rise in popularity of techniques such as habitat banking. 36 See, eg, James Andreoni, ‘Warm-Glow Versus Cold-Prickle: The Effect of Positive and Negative Framing on Cooperation in Experiments’, 60 Quarterly Journal of Economics 1 (1995). 37 Experience in the US, specifically in the Federal Trade Commission v Whole Foods Market Inc litigation (548 F 3d 1028 (D.C. Cir. 2008)), illustrates a different approach that focuses on the behaviour of a group of core consumers who, in that case, demanded premium, natural and organic supermarkets exclusively.

122  Competition, Regulation and Public Policies

Article 101 TFEU Article 101 TFEU cases raising environmental protection issues may, broadly, be grouped into two categories. The first comprises cases where environmental protection requirements are simply used as a pretext for collusive behaviour between competitors, or where collusive behaviour results by way of spillover from genuine environmentallymotivated initiatives. Examples abound of such findings in the Commission’s decisional practice including, most recently, the 2011 cartel settlement decision in the consumer detergents cartel, where detergent manufacturers were fined €315.2 million for a cartel that began when the manufacturers implemented an initiative through their trade association to improve the environmental performance of detergent products.38 On the facts, however, the Commission found that the environmental objective of the arrangement did not require them to coordinate prices or to engage in other anticompetitive practices, which the participants had done on their own initiative. The purpose of the participants’ anticompetitive conduct was, the Commission found, to stabilise the market by ensuring that none of them would use the environmental initiative to gain a competitive advantage over the others, and that market shares would remain the same following the initiative.39 A central feature of this type of case, therefore, is that anticompetitive behaviour is separable from, and not necessitated by, the environmental objective at issue. Such cases may present evidential challenges in making such a finding of fact, but do not raise any real issue of principle as to the interrelation between competition and environmental policies. Far greater difficulties may arise where an environmental aim in fact necessitates behaviour that, at least in the absence of that aim, would be viewed as anticompetitive in object or effect. The structure of Article 101 TFEU is, of course, that agreements or practices entailing competitive restrictions with an appreciable effect on inter-State trade are prohibited (Article 101(1) TFEU) unless it can be proven that they (Article 101(3) TFEU): – Contribute to ‘improving the production or distribution of goods or to promoting technical or economic progress’; – Allow consumers a ‘fair share of the resulting benefit’; – Do not contain restrictions that are not necessary to attain the relevant objective; and – Do not afford the undertakings concerned the possibility of ‘eliminating competition in respect of a substantial part of the products in question.’ This bipartite structure means that, unlike many other jurisdictions, the EU regime has an express locus where the restrictive effects of an agreement may be balanced 38 See Press Release IP/11/473 of 13 April 2011. The initiative targeted dosage and weight reduction of HDD low suds powder and corresponding packaging material (the ‘AISE’ initiative). 39 See Elina Laurinen, ‘The Consumer detergents cartel’, (2011/2) Competition Policy Newsletter 3, at 4.

Suzanne Kingston 

123  

against its benefits (or at least those benefits of the agreement deemed relevant for the purposes of EU competition law). Here, the Commission’s position taken in its 2004 Article 101(3) Guidelines is famously restrictive, confining relevant benefits under Article 101(3) TFEU to the agreement’s ‘pro-competitive’40 effects, meaning efficiency gains which may create additional value by lowering the cost of producing an output, improving the quality of the product or creating a new product.

The Commission’s view in this document is clear: relevant benefits in this sense must be ‘objective economic benefits’; other benefits, even if related to goals pursued by other Treaty provisions, must be ignored.41 Evidently, this controversial position carries implications going far beyond the environmental/competition interface to a host of other non-economic, public policy considerations. In the case of environmental protection, however, applying the economic integration argument set out above, there are strong arguments that environmental benefits that are reasonably quantifiable in economic terms, using valuation techniques developed in environmental economics, constitute (qualitative) efficiency gains within this sense. (This approach was mentioned in the 2001 Horizontal Cooperation Guidelines, but has been removed from the 2010 version.42) Whilst it is an approach that will save many agreements with genuine environmental objectives, it will require considerable creativity of application in cases where environmental benefits are difficult to quantify, or quantification is contested (such as benefits from biodiversity; benefits ensuing from future generations from, for instance, reduced carbon or other emissions). Even if environmental protection benefits were to be categorised as noneconomic in this sense, however, the Commission’s attempt to exclude any noneconomic factors from the Article 101(3) TFEU concept of benefits must be placed in perspective. As argued above, it is an effort that is at odds not only with the EU courts’ jurisprudence, but also with as the Article 11 TFEU integration obligation and principles of coherence of governance.43 As such, reports of the exclusion of genuine environmental benefits from Article 101(3) TFEU seem conspicuously premature. This is not to deny that, from the Commission’s perspective, there might have been plenty of practical reasons why, at the time the Article 101(3) Guidelines were published in 2004, such non-economic factors should be ignored. Not least among these is that, on the cusp of the largest decentralisation of competition enforcement the EU had ever seen, excluding such factors would reduce NCAs’ discretion in how they applied Article 101(3) (and with it, potentially, uncertainty). Similarly, interpreting Article 101 TFEU in line with the Sherman Act would make 2004 OJ C101/97. Para 42. 42 2011 OJ C11/1. The Commission’s view is that this does not imply any downgrading for the assessment of environmental agreements. 43 See further Kingston, Greening EU Competition Law and Policy, cited above, at 263. 40 41

124  Competition, Regulation and Public Policies things simpler for global market actors. Yet these factors were not articulated by the Commission at the time and, as noted above, find no textual basis in the Treaty. A further question is under what conditions a ‘fair share’ of environmental benefits might be considered to go to consumers. While one might be forgiven for assuming that this condition would be relatively easily satisfied in the case of environmental benefits, once again this requires going beyond the approach taken in the Commission’s Article 101(3) TFEU Guidelines, where ‘consumers’ are confined to ‘all direct or indirect users of the products covered by the agreement’.44 Logically, acceptance of the relevance of environmental benefits to Article 101(3) surely implies broadening the concept of consumer in this sense to include consumers in future generations as well as in jurisdictions outside the EU. In practice, it is likely that the principal battlefield in many Article 101 TFEU cases where an environmental protection objective is at play will be proportionality: is it really the case that such objective necessitates this level of competitive restriction, or could the same objective be achieved through less restrictive means? The environmental charges cases illustrate this issue where, in cases like VOTOB, a fixed environmental surcharge levied for storage of products pursuant to a waste management agreement harmonised the storage costs in a manner that was not necessary to achieve the agreement’s environmental aims.45 (Nonetheless, it is far from clear that VOTOB would be decided in the same manner today, given the small proportion of total costs represented by the environmental surcharge in that case.46) The proportionality issue has also been central to the line of cases on waste collection agreements in the context of national packaging waste disposal systems in Germany, the UK, France and Austria, especially in relation to the length of time of exclusivity arrangements between recycling undertakings and their waste collection partners.47 In sum, therefore, there is much scope for genuine integration between Article 101 TFEU and environmental policy, pursuant to Article 101(3) TFEU. A perhaps more audacious suggestion is that such integration could also take place within Article 101(1) TFEU itself, by extension of the ancillary restraint doctrine developed by the Court in judgments such as Gøttrup-Klim, Albany and Wouters.48 Thus, where restrictions are inherent and proportionate to an environmental agreement such that no agreement would have been reached without them (Gøttrup-Klim, Albany), or to the achievement of the environmental Para 84. Commission, XXII Report on Competition Policy 1992 (1993), paras 177–186. 46 See Kingston, Greening EU Competition Law and Policy, cited above, at 285. 47 See Valpak, XXVIII Report on Competition Policy 1998 (1999) 165; DSD 2001 OJ L319/1 (confirmed by Case T-289/01); Commission Decision of 16 October 2003 in Case COMP/35473 – ARA 2004 OJ L75/59, upheld in Case T-419/03 Altstoff Recycling Austria AG v Commission [2011] ECR II-975); Commission Decision of 15 June 2001 – Eco-Emballages 2001 OJ L233/37. 48 Case C-250/92 Gøttrup-Klim e.a. Grovvareforeninger v Dansk Landbrugs Grovvareselskab AmbA [1994] ECR I-5641; Case C-67/96 Albany International BV v Stichting Bedrijfspensioenfonds Textielindustrie [1999] ECR I-5751; Case C-309/99 J C J Wouters, JW Savelbergh and Price Waterhouse Belastingadviseurs BV v Algemene Raad van de Nederlandse Orde van Advocaten [2002] ECR I-1577 44 45

Suzanne Kingston 

125  

protection goals of a privately run environmental regulatory system (Wouters), they would fall outside Article 101(1) TFEU. While undoubtedly a controversial suggestion, the approach seems consistent not only with the spirit and ratio of the Court’s judgments in this field, but also with the Article 11 TFEU integration imperative.

Article 102 TFEU Clearly, Article 102 TFEU contains no express exception provision equivalent to Article 101(3) TFEU. The possibility of objective justification for prima facie abusive conduct is, however, now accepted by the EU Courts as well as by the Commission, which has in its 2009 Guidance paper suggested that conditions similar to those applicable to Article 101(3) TFEU must be satisfied.49 As environmental protection arguments find their natural home at the objective justification stage, this means that similar considerations will apply to the Article 102 TFEU analysis as those set out above in the Article 101(3) TFEU context, although anticompetitive behaviour will be even more difficult to justify where a dominant market position is held. In reality, of course, it will be highly risky to rely on objective justification arguments in environmental cases, for two reasons. First, in practice, objective justification arguments have been accepted on notoriously rare occasions. Second, the EU Courts’ view of the boundaries of the objective justification concept are as yet unclear, although we can be relatively sure that the concept includes economic efficiencies or ‘objective economic justification’ (British Airways50) as well as justification on grounds of legitimate public interest aims (Hilti, Tetra Pak II, Kanal 551). With regard to the latter, the Commission’s warning in its 2009 Guidance that ‘it is normally the task of public authorities to set and enforce public health and safety standards’52 should, in the environmental context, be read subject to the caveat that many formerly ‘public’ environmental protection functions are now being devolved to private operators active on a regulated market, as discussed above. In practice, environmental justifications have in themselves won the day in few Article 102 TFEU cases to date, normally because anticompetitive behaviour has been judged disproportionate to the environmental objective at issue. A classic example is DSD,53 where the EU Courts confirmed that DSD had abused its dominant position on the market for the organisation of the take-back and recovery from private final consumers of used sales packaging in Germany. In that case, the abuse consisted in the requirement that undertakings obligated under OJ 2009 C45/7. Case C-95/04 P British Airways plc v Commission [2007] ECR I-2331, paras 69 and 86. 51 Case C-30/89 Hilti AG v Commission [1991] ECR II-1439; Case T-83/91 Tetra Pak International SA v Commission (Tetra Pak II) [1993] ECR II-755; Case C-52/07 Kanal 5 Ltd e TV 4 AB v Föreningen Svenska Tonsättares Internationella Musikbyrå (STIM) upa [2008] ECR I-9275. 52 Para 29. 53 Case C-385/07 P Der Grüne Punkt - Duales System Deutschland GmbH v Commission [2009] ECR I-6155. 49 50

126  Competition, Regulation and Public Policies the German Packaging Ordinance and party to DSD’s agreements were obliged to pay DSD a fee for all sales packaging distributed within Germany which bore the green dot trademark, irrespective of whether the DSD system ultimately provided the service of collection and recycling, etc. The Commission decision, and ensuing judgments, illustrate well how the competition/environment interface cuts both ways: competition policy can benefit the operation of environmental policy in practice by working to ensure that important (and increasingly lucrative) environmental sectors such as waste management work in as efficient a manner as possible, with prima facie anticompetitive behaviour only permissible to the extent genuinely necessary to perform an environmental function.

4.2 State aid and Article 106 TFEU Even with the rise to prominence of market-based instruments in environmental policy, State action clearly retains an important role to play in environmental markets, whether by constructing the regulatory framework for the markets themselves or by (financially and otherwise) supporting the private operators active thereon. While State aid policy may be the most obvious body of rules applicable to environmentally-motivated State action, it is flanked by Article 106 TFEU and the Court’s Van Eycke doctrine, which prohibits Member States from introducing or maintaining in force measures which may render the EU’s antitrust rules ineffective.54

Article 106 TFEU and Van Eycke Article 106(1) TFEU and Van Eycke set important limits on the way in which Member States can supervise privatised environmental markets, as illustrated by the CJEU’s judgments in Dusseldorp and Sydhavnens, concerning the grant by a Member State of exclusive rights within the meaning of Article 106(1) TFEU to waste disposal undertakings, where such grant was liable to create a situation where these undertakings were induced to abuse their dominant positions.55 In Sydhavnens, Article 106(2) TFEU applied to justify exclusivity where this was necessary to ensure a sufficient flow of waste (in that case, for a new building waste facility in Copenhagen); in Dusseldorp, the Court left the Article 106(2) TFEU point to the national court, while emphasising the Dutch government’s obligation to prove that the environmental objective could not be achieved ‘equally well by other means’.56 In these cases, as well as in Commission decisions such as AVR, particular waste treatment services have been found to constitute services Case 267/86 Pascal Van Eycke v Aspa SA [1988] ECR 476. Case C-203/96 Chemische Afvalstoffen Dusseldorp BV and Others v Minister van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer [1998] ECR I-4075; Case C-209/98 Entreprenørforeningens Affalds/Miljøsektion (FFAD) v Københavns Kommune (Sydhavnens) [2000] ECR I-3743. 56 Para 67. 54 55

Suzanne Kingston 

127  

of general economic interest.57 However, it cannot be assumed that all waste treatment or environmental services automatically constitute services of general economic interest; this will depend on a case-by-case assessment on the basis of a variety of factors including the necessity of public intervention in that instance, the compatibility of the measures taken with the polluter pays principle, and whether the service benefits a broad number of users. This approach, illustrated by AVR in particular, integrates environmental and competition policies in an impressively coherent manner.

State aid A large body of law exists on environmental State aid, prompted by the popularity of financial subsidies and environmental taxation schemes as environmental policy instruments throughout the EU from the 1970s onwards. The Commission’s approach is by and large contained in its 2008 Guidelines on State Aid for Environmental Protection, which succeeds Guidelines from 2001 and 1994 and, before that, environmental aid framework positions adopted in 1974, 1980 and 1986.58 Certain types of environmental State aid are also included in the General Block Exemption.59 The 2008 Guidelines place great emphasis on the role of Member States in environmental protection policy, as well as the need to integrate the EU’s State aid, environment and energy policies. The contrast with antitrust policy is, of course, clear: non-economic policy justifications for aid are explicitly embraced in the State aid context (and provided for expressly by the Treaty), and represent a recognition that markets may not always work properly when left alone (for instance, in the case of externalities). In line with this aim, State aid law has given rise to some of the most successful examples of genuine integration between EU competition and environmental policy. At the CJEU level, the most striking example is perhaps PreussenElektra, where the CJEU interpreted the concept of State resources narrowly to hold that a German feed-in tariff for renewable energy did not constitute aid within the meaning of Article 107(1) TFEU.60 More recently, this jurisprudence has been distinguished in cases where the State has chosen to allocate emissions trading allowances for free in circumstances where it could have charged for such allowances. In UK Emissions Trading Scheme, the Commission found the UK’s scheme to constitute aid on this basis, an approach essentially confirmed by the Dutch Nitrous Oxide Emissions Trading Scheme litigation as well as by the EU Courts’ judgments on the compatibility of Member States’ allocation of allowances under the EU’s Emissions Trading Scheme.61 In practice, therefore, any free allocation 2006 OJ L84/37. Guidelines on State Aid for Environmental Protection, 2008 OJ C82/1; Community Guidelines on State Aid for Environmental Protection, 2001 OJ C37/3; Community Guidelines on State Aid for Environmental Protection, 1994 OJ C72/3. 59 Commission Regulation 800/2008 2008 OJ L214/3. 60 PreussenElektra, cited above note 19. 61 Case C-279/08 P Commission v Netherlands [2011] ECR I-7671. 57 58

128  Competition, Regulation and Public Policies of allowances by Member States will in all likelihood constitute State aid, thus reinforcing the move towards auctioning as the preferred method of allocation within the EU. This is in line with the Commission’s general approach, as recently expressed in its June 2012 Guidelines on State aid in the EU’s Emissions Trading Scheme, that aid must be necessary to achieve the environmental objective at issue (the necessity test) and must be limited to the minimum needed to achieve the environmental protection sought (the proportionality test) ‘without creating undue distortions of competition and trade in the internal market’.62 Where EU laws set out minimum environmental requirements, State aid must result in a higher level of environmental protection than that entailed by such requirements. A controversial example is the aid permissible under the EU ETS to undertakings in sectors exposed to a significant risk of carbon leakage via passing on EU ETS allowance costs in electricity prices, which aid seeks to avoid an increase in global greenhouse gas emissions due to shifts of production outside the EU in the absence of a binding international agreement on reduction of greenhouse gas emissions.63 Despite this, the EU’s State aid policy still falls short of achieving genuine integration with its environmental policy on a number of counts. Some of these comprise temporary politically-motivated exceptions from an otherwise integrated and coherent sectoral environmental aid policy, such as the transitional free allowances permitted under the EU ETS for the electricity sector in certain Member States.64 Other instances of disintegration, however, are more legal in nature. The CJEU’s new approach to material selectivity in environmental taxation, as expressed by the CJEU in its 2008 British Aggregates judgment, is a case in point. These cases concern the limits of the long-established principle that differentiation in taxation arising from the nature or overall structure of a tax system does not constitute aid, first laid down in judgments such as Adria-Wien.65 In British Aggregates, the General Court had upheld the Commission decision not to raise any objection to the British eco-tax scheme taxing the commercial exploitation of virgin aggregates, on the basis that Member States were free, in balancing the various interests involved in constructing their eco-tax systems, to set their priorities as regards environmental protection. As a result, the ecotax in question was not materially selective, and did not constitute aid within the meaning of Article 107(1) TFEU. Overturning the General Court’s judgment on this issue, the CJEU held the eco-tax to be selective on the basis that such a levy must apply to ‘all similar activities which have a comparable impact on the environment’66 in order to benefit from the Adria-Wien doctrine. 2012 OJ C158/4, at para 5. Ibid., para 8. 64 See ibid. 65 Case C-143/99 Adria-Wien Pipeline GmbH e Wietersdorfer & Peggauer Zementwerke GmbH v Finanzlandesdirektion für Kärnten [2001] ECR I-8365. 66 Case C-487/06 P British Aggregates Association v Commission and United Kingdom [2008] ECR I-10505, para 86. See also Case T-210/02 RENV British Aggregates v Commission, judgment of 7 March 2012, not yet reported (when sent back to the General Court to decide following the appeal judgment, General Court found scheme to be selective). 62 63

Suzanne Kingston 

129  

British Aggregates was followed by the CJEU’s September 2011 judgment in the Dutch Nitrous Oxide Emissions Trading Scheme judgment, where the CJEU overturned the General Court’s judgment holding the Dutch scheme not to be selective because of its ecological objective. In that case, the CJEU reasoned that the fact that the scheme only applied to large industrial facilities (with an installed total thermal capacity of more than 20 MWth) meant that it was selective, as there was no environmental justification for this cut-off (rather, the justification was one of administrative workability).67 In this line of judgments, therefore, the CJEU is clearly tightening its approach to eco-tax schemes considerably, and throwing down the gauntlet to Member States designing such schemes to ensure that their features are consistent with their environmental objective. In itself, this is a commendable aim, and one consistent with the integration imperative (especially as eco-tax schemes entailing aid may in any event be adjudged compatible with the common market by the Commission pursuant to Articles 107(2) and (3) TFEU). However, the language and reasoning used by the CJEU in coming to this conclusion in British Aggregates leaves much to be desired. If the requirement that all activities with a ‘comparable impact on the environment’ be taxed in the same way is taken seriously, it is virtually impossible to satisfy, both practically and politically. Clearly, a vast number of activities within our society have negative environmental impacts; taxing all similar activities on this basis, or even trying to catalogue all similar activities by environmental effects, would be immensely difficult (much less discharging the burden of proof, which rests in this respect on Member States, that this has been done). This part of the Court’s judgment is disappointing, therefore, and the previous ‘nature and general scheme’ formula is far preferable. Perhaps the most glaring current disconnect between EU State aid and environmental policy, however, consists in the very fact that, as the State aid rules are currently applied, aid to a project that is in breach of EU environmental law can nonetheless be approved by the Commission as compatible with the common market pursuant to Articles 107(2) or 107(3) TFEU. This remarkable position was confirmed by the General Court’s judgment in Thermenhotel Stoiser.68 There is certainly no justification for this as a matter of competition policy: these are, after all, cases where aid that is prohibited as anticompetitive pursuant to Article 107(1) TFEU is being considered for exemption. As a matter of law, there is much to support the argument that aid contrary to the Treaty’s environmental protection requirements falls outside Articles 107(2) and (3), particularly in light of the EU Courts’ consistent rulings that the Commission cannot declare aid which contravenes other provisions of the TFEU to be compatible with the common market.69 On this point, Thermenhotel Stoiser should be overruled. Netherlands v Commission, cited above note 61. Case T-158/99 Thermenhotel Stoiser Franz Gesellschaft mbH & Co. KG and Others v Commission [2004] ECR II-1. 69 See, eg, Case 73/79 Commission v Italy [1980] ECR 1533; Case C-156/98 Germany v Commission [2000] ECR I-6857; Case C-204/97 Portugal v Commission [2001] ECR I-3175. 67 68

130  Competition, Regulation and Public Policies

5. Conclusions: Unpacking old adages and overcoming institutional fragmentation It is by now an old adage that EU competition policy should be insulated from non-economic policy concerns, led exclusively by the goal of consumer welfare. Yet there are compelling arguments that this view does stand up to scrutiny in the case of environmental protection aims. These range from legal arguments as to the systematic interpretation of the Treaty, to arguments of coherence in governance, to economic arguments challenging narrow conceptions of consumer welfare. Ultimately, the debate becomes unavoidably political in nature, depending on one’s view of the correct balance in the EU’s (environmental social market) economic model and the proper place of competition policy within this model. There are strong arguments that competition policy was never intended to be an end in itself within the EU, but rather to be part of the system of economic constitutional law created by the Treaty of Rome. In this sense, the EU was destined never to embrace the Chicago espousal of antitrust purism in the manner reflected in US antitrust policy. If we take the EU Treaties and their system seriously, therefore, EU competition policy must be integrated with environmental protection objectives, including in the context of enforcement decisions in individual cases. For their part, the EU Courts have to date shown definite efforts to attain a coherence of approach to competition and environmental objectives, often relying (quite appropriately) on proportionality analysis, although there is room for greater consistency in this regard. In the Commission’s case, however, its initial receptive attitude to environmental factors has been replaced with an apparently absolute exclusionary approach outside the State aid context. As has been argued above, there are serious legal, governance-related, and economic flaws in this approach. More generally, the current means by which the Commission purports to achieve environmental integration – principally, by requiring impact assessments in the case of major policy proposals, is unacceptably weak. Substantive integration, in the sense outlined above, must occur not only in the case of policy proposals, but also in individual cases of competition enforcement raising any significant environmental issues. Integration also demands that DG Competition should liaise with DG Environment or DG Clima, and vice versa, in cases raising important issues falling within their respective policy domains. As the blurring of the public/private divide in environmental regulation continues, competition enforcers will increasingly be faced with the implications of environmental/competition integration. Aside from the substantive aspects to this integration, considered above, it will also bring considerable institutional challenges. These challenges are perhaps less daunting for the EU courts (and national courts acting within the scope of EU law), which are by and large used to using systematic techniques of interpretation, with the aim of ensuring a largely

Suzanne Kingston 

131  

coherent body of law.70 The integrated approach evident in many of the EU Courts’ judgments in competition, as well as in trade and public procurement, reflects this well.71 In the case of the Commission, however, the institutional integration challenge is perhaps greatest. This may be due partly to the natural tendency towards compartmentalisation of thought in an institution structured in separate directorates-general. Problems of institutional fragmentation of policy approach within the Commission are well-documented (although internal coordination is improving) and are by no means specific to the competition field.72 Arguments that DG Competition should be viewed as a DG apart from the rest, however, risk supporting a sense of institutional possession (hands off my policy!) that, in the environmental context, finds no basis in the Treaty.

70 See Ronald Dworkin, Law’s Empire, Fontana, 1986, at 225; Joxerramon Bengoetxea, Neil MacCormick and Leonor Moral Soriano, ‘Integration and Integrity in the Legal Reasoning of the European Court of Justice’, in Gráinne de Búrca and Joseph Weiler (eds), The European Court of Justice (Oxford University Press, 2001) p 47; David Edward, ‘Judicial Activism: Myth or Reality?’, in Angus Campbell and Meropi Voyatzi (eds), Legal Reasoning and Judicial Interpretation of European Law (Trenton Publishing, 1996) p 29; Francis Jacobs, ‘The Role of the European Court of Justice in the Protection of the Environment’, 18 Journal of Environmental Law 185 (2006). 71 See, eg, Case C-389/96 Aher-Waggon [1998] ECR I-4473; Case C-297/05 Commission v Netherlands [2007] ECR I-476; Case C-2/90 Commission v Belgium [1992] ECR I-4421; and, in the field of public procurement, Case C-513/99 Concordia Bus [2002] ECR I-7213. 72 See generally Kingston, Greening EU Competition Law and Policy, cited above, Chapter 4. See further Herwig Hofmann and Alexander Türk, EU Administrative Governance (Edward Elgar, 2006); Adriaan Schout and Andrew Jordan, ‘The European Union’s Governance Ambitions and its Administrative Capacities’, 15 Journal of European Public Policy 957 (2008); Les Metcalfe, ‘Reforming the Commission: Will Organizational Efficiency Produce Effective Governance?’, 38 Journal of Common Market Studies 817 (2000) and the paper on ‘Closer Internal Coordination’, presented at the inaugural meeting of the Prodi Commission in Aartselaar, 1999, http://ec.europa.eu/ reform/pdf/coordin_en.pdf.

Robert D Anderson and Anna Caroline Müller*

Competition Policy and the Multilateral Trading System: Three Propositions, an Observation and Some Questions for Reflection

1.  Introduction The question of whether there is a need for more explicit linkages to be developed between national competition policies and the multilateral trading system – and, if so, the question of what is the appropriate content of related disciplines – are unresolved issues in debates surrounding the future of the WTO. For the present, there is no consensus on the need for the development of such disciplines. A previous effort in the WTO from 1997 through 2003 to develop a ‘multilateral framework on competition policy’ foundered and was placed on hold in the aftermath of the Cancún Ministerial Conference of 2003, at which no consensus could be found on modalities to initiate negotiations on this topic.1 Furthermore, since that work was put on hold, international co-operation in the field of competition policy has moved ahead in various other fora, including the International Competition Network (ICN), the UNCTAD Intergovernmental Group of Experts on Competition Law and Policy and the OECD Competition Committee. Some – perhaps the majority – of competition law practitioners would take the view that sufficient international cooperation with respect to the field of competition policy is already being effectively implemented in these other fora. However, the latter view is not universally held. For example, Jenny (2003) has argued that existing co-operation mechanisms are inadequate to address the harm caused by international cartels to developing countries, which continue to suffer extensively from such arrangements (see Connor 2011 and references cited * Anderson: Counsellor (Team Leader for Government Procurement and Competition Policy), Intellectual Property Division, WTO Secretariat; Part-time Faculty Member, World Trade Institute, Bern; and Honorary Professor, School of Law, University of Nottingham (e‑mail address: robert. [email protected]). Müller: Legal Affairs Officer, Intellectual Property Division, WTO Secretariat (e-mail address: [email protected]). The views expressed are the personal responsibility of the authors and should not be attributed to the WTO or its Secretariat. 1 Subsequent to the Ministerial, in 2004 the General Council of the WTO decided, as part of its socalled ‘July package’, that no further work would be undertaken toward negotiations on competition policy (or on the separate but related issues of investment and transparency in government procurement) for the duration of the Doha Round (see WT/L/579, 2 August 2004). This situation remains in effect and, consequently, the WTO Working Group on the Interaction between Trade and Competition Policy, which was created at the Singapore Ministerial Conference, is officially designated as ‘inactive’.

134  Competition, Regulation and Public Policies therein). Others (eg, Singham 2007) take the view that, while past efforts to bring together competition policy and international trade were perhaps simplistic and premature, the extent of possible synergies and interactions between the two fields is such that strict separation is impossible and renewed dialogue is highly desirable. Such authors would argue that measures addressing barriers to trade and harmful anticompetitive practices are intrinsically inter-linked, and that they have much to contribute to each other. Still others (eg, Anderson and Evenett 2006, and Teh 2009) have made the observation that competition policy provisions have been extensively integrated into recent bilateral and regional trade agreements (RTAs) – begging the question of why this cannot or should not also be contemplated (or re-contemplated) in the multilateral context as well.2 Indeed, the extent of controversy over the past proposals for a multilateral framework on competition policy risks obscuring another significant fact, namely that the past work on competition policy in the WTO demonstrated a broad degree of agreement on key underlying points. In particular, the work done in the WTO Working Group on Trade and Competition Policy clearly identified a number of channels through which anticompetitive practices of firms – at the national and the international levels – can impede both trade and development (see WTO Working Group on the Interaction between Trade and Competition Policy 1998).3 Furthermore, it identified a number of ways in which the objectives of competition policy and international trade liberalization are inter-linked, and in which competition policy concepts and objectives are already integrated into the WTO Agreements in one way or another (see also Anderson and Jenny 2005). To a large extent, these underlying points are not controversial, and they have been mirrored extensively in the work of other international organizations (eg, UNCTAD and the OECD); in that of leading NGOs active in the field, especially CUTS; and in scholarly writings published both before and after the WTO’s work on these issues (see, eg, Singham 2007, with references). On this basis it can be argued that, while clearly there is no agreement in the international community on the objectives and ‘modalities’ for multilateral negotiations in this area – or even on the need for such a project – there is also little disagreement, at least among those who have studied the issues, that trade liberalization and competition policy are both important for development; that they are mutually supportive in important ways; and that each field can learn from and be reinforced by the other. Consistent with the above, this chapter does not seek to resolve past disagreements on the need for a multilateral framework in this area, or to support a particular path forward. The renewal of efforts to more explicitly link the fields of trade and competition policy or even to re-instate educational and exploratory work in this area in the WTO (a much more limited result) are decisions that could See also the related discussion below. The channels identified are, inter alia, international cartels that raise the price of developing country imports, including with respect to primary products; market structures that impede or depress the price of developing country exports; state monopolies and cartels that raise the costs of developing country exporters; and mixed public and private restraints that facilitate market segmentation and impede entry by alternative suppliers. For discussion, see Anderson and Jenny (2005). 2 3

Robert D Anderson and Anna Caroline Müller 

135  

be taken only by the WTO’s Member governments. Nor do we seek to provide new empirical information on particular issues. The aim, rather, is to take stock of and set out for further discussion and reflection some of the links that have been identified between competition policy and the content and objectives of the WTO Agreements, including in past WTO work in this area, and in related scholarly research. The remainder of the chapter is organized in five separate parts, which together put forward ‘Three Propositions, an Observation and Some Questions for Reflection’. In particular, Part 2 summarizes select aspects of evidence bearing on the proposition that ‘Effective national competition policies are vital to the success of the multilateral trading system, including (very much) its contribution to development’. Part 3 takes up the proposition that ‘Competition policy already figures in the WTO Agreements, in important though limited ways’. Part 4 introduces the proposition that there are additional issues at the interface of trade and competition policy which (arguably) call for further joint application of the two disciplines. Part 5 makes a related observation – namely, that explicit provisions on competition policy have been extensively incorporated in regional and bilateral agreements adopted since work on competition policy was put on hold in the WTO, including by WTO Members that had expressed significant reservations concerning such work. Part 6 offers concluding remarks and some related questions for thought. The purpose throughout is not to advocate any particular approach to the issues identified but to set out relevant points for future consideration.

2.  First proposition: effective national competition policies are vital to the success of the multilateral trading system, including (very much) its contribution to development Possibly the most basic proposition to be put forward in this area – though not one which has always been universally accepted – is that effective (national) competition policies are important to the success of the multilateral trading system including (very much) its contribution to development. This section develops three examples of such dependence, namely: (i) the importance of competition policy in deterring international cartels, which have the ability to undermine directly the gains from trade; (ii) the importance of pro-competitive structural reforms in enabling developing and transition economy businesses to take advantage of market access opportunities created by trade agreements; and (iii) the importance of competition rules as an adjunct to liberalized government procurement regimes, as embodied in the WTO Agreement on Government Procurement and the chapters on government procurement that are incorporated in many recent RTAs. It should be made clear, at the outset, that

136  Competition, Regulation and Public Policies such examples do not, by themselves, pre-suppose or establish a need for any elaborate coordination mechanism between competition policy and international trade; rather, they are intended merely to establish the (arguably very basic but nonetheless important) proposition that the success of the international trading system depends, to an important degree, on the existence of effective national competition policies.

2.1  First example: the importance of pro-competitive structural reforms in enabling developing and transition economy businesses to take advantage of market access opportunities created by trade agreements4 A first example illustrating how competition policy bears on realization of the gains from trade liberalization concerns the importance of structural reforms in developing and transition economies in ensuring that producers in such countries are well-equipped to enjoy and profit from the market access opportunities that trade liberalization opens up to them. Osakwe (2001) emphasizes that failures of trade liberalisation to generate sustained development and growth in many cases can be traced to a failure to introduce complementary domestic policy reforms. In most cases, countries (and their businesses) will not be well poised to take advantage of the potential benefits of trade liberalisation unless steps are also taken to reduce costs and enhance the efficiency of infrastructure sectors such as telecommunications, energy and transportation; to promote flexibility by eliminating artificial restrictions on entry, exit and pricing in manufacturing and other industries; and to establish and strengthen incentives for investment, innovation, the creation of efficient management structures and productivity improvement (see also Anderson and Jenny 2005). Further to the above, historically, key infrastructure services, whether in the fields of transportation, energy or telecommunications, were often provided by monopolies, in both developed and developing economies. While such monopolies have occasionally emerged through monopolistic or predatory behaviour by the firms themselves, they have most often been established through legislation or by government grant. Frequently, the combination of monopoly power and, where present, public ownership has resulted in less-than-satisfactory performance, as manifested by higher-than-competitive rates, a lack of adequate service offerings, or a lack of innovation or readiness to adapt improvements in technology as they become available. This, in turn, has directly undermined the competitiveness of developing country business users, as infrastructure services represent a key factor determining input costs for other businesses. Consequently, a major focus of efforts to improve the efficiency of public and business infrastructure services in developing and transition economies has been on efforts to ‘de-monopolize’ 4 This section of the paper draws on insights gleaned in a project undertaken for the International Trade Centre (ITC) by the two authors in addition to Frédéric Jenny. See ITC (2012).

Robert D Anderson and Anna Caroline Müller 

137  

such industries and ensure fair competition, to the benefit of users (see ITC 2012; Beato and Laffont 2002; Kessides 2004; and Francois and Manchin 2007). Addressing sub-optimal performance in infrastructure sectors due to a lack of competition can require the application of a variety of remedial measures. To be sure, an important element of the policy response to a lack of competition in public infrastructure sectors is the adoption or maintenance of a competition or antitrust statute. Such a statute is essential to prevent and/or remedy common anticompetitive practices such as abuses of a dominant position or the establishment of price-fixing cartels that raise costs to business users. Moreover, in a number of instances, competition laws, typically through application of their provisions regarding abuses of a dominant position, can be used as a platform to impose necessary re-structuring and establish competitive access regimes (see Anderson and Heimler 2007a). In many cases, however, the remedies available through competition law enforcement may not be sufficient and other measures may be needed to ‘get to the root of’ and effectively address monopoly issues in infrastructure industries. Such measures may include: (i) repeal or reform of statutes or regulations that unnecessarily limit entry to particular markets; and (ii) enactment of new legislation to re-structure (ie break up) established monopoly enterprises and permit competition to take place, for example by establishing industry-specific competitive access regimes. Such measures may be adopted either as an alternative to, in conjunction with, or even as a follow-up to competition law enforcement proceedings.5 It should also be noted that, while measures to inject competition into moribund infrastructure monopolies have most often been implemented at the national level, in many cases there is also an interface with international trade agreements and co-operation. For example, in Africa, the creation of common regional markets may be a necessary step to establish effective competition in some elements of the transportation sector, due in part to a lack of adequate demand to support multiple service providers in some individual countries (see, for related discussion, Teravantinthorn and Raballand 2009). Similarly, regional co-operation can be a key factor in facilitating competition in energy markets. Trade commitments, including in the 1997 WTO negotiations on Basic Telecommunications Services, have played an important role in reinforcing the effectiveness of pro-competitive reforms in the telecommunications sector.6 Again, the point here is not that the international trading system itself necessarily needs to grapple directly with this issue (though that might be argued), but that the success of the system in terms of creating new prosperity for developing 5 For example, the far-reaching reform of telecommunication services that took place in the U.S. was to a remarkable extent put in motion by the 1982 consent settlement in the AT&T antitrust case; nevertheless, extensive legislative action was required to complete the process. A complementary relationship between competition law enforcement actions and legislative reforms in the reform of public infrastructure sectors has also been evident in the European Union. See Anderson and Heimler (2007b). Canada’s experience in regard to these issues is discussed in Anderson et al. (1998). 6 See the discussion in Part 3 below.

138  Competition, Regulation and Public Policies country producers (and others) depends on appropriate reforms being undertaken in parallel. To that extent, the issue of pro-competitive structural reforms and their links to the exploitation of market access opportunities by developing country and other producers itself illustrates the significance of effective national competition policies for the international trading system.

2.2  Second example: the ability of international cartels to undermine the gains from trade The second example to be addressed concerns the ability of international cartels to undermine the gains from trade. This was, indeed, an important consideration taken up in the work of the WTO Working Group on the Interaction between Trade and Competition Policy when that body was active, in the period leading up to the Cancún Ministerial. In the course of that work, important note was taken of the work of Levenstein and Suslow (2001), whose policy implications were further elaborated in Evenett, Levenstein and Suslow (2001), and which indicated that many cases of such cartels involved firms headquartered in the developed world with substantial exports to developing countries. Looking at 16 ‘cartelized’ products, Levenstein and Suslow noted as follows: Examining these sixteen products – which were cartelized at some point during the 1990s and for which we were able to obtain reasonably reliable trade data – the total value of such ‘cartel-affected’ imports to developing countries was $81.1 billion. This made up 6.7% of all imports to developing countries. [The excess cost placed on developing country importers as a result of these arrangements] is equal to 1.2% of their combined GDP [annually].

To be sure, since that time, the original work done by Levenstein, Suslow and others in this area has been qualified, extended and elaborated, by the two authors and by others, in important ways (see, in addition to many other relevant sources, Connor 2011 and references cited therein). These contributions have, on the whole, buttressed and reinforced rather than weakened the original findings. In particular, the available literature on international cartels has shown beyond any doubt the ability of such arrangements to artificially raise prices and reduce output in markets. Furthermore, it has shown that, in many cases, such arrangements also deter cross-trading by one country’s suppliers into markets assigned to other countries’ suppliers. To this extent, international cartels clearly inhibit realization of the traditional gains from trade. It should be emphasized that this is not at all to suggest that the international trading system should carry the ‘front-line’ responsibility for dealing with such arrangements, but that ensuring that they are properly dealt with is an important and legitimate concern of the system. Further to the above, some would dispute that international cartels can appropriately be characterized as a ‘trade’ issue, preferring that they be dealt with as an ‘international antitrust’ issue. However, these two domains may overlap.

Robert D Anderson and Anna Caroline Müller 

139  

Recognizing that strong antitrust laws and agencies are the key to addressing this issue, the fact remains that international cartels can undermine the benefits of trade liberalization; hence, ensuring the existence of laws and institutions to address these arrangements is arguably a legitimate concern of the international trading system. Furthermore, the inducement to cartelization that may be provided by national industrial and other policies has also been noted in many studies (see, eg, International Competition Policy Advisory Committee 2000 and WTO Working Group on the Interaction between Trade and Competition Policy 1998, paragraph 87). To this extent, measures to promote trade liberalization and to limit the role of mercantilist national industrial policies are also relevant to the issue.

2.3  Third example: the importance of competition rules as an adjunct to a liberalized government procurement regime A further example of a specific application of national competition policy that is important in all economies and also to realize gains from international trade liberalization relates to the prevention of bidrigging, ie collusion between suppliers, in public procurement processes. Indeed, the enforcement of relevant competition rules can fairly be characterized as a necessary adjunct to the opening up of procurement markets internationally (see, for related discussion, Anderson, Kovacic and Müller 2011). The WTO’s Government Procurement Agreement (GPA), for example, aims at promoting the welfare of citizens by ensuring full transparency in government procurement regimes and providing market access to foreign suppliers. Provisions similar to the GPA are incorporated in many RTAs, including with Parties to the Agreement and non-Parties. Typically, these texts do not establish extensive and separate competition rules as tools to combat bidrigging. There is, however, a clear need to apply competition rules to bidders, even where international market opening has taken place: rigging bids can be made more difficult, but cannot be eliminated merely by opening procurement processes to foreign competitors, since the latter may be party to any bidrigging conspiracies. Furthermore, even after procurement rules have been liberalized, competition policy has a role to play in addressing regulatory barriers to supplier participation and ensuring that foreign bidders are not excluded by private sector anticompetitive behaviour (Anderson, Kovacic and Müller 2011). Box 1 (next page) presents examples of bidrigging schemes that have been successfully prosecuted in recent years in various jurisdictions including developed, developing and transition economies. Several of the examples (those from China, Indonesia, Peru and Chinese Taipei) are taken from inputs prepared by those countries for the 2001 OECD Global Forum on Competition. These cases demonstrate the universality of the challenge of deterring collusive practices – ie such practices are in no way limited either to developed or to developing

140  Competition, Regulation and Public Policies countries.7 The cases also illustrate a number of common characteristics of bidrigging schemes. For example, in several of the cases collusion seems to have been facilitated by restrictive regulations and/or practices of the procuring entities. The cases in Box 1 also illustrate that the mere opening of bidding processes to foreign-based suppliers may not generate effective competition, if effective rules are not in place to deter collusion. The fourth case noted in the table – a conspiracy to rig bids on construction contracts funded by the United States Agency for International Development (USAID) in Egypt – is interesting in that it shows that collusion in tendering processes can impact directly on international assistance efforts. Of course, these are but a few examples of the much larger numbers of bidrigging schemes that have been successfully prosecuted by relevant authorities. Box 1 Examples of international and domestic collusive tendering schemes that have been prosecuted in various jurisdictions a) International removal and relocation services in Belgium In 2008, the European Commission imposed fines totalling €  32,755,500 on various large firms providing international removal and relocation services in Belgium for fixing prices, sharing the market and bidrigging, in violation of the EC Treaty’s ban on cartels (Article 81 EC, now Article 101 TFEU). The cartel operated for almost nineteen years. Cartel members fixed prices, presented bogus quotes to clients and compensated each other for lost bids. Source: “Antitrust: Commission fines providers of international removal services in Belgium over €32.7 million for complex cartel.” (EC Commission, Press Release, IP/08/415, 11 March 2008). b) The International Marine Hose Case Marine hose is a flexible rubber hose used to transport oil between tankers and storage facilities and buoys. According to court papers and other documents, firms based in the United Kingdom (UK), France, Italy and Japan conspired to fix prices and rig bids for hundreds of millions of dollars worth of marine hose and related products which were sold to other firms in addition to government agencies. The conspirators met in locations such as Key Largo (Fla.), Bangkok and London. The investigation of this case involved coordinated enforcement efforts by the U.S. Department of Justice, the EC Commission and the UK Office of Fair Trading. 7 The foregoing is not at all to suggest that maintaining competition in developing countries does not involve special issues and challenges. Factors differentiating the role of competition policy in developing as compared to developed countries may include any or all of the following: (a) higher ‘natural’ entry barriers due to inadequate business infrastructure, including distribution channels and (sometimes) intrusive regulatory regimes; (b) asymmetries of information in both product and credit markets; (c) a greater proportion of local (non-tradable) markets; and (d) over-stretched/inadequately developed law enforcement and judicial systems (Kovacic 2001; Anderson and Jenny 2005). The point is simply that the problems faced by developing and transition economies in maintaining competition are not wholly dissimilar to those of developed countries and, therefore, that there is much to be gained for both sides in sharing experiences.

Robert D Anderson and Anna Caroline Müller 

141  

Source: U.S. Department of Justice, “Eight executives arrested on charges of conspiring to rig bids, fix prices, and allocate markets for sales of marine hose,” Press Release, May 2, 2007 and “Three United Kingdom Nationals Charges With Participating in Worldwide Bid-Rigging Conspiracy in the Marine Hose Industry,” U.S. Department of Justice, Press Release, December 3, 2007. c) Prosecution of bidrigging in school construction in China Ten construction companies were prosecuted for rigging bids on a contract for the construction of a school building. The ten companies agreed that one of them would get the contract in exchange for payments to the other companies. They also assigned another company to calculate the bidding prices of all candidates. The designated company won the bid at an artificially high price. The administration for industry and commerce issued a decision, declaring that the bid was invalid and the illegal gains were confiscated. Source: OECD Global Forum on Competition, Summary of cartel cases described by invitees (CCNM/GF/COMP(2001)4, 5 October 2001). d) Bidrigging on USAID contracts in Egypt Philipp Holzmann AG, a Frankfurt, Germany construction firm, pleaded guilty and was sentenced to pay a $30-million fine for its participation in a conspiracy to rig bids on construction contracts funded by the United States Agency for International Development (USAID) in Egypt. Source: U.S. Department of Justice, “German Company Pleads Guilty To Rigging Bids on USAID Construction Contracts in Egypt,” (Press Release, 18 August 2000). e) The rigging of bids for the supply of pipe and pipe-processing services in Indonesia Three pipe processors were found to have exchanged their prices with each other at a meeting in a hotel the evening before the bids were opened. Material evidence was contained in statements of a complainant, as well as in the testimony of witnesses from the respondents. As this was the first case ever brought by the Commission, no fines or other sanctions were imposed. Instead, the Commission ordered that the contract between the procuring entity and the apparent lowest bidder be dissolved and that the entire tender process be redone. Source: OECD Global Forum on Competition, Summary of cartel cases described by invitees (CCNM/GF/COMP(2001)4, 5 October 2001). f) Rigging bids for the supply of construction services in Peru Three companies were convicted of participating in bidrigging on a contract for the construction of a secondary electricity net in Puerto Maldonado City. The claim was based on evidence from the documents presented by the three bidders. The documents contained the same redaction and the same format; they also presented the same orthographic errors, the same time of construction and almost the same price bid. Following an investigation, the Free Competition Commission ordered the three companies to cease the practice and imposed fines. Source: OECD Global Forum on Competition, Summary of cartel cases described by invitees (CCNM/GF/COMP(2001)4, 5 October 2001).

142  Competition, Regulation and Public Policies g) The rigging of bids for the procurement of truck-mounted mobile cranes by the Taiwan Power Company in Chinese Taipei Six companies were prosecuted for having knowingly, and through mutual communications, apportioned the number, suppliers and amounts of the winning bids before the bid opening. These acts violated Article 14 of the Fair Trade Law, which prohibits concerted acts. The Commission ordered them to cease the concerted practices. The case also included another violation of the Law committed by Taiwan Power Company that improperly restricted the criteria to bid on its contract. The company was ordered to cease its actions. Source: OECD Global Forum on Competition, Summary of cartel cases described by invitees (CCNM/GF/COMP(2001)4, 5 October 2001).

Bidrigging in public procurement markets accounts for a striking percentage of prosecutions by competition authorities in jurisdictions where such authorities are well-established. For example, from 1972 through 1992, the U.S. Department of Justice (DOJ) obtained 1159 criminal indictments for Sherman Act violations. Some 625 of these indictments, nearly 54 per cent, attacked collusion against public procurement bodies.8 Indeed, the prevention of collusive tendering is similarly a focus of enforcement efforts for most national competition authorities. To this extent, the problem is already being vigorously addressed, at least in some jurisdictions. The point of this discussion is not to suggest otherwise, but simply to note that competition law enforcement in this area is a very important adjunct to liberalized procurement regimes, without which such regimes are unlikely to achieve their full potential benefits. The particular development dimension of the fight against bidrigging as a means to alleviate poverty and improve public health is demonstrated by the following cases in the health care sector set out in Box 2.9 Box 2 Bidrigging examples in health care procurement markets “In the South African case of The Competition Commission vs Adcock Ingram Critical Care (AICC) and four others, it was established that five pharmaceutical companies had, over a period of at least fourteen years, colluded in rigging bids for the supply of intravenous solutions to public hospitals, a tender initially issued annually and later every two years. Prices were fixed, markets worth many hundreds of millions of [South African] Rands were allocated and winners and losers were determined, as was the compensation, often in the form of a post-award sub-contract, for the losers. In a statement submitted to the hearing at which the Tribunal approved the consent decree, a representative of the national Department of Health succinctly summed up the character of bid rigging and the nature of the problems that it posed:

8 These data were collected from the Commerce Clearing House Trade Regulation Reporter for the years in question. 9 See also OECD (2013a).

Robert D Anderson and Anna Caroline Müller 

143  

‘The Department of Health purchases large volume parenterals and administration sets through the public tender system in order to secure affordable prices, which ultimately benefit the many people who use the public health system, and as has been stated, the majority of whom are poor. The advantages for the suppliers are a guaranteed market, economies of scale and a binding contract.... We are committed to giving preference to local manufacturers to promote job creation, poverty eradication and skills development. However, it is difficult to pursue these objectives of promoting local manufacture when such manufacturers act in such a manner. We find it very disturbing that SMMEs that get preferential points in the tender system to enable them to gain market share, resort to this kind of behaviour... These findings beg the question of whether this is the only case of collusion in the industry and there is a high possibility that this is not the only case of collusion in the industry. The challenge that we face is: how does one prevent such collusive practices in the future? Tender systems, by their very nature, are at risk of collusion, especially in the pharmaceutical sector where there are usually only a handful of competitors that are known to one another.’ ” A submission by Turkey to the OECD roundtable on collusion and corruption in public procurement “reveals that in 2009 most of the bidrigging investigations carried out by the [Turkish Competition Authority] TCA were in the health sector, including medicines, laboratory supplies and medical equipment. This appears to have included the prosecution of several cases of collusive boycotts of tenders issued by procurement authorities in the health sector. As with the South African public health official cited here, the Turkish submission ascribes the frequency of bidrigging in health markets to the oligopolistic structure of the suppliers’ markets. In 237 out of 310 tenders issued in Turkey for laboratory equipment, fewer than 3 suppliers responded to the tenders.” A submission by Romania to the same roundtable also “identifies the health sector as the sector most vulnerable to bidrigging practices. Thus, in 2008, the [Romanian Competition Council] RCC imposed fines totalling approximately Euro 22.6 million on four pharmaceutical companies for sharing the publicly funded section of the insulin market in the context of a national tender organised in 2003 by the Ministry of Health. The collusive practice in this case aimed at sharing the diabetes product portfolio of a drug manufacturer between 3 distributors. In another important case, 3 distributors who rigged a bid in response to a national tender issued by the Ministry of Health for the supply of dialysis products and equipment were fined Euro 1.5 million.” Source: OECD, 2010.

2.4  A further reflection As a further reflection relevant to all of the above examples, in the past, it was sometimes argued that competition policy is irrelevant in circumstances of extreme poverty. However, where incomes are severely limited, it would seem even more important than otherwise that the purchasing power of consumers

144  Competition, Regulation and Public Policies not be further diminished through anticompetitive practices.10 Also, there is substantial evidence that anticompetitive practices are particularly prevalent in regard to goods for which there are limited substitutes available in developing country markets, for example foodstuffs. Indeed, many of the major international cartels that were disclosed in the mid to late 1990s and that are believed to have been active in developing country markets concerned foodstuffs, for example those relating to the sale of vitamins, lysine and citric acid, all important inputs to agri-food production (Levenstein and Suslow 2001; Evenett, Levenstein and Suslow 2001; Anderson and Jenny 2005). Similarly, the effective prevention of bidrigging would seem to be particularly important where, as in many developing countries, governments are subject to severe fiscal constraints. A decade ago, it was also argued that developing countries had very little capacity or experience with respect to competition policy issues, and could not, therefore, be expected to grapple effectively with such issues. However, since then, many more developing countries have adopted modern competition laws (in all, more than one hundred WTO Members, including such powerhouse ‘emerging markets’ as China, India, Brazil and South Africa, now have such laws). Furthermore, extensive valuable work has been done by organizations such as UNCTAD, the OECD and CUTS to increase such countries’ capacity in this area (see, regarding CUTS’ impressive work, Mehta 2006 and, more generally, the materials available at www.cuts.org). In this context, while the control of anticompetitive practices poses significant challenges for all countries, and more assistance for developing countries grappling with these issues is urgently to be desired, the argument that these issues are fundamentally beyond the capacity of developing countries to address is increasingly unsustainable.

3.  Second proposition: competition policy already figures in the wto agreements, in important though limited ways As is well known, and this is a starting point for the present chapter, currently, the WTO Agreements do not incorporate a comprehensive body of rules on competition law and policy. Indeed, the question of the need for such a body of rules is by no means resolved, nor does this chapter suggest otherwise. It is worth noting, nonetheless, that in practice, it has proven impossible to exclude competition policy considerations altogether from the WTO Agreements. Indeed, the importance for the system of measures to ensure the competitive operation of markets is reflected in a number of provisions and subordinate instruments that have been incorporated in the various Agreements over the years, particularly with the 1994 transition from the GATT to the WTO. 10 See generally OECD 2013b, and the WTO contribution entitled ‘Competition and Poverty Reduction: A Holistic Approach’ therein.

Robert D Anderson and Anna Caroline Müller 

145  

3.1  First example: competition rules embodied in the GATS Agreement and the Reference Paper on Basic Telecommunications Services For example, Article 8 of the General Agreement on Trade in Services (GATS) obliges Members to ensure that any monopoly supplier of a service in its territory does not, in the supply of the monopoly service in the relevant market, act in a manner inconsistent with that Member’s obligations under Article II and specific commitments under GATS. Article 9 of the GATS further enlarges the application of competition principles to GATS disciplines recognizing that certain business practices of service suppliers, other than those falling under Article 8 GATS, may restrain competition and thereby restrict trade in services and obliging Members to enter into consultations with a view to eliminating such practices upon request by another Member (see World Trade Organization 1997). Another important example of the incorporation of competition policy concepts and principles into elements of the WTO Agreements concerns the so-called ‘Reference Paper’ on regulatory principles which forms part of the commitments made by most WTO Members in the context of the WTO Negotiations on Basic Telecommunications Services, conducted under the overall rubric of the General Agreement on Trade in Services (GATS), which was concluded in February 1997. The Reference Paper is intended to address, inter alia, situations where the services provided by public telecommunications networks constitute essential facilities that are exclusively or predominantly provided by a single or limited number of suppliers and for which there are no feasible substitutes – a situation which potentially constitutes an impediment to both competition and market access for service suppliers. To address this concern, the Reference Paper sets out detailed rules relating to interconnection of downstream service providers with major suppliers on non-discriminatory terms; the prevention of anticompetitive acts, including anticompetitive cross-subsidization and the making available of information needed for efficient inter-connection. These rules draw importantly on concepts of antitrust and regulatory policy such as exclusionary practices and the essential facilities doctrine (Anderson and Holmes 2002). Key elements of the Reference Paper and related provisions of Mexico’s GATS commitments were considered in the 2007 WTO Panel Decision in the Mexico: Telecoms (Telmex) case. In this case, which was brought against Mexico by the United States, the panel found that several features of Mexico’s framework for regulation of international telecommunications services were in violation of Mexico’s commitments under the Reference Paper (see Box 3 for a summary of points relating to the competition dimension of this matter).11 Rather than appealing the case to the WTO Appellate Body, Mexico chose to accept the panel’s ruling. In the view of some observers, it did so precisely because this was in the best interest of Mexico’s consumers and the long-run development of 11 For a more complete summary also covering other aspects of the case, see http://www.wto.org/ english/tratop_e/dispu_e/cases_e/ds204_e.htm.

146  Competition, Regulation and Public Policies Mexico’s telecommunications sector (see, eg, Hufbauer and Stephenson 2007; and, for useful related commentary, Fox 2006). Box 3 Competition-related elements of the WTO Panel Report in Mexico – Telecoms (DS204) a) Procedural aspects of the Dispute Complainant: United States Respondent: Mexico Establishment of Panel: 17 April 2002 Adoption of Panel Report: 1 June 2004 b) Background Telmex was, for many years, the dominant supplier of telecommunications services in Mexico. According to the WTO panel report, the applicable regulations in Mexico conferred on Telmex (as the long-distance service licensee having the greatest percentage of outgoing long-distance market share for the relevant country in the previous six months) the power to negotiate the rate to be paid by foreign carriers (including U.S. carriers such as AT&T and MCI) for the interconnection of calls terminating in Mexico. In addition, Mexican laws required all other licensed Mexican concessionaires to charge no less than the fee negotiated by Telmex for similar services. c) Summary of Competition-related Findings of the Panel The Panel ruled that Mexico violated its commitments under the Reference Paper and the GATS in that: • Mexico failed to ensure interconnection at cost-oriented rates for the cross-border supply of facilities-based basic telecom services, contrary to Article 2.2(b) of its Reference Paper; • Mexico failed to maintain appropriate measures to prevent anticompetitive practices by firms that are a major telecom supplier, contrary to Article 1.1 of its Reference Paper; and • Mexico failed to ensure reasonable and non-discriminatory access to and use of telecommunications networks, contrary to Article 5(a) and (b) of the GATS Annex on Telecommunications. Source: World Trade Organization (2006); see, for insightful commentary, Fox (2006).

As to the systemic importance of the case, Fox (2006) concludes that: The Mexican telecom case illuminates why competition rules must extend cross-border and why hybrid trade-and-competition (public/private) restraints must be treated as a unified whole, if we are to realize the good potential of globalization. […] The GATS Annex with its Reference Paper is the first instrument providing a unified vision for disciplining linked public and private restraints. The Panel Report’s interpretation of the antitrust obligation gives life to the discipline. A positive reading of the antitrust clause is a step forward on intertwined issues of trade and competition.

Robert D Anderson and Anna Caroline Müller 

147  

3.2  Second example: the competition rules of the TRIPS Agreement The area of intellectual property rights is another important example of a sphere in which the role of competition policy is already directly reflected in an existing WTO Agreement, the Agreement on Trade-Related Intellectual Property Rights (TRIPS) (Anderson 1998, Anderson and Holmes 2002). At a broad level, Article 8.2 of the Agreement stipulates that: Appropriate measures, provided that they are consistent with the provisions of this Agreement, may be needed to prevent the abuse of intellectual property rights by right holders or the resort to practices which unreasonably restrain trade or adversely affect the international transfer of technology.

In the same spirit but focusing on the specific issue of licensing practices, Article 40.1 of the Agreement notes that: Members agree that some licensing practices or conditions pertaining to intellectual property rights which restrain competition may have adverse effects on trade and may impede the transfer and dissemination of new technology.

To address this concern, Article 40.2 recognizes the right of Member governments to take measures to prevent anticompetitive abuses of intellectual property rights, provided that such measures are consistent with relevant provisions of the Agreement. Article 40.2 also contains a short illustrative list of practices which may be treated as abuses.12 It should be noted that neither Article 8.2 nor Article 40.2 indicates that specific practices shall be treated as abuses or specifies remedial measures that must be taken. In this sense, the competition provisions of the Agreement are permissive rather than mandatory (Anderson 1998). Article 40.3 of the Agreement provides that a Member considering action against an intellectual property owner that is a national or domiciliary of another Member can seek consultations with that Member. The latter Member is required to cooperate through the supply of publicly available non‑confidential information of relevance, and of other information available to that Member, subject to domestic law and to the conclusion of mutually satisfactory agreements concerning the safeguarding of its confidentiality. Competition policy considerations are also embodied in the TRIPS Agreement provisions relating to compulsory licensing in respect of patents. Article 31 of the Agreement sets out detailed conditions that must be respected in the granting by Member states of any compulsory licences. However, subparagraph (k) of Article 31 stipulates that Members are not obliged to apply certain of these conditions13 in circumstances where the compulsory licence is granted ‘to remedy a practice determined after judicial or administrative process to be anti‑competitive.’ In particular, requirements to show that a proposed user has made efforts to obtain 12 These are exclusive grant‑back conditions, conditions preventing challenges to validity and coercive package licensing. 13 Specifically, those contained in paragraphs (b) and (f) of Article 31.

148  Competition, Regulation and Public Policies voluntary authorization from the right holder on reasonable terms and conditions and that such efforts have not been successful within a reasonable period of time are not applicable in these circumstances. In addition, the requirement (in Article 31 (f)) that authorization for use of a patent under a compulsory licence be predominantly for the supply of the domestic market of the Member authorizing such use can also be rendered inapplicable by such a finding.

3.3  Third example: the WTO Agreement on Safeguards In a related vein, Article 11:3 of the Agreement on Safeguards prohibits Members from encouraging or supporting the adoption of non-governmental measures equivalent to voluntary export restraints, orderly marketing arrangements or other governmental arrangements prohibited under Article 11:1. This was considered to be necessary if the latter (explicitly governmental) arrangements were not to be replaced by ‘private’ cartels having similar effects that were encouraged or orchestrated by governments (again, see World Trade Organization 1997). Even though these and other examples represent, at best, a partial and ad hoc integration of competition policy concepts and provisions into the multilateral trading system, they demonstrate that such considerations cannot be entirely excluded from the system. As discussed below, they also raise the question of whether a vehicle for ongoing interaction between international trade and competition officials is needed to ensure appropriate application of relevant provisions.

4.  Third proposition: there are additional issues at the interface of trade and competition policy which (arguably) call for further joint application of the two disciplines Apart from the issues that have already been discussed concerning the interface between international trade and competition policy, additional issues arguably remain to be addressed. In general, these issues are not currently within the scope of application of national competition laws, as they concern areas of traditional exemption from national competition laws and/or involve an element of state action and, to that extent, enjoy an implied immunity in some jurisdictions. Nonetheless, these issues have the potential to impact negatively on both competition and international trade. The two issues discussed in the following paragraphs, export cartels and the principle of competitive neutrality, provide examples of issues that have traditionally been treated as ‘separate boxes’ from private competition restraints within the purview of national competition laws.

Robert D Anderson and Anna Caroline Müller 

149  

4.1  The disciplining of export cartels Export cartels, which focus on raising prices and extracting additional profits only in the firms’ export markets (in order to avoid prosecution in the home market), raise important issues for both competition and international trade. In many cases, export cartels operate under the umbrella of statutory exemptions for such arrangements in the home countries of the participating firms. The logic of such exemptions is that the conduct involved does not impact adversely on consumers in the home market; consequently, it is not of concern (or, at least, primary concern) from the standpoint of the home country’s competition authority. This approach, however, may raise concerns for consumers in the export markets, especially those of developing countries which may lack the resources or knowhow to address the harmful consequences of such conduct themselves, particularly where the firms involved are located abroad. Jenny and Mehta (2012) summarize the impact of such arrangements and the reasons why they typically avoid prosecution as follows: Export cartels have a significant influence on prices in general and on the swing of prices of primary products in particular. Competition authorities in the countries of origin of the export cartels do not act against them because export cartels do not affect the domestic markets of the cartelists. Competition authorities in the victimised countries do not have powers to act against the export cartels, which they suffer from for a variety of reasons. They may lack extra-territorial jurisdiction (as the litigation in India against the US-based soda ash cartel under the now repealed [Indian] Monopolies and Restrictive Trade Practices Act, 1969, showed); the sovereign compulsion doctrine may prevent them from prosecuting state-sponsored export cartels; they may not have the means to gather the evidence they would need to convict the perpetrators even if they have jurisdiction, or they can be under pressure from their government not to act against them so as not to expose the country to retaliations endangering its own economy and statesupported export cartels.

While there have been strong calls to eliminate exemptions for export cartels from national competition laws, the debate is on-going and, in practice, has not resulted in significant policy change to date (see, for discussion, Bhattacharjea 2004, Levenstein and Suslow 2005, Fröberg (undated), Sweeney 2007 and Hoekman and Saggi 2007). As a result, developing countries presently cannot simply ‘outsource’ anti-cartel enforcement in relation to export cartels to developed countries. Some argue that, eg co-operation among developing country producers may indeed have positive effects on their ability to participate in international trade. However, it should be borne in mind that the empirical literature is largely inconclusive on the overall efficiency benefits of export cartels. Examples of some cartels/other co-operation arrangements show that they have the potential to impact negatively on developing country producers by annulling the benefits that should result from international trade.14 As pointed out before, therefore, the 14

For discussion, see Jenny (2012); Taylor and Moss (2013).

150  Competition, Regulation and Public Policies potential positive and negative effects on export cartels will have to be weighed carefully depending on the particular circumstances of each case. Another important distinction that should be noted is between true export cartels and joint ventures. Whereas true cartels are concerned only with manipulating prices and outputs (to increase profits for the participating firms) and normally do not involve socially useful co-operation between the participating firms, for example in the form of joint research or marketing arrangements, joint ventures may involve such co-operation and should, in general, be judged under a different standard. As Fröberg (undated) and Sweeney (2007) point out, where true and harmful export cartels occur, they are more likely to be able to extract supra-competitive benefits or cartel overcharges in importing developing countries (compared with developed economies), as they often face limited domestic competition there. Therefore, the lack of enforcement action against such cartels can be expected to affect developing country businesses and consumers disproportionally. As a result, an adequate means to assess the effects of export cartels on international markets and to combat them when and where necessary can be considered as desirable from a development standpoint. Due to the limitations of national competition laws in regard to export cartels, international co-operation may hold some of the answers needed to find adequate solutions.

4.2  The principle of competitive neutrality and its application to stateowned enterprises (SOEs) As Fox (2006) argues, trade-restraining action by states erects some of the most serious barriers to trade and competition, harming consumers, impairing economic opportunity and growth of business, and setting back the opportunity for integration of developing countries into the world trading system. However, state measures are difficult to challenge, and the negative synergy of restraints of trade and restraints of competition have largely been entirely ignored in the debate on the interface between international trade and competition. An issue that has been receiving increasing attention, however, is the application of the principle of competitive neutrality to the economic activity of state-owned enterprises. The OECD (2009) finds in this regard that, due to their privileged position, SOEs may negatively affect competition and should, to the greatest extent possible consistent with their public service responsibilities, be subjected to similar competition disciplines as private enterprises. However, enforcing competition rules against SOEs presents enforcers with particular challenges and the application of competition laws to such enterprises may not be sufficient ensure the desired results. For example, while competition authorities may be presumed, in general, to be impartial in their investigations, the OECD study points out that ‘it is nevertheless theoretically conceivable that, in some instances,

Robert D Anderson and Anna Caroline Müller 

151  

they could be exposed to the risk of undue government influence when enforcing competition laws against SOEs. Also, competition authorities may often lack sufficient statutory power over the SOE, in particular, with respect to industries that are subject to oversight by sectoral regulatory agencies.’ Furthermore, enforcement authorities may face substantive challenges, such as difficulty in obtaining relevant information from SOEs due to lack of transparency regarding costs and insufficient standard accounting procedures, or questions as to how goals other than profit maximization should be taken into account. Overall, it may be very difficult to determine whether an SOE is cross-subsidizing, pricing below competitive levels or engaging in other forms of anticompetitive conduct. A possible solution to address these issues may be found in policies aimed at achieving competitive neutrality in markets where public and private enterprises compete. Such policies try to establish a regulatory framework (i) within which public and private enterprises face the same set of rules and (ii) where no contact with the state brings competitive advantage to any market participant, to the extent that these principles are compatible with the specific public goals pursued by the SOE in question. An aspect that should, however, not be ignored in this debate is that a competitive advantage provided to SOEs may also distort international trade, for example to the extent that the SOE in question may be a significant exporter of goods or services or that international importers of competing goods are affected by SOEs. Therefore, the application of the principle of competitive neutrality arguably is not only of crucial importance to address general competition concerns, but also to address barriers to and distortions of international trade. More research and debate in this regard will be needed in order to devise policies taking this interface into account adequately.

5.  Observation: competition policy provisions figure extensively in recent regional trade agreements A further consideration bearing on the issues and developments discussed above is that a large number of RTAs concluded between WTO Member governments contain detailed provisions dealing with competition law and policy-related matters. Two main studies have been carried out of these phenomena: Anderson and Evenett (2006) and Teh (2009). As an initial observation, Anderson and Evenett (2006) observe as follows: Perhaps the most important development is that competition language is being used increasingly in non-competition-specific chapters of RTAs, in particular in chapters on the general obligations of the parties, on state-owned enterprises, and on regulated industries. That is, whether or not an RTA has a designated chapter dealing specifically with competition policy is not necessarily a reliable guide as to potential impact of

152  Competition, Regulation and Public Policies competition language codified in [an] RTA. Nor [ ] is it simply a matter, as some fear, of trade principles contaminating competition law; there is plenty of movement in the opposite direction with competition principles influencing the interpretation of traderelated obligations.

Carrying forward this analysis, Teh (2009), who analyses 76 RTAs, finds that: Nearly three-quarters of the RTAs contain competition policy provisions. Despite the reluctance of many developing countries to enter multilateral negotiations on competition policy, fifty of the sixty-eight RTAs with developing countries as members have a competition policy provision or chapter. The principal objective of these provisions is to prevent the gains in market access arising from the RTA from being eroded by anti-competitive behaviour that is condoned or tolerated by RTA partners. The main obligations thus involve adoption or application of competition laws to curb anti-competitive behaviour and closer co-operation among competition authorities of RTA partners.

Continuing, Teh (2009) observes that: More than 40 per cent of the RTAs in the sample stipulate promotion and advancement of the conditions of fair competition as an objective of the RTA suggesting that competition considerations are not necessarily subordinate to the goal of expanding trade and investments. A large number of the RTAs in this survey included competition disciplines in the chapters on investment, services (in telecommunications, maritime transport and financial services), government procurement and intellectual property.

Another important qualitative finding is that the competition chapters of some FTAs appear to be as advanced, if not more advanced, in their provisions to foster cooperation between competition agencies on enforcement matters as second generation inter-agency agreements between competition authorities. Moreover, the competition provisions of some RTAs appear to have been deliberately structured so as to strengthen the competition agencies in at least one signatory, ensuring more resources, stature, and powers are made available to the implementing body. It is also worth noting that the competition elements in many recent RTAs go well beyond the degree of cooperation that was envisioned in the proposals for a multilateral framework on competition policy in the WTO (Anderson and Evenett 2006). A final observation concerns the implications of these developments at the regional level for the treatment of competition policy at the multilateral level – ie, in the WTO. Reflecting on this, Anderson and Evenett comment as follows: As we have seen, a large number of recently-concluded regional trade or similar agreements implicating countries in Latin America, Africa, and Asia in addition to Europe and North America contain provisions requiring the adoption of laws or policies dealing with anticompetitive practices and/or mandating co-operation between the participating countries in addressing such practices. The proliferation of these provisions calls into question directly both the idea that competition policy is unrelated to the realization of trade objectives and the perception that the subject is so complex that it cannot be meaningfully reflected in trade agreements. Indeed, in the light of the large and growing

Robert D Anderson and Anna Caroline Müller 

153  

number of provisions on competition policy in regional trade agreements, the issue is no longer whether competition policy and trade liberalization can be meaningfully linked but whether the relevant synergies will be harnessed only in regional arrangements or also in the multilateral context. (emphasis added)

6.  Summary and questions for the future Whether there is a need for more explicit linkages to be developed between national competition policies and the multilateral trading system – and, if so, the appropriate content of related disciplines – is an unresolved question in debates surrounding the future of the WTO. As is widely known, a previous effort, in the WTO, from 1997 through 2003, to develop a ‘multilateral framework on competition policy’ foundered and was placed on hold in the aftermath of the Cancún Ministerial Conference. Furthermore, since that work was put on hold, international co-operation in the field of competition policy has moved ahead in various fora, including the International Competition Network (ICN), the UNCTAD Intergovernmental Group of Experts on Competition Law and Policy and the OECD Competition Committee. Many observers would take the view that such international cooperation as is needed with respect to the field of competition policy is already being effectively implemented in these fora. The foregoing view can, however, be questioned on various grounds. For example, Jenny (2003) has argued that existing co-operation mechanisms are inadequate to address the harm caused to developing countries by international cartels, which continue to suffer extensively from such arrangements. Others (eg Singham 2007) take the view that, while past efforts to bring together competition policy and international trade were, perhaps, simplistic and premature, the extent of possible synergies and interactions between the two fields is such that strict separation is impossible and renewed dialogue highly desirable. Indeed, while it is accurate to say that the past discussions on competition policy in the WTO yielded no consensus on a way forward, the past work also showed a broad degree of agreement on key underlying points. In particular, it clearly identified a number of channels through which anticompetitive practices of firms impede both trade and development, and identified a number of ways in which the objectives of competition policy and international trade liberalization are inter-linked. Building on these insights, this chapter has not sought to resolve past disagreements in this area, or to support a particular path forward. Rather, it has sought to take stock of and set out for further discussion and reflection some of the links that have been identified between competition policy and the content and objectives of the WTO Agreements, including in past WTO work in this area and related scholarly research. As a way of organizing this material, we have put forward ‘Three Propositions, an Observation and Some Questions for Reflection’. The first proposition is that ‘Effective national competition policies are vital to the

154  Competition, Regulation and Public Policies success of the multilateral trading system, including (very much) its contribution to development’. The second proposition is that ‘Competition policy already figures in the WTO Agreements, in important ways’. The third proposition is that ‘There exist additional issues at the interface of trade and competition policy which (arguably) call for further joint application of the two disciplines’. A related observation has also been made, namely that explicit provisions on competition policy have been extensively incorporated in regional and bilateral agreements adopted since work on competition policy was put on hold in the WTO, including by WTO Members that had expressed significant reservations concerning such work. In support of the first proposition, we have put forward three examples of such interdependence between international trade and competition policy, namely: (i) the importance of pro-competitive structural reforms in enabling developing and transition economy businesses to take advantage of market access opportunities created by trade agreements; (ii) the importance of competition policy in deterring international cartels, which have the ability to undermine the gains from trade; and (iii) the importance of competition rules as an adjunct to liberalized government procurement regimes, as embodied in the WTO Agreement on Government Procurement and the chapters on government procurement that are incorporated in many recent RTAs. In regard to the second proposition, note has been taken of relevant provisions of the GATS Agreement, the TRIPS Agreement, and other WTO Agreements all of which contain references to competition policy concepts. Concerning the third proposition, we have pointed out some gaps and/ or difficulties in applying traditional competition law disciplines to (i) export cartels and (ii) state-owned enterprises. As competition restraints established by export cartels or competitive advantages provided to SOEs have the potential to negatively impact on international trade, further research and debate on the issue, including at the international level, may be warranted. In regard to the ‘related observation’ concerning RTAs, it has been noted that important competition policy provisions are incorporated in a large number of RTAs that have been entered into over the past decade. These include both provisions dealing specifically with the subject of competition law enforcement and other provisions, dealing with matters such as state monopolies, investment, services (including telecommunications, maritime transport and financial services), government procurement and intellectual property, which also implicate competition policy concepts. The foregoing raises a number of questions for reflection, only a few of which are noted here. First, given the clear importance of effective national competition policies for the success of the international trading system (and the reciprocal contribution that trade liberalization can make to ensuring competition in markets), it can be asked whether there could be benefits from more explicit recognition of this interdependence in the WTO Agreements. To be sure, one could argue (and some do) that each policy already functions effectively enough in its own sphere. Nevertheless, the extent of complementarities and possible synergies is such that the question at least seems worth pondering (see also, in this regard, Jenny 2004).

Robert D Anderson and Anna Caroline Müller 

155  

A particular question that was central to the past debates on competition policy in the WTO, and that may still be worth pondering, is whether it would be worthwhile, in the framework of the WTO, to prohibit, in principle, the operation of cartels in international trade. To be sure, the harm caused by cartels is generally recognized, and the need for effective action against such arrangements is reflected in important international instruments, for example the OECD Recommendation against Hardcore Cartels. However, notwithstanding this, the problem has hardly gone away: cartels remain a recurring and serious problem in international markets (again, see Connor 2011 and references cited therein). Moreover, as has been noted, industrial and trade policies – ie, government measures – that limit competition can themselves be a factor that encourages or facilitates cartel activity. To this extent, world welfare might be well served by a rule, embodied in the international trading system, that commits all participating countries to enforce appropriate rules against cartels and to avoid policies that encourage such arrangements. In relation to the second proposition noted above, a further question that arises is whether, given the extent of references to competition policy issues and concepts already incorporated in the WTO agreements, renewed dialogue is important to ensure that these concepts are appropriately applied. In particular, this chapter has shown that, while currently there is no comprehensive Agreement on competition policy similar to the chapters contained in many regional trade agreements, a number of ad hoc references to competition law and policy concepts are contained in the GATS Agreement, the TRIPS Agreement and various other WTO Agreements. These include provisions dealing with matters such as monopolies, exclusive supply arrangements, anticompetitive licensing practices, and other matters. As the competition community is well aware, these are specialized concepts that implicate both competition law jurisprudence and economic analysis. Arguably, this creates a need for dialogue and learning to ensure appropriate use of the provisions. Concerning the extent of competition policy provisions that have been incorporated in RTAs, as has already been noted, this, too, begs important questions. Specifically, the proliferation of these provisions calls into question both the idea that competition policy is unrelated to the realization of trade objectives and the perception that the subject is so complex that it cannot be meaningfully reflected in trade agreements. Thus, a question for thought is whether the competition policy provisions in recent RTAs are also relevant in the multilateral context, or whether the synergies they seek to capture are to be harnessed only in regional arrangements. A related question is whether an appropriate reference to competition policy as a cornerstone principle of the international trading system in the WTO Agreements could bolster the role of national competition agencies. Arguably, ensuring the existence of strong, independent competition authorities is especially crucial in times of economic crisis, and therefore today. Vigorous competitive rivalry in markets is a key determinant of innovation, productivity and consumer welfare.

156  Competition, Regulation and Public Policies Therefore, it is vital that policies adopted to combat the economic crisis do not restrict competition in markets unduly. To do so is to undermine directly the foundations of future recovery, growth and development. This is, of course, a key reason for resisting the introduction of protectionist trade measures in the current economic environment. It also highlights the role of competition law and policy as a cornerstone of a successful market economy and an essential complement to trade liberalization. However, current measures to combat the financial and economic crisis run the risk of limiting competition in various ways. For example, some measures that are currently being implemented by governments to restore confidence and ensure stability in financial markets have the potential to distort incentives and reduce competition both within the country implementing them and internationally. Furthermore, governments are also under strong pressure to intervene to alleviate the effects of the economic downturn. Governments are expected to take whatever measures are necessary to revive demand, to stimulate supply, to prevent the possible bankruptcy of firms hit by the hard economic conditions and to stem the growth of unemployment. Typical stimulus measures will include: – Direct aid to ailing business firms or small and medium size firms which are collateral victims of the credit crisis; – Subsidized interest rates; – Sectoral aid designed to boost demand in specific sectors (Jenny 2009). All these measures alter competitive relationships and can impact on trade flows, with related negative long-term economic effects. In this context, arguably, appropriate international commitments could help to ensure that the role of national competition policies is not circumscribed in times of economic crisis, as has sometimes been the case in the past (see, for an interesting discussion of historical examples, Varney 2009). A final question for the future concerns whether, at some stage, there will be a need for stronger coordination mechanisms to ensure transparent and nondiscriminatory application of laws in this area. As already noted, more than 100 countries, including such powerhouse ‘emerging markets’ as China, India, Brazil and South Africa, now have national competition laws. Perhaps 80 of these laws were adopted in the last 20–30 years. Furthermore, a growing number of such countries actually or potentially apply their laws to conduct originating abroad. These trends are unambiguously to be celebrated by all persons interested in the effective control of anticompetitive practices, for the welfare of the world’s citizens. However, it has to be said that these trends have also brought with them increased potential for inter-jurisdictional conflicts in this area. Indeed, the potential for such conflicts has already been seen. Early examples, principally involving developed jurisdictions but also potentially having spillover effects on developing countries, include the Boeing/McDonnell Douglas and GE Honeywell merger cases, in addition to the multiple cases in the U.S., the EU and other jurisdictions relating to alleged abusive conduct of the Microsoft

Robert D Anderson and Anna Caroline Müller 

157  

Corporation (see, for a review of pertinent cases, Anderson 2011). Already, major efforts are being made to avoid conflicts through efforts, involving the ICN and other fora, to promote voluntary convergence in substantive standards for the application of relevant laws. Such efforts enjoy substantial support in the international competition community, and it may be that they are all that is needed and appropriate to address the potential for conflict. On the other hand, it could also be that there will eventually be calls for the adoption of stronger coordination mechanisms to ensure the transparent and non-discriminatory application of relevant disciplines (see, for related discussions, the websites of the U.S. Council for International Business and the International Chamber of Commerce). Certainly, any such mechanisms would require careful study, as the conceptual, jurisprudential and technical issues involved are by no means simple.

Bibliography 1. Anderson, Robert D., ‘The Interface Between Competition Policy and Intellectual Property in the Context of the International Trading System’, 1 Journal of International Economic Law 655 (1998). 2. Anderson, Robert D., ‘Systemic Implications of Deeper TransAtlantic Convergence in Competition/Antitrust Policy’, in Simon Evenett and Robert Stern, eds., Systemic Implications of Transatlantic Regulatory Cooperation and Competition, World Scientific Publishing, 2011, chapter 7. 3. Anderson, Robert D. and Simon Evenett, ‘Incorporating Competition Elements into Regional Trade Agreements: Characterization and Empirical Analysis’, 2006, http://www.evenett.com/research/workingpapers/CompPrincInRTAs. pdf. 4. Anderson, Robert D. and Alberto Heimler (2007a), ‘Abuse of Dominant Position: Enforcement Issues and Approaches for Developing Countries’, in Vinod Dhall, ed., Competition Law Today: Concepts, Issues and the Law in Practice, Oxford University Press, 2007, chapter 2. 5. Anderson, Robert D. and Alberto Heimler (2007b), ‘What has Competition Done for Europe? An Inter-Disciplinary Answer’, Aussenwirtschaft (the Swiss Review of International Economic Relations), 62. Jahrgang (2007), Heft IV, pp 419 et seq. 6. Anderson, Robert D., Abraham Hollander, Joseph Monteiro and W. T. Stanbury, ‘Competition Policy and Regulatory Reform in Canada, 1986– 1997’, 13 Review of Industrial Organization 177 (1998). 7. Anderson, Robert D. and Peter Holmes, ‘Competition Policy and the Future of the Multilateral Trading System’, 5 Journal of International Economic Law 531 (2002). 8. Anderson, Robert D. and Frédéric Jenny, ‘Competition Policy, Economic Development and the Role of a Possible Multilateral Framework on

158  Competition, Regulation and Public Policies Competition Policy: Insights from the WTO Working Group on Trade and Competition Policy’, in Erlinda Medalla, ed., Competition Policy in East Asia, Routledge, 2005, chapter 4. 9. Anderson, Robert D., William E. Kovacic and Anna Caroline Müller, ‘Ensuring integrity and competition in public procurement markets: a dual challenge for good governance’, in Sue Arrowsmith and Robert D. Anderson, eds., The WTO Regime on Government Procurement: Challenge and Reform, Cambridge University Press, 2011, chapter 22. 10. Beato, Paulina and Jean-Jacques Laffont, ‘Competition in public utilities in developing countries’, Inter-American Developing Bank Washington, D.C., Sustainable Developing Department, Technical Papers Series, 2002, http:// cdi.mecon.gov.ar/biblio/docelec/MU2007.pdf. 11. Bhattacharjea, Aditya, ‘Export Cartels: A Developing Country Perspective,’ Working Paper No. 120, Centre for Development Economics, Delhi School of Economics, 2004, http://www.cdedse.org/pdf/work120.pdf. 12. Connor, John, ‘Price Effects of International Cartels in Markets for Primary Products’, paper presented at the Symposium on Trade in Primary Product Markets and Competition Policy, World Trade Organization, Geneva, 22 September 2011. 13. European Commission, ‘Antitrust: Commission fines providers of international removal services in Belgium over €32.7 million for complex cartel’, Press Release IP/08/415 of 11 March 2008. 14. Evenett, Simon J., Margaret Levenstein and Valerie Suslow, ‘International Cartel Enforcement: Lessons from the 1990s’, 24 The World Economy 1221 (2001). 15. François, Joseph and Miriam Manchin, ‘Institutions, infrastructure, and trade’, Policy research working paper No. 4152, World Bank, 2007. 16. Fox, Eleanor, ‘The WTO’s First Antitrust Case – Mexican Telecom: A Sleeping Victory for Trade and Competition’, 9 Journal of International Economic Law 271 (2006). 17. Fröberg, Martin, ‘Developing Countries and International Trade: The Occurrence and Prevention of Export Cartels’, undated, University of Maastricht, http://arno.unimaas.nl/show.cgi?fid=15750. 18. Hoekman, Bernard and Kamal Saggi, ‘Tariff bindings and bilateral cooperation on export cartels’, 83 Journal of Development Economics 141 (2007). 19. Hufbauer, Gary and Sherry Stephenson, ‘Services Trade: Past Liberalization and Future Challenges’, 10 Journal of International Economic Law 605 (2007). 20. International Competition Policy Advisory Committee, Final Report to the Attorney General and Assistant Attorney General for Antitrust, http://www. justice.gov/atr/icpac/finalreport.html. 21. International Trade Centre (R. Anderson, F. Jenny and A. Müller, principal authors), Combatting Anti-Competitive Practices: A Guide For Developing

Robert D Anderson and Anna Caroline Müller 

159  

Economy Exporters, International Trade Centre, 2012, http://www.intracen. org/layouts/three-column.aspx?Pageid=45836&id=62357. 22. Jenny, Frederic, ‘International Cooperation on Competition: Myth, Reality and Perspective’, 48 The Antitrust Bulletin 973 (2003). 23. Jenny, Frédéric, ‘Competition, Trade and Development Before and After Cancún’, in Barry Hawk, ed., International Antitrust Law & Policy: Fordham Corporate Law 2003, Juris Publishing, 2004, chapter 26. 24. Jenny, Frédéric, ‘The Economic and Financial Crisis, Regulation and Competition’, 32 Law and Economics Review 449 (2009). 25. Jenny, Frederic, ‘Export Cartels in Primary Products: The Potash Case in Perspective’, in Simon J. Evenett and Frédéric Jenny, eds., Trade, Competition, and The Pricing of Commodities, 2012, http://ssrn.com/abstract=2064686. 26. Jenny, Frédéric and Pradeep Mehta, ‘Global problems & solutions’, The Financial Express, 2012, posted online 13 January 2012, http://www. financialexpress.com/news/global-problems-&-solutions/899105/0. 27. Kessides, Ioannis, Reforming Infrastructure; Privatization, Regulation, and Competition, World Bank, 2004. 28. Levenstein, Margaret and Valerie Suslow, ‘Private International Cartels and Their Effect on Developing Countries’, Background Paper for the World Bank’s World Development Report 2001, 9 January 2001, http://www-unix. oit.umass.edu/~maggiel /WDR2001.pdf. 29. Levenstein, Margaret and Valerie Suslow, ‘The Changing International Status of Export Cartel Exemptions’, 20 American University International Law Review 785 (2005). 30. Mehta, Pradeep, Competition Regimes in the World: A Civil Society Report, CUTS International, 2006. 31. OECD Global Forum on Competition, Summary of Cartel Cases Described by Invitees, CCNM/GF/COMP(2001)4 (5 October 2001). 32. OECD (2009) (Directorate for Financial and Enterprise Affairs – Competition Committee), State-Owned Enterprises and the Principle of Competitive Neutrality, DAF/COMP(2009)37 (20 September 2010). 33. OECD (2010), Collusion and Corruption in Public Procurement, http:// www.oecd.org/competition/cartels/46235884.pdf. 34. OECD (2013a), Fighting Bid Rigging in Public Procurement in Mexico, http://www.oecd.org/daf/competition/MexicoISSSTEBidRiggingENG.pdf. 35. OECD (2013b), Competition and Poverty Reduction, http://www.oecd.org/ daf/competition/competition-and-poverty-reduction2013.pdf. 36. Osakwe, Chiedu, ‘Poverty Reduction and Development: The Interaction of Trade, Macroeconomic and Regulatory Policies’, 10th Joseph Mubiru Memorial Lecture Organized by the Bank of Uganda, 2001. 37. Singham, Shanker, ‘A General Theory of Trade and Competition: Trade Liberalization and Competitive Markets’, Cameron May, 2007. 38. Sweeney, Brendan, ‘Export Cartels: Is There a Need for Global Rules?’, 10 Journal of International Economic Law 87 (2007).

160  Competition, Regulation and Public Policies 39. Taylor, C. Robert and Diana L. Moss (2013). AAI Working Paper No. 13– 05: The Fertilizer Oligopoly: The Case for Global Antitrust Enforcement. Available at: http://www.antitrustinstitute.org/~antitrust/content/aai-workingpaper-no-13-05-fertilizer-oligopoly-case-global-antitrust-enforcement. 40. Teh, Robert, ‘Competition provisions in regional trade agreements’, in Antoni Estevadeordal, Kati Suominen and Robert Teh, eds., Regional Rules in the Global Trading System, Cambridge University Press, 2009, chapter 8. 41. Teravaninthorn, Supee and Gaël Raballand, Transport prices and costs in Africa: A review of the main international corridors, World Bank, 2009. 42. U.S. Department of Justice, ‘German Company Pleads Guilty To Rigging Bids on USAID Construction Contracts in Egypt’, Press Release of 18 August 2000. 43. U.S. Department of Justice, ‘Eight executives arrested on charges of conspiring to rig bids, fix prices, and allocate markets for sales of marine hose’, Press Release of 2 May 2007. 44. U.S. Department of Justice, ‘Three United Kingdom Nationals Charges With Participating in Worldwide Bid-Rigging Conspiracy in the Marine Hose Industry’, Press Release of 3 December 2007. 45. Varney, Christine, ‘Vigorous Antitrust Enforcement in This Challenging Era’, Remarks as Prepared for the United States Chamber of Commerce (12 May 2009). http://www.justice.gov/atr/public/speeches/245777.htm. 46. World Trade Organization, ‘Special Study on Trade and Competition Policy’, in Annual Report for 1997 (Geneva, 1997), chapter IV. 47. World Trade Organization, WTO Dispute Settlement: One-Page Case Summaries (Geneva, 2006), http://www.wto.org/english/res_e/booksp_e/ dispu_summary06_e.pdf. 48. WTO Working Group on the Interaction between Trade and Competition Policy, Annual Report of the Working Group on the Interaction between Trade and Competition Policy to the General Council (Geneva, 1998), WT/ WGTCP/2.

Alexandre de Streel*

The Antitrust Activism of the European Commission in the Telecommunications Sector

This chapter analyses the role of the European Commission in taking cases of abuse of dominance in the telecommunications sector since its liberalisation in 1998. The argument is that the Commission has played a very active antitrust role, in particular at the beginning of the liberalisation, and that its activism was blessed by the EU Courts. Indeed, the Courts see competition law as a complement, and not a substitute, to regulation. They have set a relatively low threshold for proving margin squeeze, the most common form of abuse in telecommunications; and they have allowed wide discretion to the Commission to deal with the inaction or failures of national regulatory authorities (NRAs). It is further submitted that such activism was justified in the early days of liberalisation when the telecommunications markets were still very concentrated, and when the NRAs were in their infancy and sometimes captured. However, this activism is less justified today because, on the one hand, competition in the sector has increased, and on the other hand most anticompetitive practices can be, and should be, addressed by the NRAs. Today, the Commission should instead concentrate on consolidating the expertise and the independence of the NRAs, and it should rely on its extensive oversight of NRA decisions to prevent violations of the competition law. The chapter is divided into five sections. The first section reviews the main EU cases of abuse of dominance in the various telecoms segments. The next three sections deal with the main substantive and institutional issues raised by those cases. The second section describes the approaches of the Commission and the EU Courts on the relationship between competition law and sector-specific regulation. The third section analyses the tests established by EU Courts to control margin squeeze and refusal to deal in post-liberalised markets. It also compares them with the tests proposed by the Commission in its Guidance Paper on exclusionary abuses. The fourth section analyses the different uses made by the Commission of its antitrust powers in the telecoms sector, and their institutional implications. The fifth section concludes the chapter.

* Universities of Namur and Louvain, Belgium. The author would like to thank Ekaterina Rousseva, Mel Marquis, Heike Schweitzer and the participants of the 17th Annual Competition Law and Policy Workshop for helpful comments and suggestions. He remains solely responsible for any errors.

190  Competition, Regulation and Public Policies

1. Overview of the main EU cases of abuses of dominant position in telecommunications This section reviews the main abuses of dominances cases taken at the EU level1 according the four market segments: fixed voice, leased lines, mobile services and fixed internet broadband.2

1.1 Fixed voice telephony In 1997, the Commission carried out a quasi-sector enquiry3 in the fixed voice telephony segment and collected data from all the dominant operators in Europe with regard to tariffs for international phone calls in order to determine if the underlying wholesale charges paid between operators (ie, the accounting rates) were cost-oriented.4 On that basis, in 1998 the Commission opened seven cases for excessive accounting rates. These cases were passed on to the NRAs, which secured substantial price reductions. These reductions amounted to an average of 27% in Finland, Austria and Portugal, for example.5

1.2 Leased lines In 1999, the Commission opened a sector enquiry6 in the leased lines segment, an important building block for the information highway since they were used by fixed new entrants and mobile telecom operators, Internet Service Providers and large business users.7 The Commission presented its preliminary assessment 1 For a review of the main EU cases, see Laurent Garzaniti and Matthew O’Regan, Telecommunications, Broadcasting and the Internet: EU Competition Law and Regulation (3rd edition, Sweet & Maxwell, 2010) Chapter V; ICN, Working Group on Telecommunications Services, Report to the Fifth Annual Conference in Cape Town (2006). Apart from those EU cases, a growing collection of national competition cases have been summarised in the European Competition Network Brief since 2010. 2 Section 4 of this chapter shows that EU antitrust enforcement in telecommunications has followed a cycle of three phases that may be explained in part by the different uses the Commission has made of its antitrust powers. 3 In ‘quasi-sector enquiries’, no formal sector enquiry is opened but questionnaires are sent to all operators of the market segment under review. See Dessislava Choumelova and Juan Delgado, ‘Monitoring competition in the telecommunications sector: European Commission Sector Enquiries’, in Pierre Buiges and Patrick Rey (eds), The Economics of Antitrust and Regulation in Telecommunications (Edward Elgar, 2004) pp 269 et seq. 4 Press Release of the Commission IP/97/1180 of 19 December 1997. 5 Press Release of the Commission IP/98/763 of 13 August 1998 (listing cases involving: OTE of Greece, Post & Telekom of Austria, Postes et Télécommunications of Luxembourg, Sonera of Finland, Telecom Eireann of Ireland, Telecom Italia, and Telecom Portugal); Press Release IP/99/279 of 29 April 1999. 6 Sector enquiries are now based on Article 17 of Regulation 1/2003. 7 Press Release IP/99/786 of 22 October 1999; Commission Working Document of 8 September 2000 on the initial results of the leased lines sector inquiry; Press Release IP/00/1043 of 22 September 2000; and Press Release IP/02/1852 of 11 December 2002.

Alexandre de Streel 

191  

in September 2000 and closed the enquiry in December 2002. In this context the Commission opened cases for excessive prices in short distance leased lines (in terms of prices and services provisioning) but again it passed these cases on to the NRAs. The Commission also opened five cases for excessive prices for international leased lines.8 It could not rely on the NRAs for these cases, as international leased lines were not covered by sector-specific regulation. The cases were closed, however, once the undertakings concerned had significantly reduced their prices.9

1.3 Mobile In 1998, the Commission carried out a quasi-sector enquiry in the mobile segment and collected systematic information on the high prices for fixed-to-mobile and mobile-to-fixed calls.10 This led to several cases focusing on allegedly excessive fixed termination charges,11 fixed retention charges,12 and mobile termination charges.13 The Commission was able to pass some of the cases to those NRAs that had jurisdiction to intervene under their national telecommunication laws; it dealt with the other cases itself, closing them when the operators agreed to reduce their charges from 30 to 80%.14 In January 2000, the Commission launched another sector enquiry concerning high international roaming prices.15 In November 2000, the Commission presented its preliminary findings and identified several possible instances of excessive prices. On that basis, in 2004–2005 the Commission opened several cases for excessive wholesale international roaming prices.16 However, it had to Against the incumbents of Belgium, Greece, Italy, Portugal and Spain. The average price decrease was 30% to 40% for the 2Mbits lines, the most commonly used bandwidth. See Press Release IP/02/1852 of 11 December 2002. At the time of the cases, there was no formal legal basis for the Commission to adopt commitment decisions. 10 Press Release IP/98/1036 of 26 November 1998. In the case of mobile-to-fixed calls, the fixed termination charge is the fee paid by the mobile operator to the fixed operator for terminating the call. In case of fixed-to-mobile calls, the mobile termination charge is the fee paid by the fixed operator to the mobile operator for terminating the call, and the fixed retention charge is the fee kept by the fixed operator for originating the call. 11 There were four cases involving Deutsche Telekom, KPN of the Netherlands, Telefónica of Spain, and Telecom Italia. See press Release IP/98/707 of 27 July 1998. 12 Eight cases were launched and these involved Belgacom, Telecom Eireann of Ireland, Deutsche Telekom, KPN, Telefónica, Telecom Italia, British Telecom, and P&T of Austria. See Press Release IP/98/707 of 27 July 1998. 13 Cases were opened in two countries and concerned all the mobile operators in Germany and in Italy. See Press Release IP/98/141 of 9 February 1998; Press Release IP/98/707 of 27 July 1998. 14 Press Release IP/98/1036 of 26 November 1998; Press Release IP/99/298 of 4 May 1999. 15 Press Release IP/00/111 of 4 February 2000; Commission Working Document of 13 December 2000 on the initial findings of the sector inquiry into mobile roaming charges; Commission MEMO/01/262 of 11 July 2001. International roaming tariffs are the charges that a mobile customer has to pay when giving and receiving calls abroad using a network other than the one to which he is affiliated. 16 For the case against Vodafone and O2 in the UK, see Press Release IP/04/994. For the case against Vodafone and T-Mobile in Germany, see Press Release IP/05/161. 8 9

192  Competition, Regulation and Public Policies close the cases without securing any price reductions because, as became clear, competition law tools were inadequate to deal with the market failures arising from international roaming. To address the problem, in 2007 the EU legislature adopted a widely publicised Regulation setting maximum prices for wholesale and retail international roaming prices.17

1.4 Fixed internet broadband Since the turn of the century, most of the abuse cases have related to the development of fixed internet broadband because broadband deployment is the key driver for ITC development.18 In 2000, the Commission adopted guidelines on the application of antitrust rules to restricted access to the local loop (ie, the last mile of the network, where the economies of scale and scope are the largest).19 It also opened a sector enquiry dealing with similar issues.20 On that basis, the Commission opened several individual cases and adopted four formal decisions.21 In addition to these antitrust actions, the EU legislature in 2000 adopted a specific regulation forcing the unbundling of the local loop, which proved to be effective mechanism in several Member States for increasing competition in the provision of broadband.22

Wanadoo (2003) In Wanadoo,23 the Commission imposed a fine of 10.35 million euros against Wanadoo, the Internet Service Provider belonging to Orange/France Télécom group, on the ground that its retail prices for ADSL were predatory. The Commission found first that the firm’s prices were below average variable cost between 1999 and August 2001; and second that its prices were below average total cost and part of an exclusionary plan between September 2001 and October 2002. During that time, Wanadoo’s market share rose from 46% to 72% on a 17 The recast version is Regulation 531/2012 of the European Parliament and of the Council of 13 June 2012 on roaming on public mobile communications networks, 2012 OJ L172/10. 18 More recently, see Communication of 26 August 2010, A Digital Agenda for Europe, COM(2010) 245, which also promoted high broadband connections, in particular to stimulate growth in Europe. 19 Communication of 26 April 2000 on unbundled access to the local loop, 2000 OJ C272/55. 20 Press Release IP/00/765 of 12 July 2000. 21 The vast majority of these cases related to margin squeeze for fixed broadband Internet. See Cento Vejanovski, ‘Margin squeeze: An overview of EU and national case law’, e-Competitions n° 46, 2012; Damien Geradin and Robert O’Donoghue, ‘The Concurrent Application of Competition Law and Regulation: the Case of Margin Squeeze Abuses in the Telecommunications Sector’, 1 Journal of Competition Law and Economics 355 (2005). 22 Regulation 2887/2000 of the European Parliament and of the Council of 18 December 2000 on unbundled access to the local loop, 2000 OJ L336/4. This Regulation has been repealed and carried over in Directive 2002/19/EC of the European Parliament and of the Council of 7 March 2002 on access to, and interconnection of, electronic communications networks and services, 2002 OJ L108/7, as amended by Directive 2009/140. 23 Commission Decision of 16 July 2003 in Case COMP/38.233 – Wanadoo Interactive.

Alexandre de Streel 

193  

market which experienced more than a five-fold increase in size. The abuse came to an end with a 30% decrease of wholesale charges, after which the French highspeed internet access market grew more rapidly, and in a more balanced way. The Commission’s infringement decision was upheld by the General Court in 2007,24 and then by the Court of Justice in 2009.25

Deutsche Telekom (2003) In that case,26 the Commission imposed a fine of 12.6 million euros against Deutsche Telekom, the German incumbent, for maintaining a margin squeeze between its wholesale charge for full unbundling of the local loop and its retail prices for access lines. The Commission compared the wholesale access prices that Deutsche Telekom charged its competitors for use its local loops with the retail prices the company charged end-users for its different retail offerings (analogue, ISDN and ADSL connections).27 This comparison revealed that the margin between the wholesale and retail prices was negative for the period 1998 through 2001. Although the margin then became positive it was still insufficient to cover DT’s own retail product-specific costs for the supply of end-user services. The Commission’s condemnation of DT was particularly contentious because the company’s wholesale and retail charges had been subject to the control of the German regulator. The decision to sanction DT was thus an indirect critique of the NRA’s work. However, as the retail tariffs were regulated with a price cap applying to a basket of services, DT still had some price discretion and it could have alleviated or at least diminished the squeeze. In 2008, the General Court upheld the Commission’s decision,28 and in 2010 the Court of Justice agreed.29 In 2004, the Commission found that DT had maintained another margin squeeze between the wholesale charges for shared access of the local loop and retail access lines. This time, DT settled the case by committing to petition the German NRA for a reduction of the line sharing tariffs on a lasting basis.30 In order to comply with these commitments, and following approval by the NRA for one year, DT decreased its monthly line sharing fee (from €4.77 to €2.43). Subsequently, several competitors rolled out their networks in order to provide broadband services. However, in May 2005 DT re-applied to the NRA for a shared access charge at €4.77. The Commission against found that the fee would lead to a margin squeeze. It therefore requested that DT file an application consistent with its commitments, and the company complied.31 Case T-340/03 France Télécom SA v Commission [2007] ECR II-107. Case C-202/07 P France Télécom SA v Commission [2009] ECR I-2369. 26 Commission Decision of 21 May 2003 in Case 37.451 – Deutsche Telekom, 2003 OJ L263/9. 27 In order to achieve a coherent comparison, the Commission used a weighted approach taking into account the numbers of DT’s customers for the different retail offerings. 28 Case T-271/03 Deutsche Telekom AG v Commission [2008] ECR II-477. 29 Case C-280/08 P Deutsche Telekom AG v Commission [2010] ECR I-9555. 30 Press Release of the Commission IP/04/281 of 1 March 2004. 31 Press Release of the Commission IP/05/1033 of 3 August 2005. 24 25

194  Competition, Regulation and Public Policies

Telefónica (2007) In that case,32 the Commission imposed a fine of 152 million euros against Telefónica, the Spanish incumbent, for a price squeeze between Telefónica’s national and regional wholesale charges for access to its broadband network and Telefónica’s retail prices for broadband access. The margins were insufficient to cover Telefónica retail costs between September 2001 and December 2006. Because of a lack of competition, the internet broadband prices were higher in Spain than in the other EU Member States. Wholesale regional tariffs were regulated but were based on Telefónica’s forecasts, which proved to be wrong, and wholesale national tariffs were unregulated. The abuse ended with a decrease in Telefónica’s wholesale charges. The Commission’s decision was upheld by the General Court in 2012,33 and then by the Court of Justice in 2014.34

TeliaSonera (2011) In this case,35 the Swedish competition authority condemned TeliaSonera, the Swedish incumbent, for a margin squeeze between wholesale and retail charges for ADSL products maintained between April 2000 and January 2003. Some of the wholesale charges were unregulated. When the case reached the Swedish court, the judge referred ten specific questions regarding margin squeeze to the Court of Justice. In its reply, which is analysed below in section 2, the Court established a relatively low threshold for finding an unlawful price squeeze. On that basis, the Swedish Court condemned TeliaSonera and imposed a fine of 4 million euros.

Telekomunikacja Polska (2011) In this case,36 the Commission imposed a fine of 127 million euros against Telekomunikacja Polska, the Polish incumbent which is part of the Orange/France Télécom group, for refusing to supply wholesale broadband products to competitors between August 2005 and October 2009. In particular, the Commission found that Telekomunikacja Polska was: (i) proposing to entrants unreasonable conditions for access to its wholesale broadband products; (ii) delaying the negotiation process, as 70% of the time Telekomunikacja Polska had failed to meet a 90-day regulatory deadline for concluding negotiations; (iii) limiting access to its network by, inter alia, rejecting entrants’ orders on unreasonable grounds or proposing difficult technical conditions for connecting to the company’s network; (iv) limiting access 32 Commission Decision of 4 July 2007 in Case COMP/38.784 – Wanadoo España v Telefónica, 2008 OJ C83/6. 33 Case T-336/07 Telefónica, SA and Telefónica de España, SA v Commission (Telefónica I) [2012] ECR II-000; Case T-398/07 Spain v Commission (Telefónica II) [2012] ECR II-527. 34 Case C-295/12 P Telefónica and Telefónica de España v Commission [2014] ECR I-000. 35 Case C-52/09 Konkurrensverket v TeliaSonera AB [2011] ECR I-527. 36 Commission Decision of 22 June 2011 in Case COMP/39.325 – Telekomunikacja Polska.

Alexandre de Streel 

195  

to subscriber lines by, inter alia, rejecting entrants’ orders to activate subscriber lines on unreasonable grounds or limiting the availability of subscriber lines; and (v) refusing to provide reliable general information indispensable for entrants, or providing inaccurate information. The abuse ended with an agreement between Telekomunikacja Polska and the Polish regulator to improve entrants’ access to the network. As of this writing, the Commission’s infringement decision is under appeal before the General Court.37

Statement of Objections issued to Slovak Telekom In Slovak Telekom,38 the Commission sent a Statement of Objections to Slovak Telekom, the Slovak incumbent – a part of the Deutsche Telekom group – for refusing to supply unbundled access to its local loop and wholesale services to a competitor; and for imposing a price squeeze on alternative operators. This case is currently still pending at the Commission. Having reviewed the main EU abuse of dominance cases in the telecoms sector, I now turn to the substantive and institutional issues raised by those cases. Specifically, these are: the relationship between competition law and sectoral regulation; the relevant tests to establish the most common form of abuse in telecoms; and the role of the Commission in supporting liberalisation and in controlling the NRAs.

2. The (complementary) relationship between competition law and sectoral regulation 2.1. Competition law supplements sectoral regulation The Court of Justice has consistently held that competition law applies to regulated industries unless there is a specific exemption in the law. As there is no legal exemption for telecommunications, competition law applies to the sector in its entirety.39 Moreover, competition law applies in addition to regulation unless the regulation completely removes the autonomy of the undertaking concerned. In Deutsche Telekom, the Court of Justice held that: It is only if anticompetitive conduct is required of undertakings by national legislation, or if the latter creates a legal framework which itself eliminates any possibility of competitive activity on their part, that Articles [101 and 102 TFEU] do not apply. In such a situation, 37 38 39

Case T-486/11 Orange Polska v Commission, not yet decided. Press Release IP/10/1741 of 17 December 2010; Press Release IP/12/462 of 8 May 2012. Case 41/83 Italy v Commission [1984] ECR 873.

196  Competition, Regulation and Public Policies the restriction of competition is not attributable, as those provisions implicitly require, to the autonomous conduct of the undertakings. [101 and 102 TFEU] may apply, however, if it is found that the national legislation leaves open the possibility of competition which may be prevented, restricted or distorted by the autonomous conduct of undertakings. The possibility of excluding anticompetitive conduct from the scope of Articles [101 and 102 TFEU] on the ground that it has been required of the undertakings in question by existing national legislation or that the legislation has precluded all scope for any competitive conduct on their part has thus been accepted only to a limited extent by the Court of Justice.40

Clearly, the Court sees competition law as a complement – and not a substitute – to regulation. As the Court of Justice put it in Deutsche Telekom: The competition rules laid down by the [TFEU] supplement in that regard, by an ex post review, the legislative framework adopted by the Union legislature for ex ante regulation of the telecommunications markets.41

Another more obvious point on the relationship between regulation and competition policy is that, when a competition authority investigates whether an abuse of dominance has taken place, it should account of the regulation in place. According to the General Court: Since the legislation relating to the telecommunications sector defines the legal framework applicable to it and, in so doing, contributes to the determination of the competitive conditions under which an undertaking […] carries on its business in the relevant markets, it is […] a relevant factor in the application of Article [102 TFEU] to the conduct of that undertaking, whether for the purposes of defining the relevant markets, assessing the abusive nature of such conduct or setting the amount of the fines.42

The position of the EU Courts on the relationship between competition law and regulation is famously different from that of the U.S. Supreme Court. In Trinko,43 a 40 Deutsche Telekom, cited above note 29, paras 80–81. This is well-established case law both in the telecoms sector (TeliaSonera, cited above note 35, paras 49–50; Telefónica I, cited above note 33, paras 328–329; Telefónica II, cited above note 33, paras 70–71) and in other regulated markets (Case C-359/95 P Commission and France v Ladbroke Racing Ltd [1995] ECR I-6265, para 33). On the regulated conduct defence in the EU, see, eg, Richard Wainwright and Andre Bouquet, ‘State Intervention and Action in EC Competition Law’, in Barry Hawk (ed), International Law & Policy: Fordham Corporate Law Institute 2003 (Juris Publishing, 2004); OECD, Regulated Conduct Defence (2011), DAF/COMP(2011)3. Where this defence is invoked, the regulation must of course comply with EU law. If not, it should be repealed and, in any case, to the extent that it conflicts with EU law it cannot be applied by national courts or authorities. See, eg, Case 267/86 Pascal Van Eycke v ASPA [1988] ECR 4769, para 16; Case 66/86 Ahmed Saeed Flugreisen and Silver Line Reisebüro GmbH v Zentrale zur Bekämpfung unlauteren Wettbewerbs eV [1989] ECR 838, para 48. 41 Deutsche Telekom, cited above note 29, para 92. See also Telefónica I, cited above note 33, para 293; Telefónica II, cited above note 33, para 56. 42 Deutsche Telekom, cited above note 29, para 224; Telefónica II, cited above note 33, para 55. For instance, in Deutsche Telekom, the Commission took into account the objective of tariff rebalancing between access and services prices; in Telefónica, the Commission took into account the incentives associated with the investment ladder. 43 Verizon Communications Inc v Law Offices of Curtis V Trinko, 540 U.S. 682 (2004). According to the Antitrust Modernization Commission’s Report and Recommendations (2007), at 362, ‘Trinko is best understood only as a limit on refusal-to-deal claims under Section 2 of the Sherman Act. It should

Alexandre de Streel 

197  

well-known case brought under Section 2 of the Sherman Act, the Supreme Court refused to condemn a telecoms incumbent, Verizon, for refusing to give access to its network to its competitor AT&T, since the incumbent had already been condemned by the federal and the State telecom regulators. The Supreme Court held that the regulatory duty to deal established under the 1996 U.S. Telecommunications Act did not create a new cause of action under the refusal to deal doctrine under Section 2. It is clear from the judgment that the Supreme Court regarded regulation more as a substitute than as a complement to antitrust law.

2.2. The rationale for and the risks of the parallel application of competition law and sector regulation As the General Court has clarified,44 the difference between the EU and the U.S. positions is explained by different institutional settings: EU competition law has a ‘constitutional’ value (in particular since it is enshrined in the TFEU) whose application cannot be precluded by national sector-specific regulation. By contrast, notwithstanding the importance of free enterprise in American society and law, the Sherman Act in the end has the same legal value as other federal statutes. As competition law applies in addition to regulation, it could be argued that the cumulative application of two legal instruments to the same behaviour may violate some legal principles such as legitimate expectations or ne bis in idem. With regard to the principle of legitimate expectations, the EU Courts consider that a decision adopted by a National Regulatory Authority cannot create, for a dominant operator, expectations that its behaviour complies with competition law and would not be condemned by a competition authority.45 This is because the NRA usually takes decisions on the basis of sectoral regulation, which has different objectives than competition policy.46 That has also been clarified in several sets not be read to displace the role of the antitrust laws in regulated industries as an implied immunity, nor should it be taken as a judicial rejection of a savings clause.’ See also Credit Suisse Securities (USA) v Billing, 551 U.S. 264 (2007). 44 Telefónica II, cited above note 33, para 55. This was also the view of Alexander Italianer, ‘Levelplaying field and innovation in technology markets’, speech, Palo Alto, 28 January 2013, at 6, and of Pierre Larouche, ‘Contrasting Legal Solutions and Comparability of the E.U. and U.S. Experiences’, TILEC Discussion Paper 28 (2006), at 11. 45 Telefónica I, cited above note 33, para 299; Telefónica II, cited above note 33, para 122. 46 According to Article 8 of the Directive 2002/21/EC of the European Parliament and of the Council of 7 March 2002 on a common regulatory framework for electronic communications networks and services (Framework Directive), 2002 OJ L108/33, as amended by Directive 2009/140, sector regulation should promote competition, contribute to the development of the digital internal market, and promote the interests of the European citizens. In practice, sectoral regulation aims to increase competition on the market, possibly with entry support, while EU competition policy aims to preserve the competitive structure of markets. As noted by the European Regulators Group (ie, the club of the NRAs of the EU, now BEREC) in the context of margin squeeze, ‘While competition law is intended to prevent margin squeeze as an exclusionary abuse, ex ante regulation seeks the more ambitious goal of promoting competition by facilitating entry into those markets.’ ERG Report of March 2009 on the discussion of the application of Margin Squeeze tests to bundles, ERG (09) 07, para 6.

198  Competition, Regulation and Public Policies of guidelines issued by the Commission.47 But beyond the ‘different objectives’ rationale, even if an NRA adopted its decision on the basis of competition policy, the Commission would not be bound by such national decision.48 With regard to the principle of ne bis in idem, the Court of Justice decided that three conditions need to be met for the principle to be violated: identity of facts, unity of offender and unity of the legal interest protected.49 According to the Commission, an antitrust decision which complements a regulatory decision does not violate the principle of ne bis in idem because the two decisions do not protect the same legal interest: regulation and competition law pursue, it is said, different objectives.50 The only possible defence in a competition case for a dominant operator which is regulated (in cases where it still acts with some autonomy despite the presence of regulation) is to request a reduction of the fine that would otherwise apply on the ground that the constraints imposed on it by the relevant regulation is a mitigating circumstance.51 Thus, the conception of the EU Courts with regard to the complementary relationship between regulation and competition law extends quite far the ‘special responsibility’ of the dominant undertaking, which in a sense becomes a guardian of the Treaty. Indeed, a dominant telecoms operator which is regulated has to go to the regulator to request changes to its tariffs in accordance with that regulation in order to avoid the finding of abuse.52

3. The legal tests for abuses in post-liberalised markets The next substantive issue raised by the cases reviewed above concerns the legal tests established by the Commission and by the EU Courts to address the most common forms of abuse of dominance in the telecoms sector, namely margin squeeze and refusal to deal. My claim in this regard is that the price squeeze test adopted by the Courts is relatively interventionist, and that the low threshold for 47 Commission Notice on the application of the competition rules to access agreements in the telecommunications sector, 1998 OJ C265/2 (‘Access Notice’), para 22; Commission Guidelines of 9 July 2002 on market analysis and the assessment of significant market power under the Community regulatory framework for electronic communications networks and services, 2002 OJ C165/6, para 31. See also Framework Directive 2002/21 as amended, Article 15(1). 48 See the CFI’s judgment in Deutsche Telekom, cited above note 28, para 120 (citing Case C344/98 Masterfoods and HB [2000] ECR I-11369, para 48). See also Telefónica I, cited above note 33, para 301. 49 Joined Cases C-204/00 P, C-205/00 P, C-211/00 P, C-213/00 P, C-217/00 P and C-219/00 P, Aalborg Portland A/S and Others v Commission [2004] ECR I-123, para 338. 50 Commission Decision in Telekomunikacja Polska, cited above note 36, para 139. 51 Commission Guidelines on the method of setting fines imposed pursuant to Article 23(2a) of Regulation 1/2003, 2006 OJ C210/5, para 29; Deutsche Telekom, cited above note 29, paras 279 and 286; Telefónica I, cited above note 33, paras 453–463. 52 CFI in Deutsche Telekom, cited above note 28, para 122; Telefónica I, cited above note 33, para 335.

Alexandre de Streel 

199  

a finding of abuse may be partly due to the fact that this type of abuse emerged in post-liberalised industries. I also submit that the position of the Courts and the position of the Commission are not exactly the same. The Courts apply different legal tests for price squeeze and refusal to deal, but they do not make a clear distinction between standard non-regulated markets and post-liberalised regulated markets. On the contrary, the Commission suggests similar legal tests for margin squeeze and refusal to deal, but it differentiates between non-regulated and regulated markets.

3.1. Margin squeeze From a European perspective, a price squeeze or margin squeeze arises when the margin between a (vertically integrated) dominant firm’s wholesale price and its retail price is negative or, alternatively, where it is positive but insufficient to cover the firm’s retail costs, thereby leading to the possible exclusion of equally efficient competitors. To remedy this practice, the dominant firm is expected to decrease its wholesale prices and/or increase its retail prices.53 The Court of Justice has held that a margin squeeze is abusive in itself, ie that it is an independent form of abuse.54 Consequently, in order to condemn a margin squeeze, the competition authority need not prove that the conditions for another type of abuse are satisfied. For example, it is unnecessary to show that the wholesale price is excessive, or that the retail price is predatory.55 Furthermore, the competition authority need not to prove that the conditions for an illegal refusal to deal are satisfied. That is to say, in a margin squeeze case the wholesale product or service does not have to meet the strict Bronner conditions, such as in particular the criterion of indispensability.56 According to the Court of Justice, treating margin squeeze as a form of refusal to deal would unduly reduce the effectiveness of Article 102 TFEU; it was unwilling to accept the analogy with refusal to supply.57 On the other hand, the competition authority does have to prove that the margin squeeze has potential or concrete anticompetitive effects.58 Proving such effects is facilitated where the wholesale product is essential for activities downstream,59 but it cannot be excluded that a margin squeeze can have anticompetitive effects even if that product is not essential.60 In short, it is not necessary to show that the wholesale product is an Deutsche Telekom, cited above note 29, para 181. Ibid, para 183; TeliaSonera, cited above note 35, paras 31 and 56. Deutsche Telekom, cited above note 29, para 183; TeliaSonera, cited above note 35, para 34; Telefónica I, cited above note 33, para 187. 56 TeliaSonera, cited above note 35, para 55; Telefónica I, cited above note 33, para 180; Telefónica II, cited above note 333, para 74. 57 TeliaSonera, cited above note 35, para 58. 58 Ibid, para 64; Deutsche Telekom, cited above note 29, para 252. 59 Deutsche Telekom, cited above note 29, para 255; TeliaSonera, cited above note 35, para 71; Telefónica I, cited above note 33, para 182; Telefónica II, cited above note 33, para 94. 60 TeliaSonera, cited above note 35, para 72. 53 54 55

200  Competition, Regulation and Public Policies ‘essential facility’,61 but if it is demonstrated that the product is indeed essential, this goes a long way toward establishing that the compressed margin is producing or is likely to produce anticompetitive effects. One difficult issue when the margin between the wholesale and the retail price is positive is to determine which costs – those of the dominant undertaking or the competitors – should be used to evaluate whether the margin is nevertheless unlawfully narrow. According to the EU Courts, the competition authority should in principle use the retail costs of the dominant undertaking for reasons of efficiency and legal certainty.62 Using the retail costs of the dominant undertaking ensures that a margin squeeze will be prohibited only when a competitor as efficient as the dominant firm – ie, has the same or a lower retail costs than the dominant operator – cannot enter or would be expelled from the market. Moreover, the dominant operator normally only knows (or should only know) its own retail costs and not those of its competitors. It can therefore only assess the legality of its practices on the basis of its own costs. However, the EU Courts have accepted that, in specific circumstances, the competition authority may rely on the retail costs of the dominant firm’s competitors. According to the Court of Justice, that may be the case in three circumstances: – where the cost structure of the dominant undertaking is not precisely identifiable for objective reasons; – where the service supplied to competitors consists in the mere use of infrastructure whose production cost has already been written off, so that access to that infrastructure no longer represents a cost for the dominant undertaking that is economically comparable to the cost its competitors have to incur in order to have access to it; or – where so dictated by the particular competitive conditions – for example, by reason of the fact that the level of the incumbent’s costs is specifically attributable to the competitively advantageous situation it enjoys as a result of its own dominant position.63 The Court of Justice does not impose other limiting factors on the concept of margin squeeze which have sometimes been proposed in the literature, and which were raised by the Swedish court in the TeliaSonera case. In particular, the Court held that: (i) the competition authority does not have to prove that the defendant Telefónica II, cited above 33, para 76. Deutsche Telekom, cited above 29, paras 198 and 202; TeliaSonera, cited above note 35, paras 41–44. 63 TeliaSonera, cited above note 35, paraphrasing para 45. In the Access Notice (cited above note 47, para 118), the Commission had also envisaged that, in appropriate circumstances which were not defined, the costs of a reasonably efficient operator should be used for the margin squeeze test unless the dominant company could show that its downstream operation was exceptionally efficient. In its Guidance Paper on exclusionary abuses, the Commission envisages using the costs of competitors when reliable information on the costs of the dominant undertaking is unavailable or when it is not possible to clearly allocate the dominant undertaking’s costs to downstream and upstream operations. See 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, paras 25 and 80. 61 62

Alexandre de Streel 

201  

enjoys superdominance or a quasi-monopoly on the wholesale market, although it accepts that the degree of dominance may have an impact on the effects of the contested conduct;64 (ii) the competition authority does not have to prove that the defendant enjoys a double dominance on the wholesale market and the retail market, as the dominant position on the wholesale market suffices to be able to produce anticompetitive effects on the retail market;65 (iii) there is no need to prove that the defendant will later be able to recoup losses created by the margin squeeze, as the possibility that competitors may be driven from the market depends neither on defendant suffering losses nor on its ability to recoup losses but solely on the margin between the prices applied by the defendant on the markets concerned;66 (iv) the margin squeeze may be illegal when it applies to existing customers but also to new customers;67 and (v) the margin squeeze may be illegal when it applies to mature technologies and markets but also when it applies to new and emerging technologies and markets.68 Thus, the EU Courts have adopted a rather loose legal test to determine whether a dominant undertaking has unlawfully ‘squeezed’ the margin between its wholesale and retail prices. To condemn the undertaking for such an abuse, a competition authority (or a plaintiff) merely has to prove that: (i) the margin between wholesale prices and retail prices is either negative or, if positive, insufficient to cover the undertaking’s own retail costs (or in the specific circumstances described above, the costs of a competitor); and (ii) that narrow margin produces actual or potential anticompetitive effects. The authority (plaintiff) does not have to prove the stricter conditions of excessive prices, predatory prices, or refusal to deal. Moreover, ‘double dominance’, superdominance and the possibility of recoupment are all unnecessary for the purpose of establishing liability; they may only facilitate proving the two necessary conditions described above. Remarkably, this lax legal test, which was initially developed for regulated products and services, applies equally to non-regulated products and services, as the Court of Justice does not differentiate between these circumstances.69 The European approach is in sharp contrast with the strict legal test adopted by the US Supreme Court in Linkline.70 In that case, the Supreme Court decided that margin squeeze cannot in itself be prohibited independently of whether there is a duty to deal at the wholesale level or predatory pricing at the retail level. The European approach is also in contrast with mainstream economics, according TeliaSonera, cited above note 35, para 81. Ibid, para 89; Telefónica I, cited above note 33, para 146. 66 TeliaSonera, cited above note 35, para 101. 67 Ibid, para 92. 68 Ibid, para 108, stating that: ‘particularly in a rapidly growing market, Article 102 TFEU requires action as quickly as possible, to prevent the formation and consolidation in that market of a competitive structure distorted by the abusive strategy of an undertaking which has a dominant position on that market or on a closely linked neighbouring market, in other words it requires action before the anticompetitive effects of that strategy are realised.’ In France Télécom (cited above note 24, para 107), the General Court had already considered that Article 102 applied to high growth markets. 69 TeliaSonera, cited above note 35, para 52. 70 Pacific Bell Telephone Co v LinkLine, 555 U.S. 438 (2009). 64 65

202  Competition, Regulation and Public Policies to which margin squeeze should be not an independent abuse.71 Furthermore, the Court’s position is different from that suggested by the Commission. In its Guidance Paper, the Commission applies the same legal test for margin squeeze and for refusal to deal72 but it relaxes the test in cases of regulated products and services, or where the product or service has been developed under the protection of legal monopoly or with state resources.73 The EU Courts – and needless to say, it is the Court of Justice that decides on the authoritative interpretation of Article 102 – follow a very different route. They consider that the legal test for margin squeeze is different – and looser – than for refusal to deal. And, as already mentioned, they do not differentiate as to how the test should be applied in regulated and non-regulated industries.74

3.2. Refusal to deal With regard to refusal to deal, the Commission in its Guidance Paper makes a distinction in the way it applies the legal test according to whether the refusal concerns non-regulated markets or regulated/post-liberalisation markets. According to this distinction, the Commission applies a looser test in the latter context.75 The rationale for this differentiated treatment is that the effects of the antitrust intervention on investment incentives are already taken into account by the regulator when an antitrust decision complements a regulatory decision, or that the impact on incentives is less relevant when the infrastructure concerned has been developed under conditions sheltered from normal market forces. It is not clear whether the EU Courts will embrace the above distinction advocated by the Commission in the Guidance Paper. Advocate General Jacobs alluded to such a distinction in the Bronner case.76 However, in TeliaSonera (discussed above), the Court of Justice refused to accept the possibility of differentiated treatment in the context of a margin squeeze case.77

71 William Baumol et al, Brief of Amici Curiae Professors and Scholars in Law and Economics in Support of the Petitioners, Pacific Bell Telephone Co v linkLine Communications, 2007; Geert Brunekreeft et al, ‘On the law and economics of price squeeze in telecommunications markets’, TILEC Report (2005); J Gregory Sidak, “Abolishing the Price Squeeze as a Theory of Antitrust Liability’, 4 Journal of Competition Law and Economics 279 (2008). 72 See Guidance Paper, cited above note 63, at para 80. 73 See ibid, para 82 (an application of the ‘original sin’ theory). 74 Moreover, in TeliaSonera (cited above note 35, para 92), the Court of Justice declined to follow the distinction between an abuse against existing customers and an abuse against new customers, which the Commission had introduced in the Guidance Paper at para 84. 75 Guidance Paper, cited above note 63, para 82. 76 Opinion of Advocate General Jacobs in Case C-7/97 Oscar Bronner GmbH v Mediaprint Zeitungs-und Zeitschriftenverlag GmbH [1998] ECR I-7791, para 66. 77 Geradin argues that such differentiation is not justified legally and economically. See Damien Geradin, ‘Refusal to Supply and Margin Squeeze: A Discussion of Why the ‘Telefonica Exceptions’ are Wrong’, TILEC Discussion Paper 51 (2011).

Alexandre de Streel 

203  

3.3. The rationale and the risk of a loose legal test for abuse of dominant position The loose legal test adopted by the EU Courts for margin squeeze may be explained by the fact that this type of abuse mainly takes place in the telecommunications sector, ie in a previously monopolised network industries. In this sector, the Courts may be less concerned with the possibility of a type I error (over-enforcement) and the effects of such an error on investment (dis) incentives.78 A complementary, and more speculative, explanation for the loose test is that the Courts have taken into account the effects of their decisions on sector regulation. Since 2003, the economic telecommunications regulation has been based on competition law methodologies,79 and NRAs are currently dealing with numerous price squeeze cases under sectoral regulation. The EU Courts may have feared that if they had adopted a stricter test for price squeeze under competition law, the stricter text would have ‘spilled over’ into the application of regulatory obligations, thereby substantially and inappropriately limiting the possibilities of intervention by regulators. Whatever the explanation may be, I submit that the loose legal test of the EU Courts for margin squeeze is inappropriate because it unduly increases the risk of type I errors.80 To make things worse, the costs of type I errors may be substantial in fast-moving industries such as telecommunications.81 In addition, antitrust remedies in the case of a price squeeze are difficult to design. As the U.S. Supreme Court observed in Linkline:82 Recognizing price-squeeze claims would require courts simultaneously to police both the wholesale and retail prices to ensure that rival firms are not being squeezed. Courts would be aiming at a moving target, since it is the interaction between these two prices that may result in a squeeze. Moreover, firms seeking to avoid price-squeeze liability will have no safe harbor for their pricing practices.

Worse still, the loose test is even more problematic when it is applied in sectors of the economy which, unlike telecoms, did not develop under the protection of monopoly rights or which are not extensively regulated. In those circumstances, the costs of type I errors may be even more significant. 78 This follows the same reasoning of the Commission Guidance Paper, which relaxes the legal test for markets which are regulated or did not develop under normal commercial conditions. 79 Framework Directive 2002/21 as amended, Articles 14–16. 80 Past practice in the US electricity sector shows that the risk of type I errors is already important in the network industries. See Paul Joskow, ‘Mixing Regulatory and Antitrust Policies in the Electricity Power Industry: The Price Squeeze and Retail Market Competition’, in Franklin Fisher, ed., Antitrust and Regulation: Essays in Memory of John McGowan, MIT Press, 1985, at 174, observing that ‘the great quantity of litigation motivated by concern about price squeezes in particular, and retail market competition in general, has had no positive efficiency consequences; it is at best a waste of time and litigation expense and at worst a source of inefficiency’. 81 Jerry Hausman, ‘Valuing the Effect of Regulation on New Services in Telecommunications’, Brookings Papers: Microeconomics, 1997. 82 See above.

204  Competition, Regulation and Public Policies To alleviate some of those difficulties, I suggest that margin squeeze should rely exclusively on the costs of the dominant firms and not the costs of the competitors. Otherwise, there is a risk of supporting competitors instead of competition.83 I suggest further that the degree of dominance on the wholesale level should play a more prominent role in proving the abuse, and that the competition authorities should be particularly cautious when intervening in emerging markets. As to the differentiation of the test for refusal to deal, it is intuitive economics that the costs of type I errors and the negative effects on investment incentive may be of less concern in sectors which are regulated or which have developed outside normal market conditions. However, that intuitively appealing approach has several weaknesses and may result in the contamination of orthodox economically oriented competition policy with sectoral characteristics or regulatory objectives.

4. The proliferating roles of the Commission in post-liberalised markets The last issue to be addressed here is institutional in nature. It emerges from the telecoms cases analysed above and is linked to the evolution of the cycle of EU antitrust enforcement in the telecommunications, which may be explained in part by the different uses that the Commission has made of its antitrust powers.

4.1. The cycle of EU antitrust enforcement and the different uses of antitrust power Antitrust enforcement by the Commission in the telecoms sector followed several phases. The first phase can be traced to 1998, a key year which marked the beginning of full liberalisation. At that stage, the Commission was very active. It adopted general guidelines,84 which were not based on a stock of previous cases,85 on the application of competition law to key access issues. It also carried out, as noted above, several sector enquiries or quasi-sector enquiries. And it opened many individual cases for exploitative and exclusionary abuses. None of those 83 This is also the view of the ERG, which has explained that ‘a competition authority, in assessing whether or not a dominant position has been abused, would tend to base its calculations on the costs actually incurred by the dominant undertaking. In particular, this would allow the dominant player to take credit for economies of scope or scale and consequently to be able to operate a profitable downstream service on thinner margins than would be possible for an entrant.’ Revised ERG Common Position of 2006 on the approach to Appropriate remedies in the ECNS regulatory framework, ERG (06) 33, at p 120. 84 Access Notice, cited above note 47. Prior to the Access Notice, see also Commission Guidelines on the application of EEC Competition rules in the Telecommunications sector, 1991 OJ C233/2. 85 Ibid, para 6.

Alexandre de Streel 

205  

cases led to a formal infringement decision, as all the cases were either transferred to NRAs, when they had the competence and the willingness to act, or settled by the Commission. The second phase took place after liberalisation, when oversight of the NRAs by the Commission was still weak. At that stage, the Commission adopted several decisions in cases involving exclusionary abuses, essentially forcing the NRAs to act to accommodate the Commission’s preferences. Thus, the Wanadoo case ended with a decrease of wholesale access prices imposed by the French regulator; the Deutsche Telekom case ended with a wholesale price decrease imposed by the German regulator; and Telekomunikacja Polska ended with an agreement between the incumbent and the Polish regulator on efficient access to the local loop. With those cases, the Commission used competition law to complement or correct national regulatory decisions which in the Commission’s view were inappropriate. In some of those circumstances, the Commission could also have opened infringement procedures under Article 258 TFEU against the Member States of the failing regulators. According to the EU Courts, the choice between bringing an antitrust case and an infringement procedure under Article 258 is left to the broad discretion of the Commission.86 This allows for more flexibility and efficiency, since Article 258 actions can be slower and more politicised. However, as we have seen, resolving the problem via an antitrust procedure can substantially extend the special responsibility of the dominant operator. If the Commission opts for an antitrust action, the next question is whether it should cooperate with the NRA. On that issue too, the EU Courts accord discretion to the Commission, holding that the ‘sincere cooperation’ clause of the TEU (i.e, Article 4(3)) does not require the Commission to cooperate with or consult the NRAs during an antitrust investigation.87 This flexibility can be useful, especially if the NRAs is liable to take inappropriate action and/or delay the investigation. On the other hand, since NRAs have expert knowledge of the sector, it is generally better for the Commission to cooperate with them during its investigations and when it designs remedies. This is in fact what usually happens in practice.88 At present we are in a third phase of the enforcement cycle. In this phase, the Commission opens fewer abuse of dominance cases in the telecoms sector. Meanwhile, more cases are being pursued by national competition authorities.

Telefónica I, cited above note 33, para 307; Telefónica II, cited above note 33, para 115. Telefónica I, cited above note 33, para 312; Telefónica II, cited above note 33, para 47, stating that ‘the rules for the operation of the duty of sincere cooperation which stems from Article [4(3) TEU] and which binds the Commission in its relationships with the Member States have been stated in, inter alia, Articles 11 to 16 of Regulation 1/2003, in Chapter IV headed “Cooperation”. Those provisions do not impose an obligation on the Commission to consult the NRA, nor do they provide for the Commission being able […] to undertake “joint action” with them in proceedings conducted by the Commission pursuant to Articles [101 and 102 TFEU].’ 88 At the national level, many competition and regulatory authorities have concludes cooperation agreements. See OECD, Regulated Conduct Defence, cited above note 40, at 46–48. 86 87

206  Competition, Regulation and Public Policies

4.3. The rationale for and the risk of the Commission’s proliferating roles The evolution of the cycle of EU enforcement in this sector may be explained by: increasing competition in the sector; the progressive strengthening of sectoral regulation and of regulators; the increasing importance of national competitions authorities; and by the different uses the Commission has made of its antitrust powers. The first phase described above, characterised by high antitrust activism, can be explained by the concentration of the market which had just been formally liberalised, by the weakness of sectoral regulation at that time, and by the Commission’s determination to support the liberalisation programme (which it had initiated and pushed forward against the will of many Member States). As noted by Ungerer,89 a key senior Commission official at that time: ‘The [excessive price actions] aimed particularly at passing on rapidly the advantages of liberalisation in terms of price reductions and service developments to consumers – a major objective in order to show as rapidly as possible the effective consumer benefits and to secure sustained public support for liberalisation.’ The second phase, with cases highlighting the inadequacy of national regulatory decisions, may be explained by the use that the Commission made of its antitrust power either to correct the decisions of national regulatory authorities which were new and which, in some Member States, were captured by the incumbents; or to impose on national authorities its vision of a common regulatory approach. The third phase, with fewer cases, may be explained by increasing competition in the market due to the strengthening of the regulation and the technological progress, by the increasing activity of the NCAs in the telecommunications sector, and by new powers the Commission gained, under sector regulation, to control the NRAs’ decisions.90 Thus, the analysis of the enforcement cycle and the reasons for its evolution shows that the Commission uses its antitrust powers not only to ensure that competition is not distorted on the market but also to support the liberalisation programme and to control, and if necessary correct, the decisions of NRAs. Those additional uses of antitrust power are not without danger. First of all, they lead to duplication of procedures and to the risk of conflicting decisions, and they encourage forum shopping between regulators and the Commission. Second, the political mandate and the not-fully-open procedures followed by DG Competition when dealing with a case are not well suited for such a proactive role in the market.91 Third, it upsets the EU institutional balance, as the 89 Herbert Ungerer, ‘Use of EC Competition Rules in the Liberalisation of the European Union’s Telecommunications Sector’ (2001). 90 Since 2003, in addition to the standard ex post infringement procedure, the Commission has had an additional specific tool to control the NRAs decisions and, in case of violation of competition law, veto part of those decisions. See Articles 7 and 7a of the Framework Directive (Directive 2002/21), as amended by Directive 2009/140. 91 Pierre Larouche, Competition Law and Regulation in European Telecommunications, Hart Publishing, 2000, at 353–358; Damien Geradin and J. Gregory Sidak, ‘European and American Approaches to Antitrust Remedies and the Institutional Design of Regulation in Telecommunications’,

Alexandre de Streel 

207  

Commission is sometimes behaving as a European telecoms regulator, even though the establishment of such an authority has thus far been rejected by the EU legislator. Therefore, I submit that the antitrust intervention of the Commission should be decided case-by-case on the basis of two questions: first, what are the objectives to be achieved (legal certainty, simplicity, internal market, consistency of antitrust law); and second, what are the best means to achieve them (Commission ex ante review of NRA draft decisions on the basis of Articles 7 and 7a of the Framework Directive, infringement procedure on the basis of Article 258 TFEU, or an antitrust case on the basis of Article 102). In case of inappropriate intervention by an NRA, I suggest that the Commission should signal the infringement of EU law during its review of the draft decision, and veto it when it can. If the veto is not possible (in particular because the presumed infringement relates to a remedy) and the decision is nevertheless adopted, the Commission should intervene in the most efficient way, possibly taking parallel actions against the regulated operator and against the Member State concerned.92 In those circumstances, the two actions would be at the expense of legal certainty, as all actors would be informed of the serious doubts of the Commission when reading its appraisal of the NRA’s draft decision. Another way forward would be to enable the Commission to appeal the regulatory decision before an appropriate national court.

5. Conclusion In the telecoms sector, the Commission has been very active opening cases under Article 102, especially at the beginning of the period of liberalisation. Such activism has been fully supported by the EU Courts which, in deciding on the main contentious legal issues, has extended significant power to the Commission to control unilateral conduct in the sector. in Martin Cave, Sumit Majumdar and Ingo Vogelsang, eds., Handbook of Telecommunications Economics v II, North-Holland, 2005, pp 518 et seq. 92 Some authors argue that, when multiple legal actions are possible, the Commission should prioritise the infringement action over the antitrust action in order to avoid holding the incumbent responsible for the mistakes of its regulator, to ensure more legal certainty, and to avoid multiple complaints. See Nicolas Petit, ‘The Proliferation of National Regulatory Authorities alongside Competition Authorities: A Source of Jurisdictional Confusion’, in Damien Geradin, Rodolphe Muňoz and Nicolas Petit (eds), Regulatory Authorities in the EC: A New Paradigm for European Governance (Edward Elgar, 2005) pp 194 et seq; Damien Geradin, ‘Limiting the Scope of Article 82 of the EC Treaty: What can the EU learn from the U.S. Supreme Court’s Judgment in Trinko in the wake of Microsoft, IMS, and Deutsche Telekom?’, 41 Common Market Law Review 1552 (2004); John Temple Lang, ‘European Competition Policy and Regulation: Differences, Overlaps, and Constraints’, in François Leveque and Howard Shelanski (eds), Antitrust and Regulation in the EU and U.S.: Legal and Economic Perspectives (Edward Elgar, 2008).

208  Competition, Regulation and Public Policies As this chapter has explained, the EU Courts see competition law as a complement, and not a substitute, to sector-specific regulation. They allow competition authorities in Europe to condemn practices of dominant operators even where they have been approved by regulators, provided the operators have retained some autonomy within their regulatory constraints. A second point made in this chapter is that the EU Courts have adopted a relatively loose legal test to condemn instances of margin squeeze, ie the main abusive practice in telecoms markets. The competition authority (or plaintiff) thus only has to prove that: (i) the margin between wholesale and retail price is negative or insufficient to cover the retail costs of the defendant, or in specific circumstances, insufficient to cover the costs of the defendant’s rivals; and (ii) the compressed margin produces actual or potential anticompetitive effects. There is no requirement that the wholesale price must be excessive, or that the retail price be predatory or that the wholesale service be essential. The third point is that the EU Courts allow the Commission wide freedom of action o decide how to deal with the behaviour of a dominant regulated operator, which may be explained partly by the potential for inappropriate action by the national regulator and partly by the potential for abusive exercise of the autonomy left to the operator. In such a case, the Commission has the option to bring an action against the relevant Member State for failure to comply with its obligations under EU law, and/or to bring an antitrust action against the dominant operator. If it opens an antitrust case, the Commission may, but is not obliged to, cooperate with the NRA during the investigation. The antitrust activism of the Commission in this sector, with the blessing of the Courts, was welcomed at the beginning of the liberalisation process when the sector was still very concentrated and when the NRAs were in their infancy and sometimes captured. However, this activism, which is in dramatic contrast to the application of antitrust law in the telecoms sector in the U.S., is not without its dangers. First, it increases the risk of type I errors, which may be particularly costly in fast-moving industries. Second, it may lead to the development of a sectorspecific competition law; or worse, it may contaminate the way competition law is applied more generally across the economy with sectoral reasoning. Third, it tends to multiply legal proceedings at the national and EU levels, thereby increasing transaction costs and undermining legal certainty. It is thus understandable that the Commission today concentrates on consolidating the expertise and the independence of NRAs, and on using under sectoral regulation its ex ante review power on the NRAs’ draft decisions to signal violations of the competition law. It is only when some abusive practices remain that the Commission should consider, with due caution, to open an antitrust case. In other words, the Commission has a special responsibility not to abuse the important antitrust power granted to it by the EU Courts.

Lars Kjølbye*

Enforcement of Article 102 TFEU in the Energy Sector

Introduction In the energy sector, liberalisation and enforcement of competition rules go hand in hand. The objective of liberalisation is to introduce effective competition, and the competition rules aim at protecting it. The importance of creating a competitive internal energy market has been reiterated by the Commission on numerous occasions. For instance, in its Communication on an Energy Policy for Europe, the Commission emphasised that a real internal energy market is essential to meet Europe’s three energy challenges – (a) competitiveness: a competitive market will cut costs for citizens and companies and stimulate energy efficiency and investment; (b) sustainability: a competitive market is vital to allow for the effective application of economic instruments; and (c) security of supply: an effectively functioning and competitive Internal Energy Market can provide major advantages in terms of security of supply and high standards of public service.1 More recently, the Commission reiterated that active competition policy enforcement at the European and national levels remains indispensable to foster competition and guarantee that consumers have access to energy at affordable prices.2 The enforcement of Articles 101 and 102 TFEU have played an important role in removing obstacles to competition in the internal market. In June 2005 the Commission launched an inquiry into EU gas and electricity markets due to perceived market malfunctioning.3 The energy sector inquiry marked the beginning of a period of active enforcement of EU competition law to address concerns identified in the inquiry and to support the Commission’s proposals for a Third Energy Package, including its call for effective unbundling of network and supply activities.4 Many of the competition investigations involved network access and vertical integration in this sector. * Partner, Latham & Watkins LLP, Brussels. 1 COM (2007) 1 final of 10 January 2007, SEC(2007) 12. 2 See Commission, Energy 2020 – A strategy for competitive, sustainable and secure energy, COM(2010) 639 final of 10 November 2010, p 14. 3 Commission Decision of 13 June 2005 initiating an inquiry into the gas and electricity sectors pursuant to Article 17 of Council Regulation (EC) No 1/2003 – Cases 39.172 (electricity) and 39.173 (gas). 4 The Third Package was adopted in 2009 and consists of three regulations and two directives. See http://ec.europa.eu/energy/gas_electricity/third_legislative_package_en.htm.

210  Competition, Regulation and Public Policies

Key abuse of dominance cases in the energy sector Since the launch of the energy sector inquiry the Commission has adopted no less than nine decisions under Article 102 TFEU. All the decisions are so-called ‘commitments’ decisions taken under Article 9 of Regulation 1/2003. Such decisions make commitments offered by undertakings under investigation binding upon then. Unlike an Article 7 decision, a decision to make commitments binding does not (and cannot) entail any finding of infringement. However, the decisions remain important because they set out the Commission’s enforcement policy in the sector. Several of these decisions are particularly interesting because they take Article 102 into new territory and therefore merit discussion.

Essential facilities and strategic underinvestment The Commission’s Guidance Paper on exclusionary conduct under Article 102 establishes three conditions for imposing obligations to deal: (i) access must be necessary to enable effective competition in downstream markets; (ii) the refusal must be likely to eliminate effective competition; and (iii) it must be likely to lead to consumer harm, eg because it is not objectively justified to maintain incentives to innovate.5 It is well established in EU competition law that owners of essential facilities may be required to share unused capacity with competitors. However, recent energy cases take the essential facility doctrine one step further by applying Article 102 to instances of ‘strategic limitation of investment’ in a gas pipeline system. The EU’s second liberalisation package required vertically integrated energy companies to keep the operation of transmission networks legally and functionally separate from supply activities. Transmission networks are generally considered to be natural monopolies, and unbundling requirements aim to address the risk that vertically integrated network operators may favour related supply activities instead of optimising the network in the interest of the market as a whole. In the ENI case, the Commission had concerns that the network branch of the Italian gas incumbent strategically refrained from carrying out, proposing or agreeing to investments in additional capacity in import pipelines, in order to keep import capacity tight and to maximise revenues on the related gas supply market.6 According to the Commission, ENI had failed to ascertain the feasibility and viability of capacity enhancements despite the willingness of third parties to underpin these investments with long-term capacity reservations. Since ENI 5 Guidance on the Commission’s enforcement priorities in applying Article 102 TFEU to abusive exclusionary conduct by dominant undertakings, 2009 OJ C45/7, para 81. 6 Commission Decision of 29 September 2010 in Case COMP/39.315 – ENI. Similar concerns were raised in relation to two LNG terminals in Commission Decision of 3 December 2009 in Case COMP/39.316 – GDF gas foreclosure.

Lars Kjølbye 

211  

had not taken any step to meet third parties’ demand, it was not necessary for the Commission to assess in detail whether the investments would have been technically feasible and economically viable. After the Commission sent ENI a Statement of Objections, the company offered commitments entailing divestiture of its share in the import pipelines concerned.

Long-term capacity reservations in own network The Commission arguably took the essential facility doctrine even further in the GDF and E.ON gas cases.7 In France and Germany, most of the import capacity at the entry points was reserved long-term by GDF and E.ON Ruhrgas, which also owned the infrastructure in question. The reserved capacity was actually used; there was no capacity hoarding. The Commission took the preliminary view that the long-term reservations and the resulting refusal to provide capacity to third parties were abusive. The Commission did not accept that the reservations were objectively justified because the capacity was actually used and the reservations matched long-term gas import contracts with upstream suppliers. Since, under EU competition law, gas supplies cannot be destined for a particular market, nothing obliged the importers to sell the gas in their home market. They were free to sell the gas in other markets. To resolve the case, GDF and E.ON offered commitments whereby they each released a significant share of the reserved capacity.

Withdrawal of generation capacity Experience shows that electricity markets have structural characteristics that enable electricity generators to increase prices by withdrawing generation capacity even at relatively low market share levels.8 Electricity can be produced by different technologies (hydro, coal, gas, nuclear, wind, solar, etc.) with very different cost structures. The price of electricity traded on exchanges is set by the marginal plant needed to serve demand (the ‘merit curve’). This industry pricing model is illustrated on next page:

7 GDF gas foreclosure, cited previous footnote; and Commission Decision of 4 May 2010 Case COMP/39.317 – E.ON gas foreclosure. See also Ricardo Cardoso, Sandra Kijewski, Oliver Koch, Patrick Lindberg and Károly Nagy, ‘The Commission’s GDF and E.ON Gas decisions concerning long-term capacity bookings: Use of own infrastructure as possible abuse under Article 102 TFEU’, (2010/3) Competition Policy Newsletter 8. 8 See, eg, Commission Decision of 22 December 2008 in Case COMP/M.5224 – EDF/British Energy.

212  Competition, Regulation and Public Policies

Prices are set by the marginal plant

Marginal Cost

Price formation on competitive short-term electricity markets Supply Demand

Market price

GT CCGT coal hydro

nuclear

lignite Capacity

In the German electricity wholesale market case, the Commission had concerns that E.ON held a dominant position collectively with at least RWE and might have infringed Article 102 by withholding generation capacity with a view to increasing prices.9 More specifically, the Commission had concerns that E.ON had designed and implemented a strategy to withdraw available capacity for that purpose. By limiting production of certain plants in the merit order, more expansive plants were needed to satisfy demand, resulting in higher prices. In those circumstances, the Commission decided to accept E.ON’s commitment to divest a significant amount of generation capacity to address its competition concerns. Withholding capacity was defined as a sui generis form of abuse. The Commission made the preliminary finding that the capacity E.ON had withheld was profitable to run, ie that the marginal cost of running the plant would have been covered by the price offered on the exchange. The price was considered to be ‘unfair’ because it deviated from the industry pricing model. This implies that, in the electricity sector, because of its specific features (steep demand curve, nonstorability of the good, etc.), a dominant undertaking must run all its available capacity and does not have the usual freedom to arbitrage between volume and price. The Commission’s approach has been followed by national competition authorities in the EU.10

Competition policy or energy policy? The cases summarised above each take Article 102 into new territory. They extend the essential facility doctrine so that it covers a failure to ascertain and meet demand 9 Commission Decision of 26 November 2008 in Case COMP/39.388 – German electricity wholesale market. 10 For a synthetic presentation of enforcement by national competition authorities, see Roberto Grasso and John Ratliff, ‘Unilateral conduct in the energy sector: An overview of EU and national case law’, 15 March 2012, e-Competitions, No. 44203.

Lars Kjølbye 

213  

for new capacity, and the reservation of own capacity for own use. They also suggest that the very act of exercising market power can in itself be abusive despite the fact that Article 102 prohibits abuse and not dominance.11 The cases also seem to be sector-specific. It is difficult to imagine that the Commission would take similar action in sectors that do not present the particular characteristics of the gas and electricity sectors, which are characterised by natural monopolies in networks, low short-term inelasticity of demand and (until recently) vertical integration of network and supply activities raising conflict of interest concerns. The Commission’s emphasis on sectoral specificities raises the question whether the cases reflect broader energy policy objectives that deviate from traditional antitrust orthodoxy. The cases clearly form part of an energy policy context. The Third Energy Package introduced stricter unbundling requirements with a particular emphasis on measures aimed at ensuring that conflicts of interests due to vertical integration do not lead to sub-optimal network investments. Similarly, the EU has adopted a regulation to counter the risk of market manipulation by means of capacity withdrawal.12 The fact that antitrust enforcement priorities are aligned with broader policy objectives in a given sector is natural and desirable. Competition policy, after all, forms part of internal market policy.13 The application of Article 102 should also take account of the regulatory framework that applies in a given sector. This framework forms part of the legal context in which a practice operates.14 Thus, when EU law requires vertically integrated energy companies to unbundle network and supply businesses, it is arguably justified to take this fact into account in the assessment of objective justifications under Article 102. It is submitted that, when the EU legislator has forbidden vertically integrated network operators to favour affiliated supply entities, deviation from that obligation cannot be objectively justified. This may (at least to some extent) explain the Commission’s approach in ENI (see above). Absent incentives to favour affiliated supply activities, it would have been in the ENI’s interests as a network operator to gauge demand for new capacity (and thereby for new business), which it failed to do. It is submitted, however, that the existence of regulatory obligations does not provide a basis for derogating from the normal conditions for finding an abuse.15 The EU competition rules serve to protect competition on the market – not to enforce regulatory obligations.16 Thus, although in the GDF and E.ON gas cases 11 Case 322/81 NV Nederlandsche Banden Industrie Michelin v Commission [1983] ECR 3461, para 57. 12 Regulation 1227/2011 on wholesale energy market integrity and transparency (REMIT), 2011 OJ L326/1. 13 See Protocol 27 to the Treaties, according to which the Member States consider that an internal market includes a system ensuring that competition is not distorted. 14 Case C-238/05 Asnef-Equifax [2006] ECR I-11125, para 49. 15 Contrast Guidance Paper, cited above note 6, at para 82, where the Commission takes the view that it is not necessary to meet the normal conditions for finding an abusive refusal to deal when access obligations have been imposed by law. 16 See Recital 9 to Regulation 1/2003 on the implementation of the rules on competition laid down in Article 101 and 102 TFEU, 2003 OJ L1/1.

214  Competition, Regulation and Public Policies there was a need to address long-term capacity reservations to enable effective third party access to gas networks, it is less obvious that Article 102 provided a basis for achieving this objective. With regard to Article 101(3) the Commission has made clear that ‘[g]oals pursued by other Treaty provisions can be taken into account to the extent that they can be subsumed under the four conditions of Article 101(3)’; the same principle should apply to Article 102.17 It is not clear that Article 102 provides a basis for requiring dominant firms to hand over property that they actually use. The concept of an abuse requires that the conduct in question deviate from competition ‘on the merits’.18 Similarly, it is not obvious that it is ‘unfair’ within the meaning of Article 102(a) to increase prices by withdrawing generation capacity, even if such conduct deviates from the industry pricing model. Article 102 does not prohibit dominance. It is submitted that, by implication, the exercise of market power – which is inherent in dominance – cannot be abusive in itself. For a price to be ‘unfair’, it must deviate from ‘normal’ behaviour by firms with market power. Under existing case law, a price is unfair within the meaning of Article 102 only if it has no reasonable relationship to the economic value of the product supplied.19 Conclusion The Commission has brought a number of abuse of dominance cases in the energy sector, and these cases have made a significant contribution to creating integrated and competitive energy markets in Europe. The cases form part of the broader energy policy context, and some of the principles applied are unlikely to be applicable outside the energy sector. This raises the question of whether the Commission has merely applied Article 102 in the legal, economic and factual context in which the investigated conduct occurred; or whether it has gone further, pursuing what are in fact regulatory objectives.

Commission Guidelines on the application of Article 101(3) TFEU, 2004 OJ C101/97, para 42. See, eg, Case C-52/09 Konkurrensverket v TeliaSonera Sverige AB [2011] ECR I-527, para 27. 19 Case 27/76 United Brands Company and United Brands Continentaal BV v Commission [1978] ECR 207, para 250. 17 18

Ginevra Bruzzone* and Marco Boccaccio**

The Boundaries of Legitimate Conduct in PostLiberalization Markets

1. Introduction Since the end of the 1980s, competition rules have been extensively used to promote the de-monopolization and liberalization of markets in the EU Member States. In a number of cases the joint application of Articles 106 and 102 TFEU has been a direct instrument of de-monopolization. More often, the prohibition of abuse of dominance has been used by the European Commission and competition authorities of the Member States to control and constrain the business practices of former monopolists in post-liberalization markets. In Italy, for instance, where the presence of state-owned monopolies in utilities was widespread, two-thirds of the decisions on abuse of dominance adopted by the Italian Competition Authority since its establishment in 1990 concerned exclusionary, and in a few cases also exploitative, practices in post-liberalization markets. Priorities in public enforcement are related to the different stages of the liberalization process: although in the 1990s there was a strong focus on telecommunications, in recent years there have also been several cases in the energy sector and an increasing number of decisions concerning postal services and rail transport. Some common features are relevant when applying the prohibition of abuse of dominance in post-liberalization markets. In the first place, at the very beginning of the liberalization process the incumbent holds, by definition, a very high market share and a number of elements may put it at a competitive advantage with respect to newcomers (eg, economies of scale and scope, distribution network, long-term contracts, excess of production capacity, switching costs). It may also maintain exclusive rights in some areas and/or it may be the incumbent in specific natural monopoly segments, and regulation may not be sufficient to completely address the resulting market power. Therefore, it is often the case that ascertaining that the incumbent is dominant is a straightforward exercise. However, depending on the specific features of the markets involved, things may rapidly change. It is thus good practice, especially as the liberalization process develops, not to skip the assessment of market power based on a full analysis of competitive constraints.1 * Deputy Director General, Assonime, Rome. ** Professor of Public Finance, Università di Perugia. 1 The European Commission’s Guidance Paper on exclusionary abuses states that, ‘as a general rule, the Commission will not come to a final conclusion as to whether or not a case should be

216  Competition, Regulation and Public Policies A second feature is that in these markets dominance may have been attained through state intervention instead of by competing ‘on the merits’. The issue therefore arises as to whether, and to what extent, this circumstance should influence the application of Article 102.2 It is often argued that the European idea of a ‘special responsibility’ of dominant companies not to impair undistorted competition on the common market can be more easily understood with reference to public monopolies or former monopolies, where dominance is seen as the result of public privileges instead of entrepreneurial ability. In particular, the duty to deal doctrine found a suitable habitat in an environment in which infrastructures were developed by state-owned monopolists as opposed to undertakings operating in the private sector.3 Indeed, the Court of Justice recently acknowledged that, when the existence of a dominant position has its origin in a former legal monopoly, ‘that fact has to be taken into account’.4 On the other hand, according to Article 345 TFEU, ‘[t]he Treaties shall in no way prejudice the rules in Member States governing the system of property ownership’; therefore, discrimination in the enforcement of competition rules based on the public or private ownership of undertakings is precluded.5 EU rules on State aid control contribute to prevent distortions of competition resulting from the selective use of public resources. From a substantive viewpoint, in post-liberalization markets the enforcement of Article 102 should ensure predictability and take into account the impact of the case law on the incentives of undertakings, and thereby on the competitive process. As for predictability, it is commonly acknowledged that the incumbent has the right to compete and should be enabled to know in advance the boundaries between legitimate and unlawful competitive practices. A corollary is that the assessment of competition on the merits as the dividing line between lawful and abusive practices cannot be merely based on the origin of dominance (which in these markets may not depend on merit) and on alleged harm to competitors, since in principle any competitive conduct by a dominant company can be challenged in that it may harm competitors. Therefore, the assessment of an abuse within the meaning of Article 102 has to be based on objective criteria relating to the conduct itself and its impact on competition (as distinct from its impact on competitors).6 pursued without examining all the factors which may be sufficient to constrain the behaviour of the undertaking’. 2009 OJ C45/02, para 15. 2 See, eg, the Opinion of Advocate General Maduro in Case C-109/03 KPN Telecom BV v Onafhankelijke Post en Telecommunicatie Autoriteit (OPTA) [2004] ECR I-11273, para 41. 3 George Priest and Franco Romani, ‘L’antitrust negli Stati Uniti e in Europa. Analisi e psicoanalisi di una divergenza’, 4(1) Mercato concorrenza regole 151 (2002). See also the Opinion of Advocate General Mazák in Case C-52/09 Konkurrensverket v TeliaSonera Sverige AB [2011] ECR I-527, paras 24–25 and the case law cited therein. 4 Case C-209/10 Post Danmark A/S v Konkurrencerådet [2012] ECR I-000, para 23. 5 Opinion of Advocate General Mazák in TeliaSonera, cited above note 3, para 27. 6 On the irrelevance of the origin of dominance and the need to focus on the operator’s conduct on the market, see Commission Decision of 14 January 1998 in Case IV/34.80 – Flughafen Frankfurt/ Main AG, 1998 OJ L72/30.

Ginevra Bruzzone and Marco Boccaccio 

217  

The challenge for the case law is to establish criteria which can be taken as guidance by dominant companies for their future competitive conduct. In this perspective, the attempt by the Commission to base its decisions on clearly defined theories of harm and to follow an effect-based approach – carefully assessing the actual or likely negative impact of the conduct on the parameters of competition (such as price, output, innovation, the variety and quality of products) and therefore, at least indirectly, on consumers7 – represents a positive development since it clarifies the benchmark for the assessment of abuses of dominance. Furthermore, the ECJ in a number of significant judgments has acknowledged the relevance of the analysis of the actual or potential impact the conduct has (or may have) on the market for ascertaining whether it can be considered an abuse under Article 102.8 A further positive development from the point of view of legal certainty is the endorsement by the EU Courts, both for margin squeeze and predation, of the principle that in the assessment of allegedly exclusionary pricing practices reference should be made to the ability to exclude an undertaking at least as efficient as the dominant company,9 and therefore that the costs of the dominant company are the benchmark when assessing whether a given price can be considered abusive.10 Predictability is also enhanced by the acknowledgement in TeliaSonera that the degree of dominance (the level of market shares) is relevant in the application of Article 102 only because of its influence on the potential effects of the conduct.11 Therefore, the ECJ endorses an effects-based approach to the assessment of the conduct of super-dominant companies, shedding some light on the foggy area of their special responsibility. Besides predictability, a second requirement for a proper application of Article 102 in post-liberalization markets is that competition authorities should take into account the consequences of their enforcement choices on the incentives of both the incumbent and the entrant(s) in these markets, as well as of undertakings operating in other markets. Indeed, an ad hoc approach aimed at attaining a specific objective in a case may easily acquire a life of its own and produce principles with broader consequences.12 Although several cases of application of Article 102 in post-liberalization markets, both by the Commission and by competition authorities of the Member 7 See, eg, Commission Decision of 24 March 2004 in Case COMP/37.792 – Microsoft, 2007 OJ L32/6, upheld (on the substantive aspects): Case T-201/04, Microsoft Corp v Commission [2007] ECR II-3601. See also Commission Decision of 13 May 2009 in Case COMP/37.990 – Intel, 2009 OJ C227/13. 8 Case C-280/08 P Deutsche Telekom [2010] ECR I-9555, paras 252–254 and 259; TeliaSonera, cited above note 5, paras 66–74; Post Danmark, cited above note 4, para 44; Case T-336/07, Telefónica, SA and Telefónica de España, SA v Commission [2012] ECR II-000, paras 201 and 275–276. 9 Deutsche Telekom, cited previous footnote, para 177; TeliaSonera, cited above note 5, paras 131– 133; Post Danmark, cited above note 6, paras 21–22, 25 and 38; Telefόnica, cited previous footnote, paras 189–191. 10 TeliaSonera, cited above note 5, para 44. 11 Ibid, para 81. 12 Opinion of Advocate General Mazák in TeliaSonera, para 26.

218  Competition, Regulation and Public Policies States, are based on established principles (such as the need to avoid longterm exclusive dealing arrangements which would protect the incumbent from competitive challenges), there remain a number of controversial issues. In this chapter we will focus on two topics which have been crucial in cases applying Article 102 both at EU level and at the level of the Member States. The first one is the extent to which a multi-product dominant undertaking can compete aggressively on price in post-liberalization markets; this issue was addressed by the ECJ in its Post Danmark judgment of 2012. The second issue, raised by certain cases in the energy sector, is whether Article 102 implies not only a duty of dominant companies to provide access to essential facilities, but also, and under what circumstances, a duty to ensure a sufficient level of investment in infrastructure.

2. Price competition by multi-product dominant undertakings The Commission’s approach in Deutsche Post and in the 2009 Guidance Paper Under the traditional AKZO approach to predation, the ECJ considers prices below average variable costs to be presumptively abusive since each sale generates a loss, whereas prices below average total costs (ie, fixed costs plus variable costs) but above average variable costs are regarded as abusive only if they are adopted as part of a plan for eliminating a competitor.13 The issue of alleged predation by multi-product undertakings in postliberalization markets was dealt with by the Commission in Deutsche Post AG in 2001,14 with reference to a situation where the postal operator still benefited from a reserved area; it still had a statutory universal service obligation for specific services, and it was said to use revenue from its profitable reserved area to crosssubsidize liberalized services, which were allegedly offered below cost. The Commission argued that cross subsidization occurs when the earnings from the liberalized service do not suffice to cover its incremental cost and there is another service, or bundle of services, the earnings of which exceed the stand-alone costs. Moreover, the Commission stressed that the incremental costs comprise only the costs incurred in providing the service at stake. In particular, they include the fixed costs incurred as a result of providing the service, but not the fixed costs which cannot be attributed solely to that service (common fixed costs). Following this 13 Case C-62/86, AKZO Chemie BV v Commission [1991] ECR I-3359, paras 71–72. In particular, the ECJ stated that prices below average total costs ‘can drive from the market undertakings which are perhaps as efficient as the dominant undertaking but which because of their smaller financial resources are incapable of withstanding the competition waged against them’. 14 Commission Decision of 20 March 2001 in Case COMP/35.141 – Deutsche Post AG, 2001 OJ L25/27.

Ginevra Bruzzone and Marco Boccaccio 

219  

approach, the Commission analyzed, at each of the stages through which mail order parcels were processed, the share of costs which were attributable to mail order parcel services. Collection costs, for instance, were fully attributable to the service (ie, they would have been avoided if the service was stopped); the capital cost of setting up freight centers and delivery points could not be attributed to a particular service; the staffing and equipment costs of sorting could be attributed in direct proportion to the volume of parcels to be conveyed; the cost of long distance transport in terms of staffing, equipment and capital was not attributable, since it would have been borne anyway. The Commission, which expressly referred to the writings of Baumol, Sidak and Spulber,15 acknowledged that the presence of a statutory universal service obligation affects the share of common non-attributable fixed costs. Indeed, the postal operator was required by law to maintain a reserve capacity to cover any peak demand which might arise, while meeting statutory service quality standard obligations. Even if Deutsche Post were to stop offering a specific parcel service, it would not be free to cut back on staff and equipment in perfect proportion, since it was the ‘carrier of last resort’. In Deutsche Post the incumbent undertaking was found to have applied prices below average incremental costs, thereby abusing its dominant position, in the years from 1990 to 1995 but not later. To resolve the investigation, Deutsche Post undertook to ensure complete transparency of the financial relations between the reserved area and liberalized services. No fines were imposed because the relevant cost measures that a multi-service operator had to meet had not previously been clarified. Therefore, in Deutsche Post the Commission took incremental costs as the relevant benchmark that a multi-service postal operator must meet in its competitive activities in order to comply with the prohibition of predatory pricing under Article 102. This approach was not meant to burden these operations with the common fixed costs resulting from statutory universal service obligations, which would have implied setting ‘artificial’ limits to the application of low prices. A problem, under this approach, is that in the presence of a very high share of common fixed costs it may be difficult for a competitor to profitably compete at prices just covering the average incremental costs of the incumbent, and in principle the incumbent may adopt strategic behaviour aimed at artificially increasing common fixed costs with an exclusionary intent.16 On the other hand, the practical relevance of this scenario should not be presumed but should be evaluated on a case-by-case basis, since in several sectors the costs resulting from public service obligations are to a large extent exogenous. In the Guidance Paper of 2009, the Commission illustrates its general approach to these issues in the section on price-based exclusionary abuses. According to 15 William Baumol and Gregory Sidak, Toward Competition in Local Telephony (MIT Press, 1994); Gregory Sidak and David Spulber, Protecting Competition from the Postal Monopoly (American Enterprise Institute Press, 1996). 16 See, eg, Frédéric Jenny, ‘Recent Developments in the Enforcement of Article 102: Predation’, presentation at the IVth Intertic Conference on Antitrust, Rome, 24 May 2012.

220  Competition, Regulation and Public Policies the Guidance Paper, the Commission will normally intervene only in case of foreclosure of an as efficient competitor (in line with AKZO) and if, in addition, foreclosure leads to likely consumer harm (anticompetitive foreclosure), taking into account all relevant quantitative or qualitative evidence. As for cost benchmarks, the Commission explains that it will normally refer to average avoidable costs (variable and fixed costs which could have been avoided if the company had not produced the amount of output allegedly subject of abusive conduct in the period under consideration) and long run incremental costs (ie, all the attributable variable and fixed costs, including sunk costs, that the company occurs to provide the product). For multi-product undertakings, long-run incremental costs may be below average total costs, since they do not include true common costs. However, any costs that could have been avoided by not providing the product are not considered to be common costs.17 The Commission emphasizes that failure to cover average avoidable costs means that the undertaking is incurring avoidable losses and that an equally efficient competitor would not be able to serve the customers without incurring a loss. On the other hand, failure to cover long run incremental costs means that the undertaking is not recovering all the attributable fixed costs, and that an equally efficient competitor could be foreclosed from the market. Interestingly, taking into account the debate which followed the Deutsche Post decision, the Commission adds that, in some situations where common costs are significant, those costs ‘may have to be taken into account when assessing the ability to foreclose an equally efficient competitor’.18 Moreover, as anticipated, according to the Guidance Paper the assessment of foreclosure or likely foreclosure of an ‘as efficient competitor’ based on the cost benchmark will have to be followed by an assessment of its anticompetitive consequences. As the Guidance Paper states, ‘likely consumer harm may be demonstrated by assessing the likely foreclosure effect of the conduct combined with consideration of other factors, such as entry barriers. In this context the Commission will also consider possibilities of re-entry.’19

The Post Danmark preliminary ruling In the Post Danmark preliminary ruling of 2012, the ECJ had the opportunity to express its view on how to assess the pricing conduct of a multi-product dominant undertaking under Article 102. The referring court wished to know whether a postal undertaking abused its dominant position on the ground of selective application of low prices in concluding contracts with three major customers of its principal competitor, Forbruger Kontakt, in the Danish market for the distribution of unaddressed mail. The market for unaddressed mail was fully liberalized. However, at the time, Post Danmark, which had a universal service obligation, 17 18 19

Guidance Paper, cited above note 3, paras 25–27. Ibid, footnote 18. Ibid, para 71.

Ginevra Bruzzone and Marco Boccaccio 

221  

benefited from a monopoly in the distribution of addressed mail not exceeding a certain weight. The undertaking had a distribution system covering the whole of Denmark, which it also used for the liberalized distribution of unaddressed mail.20 The question referred to the Court was whether selective price reductions by a dominant undertaking to a level lower than average total costs but higher than average incremental costs – with no evidence of predatory intent – was an exclusionary abuse; and in case of a positive answer, under what conditions. Advocate General Mengozzi shared concerns about a test based only on incremental costs in a case where the fixed common costs of the distribution network were high: in that situation, ‘the dominant undertaking could […] cause the customers of the partly reserved operations in the market in which it has the universal service obligation to bear the cost of the distribution network capacity mobilized by the dominant undertaking for its unaddressed mail business’. The Advocate General continued, stating: [I]t is perfectly possible, taking the average incremental cost as the yardstick, for the dominant undertaking to be able to charge a price slightly higher than the average (thus preventing the price from being automatically regarded as predatory) by causing all or part of the common fixed costs to be borne by the partly reserved operations in the market in which it has a universal service obligation and those operations therefore subsidize the price offered on the market which is open to competition. Whether selective or not, such practice could eventually lead to driving competitors from the liberalized market.21

In the first part of the Post Danmark judgment, the Court of Justice discusses the use of average incremental costs as a cost benchmark. The Court acknowledges that ‘a notable feature of the case in the main proceedings is that there are considerable costs related to both the activities within the ambit of Post Danmark’s universal service obligation and to its activity of distributing unaddressed mail’. These costs seemed at least in part attributable to the liberalized service since, according to the Danish Competition Authority, ‘because Post Danmark’s unaddressed mail used the undertaking’s “common distribution network resources”, the costs of its universal service obligation activities could be reduced over a period of three to five years if Post Danmark were to give up distributing unaddressed mail’.22 With reference to the cost benchmarks, the Court of Justice underlines that the exercise of attribution of costs in order to estimate the average incremental cost remains relevant where formally common costs are incurred as a result of providing a number of services. The AIC notion includes those ‘common costs’ which would be avoided if the company were to stop the provision of the service. From a conceptual point of view, the Danish authorities, consistently with the Commission’s approach in Deutsche Post and in the Guidance Paper, defined incremental costs as being ‘those costs destined to disappear in the short or medium 20 On the main features of the national case from which the preliminary ruling originated, see Christian Bergqvist, ‘Final Curtain or Another Around on Post Danmark?’, 34 European Competition Law Review 287 (2013). 21 Opinion of Advocate General Mengozzi in Post Danmark, cited above note 8, paras 111–112. 22 Post Danmark, para 32.

222  Competition, Regulation and Public Policies term (three to five years) if Post Danmark were to give up its business activity of distributing unaddressed mail’.23 A more controversial issue is whether the way in which the Danish competition authority estimated average incremental costs was a correct application of this approach. The preliminary ruling just mentions that, for the purpose of estimating average incremental costs, the competition authority included not only those fixed and variable costs attributable solely to the activity of distributing unaddressed mail but also elements described as ‘common variable costs’, ‘75% of the attributable common costs of logistical capacity’ and ‘25% of non-attributable common costs’,24 and that, in the specific circumstances of the case, ‘such a method of attribution would seem to identify the great bulk of the costs attributable to the activity of distributing unaddressed mail’. This approach does not seem to correspond to the notion of AIC but to a more ambiguous mix that includes a share of non-attributable common costs. Be that as it may, the price applied by Post Danmark covered the average incremental costs as estimated by the Danish competition authorities. Therefore, the question put by the Danish court to the ECJ was how to assess, under Article 102, prices above average incremental costs but lower than average total costs (which were defined as average incremental costs plus a portion, determined by estimation, of Post Danmark’s common costs connected to activities other than those covered by the universal service obligation). With reference to this question, the ECJ states in its judgment that prices higher than the average incremental costs pertaining to the activity (as estimated in proceeding) may not be considered to amount to an exclusionary abuse merely because they are below the average total costs attributed to the activity concerned. In order to assess the existence of anticompetitive effects in circumstances such as those of the case it is necessary to consider ‘whether that pricing policy produces an actual or likely exclusionary effect, to the detriment of competition and thereby of consumers’ interests’. The Court adds that it is open to the dominant undertaking to provide a justification for behaviour liable to be caught by the prohibition under Article 102.25 In particular, the undertaking may demonstrate either that its conduct is objectively necessary, or that the exclusionary effect is counterbalanced or outweighed by advantages in terms of efficiency that also benefit consumers.26

Ibid, para 31. Ibid, para 33. 25 TeliaSonera, cited above note 5, para 76. 26 With regard to efficiencies, the Court indicates strict requirements on the model of Article 101(3): the efficiency gains should counteract any likely negative effects on competition and consumer welfare in the affected markets; those gains must have been, or must be likely to be, brought about as a result of the conduct; the conduct must be necessary for the achievement of those efficiencies; and the conduct must not eliminate effective competition by removing all or most existing sources of actual or potential competition. 23 24

Ginevra Bruzzone and Marco Boccaccio 

223  

Some comments on Post Danmark In several respects, Post Danmark is a landmark judgment. Notably, the Court endorses the effects-based approach as well as the ‘as efficient competitor’ test and the approach suggested in the Commission’s Guidance Paper as to how to frame an efficiency defense under Article 102.27 What the judgment leaves unclear is whether, as a general rule, prices above long-run average incremental costs but lower than average total costs, including a share of truly common costs, may represent an exclusionary abuse. Following the approach of the Guidance Paper, the cases in which it is justified to consider a price above long-run average incremental costs as potentially exclusionary are only a subset of all possible cases. This subset should be circumscribed as clearly as possible, since a broad application of an approach requiring the artificial attribution of truly common costs could have negative side-effects. As the CEPS report on the 2009 Guidance Paper pointed out, ‘when a firm decides whether to produce an additional good, it normally compares incremental costs and incremental revenues. If the firm concerned already covers common costs through the sale of other products whose demand is less elastic, a rule requiring that each product makes an appropriate contribution to general overheads could result in higher prices. Dominant firms should not be obliged to behave as if they were less efficient.’28 In other words, dominant firms should not be prevented from taking advantage of economies of scope, and competitors should maintain an incentive to expand their product range, also in ways different from the one chosen by the dominant company, if it is efficient to do so. These remarks also indicate why a competition perspective, focusing on incentives and on the dynamic impact of the market process on competitive variables, may differ from a regulatory/accounting perspective, whereby truly common costs are attributed to all activities. In order to circumscribe the set of conditions which can justify the inclusion of a share of truly common costs in the thresholds relevant for the application of Article 102 to prices ‘below cost’, a possibility would be to include only those situations in which three requirements are met: first, common costs are significant; second, the dominant company benefits from artificial advantages; and third, as suggested by the CEPS report, competitors cannot realize similar economies of scope by expanding their product range. As for artificial advantages, the EU’s control of State aid plays a complementary role with respect to the application of Article 102 since it aims at preventing situations of unfair advantages which could lead to abusive pricing conduct in 27 For a complete overview, see Ekaterina Rousseva and Mel Marquis, ‘Hell Freezes Over: A Climate Change for Assessing Exclusionary Conduct under Article 102 TFEU’, 4 Journal of European Competition Law and Practice 32 (2013). See also Stefano Barazza, ‘Post Danmark: the CJEU Calls for an Effect-based Assessment of Pricing Policies’, 5 Journal of European Competition Law and Practice 466 (2012). 28 CEPS, Treatment of Exclusionary Abuses under Article 82 of the EC Treaty. Comments on the European Commission’s Guidance Paper, 2009, especially pp 35–37.

224  Competition, Regulation and Public Policies markets open to competition. In particular, the rules on State aid in support of services of general economic interest (SGEI), recently revised by the so-called Almunia package,29 aim at avoiding overcompensation of operators charged with public service obligations and the related risk of cross-subsidies.30 These considerations notwithstanding, the risk that the application of Article 102 unduly discourages dominant firms from charging low prices is mitigated by the requirement, set in Post Danmark, of an effects-based approach. As noted earlier, the ECJ indicates that the assessment of the potentially negative effect of prices above average incremental costs but below average total costs should not hinge on whether competitors are harmed: it is necessary to consider whether the pricing policy, without objective justification, has a negative impact on the parameters of competition and thereby on consumer interests.

Developments in the decisional practice and case law of the Member States In the light of Post Danmark, it is interesting to monitor how Article 102 is being applied at national level with respect to the pricing policies of dominant companies subject to universal service obligations. In Italy, in 2012 the system of proof for the application of Article 102 to allegedly exclusionary pricing practices in liberalized postal services was at the center of a judgment of the Italian Administrative Court (TAR Lazio) reviewing the decision by the Italian Competition Authority in the TNT Post Italia/Poste 29 The Almunia package includes: the Communication from the Commission on the application of the EU State aid rules to compensation granted for the provision of SGEI, 2012 OJ C8/4; Commission Decision of 20 December 2011 on the application of Article 106(2) TFEU to State aid in the form of public service compensation granted to certain undertakings entrusted with the operation of SGEI, 2012 OJ L7/3; Communication, EU framework for State aid in the form of public service compensation (the ‘Framework’), 2012 OJ C8/4; Commission Regulation 360/2012 of 25 April 2012 on the application of Articles 107 and 108 TFEU to de minimis aid granted to undertakings providing services of general economic interest, 2012 OJ L114/8. For land transport, see Regulation 1370/2007 of the European Parliament and the Council of 23 October 2007 on public passenger transport services by rail and by road, 2007 OJ L315/1. 30 The criteria set out by the ECJ in Altmark which, if satisfied exclude a finding that public service compensation constitutes State aid, requires that the public service obligation and the parameters for calculating compensation be clearly defined ex ante and that either the provider be chosen through a public procurement procedure or that the level of compensation be calculated with reference to the costs of a typical, ‘well run’ and adequately equipped company. See Case C-280/00 Altmark Trans GmbH [2003] ECR I-7747. Accounting separation for undertakings in charge of SGEI is required by EU rules independently of the presence of State aid (Commission Directive 2006/111/EC of 16 November 2006 on the transparency of financial relations between Member States and public undertakings as well as on financial transparency within certain undertakings, 2006 OJ L318/17). Moreover, whenever the Altmark conditions are not met and aid exceeds the de minimis thresholds, stricter rules on accounting separation apply. Interestingly, for large amounts of compensation which have to be notified to the Commission and which will be assessed under the Framework, the Commission indicates a preference for a calculation of compensation based on an avoidable cost approach compared to a full-cost allocation approach. In other words, whenever possible, the costs of public service obligations should be calculated with reference to the long run costs which could have been avoided by the undertaking in the absence of public service obligations, following an approach based on incentives instead of an accounting approach.

Ginevra Bruzzone and Marco Boccaccio 

225  

italiane case.31 In its decision, the Authority had held that Poste Italiane infringed the prohibition of abuse of dominance through a unitary strategy based on different practices aimed at impeding the development of competition in the liberalized markets for guaranteed mail delivery by a specific date and time, and for notification via messengers. The practices included allegedly selective predatory pricing in the provision of PostaTime Service (delivery by a specific date and time) and allegedly predatory conduct in the bids for two contracts (with the Milan municipality and Equitalia). The review court quashed the decision on the ground that the Authority had not established to a sufficient legal standard that the practices were abusive. In particular, with reference to the application of prices below incremental costs in the provision of Posta Time service the court found that: some costs had been regarded as incremental costs without a proper justification; profit sacrifice had not been proved; and the competition authority had not taken sufficient account of the absence of an exclusionary impact (indeed, the competitor maintained a market share above 90% on the relevant market). As for alleged predation in the bids for specific public contracts, the court held that the use of the average national cost borne by the postal operator subject to universal service obligations as a cost benchmark for specific bids in areas which may present more favorable cost conditions was unjustified. Moreover, the court found that the economic importance of the bids considered in the decision was insufficient to entail significant foreclosure on the relevant market. More generally, the court emphasized that the decision was apparently based on the assumption that a system in which a postal operator subject to universal service obligations can compete on the markets for liberalized services necessarily results in distortions to competition. Although not innovative with reference to substantive criteria emerging from the EU case law, this judgment underscores the importance of an adequate standard of proof in cases of alleged exclusionary pricing conduct by multi-product dominant undertakings operating in post-liberalization markets. In another case of 2012, this one arising in the Netherlands (Sandd v Post NL), the Dutch Competition Authority took a decision under national competition law on alleged predation concerning a liberalized postal service.32 In determining whether PostNL had engaged in predatory pricing, the Authority compared the relevant prices with long-run average incremental costs. It took into account that PostNL had a statutory universal service obligation and that the volume of the mail flow resulting from the service was very low compared to PostNL’s total mail flow. Moreover, the service did not imply strict time constraints. The Authority considered it unlikely that if PostNL ceased to provide the service this would make it possible to significantly reduce the costs of the postal network: therefore, incremental costs were low. The prices applied for the service exceeded the estimated long-run average incremental costs and were not considered predatory. Interestingly, the Authority also took the view that the allocation of costs to 31 Autorità garante della concorrenza e del mercato, decision of 30 November 2011, TNT Post Italia/Poste italiane, Bulletin No 48/2011; Tar Lazio, judgment of 25 June 2012, No 5769. 32 Dutch Competition Authority, No 6207/476, Sandd v PostNL.

226  Competition, Regulation and Public Policies products or services in the different market segments is not in itself a form of cross-subsidization that falls within the prohibition of abuse of dominance, and that the antitrust rules were not, as such, an instrument to promote the liberalization of postal services and to create a level playing field on these markets. A further interesting case concerns public transport in France. In December 2012, the Paris Court of Appeal annulled a 2004 decision of the then-Conseil de la Concurrence which had dismissed a case of alleged predation by an undertaking charged with a public service obligation in transport services between the Ile d’Yeu and the European Continent in the Vendée region. The allegation was that the undertaking had abusively applied low prices to exclude competitors from the market for transport of passengers during the summer season.33 Making reference to the criteria set by EU case law and, in particular to Post Danmark, the French court argued that the way average incremental costs had been assessed in the Conseil’s decision was not based on sufficient evidence.34 It also stressed that, when prices are lower than the average total costs attributable to the competitive activity, the finding of an abuse does not require the proof of intent: it is sufficient to demonstrate that the conduct is able to exclude as efficient competitors with a negative impact on competition and, therefore, on consumers. As for Post Danmark, in 2013 the Danish Supreme Court annulled the decision of the Danish Competition Authority which was at the origin of the request of a preliminary ruling, on the ground that a materially incorrect test had been applied: the Authority had limited itself to arguing that selective price cutting constituted an exclusionary abuse.35 These examples clearly indicate that Post Danmark has paved the way for a more uniform and better articulated application of Article 102 in cases of alleged exclusionary prices by multi-product dominant undertakings in the presence of universal service obligations.

3. Refusal to deal and duty to invest Since its first applications, the duty to deal doctrine for dominant companies has been widely debated, and the importance of limiting principles has been pointed out. The Commission’s Guidance Paper recalls the main arguments in the section on refusal to supply: Any undertaking, whether dominant or not, should have the right to choose its trading parties and to dispose freely of its property. […] An obligation to supply, even for a 33 Cour d’Appel de Paris, judgment of 20 December 2012, 2011/05667, Vedettes Inter-Iles Vendéennes/Régie Départementale des Passages d’Eau de la Vendée. 34 The issue was whether the fixed costs of the fast boat Amporelle, which was used by the dominant undertaking both to discharge public service obligations and to operate on the competitive market, should be fully attributed to the public service, since the boat was necessary to provide it, or whether they were partly attributable to the competitive service. 35 See Bergqvist, cited note 22.

Ginevra Bruzzone and Marco Boccaccio 

227  

fair remuneration, may undermine undertakings’ incentives to invest and innovate and, thereby, possibly harm consumers […]. Also, competitors may be tempted to free ride on investments made by the dominant undertaking instead of investing themselves. Neither of these consequences would, in the long run, be in the interest of consumers.36

As for post-liberalization markets, the Commission acknowledges that, in sectors where the upstream position of the dominant undertaking has been developed under the protection of special or exclusive rights or has been financed by state resources, the negative impact of an obligation to supply on the incentives to invest may not be significant.37 This argument has often been referred to in the case law.38 However, it should not be employed mechanically. In particular, it should not be understood as indicating that, in post-liberalization markets, the application of Article 102 entails a generalized duty to deal for dominant companies. In the first place, the alleged absence of a negative impact on the incumbent’s incentives should be verified on a case-by-case basis taking into account, inter alia, the stage of development of the liberalization process and the motivations for the ‘refusal to deal’.39 Secondly, the impact on the incentives of third parties should also be considered, so as not to unduly distort the competitive process. Moreover, as recalled in the Guidance Paper, a refusal to deal should be considered an exclusionary abuse only when foreclosure is anticompetitive, ie when it is likely to result in harm to competition and, therefore, to consumers.40 In the last decade, a number of cases in which Article 102 was applied in post-liberalization network industries raised a new issue in the area of refusal to supply: namely, the issue of whether the prohibition of exclusionary abuses can be applied not only to the refusal to provide access to an existing facility but also to alleged ‘anticompetitive underinvestment’ by the dominant company. Concerns of strategic underinvestment rose, in particular, in the energy sector where there Guidance Paper, para 75. Ibid, para 82. In this paragraph the Commission also stresses that when EU law already imposes an obligation to supply, the necessary balancing of incentives may already have been made. On the relationship between regulation and the application of competition rules, see, eg, Mario Siragusa, ‘Gli obblighi di non discriminazione nella regolazione settoriale e nella disciplina antitrust’, in Giulio Napolitano and Andrea Zoppini, eds., Annuario di diritto dell’energia 2012. Il regime giuridico delle infrastrutture di energia, Il Mulino, 2012, pp 147 et seq. 38 See the Opinion of Advocate General Maduro in KPN Telecom, cited above note 2, para 41. See also Joined Cases C-544/03 and C-545/03 Mobistar and Belgacom Mobile [2005] ECR I-7723, para 49; Joined Cases C-327/03 and C-328/03 ISIS Multimedia and Firma O2 [2005] ECR I-8877, paras 45–46. 39 For instance, in his Opinion in TeliaSonera, Advocate General Mazák states at paragraph 27 that ‘it will not always be easy to say that the source of funds is unambiguously public in nature. Much of the infrastructure of former State monopolies has been the subject of significant improvements following privatisation with the result that sources of funding are now substantially mixed. Moreover […] the vertically integrated undertakings may also find themselves burdened with old infrastructure which requires maintenance and often one is dealing with industries with considerable technological innovation where the incumbent must innovate in order to compete.’ The Advocate General also stresses that, in the case he was considering, ‘there was no legal monopoly on internet access services in Sweden during the 1990s or 2000s’. 40 See, eg, Robert O’Donoghue and Jorge Padilla, The Law and Economics of Article 102 (Oxford University Press, 2013) p 510. 36 37

228  Competition, Regulation and Public Policies are undoubtedly still important challenges which must be met in order to create open and competitive European markets. Notably, the European Commission sector inquiry of 2007 concluded that: energy markets were often too highly concentrated and insufficiently liquid; a high degree of vertical integration, or rather insufficient unbundling of network and supplier activities, prevailed; and there was insufficient cross-border integration and cross-border competition between energy companies. The inquiry also highlighted a lack of transparency, for example with respect to available transport capacity, leading to lack of trust in pricing mechanisms.41 It is in this context that competition authorities undertook to enforce the competition rules, somehow stretching the application of Article 102 so as to complement actions taken in the sphere of regulatory policy and to promote competition and market openness in this sector.42 Article 102 was thus used as an instrument of a broader public policy strategy.43 In the Italian TTPC/ENI case, ENI had undertaken to increase the transport capacity of the Trans Tunisian natural gas pipeline connecting Algeria with Italy as a remedial action to comply with a prior finding of abuse of dominance in the gas sector (Blugas-Snam).44 TTPC, a company wholly controlled by ENI, had signed transport contracts with potential shippers. The entry into force of the contracts was conditional on certain events which did not occur within the contractual terms. In this context, the controlling company convinced TTPC to discontinue the project, allegedly to avoid an excess of supply in the Italian market. Although the undertaking proposed to the Italian Competition Authority alternative remedies to comply with the Blugas-Snam infringement decision, it was fined for non-compliance. More interestingly, the Authority opened a new investigation alleging that the decision to discontinue the planned expansion of the pipeline was a new, distinct abuse of dominance. Although it did not consider the pipeline to be an essential facility, the Authority stated that ENI had abusively interfered with the market strategy of its subsidiary by inducing it to cease the planned expansion of the pipeline as part of a strategy aimed at maintaining and strengthening the company’s dominant position in the wholesale market for the supply of gas in Italy. The fine of 290 million euros was subsequently reduced by the review courts, which held that there was insufficient evidence of any intent to exclude competitors.45 Separately, the European Commission launched the ENI foreclosure case ex officio due to 41 Philip Lowe, ‘Can EU competition policy create competition in the energy sector?’, The Beesley Lecture, London, 6 November 2008. 42 Autorità garante della concorrenza e del mercato, decision No 15174 of 15 February 2006, A358, ENI-Trans Tunisian Pipeline (ENI-TTPC), Bulletin No 5/2006. The decision was reviewed by TAR Lazio, 30 March 2007, No 2798 and Consiglio di Stato, 20 December 2010, No 9306. See also Commission Decision of 29 September 2010 in Case COMP/39.315 – ENI; Commission Decision of 3 December 2009 in Case COMP/39.316 – GDF gas foreclosure. 43 For a discussion, see, eg, Philip Lowe, cited above note 43; Alessandro Noce, ‘Antitrust e regolazione nelle decisioni con impegni in materia di energia’, 13(2) Mercato concorrenza regole 333 (2011). 44 Autorità garante della concorrenza e del mercato, decision No 11421 of 21 November 2001, A329, Blugas-Snam, Bulletin No 47/2002. 45 Decision A358, ENI-Trans Tunisian Pipeline, Bulletin No 5/2006; TAR Lazio, 30 March 2007, No 2798; Consiglio di Stato, 20 December 2010, No 9306.

Ginevra Bruzzone and Marco Boccaccio 

229  

concerns that ENI was: hoarding pipeline capacity that could have been profitably offered to third parties; making available capacity more difficult to purchase and less valuable to third parties by various means; and deliberately under-investing in transport capacity in three specific pipelines despite significant demand from third party shippers. The Commission considered the whole infrastructure for gas transport into Italy to be an essential facility, and it accepted a set of commitments based on the divestiture of all ENI’s shares in the three above-mentioned pipelines.46 In the GDF foreclosure case, the initial allegations of foreclosure of competitors from gas import in France included both long-term capacity reservation for most of the gas import capacity and underinvestment in liquefied natural gas terminals. However, the commitment by the dominant undertaking to release a large share of its long-term reservations of gas import capacity was considered by the Commission sufficient to address its concerns. A full analysis of these cases is beyond the scope of this chapter. We will focus only on one aspect, ie the allegation of strategic underinvestment as a possible abuse of dominance. The theory is clear: a vertically integrated undertaking may have an incentive to invest less in the infrastructure than a non-vertically integrated transport operator. When this risk is significant enough, it can justify a public policy measure imposing a structural separation. Indeed, several EU directives in network industries require at least some kind of vertical unbundling, although they allow undertakings to adopt solutions less radical than full ownership separation.47 46 Frank Maier-Rigaud, Federica Manca and Ulrich von Koppenfels, ‘Strategic underinvestment and gas network foreclosure – the ENI case’, (2011/1) Competition Policy Newsletter 18. 47 See, eg, Directive 2009/73/EC of the European Parliament and the Council of 13 July 2009, 2009 OJ L211/94. This Directive establishes common rules for the internal market in natural gas and underlines that any system of unbundling should be effective in removing any conflicts of interest between producers, suppliers and transmission system operators, in order to create incentives for the necessary investments and to guarantee access for new market entrants (Recitals 6, 8 and 9). The Directive contemplates three unbundling options for the transmission operator: full ownership unbundling; an Independent System Operator (ISO), with the network assets still belonging to the vertically integrated company; or an Independent Transmission Operator that owns the assets and belongs to the vertically integrated undertaking, with special rules to guarantee its independence in decision-making. The three unbundling options at the transmission level are also provided for by Directive 2009/72/EC of the European Parliament and the Council of 13 July 2009, 2009 OJ L211/55, on common rules for the internal market in electricity. For detailed discussion, see Dirk Buschle, ‘Unbundling of State-Owned Transmission System Operators – Effective Remedy or Eyewash?’, 1 European Networks Law And Regulation Quarterly 49 (2013). For electronic communications, Directive 2009/140/EC of the European Parliament and the Council of 25 November 2009, 2009/OJ L337/37, amended the ‘Access Directive’, ie Directive 2002/19/EC (which contemplated accounting separation for operators designated as having significant market power in relation to specified activities related to interconnection and/or access), adding provisions on mandatory and voluntary functional separation (ie the establishment of operationally separate business units). Mandatory functional separation for the wholesale provision of access products is considered an exceptional measure which can be adopted by the national regulatory authority only when obligations on transparency, non-discrimination and accounting separation have failed to achieve effective competition, and it is subject to the prior approval by the Commission. The Directive stresses that it is important to ensure that functional separation preserves the incentives of the undertaking to invest in its network (Recital 61 and Articles 13(a) and 13(b)). For rail transport, Directive 2012/34/EC of the European Parliament and the Council of 21 November 2012 establishing a single European railway area, 2012 OJ L343/32 (recasting the pre-

230  Competition, Regulation and Public Policies Regulatory policy choices on this issue may be complex and politically sensitive, but the assessment of the costs and benefits of the different alternatives does not raise special methodological issues. By contrast, the use of Article 102 in cases of alleged underinvestment does entail specific and very serious problems. The main reason is that investment decisions in a market economy depend on expectations and on the risk propensity of entrepreneurs. They therefore entail highly subjective elements. In the TTPC/ENI case described above, the benchmark for assessing whether the company had abusively underinvested was the originally planned expansion of the pipeline. The alleged reason for modifying the plan was a change in the company’s expectations with regard to the risk of an excess of supply in the Italian market, which led it to revise its investment choices. The Italian Competition Authority alleged that the change had to be viewed as part of a strategy aimed at anticompetitive foreclosure and argued that, if the excess of supply were to materialize, the increase in the transport capacity of the pipeline would not entail any risk of economic losses for ENI, since the undertaking could resort to alternative remedies. Interestingly, however, the defendant argued that the options suggested by the Authority in case of excess of supply implied meaningful legal risks for the undertaking. This example shows that, even in a case with a simple benchmark (the original investment plan), it may be difficult to prove to a sufficient legal standard that a change in investment decisions by the dominant company must necessarily be regarded, from an economic viewpoint, as part of an anticompetitive foreclosure strategy. Considering the abusive underinvestment doctrine from a broader perspective, the definition of the relevant investment benchmark becomes crucial. In the ENI foreclosure case, the Commission referred to the conduct which would have been adopted by a non-vertically integrated transport system operator. This approach recalls the market economy investor principle (MEIP) for assessing the existence of State aid, according to which the existence of aid, ie the existence of an artificial advantage for an undertaking can be excluded if public resources are used in a way which might also have been resorted to by a private operator in a market economy.48 The case law on State aid usefully suggests that reference should be made to the reasonable conduct of a private undertaking operating under normal market conditions, and not to an abstract, narrowly-defined profit maximizing operator.49 The application of the non-vertically integrated investor principle (‘N-VIIP’), like the application of the MEIP, raises issues of proof related to the evidence existing directives) requires legal, organizational and decision-making independence of infrastructure managers from railway transport undertakings, but only for essential functions. The proposal for a new directive contained in the fourth railway package extends the requirement of decision-making independence to all the functions of the infrastructure manager. 48 Commission Decision 2320/81/ECSC of 7 August 1981 on rules for aids to steel, 1981 OJ L228/14; Council Directive EEC/81/363 of 28 April 1981 on aid to shipbuilding, 1981 OJ L137/39; Case 234/84 Belgium v Commission [1986] ECR 2263, para 2; Case 40/85 Belgium v Commission [1986] ECR 2321, para 13. 49 See, eg, 1993 OJ C307/3.

Ginevra Bruzzone and Marco Boccaccio 

231  

which must be collected to demonstrate that the vertically integrated company has not invested in the same way as a hypothetical non-vertically integrated operator. For instance, in the ENI foreclosure case the Commission argued that some operators had expressed an interest in co-financing an expansion of the pipelines. However, no investor had expressed an interest in bearing the whole cost of an increase in the transport capacity of any of the three pipelines. Apart from the difficulties which may arise when trying to prove an alleged underinvestment with respect to a benchmark, the most serious problems are related to the legal consequences which should be drawn from such conduct. In the context of State aid control, if public resources are used in a way which does not satisfy the MEIP, their use may be qualified as aid, and compatibility with the common market under Article 107 TFEU will have to be ascertained. By contrast, an endorsement of the underinvestment doctrine in the application of Article 102 would imply that the conduct is prohibited, ie that the undertaking is infringing the prohibition of abuse of dominance and is liable to be fined. As is widely known, the case law on refusal to deal does not go beyond imposing a duty to grant access to an existing facility and clarifying that this duty cannot be avoided through inefficient organization that causes an artificial shortage of capacity, or if a third party is ready to fully finance the expansion of existing capacity.50 This is significantly different from a positive obligation to invest additional financial resources to expand infrastructures. Notably, the implicit assumption of a duty to invest on the dominant company would impinge on property rights and freedom of enterprise much more than the traditional essential facility doctrine. The use by competition authorities of soft enforcement tools (abstention from imposing fines, and/or the use of commitment decisions) is insufficient to avoid these problems. An infringement decision, even if not accompanied by fines,51 establishes a rule of conduct to which undertakings will be held, and therefore affects the future evolution of markets. In addition, such a decision may be followed by actions for damages. As for commitment decisions, several commentators have underscored the need for caution in respect of their precedential value.52 However, if strategic underinvestment is a source of concern for competition authorities leading them to adopt commitment decisions, any undertaking seeking to comply 50 See Pietro Merlino and Gianluca Faella, ‘L’obbligo di investire nello sviluppo di infrastrutture: note a margine di due storie parallele’, 14(1) Mercato concorrenza regole 117, 135 (2012), and the cases cited therein. For instance, in Commission Decision of 21 December 1993 – Port of Rødby, 1994 OJ L55/52, the refusal to allow competitors to build a new port in the immediate proximity of the Port of Rødby and to operate from the existing port facilities was found to be abusive. 51 According to the case law, when the criteria which establish the unlawful nature of the conduct are uncertain or have not been clarified previously, competition authorities should refrain from imposing fines. See, eg, AKZO, cited above note 13; Joined Cases 40–48, 50, 54–56, 111, 113, 114/73 Suiker Unie v Commission [1975] ECR 1663. The principle is recalled by the Commission in several decisions. See, eg, Commission Decision of 20 July 1999 in Case IV/26.888 – 1998 Football World Cup, 2000 OJ L5/55; Deutsche Post, cited above note 14; Commission Decision of 21 May 2003 in Case COMP/37.451 – Deutsche Telekom AG, 2003 OJ L263/9 (upheld on appeal – see above note 10); Commission Decision of 27 August 2003 in Case COMP/37.685 – GVG/FS, 2004 OJ L11/17. 52 See, eg, O’Donoghue and Padilla, cited above note 42, at 914–915.

232  Competition, Regulation and Public Policies with antitrust rules will necessarily have to consider how to avoid raising this kind of competition concern. Moreover, even commitment decisions may encourage follow-on damages actions, as they can be used as circumstantial evidence.53 The need for predictability with regard to the boundaries of unlawful conduct under Article 102 and the need to avoid excessive impingement on property rights even for dominant companies suggest that a systematic use of the prohibition of abuse of dominance to promote investment falling short of desirable targets would be both impractical and undesirable.54 Interestingly, the most recent EU measures aimed at promoting investment in network infrastructures in energy, transport and communication, within the Europe 2020 strategy, are based on more direct and less controversial policy instruments, including for instance the simplification of administrative proceedings. Sectoral regulation, which already provides detailed access obligations in a wide range of situations, often in association with forms of vertical unbundling,55 include, in some cases, rules that require undertakings to maintain and develop infrastructures. A prominent example is provided by Directive 2009/73/EC on the internal market for natural gas, contained in the Third Energy Package. Under the previous Directive 2003/55/EC, transmission system operators and distribution system operators had a generic duty to maintain and develop under economic conditions secure, reliable and efficient facilities, and Member States could require transmission system operators to comply with minimum requirements to that end.56 The 2009 Directive is more prescriptive. In particular, each transmission operator is required to build sufficient capacity to integrate European transmission infrastructure ‘accommodating all economically reasonable and technically feasible demands for capacity and taking into account security of gas supply. The regulatory authorities may require operators to comply with minimum standards for the maintenance and development of the transmission system, including interconnection capacity.’57 Moreover, in cases of vertical unbundling less clearcut than full ownership separation,58 the Directive sets detailed requirements on investment planning so as to ensure the long-term ability of the system to meet reasonable demand and to guarantee security of supply, and entrusts national regulatory authorities with monitoring and enforcement powers.59 53 See, eg, Alberto Pera and Giulia Codacci Pisanelli, ‘Decisioni con impegni e private enforcement del diritto antitrust’, 14(1) Mercato concorrenza regole 69 (2012). 54 For critical views on an application of Article 102 entailing a duty to invest for dominant companies, see Pietro Merlino and Gianluca Faella, cited above note 52; Ulrich Scholz and Stephan Pups, ‘The Application of EC Competition Law in the Energy Sector’, 1 Journal of European Competition Law and Practice 37 (2010); Malgorzata Sadowska, ‘Energy Liberalization in Antitrust Straightjacket: A Plant Too Far?’, 3 World Competition 449 (2011). 55 See above note 49. 56 Directive 2003/55/EC, Articles 8 and 12. 57 Directive 2009/73/EC, Article 13. 58 In the form of either an Independent System Operator (ISO) or an Independent Transmission Operator. See Directive 2009/73/EC, Recitals 13 and 16, and Articles 9(8), 14 and 17–23. 59 Ibid, Articles 13, 14, 17, 22 and 41. On regulatory holidays and their use in the energy sector, see Giorgio Monti, ‘Regulatory Holidays in Utilities Regulation and EU Competition Law: A Case Study on the Role of Efficiency Considerations in Economic Law’, this volume.

Ginevra Bruzzone and Marco Boccaccio 

233  

In telecommunications as well, the evolution of sectoral regulation shows an increasing emphasis on the incentives to invest in infrastructures. In particular, Directive 2009/140 requires national regulatory authorities to promote efficient investment and innovation in new and enhanced infrastructures, including by ensuring that any access obligation takes appropriate account of the risk incurred by the investing undertakings and by permitting various cooperative agreements between investors and parties seeking access to diversify the risk of investment, while ensuring that competition in the market and the principle of non-discrimination are preserved.60 More generally, specific requirements regarding investment duties can be contemplated in contractual relations between public authorities and undertakings, also in the form of public service obligations. As for financial incentives, due to the currently stringent public budget constraints the use of public funds to finance infrastructures is increasingly seen as a second-best solution which should adopted only in a limited set of situations and on the basis of clearly pre-defined criteria. For the use of EU funds, a notable example of this approach is given by Regulation 347/2013 on guidelines for trans-European energy infrastructure.61 For the use of State resources, the 2013 Commission guidelines on State aid to finance broadband infrastructure provide useful insights: in order to be compatible with the common market, State aid should be necessary and proportionate to correct a market failure or address a clearly defined social objective. Moreover, State aid should not crowd out, but on the contrary should promote private investment through an incentive effect.62 The efforts to create in the internal market an environment favourable to private long-term investment in infrastructures, including by way of publicprivate partnerships,63 can be seen as a corollary of the approach whereby public financing of infrastructures should be limited to cases where it is strictly justified from a policy perspective.

4. Concluding remarks Strong enforcement of Article 102 in post-liberalization markets is needed in order to ensure that dominant undertakings do not abusively hinder the development of competition and thereby make measures to deregulate or liberalize industrial sectors ‘of little value’.64 At the same time, the criteria for the application of 60 Directive 2009/140/EC amending the Framework, Access and Authorization Directives. See, in particular, the new paragraph 5(d) of Article 8 of Framework Directive 2002/21/EC. 61 Regulation (EU) No 347/2013 of the European Parliament and the Council of 17 April 2013 on guidelines for trans-European energy infrastructure, 2013 OJ L115/39. 62 Commission, EU Guidelines for the application of State aid rules in relation to the rapid deployment of broadband networks, 2013 OJ C25/01. 63 See, eg, High Level Expert Group on SME and Infrastructure Financing, Finance for Growth, 2013, http://europa.eu/efc/working_groups/hleg_report_2013.pdf. 64 See, eg, John Temple Lang, ‘Defining Legitimate Competition: Companies’ Duties to Supply Competitors and Access to Essential Facilities’, 18 Fordham International Law Journal 437, 483 (1994).

234  Competition, Regulation and Public Policies Article 102 must be defined in a way that does not unduly limit the right of the incumbent to compete. Two important and potentially controversial issues in the recent debate on the application of Article 102 concern, respectively, the conditions under which a multi-product dominant company subject to public service obligations can legitimately compete on prices in liberalized markets, and whether Article 102 implies, for a dominant company, the duty to invest in infrastructures which are needed to foster the development of competition. With regard to pricing policy, the ECJ judgment in Post Danmark paves the way for a more uniform and systematic approach by competition authorities within the European Competition Network. Importantly, the ECJ has endorsed the ‘as efficient competitor’ criterion and an approach based on real competitive impact. As for cost benchmarks, reference should normally be made to average avoidable costs and long run average incremental costs, as indicated by the Commission’s Guidance Paper of 2009. Attribution of a share of truly common costs, which would be borne anyway even if the undertaking were to stop the provision of the service, should be limited to exceptional circumstances based on clearly predefined criteria, in order not to unduly discourage price competition. As for incentives to invest, during the last decade the aim to promote investments in infrastructures has been pursued by public authorities in the EU through a wide range of public policy instruments, including the EU competition rules. In this context, Article 102 has been used in a small number of cases, through the allegation of abusive underinvestment by dominant companies in post-liberalization markets, to overcome bottlenecks which could foreclose competitors. However, the use of the prohibition of abuse of dominance to modify investment plans is an awkward task, heavily impinging on property rights, and it is doubtful that it can become a standard approach in the application of EU antitrust rules. The recent evolution of EU public policy for the development of infrastructures in the energy, telecommunications and transport sectors indicates an increasingly systematic approach to investment in infrastructures, with a focus on more direct and less controversial instruments. An example in sectoral regulation is provided by Directive 2009/73 on the internal market for natural gas, which establishes detailed rules on investment duties in cases of vertical unbundling falling short of full ownership separation. On the other hand, Directive 2009/140 on electronic communications stresses that access obligations and functional separation should preserve the incentive of the undertaking to invest in networks. The approach to financial incentives to invest in infrastructure is strongly affected by the modernization of the rules on State aid control and by the scarcity of public resources: public financing should be limited to a narrowly defined set of circumstances, whereas measures aimed at promoting long-term private investment, also in the form of public-private partnerships, play an increasingly important role in the internal market.

Calvin S Goldman, Julie Soloway, and David Dueck*

Competition Policy and Technology Markets: A Canadian Perspective The fundamental impulse that sets and keeps the capitalist engine in motion comes from the new consumers’ goods, the new methods of production or transportation, the new markets, the new forms of industrial organization that capitalist enterprise creates...[It] incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one. This process of Creative Destruction is the essential fact about capitalism.1

One key objective in the general application of competition law is finding a proper balance between intervening in markets when necessary while avoiding counterproductive and excessive interference. Markets are generally recognized as the best means of creating economic growth and prosperity. This balancing act becomes all the more challenging in markets which have rapidly changing technologies with significant innovation. A potential for sudden innovation requires that a long-term perspective be taken in areas where technology can rapidly change. Even if two technology firms were to accumulate a significant share of a market after merging, for example, this may have little bearing on the market power the merged entity would actually possess a year or two later if the pace of innovation in the industry is sufficient.2 Therefore, what may appear to be the best policy in the short run could be unnecessary or even counterproductive in the long run. Joseph Schumpeter referred to the ‘Creative Destruction’ of the market, whereby obsolete goods and services (often along with the companies providing them) are continually replaced by newer and better innovations.3 In particular, Schumpeter had in mind the technological revolutions we have seen in diverse industries like agriculture, manufacturing, power generation, and transportation.4 However, the pace of innovation we have recently seen in high technology industries since then * At the time of writing, Calvin S. Goldman was a Partner at Blake, Cassels & Graydon LLP, Toronto. He is currently a Partner at Goodmans LLP, Toronto. Julie Soloway is a Partner and David Dueck is a Student-at-Law at Blake, Cassels & Graydon LLP, Toronto. 1 Joseph Schumpeter, Capitalism, Socialism and Democracy (Routledge, 1942) pp 82–83, http:// books.google.com/books?id=6eM6YrMj46sC. 2 Ilene Knable Gotts, Scott Sher, and Michelle Lee, ‘Antitrust Merger Analysis in High-Technology Markets’, 4 European Competition Journal 463 (2008). 3 Schumpeter, Capitalism, Socialism and Democracy, cited above note 1, at 82–83. 4 Ibid.

262  Competition, Regulation and Public Policies has greatly exceeded even these remarkable advances. Moore’s Law predicts that the number of transistors per chip will double approximately every 18 months, and this astonishing rate of progress has so far been maintained for four and a half decades.5 Many companies that were once leaders in their technology markets are no longer in that position. For example, the PalmPilot, which used to be a leader, does not play any leading role today. The BlackBerry is also facing significant new challengers. A key factor to consider is the contestability of the market. Another is whether barriers protect dominance and the degree to which anticompetitive conduct has occurred or is likely to occur. Because there are many other relevant factors to consider, not just market share, caution is in order. Looking back at past cases where governments have intervened in rapidly changing industries illustrates the importance of keeping a long-term perspective. The concern surrounding IBM’s dominance of the punch cards used in early tabulating and computing machines is a classic example.6 This concern resulted in IBM signing an antitrust consent decree against the company in 1956 that was only recently rescinded.7 Then there were the antitrust proceedings brought because of the pricing of compact discs (CDs) in the music industry.8 More recently we saw a lengthy investigation in the US resulting in restrictive conditions placed on the merger between Time Warner and AOL, which was spurred by fears that the merger would lead to an ‘ability to exercise unilateral market power’.9 Although ‘AOL once defined the web for millions of people’, its provision of dial-up access was soon marginalized by the rise of broadband internet access, leaving both companies ‘in a weaker position’ than before the merger.10 In addition, just as the internet was rising in popularity,11 Canada saw antitrust proceedings brought against Tele-Direct for engaging in tied-selling and other anticompetitive acts with respect to advertising in its printed Yellow Pages telephone directory.12 Around the same time, Canada also saw antitrust proceedings 5 John Markoff, ‘Advances Offer Path to Further Shrink Computer Chips’, NY Times (30 August 2010), http://www.nytimes.com/2010/08/31/science/31compute.html?_r=1&hp. 6 International Business Machines Corp v United States, 298 US 131 (1936). 7 United States Department of Justice, ‘Justice Department Agrees to Terminate Last Provisions of IBM Consent Decree in Stages Ending 5 Years From Today’, Press Release of 2 July 1996, http:// www.justice.gov/atr/public/press_releases/1996/0715.htm. 8 In Re Compact Disc Minimum Advertised Price Antitrust Litigation, 216 FRD 197; 2003 US Dist LEXIS 11257 (Maine District Court, 2003). 9 United States of America Before Federal Trade Commission, ‘In the Matter of America Online, Inc and Time Warner Inc: Complaint’, Docket No C-3989, at para 25, http://www.ftc.gov/os/2000/12/ aolcomplaint.pdf. See also Federal Trade Commission, ‘FTC Approves AOL/Time Warner Merger with Conditions’, Press Release of 14 December 2000, http://www.ftc.gov/opa/2000/12/aol.shtm. 10 ‘Time Warner to Spin Off AOL, Ending Ill-Fated Deal’, US News (28 May 2009), http://money. usnews.com/money/business-economy/technology/articles/2009/05/28/time-warner-to-spin-off-aolending-ill-fated-deal. 11 KG Coffman and A M Odlyzko, ‘The Size and Growth Rate of the Internet’, AT&T Labs Research (2 October 1998), http://www.dtc.umn.edu/~odlyzko/doc/internet.size.pdf, at 15. 12 Canada (Competition Act, Director of Investigation and Research) v Tele-Direct (Publications) Inc, [1997] CCTD No 8, 73 CPR (3d) 1, paras 1, 18, 527–534, and 761–765.

Calvin S Goldman, Julie Soloway, and David Dueck  263   brought against Warner Music Canada for an alleged refusal to deal in the nowarchaic ‘mail-order record club business’.13 Such potential for rapid technological change was explicitly taken into account by the Canadian Competition Bureau (the ‘Bureau’) when it allowed the acquisition of Microcell Telecommunications by Rogers Wireless in 2004: The role of change and innovation had an important impact on the Bureau’s conclusions in this matter. The rate of growth in the mobile telecommunications market in the next six to seven years is expected to be significant [...]. Advances in mobile handset technology are rapidly bringing newer and more advanced services to market [...]. Change and innovation will, in the Bureau’s view, continue to play an important, positive role in the future evolution of competition in this market.14 The Bureau also noted in its decision that markets with “rapid and frequent product or service innovations are less conducive to coordinated behaviour.’15 Likewise, this potential for rapid technological change was taken into account by the US Department of Justice (DOJ) in its decision not to challenge a merger between XM Satellite Radio and Sirius Satellite Radio. Although the merger was criticized by some as a ‘merger to monopoly’, XM and Sirius successfully argued that both actual and imminent potential substitutes would prevent them from having actual market power.16 The DOJ ultimately found that the parties would be unable to profitably increase prices based in part on ‘technological change that is expected to make [alternative services] increasingly attractive to consumers over time’.17 A number of technology platforms under development were seen as likely to offer ‘new or improved alternatives to satellite radio’.18 Former Canadian Commissioner of Competition, Melanie Aitken has emphasized the importance of encouraging innovation,19 expressing the view that the Canadian Competition Bureau’s mission is to ensure that Canada has ‘markets where efficiencies and innovation are fostered’.20 In the midst of such innovation, George Addy, formerly the head of the Competition Bureau, warned in the 1990s that: Rapidly changing technologies together with global competition and the risk and uncertainty associated with the development of the information economy are undermining 13 Canada (Competition Act, Director of Investigation and Research) v Warner Music Canada Ltd, [1997] CCTD No 53, 78 CPR (3d) 321, para 10. 14 Competition Bureau, ‘Acquisition of Microcell Telecommunications Inc by Rogers Wireless Communications Inc’, Technical Background (April 2005), http://www.competitionbureau.gc.ca/eic/ site/cb-bc.nsf/eng/00257.html. 15 Ibid. 16 Gotts, Sher and Lee, ‘Antitrust Merger Analysis in High-Technology Markets’, cited above note 2, at 471. 17 US Department of Justice, ‘Statement of the Department of Justice Antitrust Division on its Decision to Close its Investigation of XM Satellite Radio Holdings Inc’s Merger with Sirius Satellite Radio Inc’, Press Release of 24 March 2008, http://www.usdoj.gov/atr/public/press_ releases/2008/231467.htm. 18 Ibid. 19 Melanie Aitken, ‘Remarks to the Economic Club of Canada’, speech, 4 May 2010. 20 Melanie Aitken, ‘The Senate Banking, Trade and Commerce Committee Hearings Regarding Competition Act Amendments’, speech, 13 May 2009.

264  Competition, Regulation and Public Policies the need for the economic regulation and the ability of regulators to regulate. Regulators or policy makers must think very carefully before implementing regulatory policies for the Information Highway because the cost associated with limiting choice can be great.21

Under Canadian competition law, there are a number of ways in which future technological changes in an industry can be brought into consideration. Section 93(g) of the Competition Act states that in determining whether a merger is likely to lessen or prevent competition, the Competition Tribunal hearing the application may take into account ‘the nature and extent of change and innovation in a relevant market’.22 It also may take into account the extent to which acceptable substitutes for the products supplied by the parties ‘are or are likely’ to be available.23 Canadian competition law also provides an efficiencies defence for mergers under section 96 of the Competition Act. Where the ‘gains in efficiency’ from the merger ‘will be greater than, and will offset’ the effects of any prevention or lessening of competition, section 96 prevents the Tribunal from interfering.24 According to the Bureau’s Merger Enforcement Guidelines, allocative efficiency, productive efficiency, and dynamic efficiency are all relevant to the analysis.25 When this provision was interpreted in Superior Propane, a ‘balancing weights approach’ to the trade-off was ultimately adopted.26 This approach takes into account a number of factors in assessing the potential anticompetitive effects of the merger, including any ‘loss of potential dynamic efficiency gains’.27 Although allocative, productive, and dynamic efficiencies are all relevant, the weight placed upon each type of efficiency will vary under the circumstances and must be determined by the Tribunal. It has been noted that when transactions involve high-technology industries, the significant rates of innovation and change mean that dynamic efficiencies are likely to be more relevant.28 As a result, the Bureau has indicated that it will place a high degree of importance on dynamic efficiencies in these circumstances.29 In fact, the major reason that intellectual property rights exist and purposefully create market power in the first place is because of the beneficial long-term consequences that this can have on dynamic efficiency.30 21 George Addy, ‘Appearance before the Canadian Radio-Television and Telecommunications Commission’, speech, 16 March 1995. 22 Competition Act, RSC 1985, c 19 (2nd Supp), s 93(g). 23 Ibid, s 93(c). 24 Competition Act, cited above note 22, at s 96(1). 25 Competition Bureau, Merger Enforcement Guidelines (11 January 2012), http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/vwapj/cb-meg-2011-e.pdf/$FILE/cb-meg-2011-e.pdf, para 12.4. 26 Canada (Commissioner of Competition) v Superior Propane Inc, [2003] FCJ No 151, para 20. 27 Ibid, para 21. 28 See Calvin S. Goldman, Richard F.D. Corley and Crystal L. Witterick, ‘A Canadian Perspective on Intellectual Property Rights and Competition Policy: Striving for Balance and Related Comity Considerations’, paper presented at the 31st Annual Conference on International Antitrust Law & Policy, Fordham Corporate Law Institute, October 2004. 29 Competition Bureau, Intellectual Property Enforcement Guidelines (21 September 2000), http:// www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/vwapj/ipege.pdf/$FILE/ipege.pdf, para 5.4. 30 Ibid.

Calvin S Goldman, Julie Soloway, and David Dueck  265   Therefore, in the context of rapid technological changes, the approach taken by competition authorities towards IPRs may also be of key importance, and this makes it necessary to also find a proper balance in this regard. Consumers benefit from competition, which leads to the development of new products or more efficient production methods. Of course, the protections offered by intellectual property laws also provide incentives to innovate. Recognizing this, Canada’s Competition Act has taken into account specific considerations relating to IPRs, and the Canadian Competition Bureau has issued Intellectual Property Enforcement Guidelines that provide its views as to the general manner in which conduct involving IPRs will be assessed under Canada’s competition laws. Market power is generally assessed on the basis of direct and indirect evidence. However, in the context of technology markets, the traditional indicators of market power in final products may not be relevant; in fact, these factors may distort the assessment.31 Therefore, the Bureau’s assessment of market power in the case of IP is likely to focus more on qualitative factors, including ‘conditions of entry into the relevant market, whether IP development is resulting in a rapid pace of technological change, the views of buyers and market participants, and industry and technology experts’.32 In addition, the IP Enforcement Guidelines expressly recognize that mere ownership of IPRs is not necessarily synonymous with possession of market power.33 Nor will a high degree of concentration necessarily create market power concerns, especially in industries with ‘low barriers to entry, a high rate of technological change and a pattern of firms “innovating around” or “leap-frogging over” technologies that had previously controlled a dominant share of a market’.34 However, it is also critical to recognize that high technology does not preclude high barriers to entry, nor does it preclude conduct that raises and entrenches such barriers. Technology encompasses both physical infrastructure and intangibles, such as software, which are subject to quite different constraints. While physical infrastructure can have high marginal costs and require substantial investments for both the acquisition and maintenance of such assets, intangibles such as software have a marginal cost that is close to zero and are generally unlimited in volume. For example, services which require substantial physical infrastructure (such as the data centres and hundreds of thousands of servers required to index the entire Internet for searching) may in fact be protected by a high barrier to entry. The combination of high barriers to entry with high market shares creates threats to competition in such markets which are much more serious than would be the case absence the requirement for costly physical infrastructure. This is especially so if the dominant entity is entrenching its position by its anticompetitive conduct such as exclusionary acts. 31 For example, factors such as high profit margins, excessive prices and excess capacity would not necessarily assist in defining a technology market. See Richard Posner, ‘Antitrust in the New Economy’, 68 Antitrust Law Journal 925, 926–927 (2001). 32 Ibid. 33 IP Enforcement Guidelines, cited above note 29, para 4.1. 34 Ibid, para 5.2.1.

266  Competition, Regulation and Public Policies Under section 79(4) of the Competition Act, conduct will not be considered anticompetitive merely because of a dominant firm’s superior competitive performance.35 This is an important Canadian statutory qualification, the breadth of which is still to be judicially determined. To this end, the IP Enforcement Guidelines state that ‘a firm would not contravene the Competition Act if it attains its market power solely by possessing a superior product or process, introducing an innovative business practice or other reasons for exceptional performance’.36 In addition, section 79(5) of the Competition Act, dealing with abuse of dominance, has the following exception: ‘an act engaged in pursuant only to the exercise of any right or enjoyment of any interest derived under the Copyright Act, Industrial Design Act, Integrated Circuit Topography Act, Patent Act, Trademarks Act or any other Act of Parliament pertaining to intellectual or industrial property is not an anti-competitive act’.37 There must be more than a mere refusal to license; there has to be further anticompetitive conduct. Consequently, the Competition Tribunal ruled in Tele-Direct that the decision whether or not to license intellectual property in a trademark lies solely within the prerogative of its owner.38 Therefore, a refusal to license cannot be an anticompetitive act.39 The Tribunal also held in Warner that licenses of intellectual property are not a ‘product’ for the purpose of an alleged refusal do deal.40 However, section 79(5) is balanced by an important counterweight found in section 32 of the Competition Act. This section applies where a dominant firm ‘unduly’ limits the supply of IP, and it gives the courts the power to restrain the use of the IP, or even declare it altogether void, where circumstances warrant.41 According to the IP Enforcement Guidelines, it is only in certain ‘very rare’ and ‘narrowly defined circumstances’ that the Bureau would make a recommendation to the Attorney General for a section 32 application.42 Such circumstances would require that a dominant firm refuse to license IP that is an essential input in the market where the firm is dominant, with the result that further investment in innovation is stifled rather than promoted.43 In sum, a delicate balancing act between non-interference and regulation or intervention must be maintained when dealing with industries undergoing rapid technological changes. For this reason, Canadian competition law acknowledges that superior competitive performance is not of itself inherently anticompetitive, particularly in areas of rapid technological change involving intellectual property. It also recognizes that a high degree of concentration does not necessarily create market power concerns in industries with low barriers to entry and high levels Competition Act, cited above note 22, s 79(4). IP Enforcement Guidelines, cited above note 29, para 2.2. 37 Competition Act, s 79(5). 38 Tele-Direct, cited above note 12, para 66. 39 Ibid, paras 59 and 66–68. 40 Warner, cited above note 13, para 30. 41 Competition Act, s 32. 42 IP Enforcement Guidelines, cited above note 29, para 4.2.2. 43 Ibid. 35 36

Calvin S Goldman, Julie Soloway, and David Dueck  267   of innovation and change. After all, dominance in any high-technology industry can be temporary unless the dominant entity has erected barriers or taken steps to thereby entrench its dominance. Therefore, it is important to always be looking towards the future in competition policy, because what may appear to be the best policy in the short run may, in some cases, be counterproductive in the long run.

Paul Lugard*

Innovation in the Domain of EU Competition Law Enforcement: Blowing in the Wind?

Introduction It is undisputed that innovation is a key driver for economic growth and societal welfare. New products and services and improved production methods may on balance produce more consumer benefits than price decreases of existing products. However, competition law and policy target, first and foremost, cartels, market allocation and other static market inefficiencies. Against this background, it is not surprising that at regular intervals the question is raised whether competition policy should not be drastically reoriented and expressly focus on stimulating innovation, or, more correctly, on combatting practices that limit or endanger innovation.1 In a similar vein, it is sometimes argued that a disproportionate emphasis on price competition may have a negative impact on innovation. This chapter does not seek to formulate a definite answer to the question of how competition law can best be applied at a macroeconomic level to bring about the largest consumer gains. However, it does attempt to touch on a few trends and peculiarities in relation to the treatment under competition law of business transactions involving innovation. Underlying this discussion are the questions of (i) whether under EU competition law a comprehensive vision on the nature and the importance of innovation has been developed, and (ii) which analytical framework, if any, applies to protect innovative business activities. In the discussion of innovation and competition law, the relationship and interaction between competition law and intellectual property rights obviously play an important role. The traditional view on the relationship between these bodies of law is based on the thought that both areas are complementary. Intellectual property law aims to incentivize companies to innovate by establishing a prospective temporary legal monopoly – especially in the form of patents – which enables companies to commercialize their inventions and to recuperate * Partner, Baker Botts LLP, Brussels. 1 Among the many contributions, see, eg, Herbert Hovenkamp, ‘Signposts of Anticompetitive Exclusion: Restraints on Innovation and Economies of Scale’, in Barry Hawk (eds), International Antitrust Law and Policy 2006: Fordham Competition Law (Juris Publishing, 2007) chapter 18, also noting that ‘[t]he often underappreciated corollary of Schumpeter’s and Salow’s work is that, because innovation is such an important engine of economic growth, restraints on innovation very like cause considerable greater social harm, than restraints on price. Such restraints may be more harmful even than cartels, but they are almost certainly more harmful than the price effects of exclusionary practices.’

270  Competition, Regulation and Public Policies their investments. In this view, competition law seeks to safeguard competition on product and service markets, which subsequently motivates companies to invest in innovation. Much could be said about the supposed linear relationship between competition and innovation, and the way in which competition supports innovation.2 But leaving aside that debate, the protection of intellectual property rights unquestionably plays an important role in the promotion of innovation. In this context, it can thus be argued that competition law should also serve to safeguard the system of intellectual property rights against attacks which may endanger that very system of legal protection. In short, from this perspective, competition law not only plays a role in directly stimulating innovative activities but also serves to safeguard the integrity of the system of intellectual property rights as a carrier or pillar of innovation. The Commission’s action against the attempts by AstraZeneca to extend the duration of patent protection for its products by providing false information to patent agencies fits squarely within this heading.3 Intellectual property rights are however by no means the only pillars of innovation. Technology platforms constitute a second pillar of innovative activities. In many high-tech sectors, technology platforms are often of great importance for innovation. Aside from the fact that competition in those sectors often manifests itself as competition between different platforms, (for example between the Android, Apple and Windows platforms for smartphones), technology platforms also lower entry barriers and costs for many market participants.4 For instance, a developer of apps does not have to develop and market an operating system or smartphone itself, but can simply offer the developed application to Google, Apple or Microsoft and can, via the respective platform, reach a large number of users. The existence of and access to such platforms are therefore important conditions for developing innovative activities. Competition law can play a role in the preservation of those platforms and access to them.5 Finally, technical standards form a third innovation pillar. Many technological markets are characterised by the existence of technical standards. The purpose of such standards is to set technical requirements with which current or future products may comply. Technical standards often aim at ensuring interoperability and compatibility between different products that are part of the same system. For example, the ‘USB’ (universal serial bus) standard sets the technical specification 2 For a brief overview of the discussion on the relationship between market structure and innovation, see Paul Lugard, ‘Mededingingsrechtelijke Known Unknowns. Of toch Unknown Unknowns?’, Markt & Mededinging, 2012(5). Empirical research suggests an inverted U relationship between market concentration and the ratio of industry R&D to industry sales. See Philippe Aghion et al, ‘Competition and Innovation: An Inverted-U Relationship’, 120 Quarterly Journal of Economics 701 (2005). 3 See Commission Decision of 15 June 2005 in Case COMP/A.37.507 – AstraZeneca. 4 Platform technologies are products or services that form a basis for the offering of complementary products such as software, films or music services. See, eg, Fiona Scott-Morton, ‘Antitrust Enforcement in High Technology Industries: Protecting Innovation and Competition’, speech, New York, 7 December 2012, www.justice.gov/atr/public/speeches/290876.pdf. 5 See Commission Decision of 24 March 2004 in Case COMP/37.792 – Microsoft; and Cases COMP/39.740, 39.775 and 39.768 – Google.

Paul Lugard 

271  

based upon which peripheral devices can be connected easily to computer equipment. Like platforms, technical standards may result in a decrease of production and sales costs, and may therefore reduce entry barriers.6 Technical standards are set either by companies themselves or by a number of market participants together, often (but not always) in the context of standardsetting organisations (SSOs), such as ETSI and the ITU.7 An example of a joint standard is the DVD (Digital Versatile Disc) standard established by ETSI, in which Philips, Sony, Pioneer and other companies had cooperated. In contrast, Apple’s FaceTime application for example was partially based on private Apple standards. One feature associated with the development and setting of technical standards by SSOs is the fact that competition between alternative technologies for inclusion in the standard takes place while the standard is being developed. Once a certain technology has been chosen and the standard has been set, competing technologies may potentially be excluded from the market. Obviously, this risk only exists if the standard actually becomes a success and is accepted by the market. However, there is no certainty in advance: technical standards and standardised technologies and products often compete with other standards, and in many cases standards are not successful.8 A well-known example of ex ante competition between standards was the ‘battle’ between HD DVD and Blu-ray in the years 2006–2008, which was ultimately won by Blu-ray. This was mainly due to the support of large Hollywood movie studios, which believed that Blu-ray offered better picture and sound quality, and better content protection. It is important to appreciate that many technical standards involve intellectual property rights. This is not surprising, given the fact that the technologies suggested by companies for inclusion in a standard are often protected by such rights. Once the respective technical solution, protected by intellectual property, is included in a standard, the patents or other rights are considered to be ‘essential’: a manufacturer needs access to all essential patents, often in the form of a (cross-) licence in order to manufacture and market a product that complies with the respective technical standard. A licence under essential patents is of course especially important if the standard is widely accepted in the market. For example, a company that manufactures smartphones has no other choice but to obtain a license under the patents of Samsung, Apple, Huawei, Qualcomm, and other holders of essential UMTS patents. If standards are considered an important pillar of innovation, it is not surprising that attempts are made to apply competition law first to safeguard access to standardsetting discussions and, second, to ensure that the fruits of such standardisation discussions – the essential technology protected by intellectual property rights 6 See Tim Wu, ‘Taking Innovation Seriously: Antitrust Enforcement. If Innovation Mattered Most’, 78 Antitrust Law Journal 313 (2012). 7 The two main standard-setting organisations in the telecommunications sector are the European Telecommunications Standard Institute (ETSI) and the International Telecommunication Union (ITU). 8 An example of a standard that did not become a success is DAT (Digital Audio Tape), which was launched as the successor of the music cassette tape. See also OECD Competition Committee, Background Note on Standard Setting (1 June 2010), DAF/COMP/WP2(2010)4.

272  Competition, Regulation and Public Policies – are accessible for users of those technologies – under fair, reasonable and nondiscriminatory (FRAND) terms. The three pillars of innovation touched upon above, ie intellectual property rights, platforms and technical standards, and the nature of the intervention of competition authorities in markets where these phenomena occur, are the central theme of this chapter. In particular, it appears that the European Commission is increasingly relying on competition law to correct certain effects of intellectual property protection that it deems undesirable. This trend is visible in individual investigations under Article 102 TFEU, such as in the Microsoft and Samsung cases in the high-tech sector and the (Article 101 TFEU) ‘pay for delay’ cases in the pharmaceutical sector, especially Lundbeck and Les Laboratoires Servier. The trend can also be seen in the Commission’s recently proposed changes to the regulatory framework for technology transfer agreements. The Commission’s policy with regard to technical standards is particularly evident in the Qualcomm and Rambus cases, as well as in its investigations into the use of essential patents by Samsung and Motorola. The approach the Commission has taken until now in those cases is partly reflected in the section on standardisation agreements in the EU Guidelines on horizontal cooperation agreements of 2011. Before addressing these trends in some further detail, we briefly look at the phenomenon of innovation and then at a number of developments, particularly in the high-tech and pharmaceutical sectors, which seem to co-determine the policy of the Commission. Thereafter the discussion turns to a number of innovation-related merger cases, and finally to a number of Article 101 and Article 102 matters.

The innovation phenomenon Innovation is a multifaceted concept that is difficult to circumscribe adequately. Considering the importance that the Commission attaches to innovation and especially the essential role of innovation in the Europe 2020 Agenda, one would expect that over time a clear framework would have developed under EU competition law regarding the nature and the importance of innovation and the way in which innovation should be incorporated in the competitive assessment of business transactions. One would also expect such framework to apply both when innovation results in efficiencies and when transactions may have innovationlimiting effects. However, nothing could be further from the truth. The absence of such a general analytical framework is even more surprising if one realizes that in recent years the Commission has taken enforcement actions in a number of Article 102 cases which specifically concerned innovation in the relevant sector: Qualcomm, Intel, Microsoft, Google, and more recently, Samsung and Motorola. The lack of a clearly articulated analytical framework is due to a number of factors. First, it is difficult to predict whether innovation will occur in a specific

Paul Lugard 

273  

market or transaction context. Second, the conditions for recognition of dynamic efficiencies under Article 101(3) and under the EU Merger Regulation are strict. And third, the perception of the Commission is at least in part shaped by the specific cases with which it is confronted and must deal with in a reactive manner, either summarily or in a more elaborate fashion. These cases mainly consist of merger filings and complaints under Article 102. Many of these cases do not require an in-depth analysis of the specific innovation efforts of the companies concerned. This may limit the Commission’s ability to develop a generally applicable approach. In addition, it seems plausible that, as part of this reactive approach, the Commission tends to perceive innovation as originating from within companies and thus emphasises existing business models and ‘value networks’, ie the current ecosystems in which companies operate. This somewhat limited vision is inconsistent with views developed in economic literature which do not exclusively concentrate on individual companies as the originators of innovation, but also attach importance to the factors linked to the commercial and economic environment that form the basis of innovation.9 They point out that, in some cases, innovation builds on earlier inventions (cumulative innovation),10 while in other cases innovation results from interaction between companies and users of services and products (user-driven innovation)11 or is a result of cooperation with other organisations (open innovation).12 The difference between ‘sustaining’ innovation on the one hand, and ‘breakthrough’, ‘radical’ and ‘disruptive’ innovation on the other hand, is equally relevant. Sustaining innovation does not generate new markets or ‘value networks’ but is aimed at enhancing existing products and services. An example of such innovation is the expansion of storage capacity and the speed of micro-processors or the transition from DVD technology to Blu-ray. In many cases, innovation that can be considered ‘sustaining’ is also ‘incremental’.13 In contrast, ‘breakthrough’ or ‘radical’ innovation has major consequences for existing markets. Examples of this type of innovation include the development of electric watches and the use of bio fuels. If this type of innovation leads to the elimination or the disruption of existing markets and the creation of new markets, it is generally referred to as ‘disruptive’ innovation. For example, for many applications, smartphones have replaced the personal computer digital photography has made the market for traditional photography disappear; light emitting diodes (LEDs) are replacing regular light bulbs; and USB flash drives are becoming obsolete because of cloud computing. It is striking that the most prominent Article 102 cases, Microsoft, Intel, Samsung and Google, related to existing platforms and value networks of the relevant 9 See Vijay Govindarajan and Chris Trimble, ‘Organizational DNA for Strategic Innovation’, 47 California Management Review 47 (2005). 10 See Suzanne Scotchmer, Innovation and Incentives (MIT Press, 2004). 11 See Eric Von Hippel, The Sources of Innovation (Oxford University Press, 1988). 12 See also Roger Miller and Marcel Côté, ‘The Faces of Innovation’ (2008), http://ssrn.com/ abstract=1106740. 13 Incremental innovation is not necessarily less important than breakthrough innovation given the possibility that incremental innovation may occur more often.

274  Competition, Regulation and Public Policies dominant companies and concerned, in particular, ‘sustaining’ innovation. In the Microsoft case, the Commission attempted to safeguard competition from Sun and Novell (in the area of operation systems) and from RealNetworks (in the market of media players) while in Intel the Commission aimed to secure competition between Intel and AMD in the market for CPUs for personal computers.14 At the same time, it is clear that the arrival of disruptive innovation in the form of the internet and Google’s internet search technology has significantly reduced the significance and importance of Microsoft’s operating system.15 The same applies to Intel’s market position on the market for CPUs for desktops: trends in the use of tablets and smartphones have created room for CPU manufacturers such as ARM and Qualcomm. It cannot be excluded that the same trend will occur in the Google case, where the Commission is attempting to maintain competition between search engines but where the future could very well lie in specific applications for mobile devices or social networks such as Facebook. Furthermore, the investigation of the Commission in the Samsung case focuses on the use of standard essential patents for the fourth generation of mobile telephony, but that technology will not continue to exist forever. It is not surprising that the Commission focuses on existing platforms and seems to have less affinity with disruptive innovation. After all, disruptive innovation is by definition hard to predict and therefore difficult to factor into an analysis of competitive effects. Moreover, many forms of disruptive innovation – in contrast to incremental and follow-on innovation, which often come to fruition as a result of the existence of an existing platform such as Google’s Android OS or Microsoft’s operating system – seem by their nature less likely to give cause for exclusionary abuse under Article 102. Nevertheless, it is noteworthy that the Commission considers incremental innovation to be an independent phenomenon worthy of protection under competition law, even if such incremental innovation builds on and is dependent on earlier breakthrough (and disruptive) innovation.16

Trends In sectors where innovation is important there are a number of trends that seem to partially explain the increasing intervention of competition authorities. The summary below mainly looks at the ICT and pharmaceutical sectors. 14 See Microsoft, cited above note 5; and Commission Decision of 13 May in Case COMP/37.990 – Intel. 15 See, eg, www.becker-posner-blog.com/2013/09/dynamic-competition-and-anti-trust-policybecker.html. See also Case T-79/12, Cisco Systems Inc., Messagenet SPA v Commission and Microsoft Corp., judgment of the General Court of 11 December 2013, not yet reported. 16 In the Commission’s decisional practice, one will not find any cases in which a balancing takes places between the importance of innovation of the original innovator and incremental innovation that builds thereupon, or a discussion of whether or not incremental innovation will occur at the expense of the original innovation. In this regard, see Pierre Larouche, ‘The European Microsoft Case at the Crossroads of Competition Policy and Innovation’, 75 Antitrust Law Journal 933 (2008).

Paul Lugard 

275  

Firstly, especially in the ITC sector, in the past twenty years there has been a significant increase of patent applications and patents actually granted. This trend manifests itself in Europe and the United States, but especially also in Asia. For example, the number of patent applications in China increased from approximately 50,000 in 2000 to 250,000 in 2007, while the number of European applications increased from 150,000 in 2000 to 250,000 in 2012.17 This growth is fuelled not only by changes in patent legislation but especially by the economic growth in Asia and the globalisation of the world economy, as a result of which companies in more jurisdictions are seeking to obtain patent protection. In this respect is also important that the backlog in the processing of patent applications has increased significantly, creating greater uncertainty regarding the validity of patents.18 Secondly, products are becoming more technically complicated, and more patent applications per product are being submitted.19 This trend can be illustrated by the number of successive generations of products and the number of essential patents involved in the manufacture of these subsequent products. For example, in the mobile telephony sector, approximately 100 essential ‘first generation’ GSM patents initially existed, but that number rose to 1,000 3G/UMTS patents and over 1,500 4G/LTE essential patents. The number of essential CD patents was approximately 100, while for DVD technology that number was approximately 400, and 2,000 for Blu-ray. In sum, nowadays in the ICT sector a much larger number of essential patents is in the hands of a larger number of (new) patent holders with varying strategic considerations. On a price-per-patent basis, patents in this sector have on the one hand lost value, but they also function increasingly as strategic negotiation tools, or ‘bargaining chips’. Moreover, companies are submitting an increasing number of patent applications to maintain their negotiating position. In many ICT sectors, these trends have led to patent clusters or ‘patent thickets’, a tangle or web of patents through which licence holders must find a path in order to manufacture a product that meets a certain technical standard.20 The number of patent applications is also rising in the pharmaceutical sector, and patent thickets are occurring more frequently. The number of European patent applications increased by 40% between 2000 and 2007. The number of patents 17 See European Patent Office, Fact and Figures 2013, www.epo.org/ service-support/publications/ general-information/facts-figures/2013.html. Incidentally, approximately sixty percent of the increase of patent applications is accounted for by Chinese, Japanese and Korean companies. Samsung, Sony, LG, Panasonic, Honda, Mitsubishi and Toyota are high in the top ten of European patent applications. 18 The U.S. Patent and Trademark Office estimated that in 2008 there were more applications than granted patents. See also Intellectual Property Office, Patent backlogs, Inventories and pendency: an international framework, www.ipo.gov.uk/lpresearch-uspatlog-201306.pdf. 19 It can also be assumed that more technical standards are being developed, that standardisation activities are more fragmented and that more parties are participating in standardisation discussions. 20 See Pierre Régibeau and Katharine Rockett, ‘Assessment of potential anticompetitive conduct in the field of intellectual property rights and assessment of the interplay between competition policy and IPR protection’, Report for the European Commission (November 2011), http://ec.europa.eu/ competition/consultations/2012_technology_transfer/study_ipr_en.pdf.

276  Competition, Regulation and Public Policies and patent applications for medicines in Europe is estimated at 40,000.21 In this sector there is also an increase of voluntary divisionals of patent applications, as a result of which the duration of the investigation by patent agencies is being extended and publication of the patent is being postponed, which may be part of a strategy to restrict competition.22 In its conclusions of 2009 after it had finished its pharmaceutical sector investigation, the Commission observed, more in general, that market access of generic medications may be delayed or blocked in a number of ways, which may lead to less competition and higher prices. In this context, the Commission was particularly concerned about a significant number of settlement agreements between originator and generic companies. Such settlement agreements are often a result of patent disputes and they frequently limit the possibilities of generic companies to bring their medications to the market. In many cases, such settlement agreements are accompanied by a payment or other value transfer by the originator to the generic company. These ‘reverse payment’ arrangements are being attacked by the Commission.23 The picture that emerges is that there are now more possibilities to use intellectual property rights strategically than there were twenty years ago, and this phenomenon has increased the potential to restrict competition. This situation has subsequently given rise to discussions regarding necessary amendments to intellectual property law so that patents cannot be obtained as easily, or to limit the duration of patent protection. So far, however, those discussions have not led to drastic reforms of intellectual property law. The remainder of this chapter addresses how the Commission seeks to apply European competition law against the (inappropriate) use by market participants of their intellectual property rights. Before doing this, a few comments are made regarding the way in which innovation plays a role in concentration control.

Innovation in merger control Just like in cases under Articles 101 and 102, innovation-related considerations in concentration cases may impact the competition law analysis under merger control law in a number of ways. In broad terms the following scenarios can be distinguished. 21 See Pharmaceutical Sector Inquiry Final Report, http://ec.europe.eu/competition/sectors/ pharmaceuticals/inquiry/staff_working_paper_part1.pdf, p. 161. See also Summary of the report about the sectorial investigation into the pharmaceutical sector, http://ec.europa.eu/competition/ sectors/pharmaceuticals/inquiry/communication_en.pdf, p. 11 (‘Filing numerous patent applications for the same medicine (forming so called “patent clusters” or “patent thickets”) is a common practice. Documents gathered in the course of the inquiry confirm that an important objective of this approach is to delay or block the market entry of generic medicines.’). 22 See OECD, Competition, Patents and Innovation II (1 April 2010), DAF/COMP(2009), p. 20. 23 In further detail, see below notes 36-39 and accompanying text.

Paul Lugard 

277  

In the first place, innovation may affect the definition of the relevant product or service market. In this regard, the argument has been made in a number of cases that innovation leads to larger product markets.24 In addition, innovation can play a role in the analysis of market structure, especially when analysing market power. The presence or likelihood of innovation can then be an argument for the position that entry barriers are low and that market power is only temporary, or, put differently, that it is not durable.25 Finally, where appropriate the Commission reviews whether a concentration will negatively affect innovation, either because the concentration relates to one or more companies that innovate and it can be expected that the parties involved in the concentration will in the future themselves innovate less as a result of the transaction, or because the transaction would impact the innovation potential or incentives of third parties. In Case M.6214, Seagate Technology/The HDD Business of Samsung Electronics26 and Case M.5727, Microsoft/Yahoo! Search Business, the Commission concluded that the transaction would not result in less innovation, but in three other cases it believed that innovation would be restricted.27 Clear examples of concentrations that were deemed to negatively impact the future innovation of third parties are: Case M.5984, Intel/McAfee,28 in which the Commission considered that the combination of Intel’s dominant position in the chipset/x86 CPU market with McAfee’s (modest) position in the market for security software would eliminate innovative competitors of McAfee; and Case M.5669, Cisco/Tandberg, in which the acquisition by Cisco of Tandberg, a competitor in the market for video communication technology, would encourage Cisco to refrain from enabling other competitors to manufacture interoperable products.29 Concentrations involving essential patents constitute a separate category. In the United States, three such transactions have drawn particular attention:30 the takeover by Google of Motorola Mobility, the acquisition by Apple, Microsoft and RIM of 6,000 Nortel patents, and the acquisition by Apple of a patent portfolio belonging to Nortel.31 The central issue in these cases was the question of how the acquisition of intellectual property rights may affect the economic incentives to license essential patents in the future. In its assessment of the Google/Motorola transaction, the Commission investigated whether the acquisition of Motorola See, eg, Case COMP/M.6281 – Microsoft/Skype; Case M.4941 – Nokia/Navteq. See, eg, Case COMP/M.2537 – Philips/Marconi Medical Systems; Case COMP/M.5727 – Microsoft/Yahoo! Search Business; cited above note 15. 26 Case COMP/M.6214 – Seagate Technology/The HDD Business of Samsung Electronics. 27 Case COMP/M.5658 – Unilever/Sarah Lee; Case COMP/M. 6106 – Caterpillar/MWM; Case COMP/M.5675 – Syngenta/Monsanto. 28 Case COMP/M.5984 – Intel/McAfee. 29 Case COMP/M.5669 – Cisco/Tandberg. The Commission’s findings are however based in particular on the static effects of the concentration. 30 See Paul Lugard, ‘Technology Licensing: Evolving Antitrust Standards in the Smartphone and Other Sectors’, CPI Antitrust Chronicle 2012(2). 31 See U.S. Dept. of Justice, Statement of Department of Justice’s Antitrust Division on its Decision to Close its Investigation of Google Inc.’s Acquisition of Motorola Mobility Holdings Inc. and the Acquisitions of Certain Patents by Apple Incl., Microsoft Corp., and Research in Motion.Ltd., at 3 (13 February 2012), www.justice.gov/opa/pr/2012/February/12at-210.html. 24 25

278  Competition, Regulation and Public Policies Mobility would incentivize Google to hinder competitors by commencing infringement actions based on essential smartphone patents, or by seeking to impose unreasonable, non-FRAND licence conditions, which would threaten innovation.32 This question also came up in the recently approved acquisition by Microsoft of Nokia’s smartphone activities.33 Finally, it is striking that, so far, not a single (second phase) concentration has been approved based on dynamic efficiency improvements. Although claims of such efficiencies were raised in four cases, the argument was deemed plausible in only one case, Case M.4187, Metso/Ake Kvaerner.34 However, in that case, the dynamic efficiencies were held not to outweigh the observed negative effects.

Competition Policy under Article 101 TFEU In recent years the Commission, in its enforcement of Article 101, has made significant investments in the fight against innovation-limiting practices. The Commission’s approach in these matters reflects an increasing scepticism regarding commercial behaviour that relates to intellectual property rights and technical standards. This stricter policy relates in particular to ‘pay for delay’ agreements in the pharmaceutical sector, standardisation agreements and technology transfer agreements.35 With regard to the first type of agreements, in response to its findings in the pharmaceutical sector the Commission has initiated infringement proceedings against a number of pay for delay agreements. Although the Commission acknowledges that such agreements may be a legitimate instrument to resolve patent disputes, it believes that such agreements can also constitute a violation of Article 101, particularly if the licensor provides an inducement, financially or otherwise, for the generic company to accept more restrictive settlement terms than it would otherwise have accepted based on the merits of the licensor’s technology.36 In this context, the Commission imposed a fine for the first time in June 2013 on Lundbeck and a number of producers of generic medicines,37 and it opened pay-for-delay investigations involving Cephalon and Case COMP/M.6381 – Google/Motorola Mobility. Case COMP/M.7047 – Microsoft/Nokia. 34 Case COMP/M.4942 – Nokia/Navteq; Case COMP/M.4854 – TomTom/Tele Atlas; Case No. COMP/M.3916 – T-Mobile, Tele.ring; Case COMP/M.4187 – Metso/Ake Kvaerner. 35 It is striking that few cartel cases are directly related to innovation of the cartel participants. However, according to Commissioner Almunia, this did occur in the TV and computer monitors cartel (Commission Decision of 5 December 2012 in Case COMP/39.437 – TV and computer monitors). See Joaquín Almunia, ‘EU competition policy and innovation’, SPEECH/13/697 of 13 September 2013¸ Florence. 36 See Third Report on the monitoring of patent settlements, http://ec. europa.eu/competition/ sectors/pharmaceuticals/inquiry/patent_ settlement_report3_en.pdf. 37 See http://europa.eu/rapid/press-release_IP-13-563_en.htm?locale=. en. 32 33

Paul Lugard 

279  

Teva Pharmaceutical Industries Ltd.,38 Servier,39 and Johnson & Johnson and Novartis.40 It is expected that the Commission will decide to tighten is policy with regard to pay for delay agreements in the 2014 version of the Block Exemption for technology transfer agreements. In its draft proposal for a revised exemption and revised guidelines, the Commission states that pay for delay agreements are regarded as violating Article 101 if the licensee accepts more restrictive licence terms in than it would otherwise have accepted based on the merits of the licensor’s technology in return for a financial or other inducement.41 Secondly, in 2011 the Commission significantly toughened its policy regarding standardisation agreements in order to better respond to the strategic use of essential patents that limit competition and innovation. This stricter approach is currently reflected in the Guidelines on horizontal cooperation agreements.42 In contrast to the previous guidelines, the current version includes detailed provisions regarding participation in standardisation discussions, and an entirely new section which aims to force standard-setting organisations to require participants of those meetings to license their intellectual property rights on FRAND conditions if their technology is included in a technical standard.43 Guidance and instructions for the determination of FRAND conditions are also included. A third initiative to promote innovation relates to the conditions that licence agreements and other technology transfer agreements must meet in order not to violate Article 101. Here again, the Commission’s proposed Block Exemption and accompanying guidelines for technology transfer agreements are relevant.44 A number of the proposed changes are explicitly based on the Commission’s desire to eliminate or ‘weed out’ invalid intellectual property rights.45 Interestingly, to achieve that objective the Commission specifically aims to involve licensees. For example, Article 5(1)(b) of the draft Block Exemption provides that the right See http://europa.eu/rapid/press-release_IP-11-511_en.htm?locale=. fr. See http://europa.eu/rapid/press-release_IP-12-835_en.htm?locale=. en. 40 See http://europa.eu/rapid/press-release_IP-13-81_en.htm. On 10 December 2013, the Commission imposed a €16 million fine in this case. 41 See Draft Communication from the Commission - Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements, paragraph 223, http://ec.europa.eu/competition/consultations/2013_technology_transfer/guidelines_en.pdf. Following the completion of this chapter, the wording was changed: see paras 238–239 of the final version of the Guidelines (citing the Lundbeck decision (see above note 37) and emphasizing the risk of market sharing where a pay-for-delay involve a significant value transfer between competitors). 42 See Guidelines regarding the applicability of Article 101 of the Treaty on the working of the European Union in horizontal cooperation agreements, paras 257–335, http://eur-lex.europa.eu/ LexUrlServ/LexUrl Serv.do?url=OJ:C:2011:011:0001:0072:NL:PDF. 43 See ibid, paras 283–286. 44 See above note 41. 45 See, eg, draft paragraph 123 (’In the interest of undistorted competition and in conformity with the principle underlying the protection of intellectual property, invalid intellectual property rights should be eliminated. Invalid intellectual property stifles innovation rather than promoting it.’ This language is retained in paragraph 134 of the final version of the Guidelines.) See also draft paragraphs 226–227 (and now paragraphs 242–243 of the final version). The Commission also proposes new rules with regard to technology pools and licences. 38 39

280  Competition, Regulation and Public Policies of the licensor to terminate a licence agreement in case the licensee challenges the licensed intellectual property rights is not automatically exempted from the prohibition contained in Article 101. These types of clauses are generally considered to safeguard the contractual balance between the parties to a licence agreement, and until now they have been covered by the exemption. The proposed amendment aims to make it easier for licensees to challenge the validity of intellectual property rights without running the risk that the licence agreement will be terminated for that reason. However, this element of the Commission’s proposal has also given rise to much criticism, especially because the proposed new rule may give rise to opportunistic behaviour on the part of licensees.

Competition Policy under Article 102 TFEU The Commission’s scepticism regarding the use of intellectual property rights is also reflected in its enforcement policy under Article 102. In the pharmaceutical sector, AstraZeneca was the first case in which the Commission determined that a pharmaceutical company had abused its dominant position by providing misleading information to patent offices in order to extend its patent protection for its products – specifically, its ulcer drug Losec – and to delay the entry of generic medicines.46 However, the Commission’s policy under Article 102 has mainly focused on the ICT sector. As mentioned above, in this sector, standard setting and essential patents play an important role. Standardisation normally produces significant positive economic effects, for example by stimulating the development of new products, enhancing competition between products that comply with technical standards and promoting interoperability and compatibility between products. However, standards that set detailed technical specifications may also give rise to restrictive effects on competition. For example, it is conceivable that valuable technology may be wrongly excluded from the market, or that standardisation may extend to price agreements regarding products that use the standardised technology. Recently, however, the concerns of competition agencies seem to be focusing particularly on the use of essential patents. Patents are deemed to be standard essential patents if the manufacture of a product in accordance with the technical specifications of the standard necessarily implies an infringement of those patents. During the standard-setting process, the parties that take part in the relevant discussions are usually aware that the selection of a specific technical solution also implies the use of the intellectual property rights of the owner of the technology. 46 AstraZeneca, cited above note 3. The Commission also considered the selective de-registration of market authorisations for Losec capsules as abusive. The Commission’s decision was upheld on appeal in Case C-457/10 AstraZeneca v Commission [2012] ECR I-000.

Paul Lugard 

281  

One issue of concern is that firms participating in standardisation discussions might fail to timely disclose intellectual property rights which, once the standard has been set, turn out to be essential. In such a case, the holder of the essential patents may be able to negotiate excessively high royalties or other unreasonable licensing conditions, or it may refuse to grant a licence altogether. This risk is particularly manifest when market participants have made investments in the production of the standardised products. These types of patent ambushes do not occur often, but in theory they may lead to a scenario where producers of the respective products are seriously disadvantaged. Incidentally, in EU competition law there only one example of a patent ambush case: the Rambus case, which ended in 2009 with a commitment decision.47 The Commission seeks to remove the risk of patent ambushes in two ways: (1) via obligations imposed on the participants of standardisation discussions to make good faith disclosures of their intellectual property rights and of the conditions under which they are willing to grant licences; and (2) via ex ante requirements to license patents under FRAND conditions if those patents later turn out to be essential. However, practice has shown that the determination of FRAND conditions, and in particular the amount of royalties, can be complex. Over time, a number of methods have been suggested for the determination of FRAND conditions but there is no consensus over the correct methodology.48 The Qualcomm case provides a telling illustration of this problem, but it did not result in either a prohibition or commitment decision.49 In recent months, attention has shifted to a specific problem closely related to FRAND licence terms, ie the question of whether, and if so, under which circumstances, holders of essential patents may obtain injunctive relief against another party without violating Article 102 if no licence agreement between the holder of the intellectual property right and the infringing party has been concluded. This question has come up particularly in the ‘smartphone war’ between Apple and Samsung and other manufacturers of smartphones. As addressed elsewhere, it is argued that such injunctions may result in undue pressure in the negotiation of licence agreements (and hence onerous conditions).50 That pressure is believed to stem from the very nature of injunction proceedings, which is to secure a ban on the continued marketing of the infringing products.51 This discussion has gained momentum with the advent of ‘non-practicing entities’ (NPEs), or ‘patent trolls’, ie companies that have fewer economic Commission Decision of 9 December 2009 in Case COMP/88.636 – Rambus. See, eg, J. Gregory Sidak, ‘The Meaning of Frand, Part I: Royalties’, 9 Journal of Competition Law and Economics 931 (2013). 49 See http://europa.eu/rapid/press-release_MEMO-09-516_en.htm? locale=en. 50 See Case COMP/M.6381 – Google/Motorola Mobility, para 107 (‘Depending on the circumstances, it may be that the threat of injunction, the seeking of an injunction or indeed the actual enforcement of an injunction granted against a good faith potential licensee, may significantly impede effective competition by, for example, forcing the potential licensee into agreeing to potentially onerous licensing terms which it would otherwise not have agreed to.’). 51 See Paul Lugard, ‘Past, Present and Future’, Markt & Mededinging, 2013(2). 47 48

282  Competition, Regulation and Public Policies incentives to offer patent licences on FRAND terms. It is argued that this is the case because these firms do not themselves manufacture products that implicate the use of the essential patents of other parties and are therefore immune against counter-attacks from other holders of essential patents. As a consequence, it is argued that NPEs are more likely to demand unreasonable licensing terms, possibly through the use of injunctive relief proceedings.52 Similar observations apply to ‘patent privateers’, NPEs which are (covertly) supported financially by manufacturers that may have a strategic interest in obstructing their competitors on product markets.53 The Commission takes the position that Article 102 severely limits the right to oppose infringements of essential patents (in summary proceedings), and that a holder of such patents may only do so if it has been shown that the infringing party is obviously not willing to negotiate in good faith on FRAND terms. Not surprisingly, there are widely diverging views on when this is the case. Meanwhile, the Commission has received a number of complaints regarding the imposition of non-FRAND terms, and following complaints received from Apple the Commission issued, in December 2012, a Statement of Objections against Samsung for seeking injunctions against that company, in particular in Germany.54 Recently, the Commission published draft commitments offered by Samsung pursuant to Article 9 of Regulation 1/2003 in order to meet the Commission’s concerns. The proposed commitments would provide for a procedure for the negotiation of licence agreements on FRAND conditions. Much could be said about the benefits and drawbacks of this procedure and its contents,55 but the key element is an obligation on the licensor not to seek an injunction before any court or tribunal in the EEA of its SEPs implemented in smartphones against a potential licensee that agrees to and complies with a particular negotiating framework for the determination of FRAND terms. This framework includes a negotiation period of up to twelve months, and provides for a third party (arbitration or court adjudication) determination of FRAND terms in the event that no licence agreement or alternative process for determining FRAND terms has been agreed by the end of the negotiation period. We must wait and see what the result of the Samsung case will be. However, it can be expected that a future commitment decision in this case will constitute a template for contract negotiations involving holders of essential patents, and for users of essential technology more generally. In addition, the Court of Justice can still complicate matters: on 5 April 2013, the District Court of Dusseldorf referred 52 See, eg, Joachim Henkel, Markus Reitzig and Christopher Heath, ‘On sharks, trolls, and their patent prey. Unrealistic damage awards and firms’ strategies of “being infringed”’, 36 Research Policy 134 (2007). 53 See, eg, Tom Ewing, ‘Introducing the patent privateers’, Intellectual Asset Management Magazine (January/February 2011), pp. 31 et seq. 54 See http://ec.europa.eu/competition/elojade/isef/case_details.cfm?proc _code=1_39939. A comparable investigation pertains to Motorola. See http://europa.eu/rapid/press-release_IP-13-406_ en.htm. 55 See Paul Lugard and Martin Mollmann, ‘The European Commission’s Practice Under Article 9 Regulation 1/2003: A Commitment a Day Keeps the Court Away?’, CPI Antitrust Chronicle 2013(3).

Paul Lugard 

283  

a number of preliminary questions about the conditions in German law under which injunctions can be requested under the Orange Book doctrine.56 It cannot be excluded that the Court of Justice may arrive at conclusions that differ from the Commission’s assumptions that underlie the commitments the Commission is aiming for in Samsung.

Conclusion The above survey attempts to provide an impression of the way in which innovation-related conduct is evaluated under EU competition law. It can be concluded that EU competition law lacks a well-articulated, comprehensive concept of innovation and lacks an analytical framework that could be used for the evaluation of innovation in relevant cases. However, in a number of specific settings, innovation plays a significant role in the evaluation of transactions. This is not surprising considering that standardisation, access to technology platforms and (essential) intellectual property rights play a key role in innovation in many high-tech sectors.

56 See Case C-170/13 Huawei Technologies Co. Ltd v ZTE Corp., ZTE Deutschland GmbH, not yet decided.

Sven B Völcker*

Promises, Promises … Law and Policy Considerations with Respect to the Application of Article 102 TFEU to FRAND-Encumbered Standard-Essential Patents

1. Introduction In the 1968 musical ‘Promises, Promises’ (based on the 1960s film ‘The Apartment’), a junior executive at an insurance company allows his married superiors to use his apartment for trysts in exchange for promises that this will advance his career. As it turns out, this is one of many promises that are not kept as things get complicated. The Commission, the European Court of Justice, and national courts are currently faced with the competition law implications of a promise that one would have thought more credible and far-reaching in its implications, namely the promise given by patent holders having declared patents to be essential to a given industry standard (in short, standard-essential patents, or ‘SEPs’),1 to license those patents on ‘fair, reasonable and non-discriminatory’ (FRAND) terms to any implementer of the standard. The Commission has initiated proceedings and adopted Statements of Objections against Samsung and Motorola Mobility with respect to their assertion of FRAND-encumbered SEPs against Apple.2 In each case, the Commission has emphasized the centrality of the ‘irrevocable’ FRAND commitment given by SEP holders and its concerns with SEP holders seeking (and in Motorola’s case enforcing) injunctions notwithstanding their promise to license on FRAND terms. In October 2013, Samsung proposed commitments, which it is currently revising in light of the results of the Commission’s market test.3 The Commission’s investigations also triggered a reference for a preliminary ruling * Partner, Latham & Watkins, Brussels, and Honorary Professor, Göttingen University. The views expressed in this paper are entirely personal and cannot be attributed to Latham & Watkins or any of its clients. 1 A better term is ‘declared essential patents’, as the true essentiality of a patent to a standard is rarely tested in patent litigation, and over-declaration is a widely acknowledged problem. H owever, the ‘SEP’ terminology seems to have taken hold, and it is thus used throughout this discussion. 2 Commission Press Release IP/12/1448 of 21 December 2012 (Commission sends Statement of Objections to Samsung on potential misuse of mobile phone standard-essential patents); Commission Press Release IP/13/496 of 6 May 2013 (Commission sends Statement of Objections to Motorola Mobility on potential misuse of mobile phone standard-essential patents). 3 Commission Press Release IP/13/971 of 17 October 2013 (Commission consults on commitments offered by Samsung Electronics regarding use of standard essential patents).

286  Competition, Regulation and Public Policies by the Landgericht Düsseldorf in the case of Huawei v ZTE, which focuses on the circumstances under which a standards-implementer could be characterized as an ‘unwilling’ licensee and under which an SEP-holder could thus be justified in seeking an injunction.4 Most recently, the Landgericht Mannheim posed a series of questions to the Commission seeking guidance on how to determine the FRAND rate for a comprehensive SEP portfolio given the inherent uncertainty about the validity and essentiality of many of those patents.5 Even before the more recent wave of cases, the Commission was no stranger to FRAND commitments. It spent many years investigating complaints that Qualcomm gave FRAND commitments that it allegedly never intended to honor in order to have its technology included in the UMTS standard,6 and obtained a commitment from IPCom to respect the FRAND commitment given by Bosch, the previous owner of the SEPs asserted by IPCom.7 The Commission’s 2010 Guidelines on Horizontal Cooperation devote several paragraphs to explaining aspects of a FRAND commitment that would allow a standard-setting organization to claim a safe harbor under Article 101 TFEU.8 Several years earlier, in its 2004 Microsoft decision – a de facto standards case in which Microsoft had never committed to make its technology available to anyone – the Commission imposed the obligation to make interoperability information available to rival server operating system suppliers on ‘reasonable and non-discriminatory terms’.9 The Commission deemed this obligation sufficiently precise to impose substantial periodic penalty payments in 2008,10 and its assessment was confirmed by the General Court.11 And yet the recent surge in disputes over the assertions of SEPs has revealed that virtually every aspect of the FRAND promise, and its significance for the application of Article 102 TFEU, is still in dispute. Does the FRAND promise imply a promise not to seek injunctions against standards-compliant products, and if so, when is the breach of such a promise an abuse under Article 102 TFEU? What limits, if any, does Article 102 impose on the royalties that an SEP holder can demand for its FRAND-encumbered patents? For instance, does the prohibition of imposing ‘unfair selling prices or other unfair trading conditions’ affect the 4 Request for a preliminary ruling from the Landgericht Düsseldorf lodged on 5 April 2013 – Case C-170/13 Huawei Technologies Co Ltd v ZTE Corp, ZTE Deutschland GmbH, 2013 OJ C215/5, not yet decided. 5 See ‘German court seeks EU Commission views on patent rates, portfolios in Apple, Motorola dispute’, Mlex News service (20 November 2013). 6 In the Qualcomm case, the Commission investigated whether Qualcomm’s royalties for its patents declared essential to the 3G standard were unreasonably high. The Commission eventually decided to drop a four-year investigation (officially opened in October 2007), without taking any action against Qualcomm. The Commission explained that the companies that had complained about Qualcomm’s behavior had withdrawn their objections, or were planning to, and that it would not be ‘appropriate to invest further resources in this case’. See Commission MEMO/09/516 of 24 November 2009. 7 Commission MEMO/09/549 of 10 December 2009. 8 Commission, Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements, 2011 OJ C11/1, paras 277–291. 9 Commission Decision of 24 March 2004 in Case COMP/37.792 – Microsoft, 2007 OJ L32/23. 10 Commission Decision of 27 February 2008 in Case COMP/37.392 – Microsoft, 2009 OJ C166/20. 11 Case T-167/08 Microsoft Corp. v Commission [2012] ECR II-000.

Sven B Völcker 

287  

royalty base and methodology used to determine the actual royalty rate to be paid by the licensee? This chapter argues that there are sound policy reasons for the Commission to investigate the use of SEP-based injunctions as a priority. The Commission can order effective relief in the form of a ‘no-injunction’ remedy, and it is better placed than national courts to give meaning to the FRAND promise in the context of coherent competition policy. The chapter then explains that the pursuit of injunctions by SEP-holders should not be analyzed exclusively under the strict standard developed for when ‘vexatious litigation’ amounts to an abuse contrary to Article 102, but rather should be seen as: an abusive refusal to deal; an attempt to impose unfairly high royalties on implementers (Article 102(a)); and typically also as discrimination (Article 102(c)). Consequently, the Commission would be on firm legal ground if it were to find injunctions based on FRAND-encumbered SEPs abusive, at least barring exceptional circumstances.

2. Policy issues 2.1 Is there a need for intervention? It has been argued in the past that the abuse of SEPs and related disputes over the interpretation of FRAND commitments were not widespread or serious enough to warrant the Commission’s intervention. Whatever may have been the situation in the past, that position appears difficult to defend today. While global litigation battles between well-resourced technology giants such as Apple and Samsung tend to receive the most media attention, many other companies are aggressively maximizing the value of their SEP portfolio by threatening or seeking court injunctions against implementers unwilling to meet their licensing demands. These companies include not only non-practicing entities (NPEs) such as IPCom, whose very purpose is to maximize royalties based on acquired patents; increasingly, they also include once market-leading vertically integrated companies such as Ericsson and Nokia, whose fortunes on downstream product markets have declined, leading them to seek to replace lost product revenue by ‘monetizing’ their considerable patent portfolios, including SEPs. A strong manifestation of market failure consists of the exorbitant price tags that SEP portfolios have commanded of late, several billion dollars in the case of the sale of the SEP-heavy Nortel patent portfolio,12 and Google’s overtly SEP-motivated acquisition of Motorola Mobility.13 The extent to which these 12 See US Department of Justice, Press Release of 13 February 2012, http://www.justice.gov/opa/ pr/2012/February/12-at-210.html. 13 See, among the others, http://www.nytimes.com/2012/10/10/technology/widening-scrutiny-ofgoogles-smartphone-patents.html?_r=0 (9 October 2012).

288  Competition, Regulation and Public Policies acquisitions may have been driven by ‘offensive’ or ‘defensive’ purposes is largely irrelevant in this context: what matters is the SEPs’ perceived value as a bargaining chip, and that value has skyrocketed without any underlying increase in technological innovation. Indeed, the reported royalty demands by Samsung and Motorola – 2.4% and 2.25% based on the average selling price (ASP) for any device containing 3G functionality sold worldwide, suggest far higher valuations of existing 3G SEP-portfolios – probably in excess of €100 billion if one considers the enormous volume of smartphone sales and the duration of patent protection in light of the need for backwards-compatibility even as new generations of wireless air interface standards emerge.14 The apparent onset of hyperinflation for SEP-portfolios suggests that the market has only limited faith that the current legal process will guarantee that compensation for FRAND-encumbered patents is based only on the prestandardization (ex ante) value, as the Commission’s Horizontal Cooperation Guidelines would have it. Instead, the market seems to recognize the vastly increased strategic value of those patents in today’s commercial and legal environment. As the Commission concedes in its Google/MMI merger decision, ‘a FRAND commitment cannot be considered as a guarantee that the SEP holder will not abuse its market power’.15 It is difficult to ascribe any welfare-enhancing effect to the vastly inflated price tags for SEP portfolios. Most litigated SEPs relate to standards that have long been frozen, meaning that an improved prospect of monetizing them can no longer incentivize the contribution of IP to a standard. With respect to ongoing standardization processes, the theoretical possibility that innovators will more readily contribute their technology to standards because of higher returns needs to be balanced against the increasing risks of opportunistic patenting in parallel to the elaboration of the standard, as well as over-declaration of patents as essential. Indeed, both of these risks are well-documented, in particular for the 3GPP standardization process, which took place at a time when SEP-portfolios were predominantly cross-licensed rather than monetized.16 The implications for businesses and consumers are substantial. In particular, SEP-holders’ royalty demands and implementers’ offers can be billions of Euros apart – billions that will, at least in part, ultimately be passed on to consumers if SEP-holders prevail in such disputes. Commission guidance on key issues such as when an SEP-holder may have recourse to injunctions, the appropriate royalty base, and any cumulative limits on the overall standards-related ‘royalty stack’ could dramatically affect the way that these disputes are resolved. The potential amounts of overcharge – and, consequently, the risk of consumer 14 For instance, Motorola has claimed in court filings that Microsoft alone owes it 4 billion euros per year for the use of its standard-essential patents. See http://www.bloomberg.com/news/2012-04-02/ motorola-mobility-wants-4-billion-from-microsoft-over-patents.html (2 April 2012). 15 Commission Decision of 13 February 2012 in Case COMP/M.6381 – Google/Motorola Mobility, para 113. 16 See, eg, Rudi Bekkers and Joel West, ‘Standards, Patents and Mobile Phones: Lessons from ETSI’s Handling of UMTS’, 7 Journal of IT Standards and Standardization Research 13 (2009).

Sven B Völcker 

289  

injury – vastly exceed those that typically prompt Commission intervention in merger or cartel cases.

2.2 Can the Commission provide effective relief under Regulation 1/2003? The claims that there is no empirical evidence of any FRAND-related problem that needs to be remedied are often accompanied with stern warnings that the Commission or other competition authorities are ill-equipped to intervene in such situations, as they would necessarily have to act as price regulators. It is of course true that, for sound policy and practical reasons, the Commission has generally shied away from intervening in cases of purely exploitative abuses. Excessive prices often encourage market entry; a maximum-price remedy is often difficult to formulate, and may require ongoing monitoring and adjustment to be effective. However relevant those objections may be as a general matter,17 they do not apply to the most straightforward remedy to address an SEP-holder’s abusive recourse to injunctions to influence royalty negotiations: a simple ‘no-injunction’ remedy requiring that the dominant SEP holder withdraw existing injunction applications and refrain from bringing equivalent actions in the future. Article 7 of Regulation 1/2003 gives the Commission broad powers to order a remedy that is adapted to the specific abusive conduct in question. Indeed, the provision empowers the Commission to impose ‘any behavioural or structural remedies which are proportionate to the infringement committed and necessary to bring the infringement effectively to an end’. In Magill the ECJ confirmed that Article 3 of Regulation 17 (the predecessor to Article 7 of Regulation 1/2003) is ‘to be applied according to the nature of the infringement found and may include an order to do certain acts or things which, unlawfully, have not been done as well as an order to bring an end to certain acts, practices or situations which are contrary to the Treaty’.18 In other words, the remedy can concern any conduct that constitutes an abuse, and any conduct that is necessary to re-establish the status quo ante existing before the abusive conduct. To the extent that a dominant SEP-holder’s pursuit of injunctions is abusive, the Commission must have the power to order the SEP-holder to cease bringing such injunctions and to withdraw any pending applications for injunctions to restore the status quo ante. A different question is how prescriptive the Commission can and should be in terms of what constitutes FRAND terms.19 It is submitted that the Commission 17 Recent Commission commitment cases in the area of financial services show a renewed focus on excessive pricing in areas where barriers to entry are high due to de facto standardization, and suggest confidence that commitment decisions are an appropriate means to address such excesses. See, eg, Commission Decision of 15 November 2011 in Case COMP/39.592 – Standard & Poor’s, (Press Release IP/11/1354 of 15 November 2011; and Commission Decision of 20 December 2012 in Case COMP/39.654 – Reuters Instrument Codes (RICs) (Press Release IP/12/1433 of 20 December 2012). 18 Joined Cases C-241/91 P and C-242/91 P Radio Telefís Éireann (RTE) and Independent Television Publications Ltd (ITP) v Commission (Magill) [1995] ECR I-743, para 90. 19 The General Court’s 2012 Microsoft judgment could be read as imposing some limits, but it refers to the margin of discretion that a company may have in implementing remedies ordered in a

290  Competition, Regulation and Public Policies could, and indeed should, give guidance as to what a (non-abusive) FRAND-offer could look like, without actually setting the royalty rate in any given case. For example, the Commission could stipulate the appropriate base for calculating a royalty, indicating in particular whether this should be the average sales price of the full device or the price of the component practicing the patented technology in question. Indeed, the Commission did exactly that when, in response to concerns expressed during the market test, it insisted that Rambus clarify that the royalty caps it offered be calculated based on the sales price of DRAM-chips and not on the sales price of devices incorporating DRAM-chips, or the sales price of multifunction chips incorporating DRAM functionality.20 It is difficult to see why the Commission could not impose a similar obligation as part of a remedy under Article 7 of Regulation 1/2003. The same applies to the general principles for the calculation of the royalty rate itself. For example, according to the so-called ART (aggregate reasonable terms) principle, royalty rates for a given SEP-portfolio must consider the costs of obtaining necessary licenses from other SEP-holders for the same standard, and how a given SEP-holder’s portfolio would be evaluated under such a rule.21 Finally, one could consider the benefits of the Commission requiring an SEPholder to offer dispute resolution mechanisms that are designed to determine a FRAND rate for the SEP-holder’s portfolio free from the threat of injunctions. Samsung’s recently offered proposed commitments are based on such an approach. While a detailed discussion of such a process-focused remedy is beyond the scope of this chapter, great care must be taken to ensure that it does relieve the patent holder of the burdens of proof it must discharge in ordinary patent litigation and thus abrogates the standards-implementer’s legal defenses rather than impose an effective remedy to curb abuses by the SEP-holder. In particular, the patent holder should have to respond to challenges of validity of its patents, and to prove essentiality and infringement for each patent or at least a sufficiently representative set of such patents before the adjudicator moves to the setting of a FRAND rate so as to ensure that the standards-implementer is not forced to pay for patents it does not practice, or for patents likely to be found invalid. prior decision, and does not directly bear on the Commission’s ability to formulate such remedies in the first place. See Microsoft, cited above note 11, para 95. The judgment’s reference to Case T-24/90 Automec v Commission [1992] ECR II-2223, para 52, does not appear convincing, as the latter refers to the consequences of an infringement of Article 101, rather than Article 102. 20 See Commission Decision of 9 December 2009 in Case COMP/38.636 – Rambus, para 66. Indeed, using the component price as a royalty base appears to be the only method that is clearly compatible with Article 102. Using the ASP of the final device incorporating the component ultimately amounts to demanding royalties for functionalities that the patent holder can claim no credit for, such as other components provided by third-party suppliers, or the device manufacturer’s own efforts to distinguish its product through R&D, design and marketing. As the General Court confirmed in Case T-151/01 Der Grüne Punkt – Duales System Deutschland GmbH [2007] ECR II-1607, para 121, regardless of the level of the royalty rate, a dominant firm requiring payment for services not actually provided imposes an unfair trading term within the meaning Article 102(a). 21 See, eg, Nokia’s presentation on Standards & Patents, World Intellectual Property Organization (WIPO), November 2006, http://www.wipo.int/meetings/en/2006/patent_colloquia/11/pdf/frain_ presentation.pdf.

Sven B Völcker 

291  

2.3 Are national courts better placed to resolve the question of injunctions and FRAND levels? It is frequently argued that national courts are better placed to resolve FRANDrelated disputes, with respect to both the granting injunctions and the issue of what license terms could be considered FRAND. Without questioning in any way national courts’ competence to adjudicate the technical patent and commercial issues in such disputes, it seems highly unlikely that courts could be guided by an ‘invisible hand’ to develop the kind of coherent EU competition policy that is called for in this important area. With respect to injunctions, as discussed further below, national courts in the EU are bound to ensure the practical effectiveness of Article 102, and should thus be able to incorporate such considerations into their assessment when called upon to grant injunctions based on FRAND-encumbered patents. However, in practice, national courts have given only very limited consideration to EU competition law issues – at least until the Commission began its recent wave of enforcement actions. Part of the explanation may be that injunctions are often decided by specialized patent courts or chambers with limited competition law experience. Indeed, it appears that some courts, in particular in Germany, have been vying to establish themselves as particularly patentee-friendly fora by granting injunctions quickly. Competition law defenses take patent judges out of their ‘comfort zone’ and open up their rulings to scrutiny by other stakeholders and institutions. While the German Supreme Court has recognized principles under which the implementer of a standard can avoid an injunction by an SEP-holder (the so-called Orange Book defense),22 that judgment does not make any reference to Article 102, and lower courts have applied the Orange Book principles so narrowly that it is doubtful whether they ensure Article 102’s practical effectiveness as they are required to do under the principle of sincere cooperation enshrined in Article 4(3) TEU.23 Moreover, no EU national court has access to the vast body of institutional knowledge on the practice of standard setting and standard-essential patents that the Commission has accumulated over the years as a result of its investigations, its consultation preceding the revision of its Horizontal Cooperation Guidelines,24 and its active involvement in ongoing discussions within ETSI and ITU about possible revisions of their IPR policies.25 Last, the Commission can also consider the extent to which the effective application of Article 102 requires an assessment of SEP-holders’ pursuit of 22 Case KZR 39/06, [2009] 180 BGHZ 312, 316 – Orange-Book-Standard. An English translation is available in 41 IIC International Review of Intellectual Property and Competition Law 369 (2010). 23 See, among the others, Torsten Körber, Standard Essential Patents, FRAND Commitments, and Competition Law, Nomos, 2013. 24 Consultation materials are available on the Commission’s website at http://ec.europa.eu/ competition/consultations/2010_horizontals/index.html. 25 Representatives of the European Commission regularly attend the ETSI and ITU’s meetings and roundtables.

292  Competition, Regulation and Public Policies injunctions or exclusion orders from non-EU courts or similar bodies such as the US International Trade Commission – bodies that are obviously not themselves bound by the European Treaties. Indeed, as discussed below,26 SEP-holders often justify the need for injunctive relief in those jurisdictions where it is available with the need to force implementers to sign worldwide portfolio licenses. This means that the Commission should consider such conduct outside the EEA under the effects doctrine, as it has done in other cases such as Rambus and Intel/AMD. Similar considerations apply to the determination of whether a given royalty demanded by an SEP holder is actually FRAND. The only national court in the EU that appears to have addressed the issue thus far is a Dutch court that determined, albeit provisionally in the context of summary proceedings, that Samsung’s demand to Apple of a royalty of 2.4% of the average selling price of Apple 3G-compliant mobile devices was clearly outside the bounds of FRAND.27 In the case of Motorola v Apple, the Landgericht Mannheim has begun a ratesetting proceeding based on the ‘Orange Book’ offer that Apple had to make to avoid an injunction, but it decided to stay the case pending a decision by the European Commission and pose questions to the Commission that illustrate the need for guidance. It is only in the United States,28 and more recently in China,29 that courts have set FRAND rates – in all cases, well below the amounts sought by SEP-holders. One of many difficulties is that, in determining the ‘nondiscriminatory’ nature of the offer, terms of other license agreements concluded by the SEP-holder are typically covered by confidentiality obligations, and most European jurisdictions do not have the tools to compel discovery, including from third parties, and to deal with the resulting confidentiality concerns. It is thus not surprising that national judges dealing with these issues in turn recommend arbitration or mediation by specialist bodies.30 However, even if alternative dispute resolution mechanisms were to develop guiding cohesive principles for FRAND rates that are consistent with sound competition policy (and there is no reason to assume that they would, given the lack of procedural instruments that are at least in theory available to courts, such as Article 15 of Regulation 1/2003 and Article 267 TFEU), the confidentiality restrictions that typically govern alternative forms of dispute resolution mean that they could not play a meaningful role in achieving any lasting ‘patent peace’. There is thus an excellent case to be made for the Commission establishing guiding principles both on the use of injunctions and for the calculation of FRAND royalties. This would still leave the task of setting actual FRAND rates in specific See Section 3.2.3 below. See District Court of The Hague, Judgment in Summary Proceedings of 14 October 2011, paras 4.30–4.41. 28 See Microsoft Corp. v Motorola Inc., 2013 US Dist. LEXIS 60233 (W.D. Wash. 2013); In re Innovatio IP Ventures, LLC Patent Litig., 213 WL 5593609 (N.D. Ill. 2013) 29 See ‘Guangdong High Court rules against InterDigital in landmark antitrust lawsuit’, Mlex News service (29 October 2013). 30 See ‘Judges point to benefits of arbitration in FRAND rate disputes, raise questions over form’, MLex News service (9 November 2012). 26 27

Sven B Völcker 

293  

cases to courts or arbitral tribunals, as anticipated in the Commission’s Horizontal Cooperation Guidelines.31

3. Can an SEP-holder seek injunctions without violating Article 102? As explained in the introduction, a key issue in the current disputes is the question of whether, and under what circumstances, an SEP-holder may seek an injunction against the implementer of a standard without violating Article 102 TFEU. The following sections briefly discuss the question of dominance (3.1), before turning to the various theories of abuse that could apply (3.2), namely vexatious litigation (3.2.1), refusal to deal (3.2.2), excessive pricing (3.2.3) and discrimination (3.2.4). In conclusion, it should be only in rare situations that an SEP-holder’s recourse to injunctions is not abusive.

3.1 Dominance At the very least for ubiquitous standards such as ETSI’s standards for mobile telephony or IEEE’s 802.11 (WiFi) standard, there should be little doubt that a company claiming to hold patents essential to that standard occupies a dominant position. As the Commission states in Google/MMI, ‘the specificity of SEPs is that they have to be implemented in order to comply with a standard and thus cannot be designed around, ie there is by definition no alternative or substitute for each patent. Therefore, each SEP constitutes a technology market of its own.’32 The holder of an SEP is by definition a monopolist on a market thus delineated. The Commission’s findings are consistent with those of national courts that have considered the issue in the past.33 In defending against SEP-based injunctions, the implementer of a standard will often assert non-competition defenses such as invalidity, non-infringement, or exhaustion. The possibility of ultimately succeeding with such defenses does not detract from the hold-up power that such a patent confers. As the General Court recently held in AstraZeneca, ‘the existence of remedies specific to the patent system is not capable of altering the conditions of application of the prohibitions laid down in competition law’.34 Indeed, ‘[w]hen granted by a public authority, an intellectual property right is normally assumed to be valid and an undertaking’s Horizontal Cooperation Guidelines, cited above note 8, para 291. Google/Motorola, cited above note 15, para 54. See, eg, BGH (German Supreme Court), judgment of 13 July 2004, KZR 40/02, GRUR 2004, 966, 967 – Standard-Spundfass (referring to Section 20 GWB). 34 Case T-321/05 AstraZeneca AB and AstraZeneca plc v Commission [2010] ECR II-2805, para 366. This decision was upheld on appeal in Case C-457/10, judgment of the ECJ of 6 December 2012, not yet reported. 31 32 33

294  Competition, Regulation and Public Policies ownership of that right is assumed to be lawful. The mere possession by an undertaking of an exclusive right normally results in keeping competitors away, since public regulations require them to respect that exclusive right.’35 The SEPowner enjoys hold-up power because, absent a license, the implementer must risk an injunction in an infringement action or bring a costly legal challenge to the validity or essentiality of the declared-essential patent.

3.2 Possible legal theories supporting a finding of abuse Prima facie, there are a number of ways in which the pursuit of injunctions by an SEP-holder notwithstanding a FRAND-commitment could be analyzed under Article 102. As discussed below, it would be wrong to assess such conduct exclusively under the strict criteria developed for vexatious litigation (see below section 3.2.1). More appropriately, the pursuit of injunctions should be seen as what it is, namely a refusal to license, and a refusal that arguably satisfies even the narrow criteria developed by the European Courts for a de novo refusal to license intellectual property rights for de facto standards in the absence of a FRAND commitment (section 3.2.2). The pursuit of injunctions will typically also represent an exploitative abuse, as it is aimed at ‘imposing unfair selling prices’ under Article 102(a) (section 3.2.3), and discrimination contrary to Article 102(c) (section 3.2.4).

3.2.1 Vexatious litigation? To state the obvious, an SEP-holder pursuing injunctive relief has recourse to national courts. The question has therefore been raised whether such actions should be analyzed as ‘vexatious litigation’ and thus under the standard enunciated by the General Court in ITT Promedia36 and more recently in Protégé.37 In those cases, the General Court emphasized that access to courts is a fundamental right and a general principle ensuring the rule of the law. It found that only ‘in wholly exceptional circumstances [is] the fact that legal proceedings are brought […] capable of constituting an abuse of a dominant position’.38 In ITT/Promedia, the Commission had stipulated two proposed cumulative criteria for such ‘wholly exceptional circumstances’. According to these proposed criteria, the action (i) cannot reasonably be considered as an attempt to establish the rights of the undertaking concerned and can therefore only serve to harass the opposing party, and (ii) must be conceived in the framework of a plan whose goal is to eliminate competition.39 After having left open in ITT/Promedia whether those cumulative Ibid, para 362. Case T-111/96 ITT Promedia v Commission [1998] ECR II-2937. 37 Case T-119/09 Protégé International v Commission [2012] ECR II-000. 38 ITT Promedia, cited above note 36, para 60; Protégé International, cited previous footnote, para 48. 39 ITT Promedia, cited above note 36, para 55. 35 36

Sven B Völcker 

295  

criteria were compatible with Article 102,40 the General Court confirmed in Protégé that they are. Of course, the application of the cumulative criteria would not exclude a finding that an injunction for SEPs is abusive in any given case. For example, internal documents may confirm that the patentee knew that the asserted patent was not valid or was not actually essential to the standard, or that the implementer was already licensed by virtue of an agreement between the patentee and a component supplier. Still, accepting that SEP-based injunctions should be analyzed exclusively as vexatious litigation would likely imply a significant limitation of the Commission’s ability to intervene in SEP cases. But can this be the right legal standard? In its Google/MMI decision, the Commission notes that ITT/Promedia was ‘a case different from the one at hand’,41 and also points to the Scarlet Extended judgment, in which the Court of Justice found that the rights to intellectual property must be balanced against other fundamental rights, holding on that basis that an injunction requiring an Internet Service Provider to install an onerous filtering system was contrary to EU law.42 Indeed, it would be a serious misunderstanding of ITT/Promedia and Protégé to shield otherwise abusive conduct from scrutiny simply because it involves recourse to judicial remedies.43 As discussed below, an SEP-holder’s pursuit of injunctions is the clearest imaginable manifestation of a refusal to license an intellectual property right, and such action can also be analyzed as both an exploitative abuse and discrimination – all of which are well-established categories of abuse under Article 102 even in the absence of a FRAND commitment. The fact that certain conduct does not meet all of the criteria of a specific type of abuse (in this case, vexatious litigation) does not mean that it cannot fulfill the criteria for another type of abuse.44 Vexatious litigation is not somehow lex specialis to other types of abuses. The situations in ITT Promedia and Protégé were peculiar because the specific allegations concerned conduct that did not fit any category of abuse other than vexatious litigation: in the case of ITT Promedia it was claimed that the would-be competitor was making inappropriate use of a commercial database; in Protégé the maligned conduct was the recourse to trademark offices to oppose trademark registrations. Ibid, paras 57–58. Google/Motorola, cited above note 15, at footnote 55. Case C-70/10 Scarlet Extended SA v Société belge des auteurs, compositeurs et éditeurs SCRL (SABAM) [2011] ECR I-11959, paras 41–46. 43 See also Maurits Dolmans, ‘A Tale of Two Tragedies – A Plea for Open Standards, and Some Comments on the RAND Report, Open Standards & Antitrust’, Concurrences no. 1, 2010, at footnote 6 (‘It has been argued that injunctions should be allowed in standards context subject only to the criteria of abusive litigation [...] But that ignores the crucial element distinguishing standard setting from a normal situation, namely, that the IPR owners have promised to license on FRAND terms, the standards organization has relied on it leading to an agreement to limit inter-technology competition that would otherwise have existed, and the industry has relied on it by making investments in innovation. Having made such a promise and obtained monopoly as a result, it should be an abuse of dominance to seek injunctive relief to extract royalties higher than those that would have pertained in ex ante inter-technology competition.’) 44 See, eg, Case C-52/09 Konkurrensverket v TeliaSonera Sverige AB [2011] ECR I-527, paras 54–58; Case T-201/04 Microsoft v Commission [2007] ECR II-3601, paras 860–861. 40 41 42

296  Competition, Regulation and Public Policies A dominant undertaking cannot be subject to a lesser degree of scrutiny under Article 102 simply because it enlists national courts to render its abuse more effective. For instance, a dominant company that imposes exclusive purchasing obligations on its distributors and enforces these obligations through national courts cannot argue that it is no longer subject to the normal rules governing the use of exclusive agreements.45 A dominant undertaking is not entitled to judicial protection in such circumstances, as the national court in question is itself under an obligation not to grant an injunction that would aid an abusive refusal to license or other conduct that would infringe Article 102. Like other Member State entities, national courts are subject to Article 4(3) TEU’s principle of sincere cooperation and they must not undermine the effective enforcement of Article 102. As the ECJ stated in Inno v ATAB, with particular reference to national courts, ‘while it is true that Article [102] is directed at undertakings, nonetheless it is also true that the treaty imposes a duty on Member States not to adopt or maintain in force any measure which could deprive that provision of its effectiveness’.46 Indeed, the Court has found on numerous occasions that national procedural rules should be disregarded where they risk undermining the effective application of the EU competition rules.47 Applying a higher legal standard for what constitutes an abuse and thus granting privileged treatment to abuses of a dominant position that involve recourse to courts would be particularly inappropriate in the context of intellectual property rights such as patents and copyright, which by their very nature can only be enforced by judicial means. Indeed, cases such as Magill, IMS, and Rambus all involved situations in which the right-holder had attempted to enforce its rights through the courts. In those cases, neither the Commission nor the EU Courts appear to have ever considered whether these undertakings’ actions should be subject to the particularly strict test for abusive vexatious litigation. To be sure, the EU Courts have emphasized that refusals to license IP rights can be abusive only in ‘exceptional circumstances’, but they have not seen any need to limit a finding of abuse to the ‘wholly exceptional circumstances’ referred to in ITT/Promedia just because the dominant undertaking enlisted national courts in rendering its refusal to license more effective. For the avoidance of doubt, the binding force now given to Article 47 of the Charter of Fundamental Rights of the European Union48 has not changed the legal situation. 45 See Case C-344/98 Masterfoods Ltd v HB Ice Cream Ltd [2000] ECR I-11369, which involved an attempt to enforce ‘freezer exclusivity’ by means of injunctions. 46 Case 13/77 GB-Inno-BM [1977] ECR 2115, para 31. See also Case C-280/08 P Deutsche Telekom v Commission, para 45 (‘[I]t is for each Member State to take all appropriate measures, whether general or particular, to ensure the fulfilment by the national regulatory authorities of the obligations which are binding under EU law [and that] Articles [101 and 102 TFEU], in conjunction with Article 10 EC [now Article 4(3) TEU], require the Member States not to introduce or maintain in force measures, even of a legislative or regulatory nature, which may render ineffective the competition rules applicable to undertakings.’). 47 See, eg, Case C-453/99 Courage Ltd v Bernard Crehan [2001] ECR I-6297; Joined Cases C-295/04 to C-298/04 Vincenzo Manfredi v Lloyd Adriatico Assicurazioni SpA [2006 ECR I-6619. 48 As Article 47 states: ‘Everyone whose rights and freedoms guaranteed by the law of the Union are violated has the right to an effective remedy before a tribunal in compliance with the conditions laid down in this Article’.

Sven B Völcker 

297  

First, Article 47 of the Charter guarantees ‘an effective’ remedy, not any particular remedy that the plaintiff may wish to pursue. A finding that a dominant SEP-holder may not be entitled to injunctions in certain circumstances in no way detracts from its ability to pursue damages actions or to seek a declaratory judgment as to the FRAND rates the SEP-holder believes to be entitled to. Second, like all fundamental rights, the right to effective judicial review is subject to limits, as long as those limits are provided for by law and are proportionate, and as long as they respect the right’s essence (Article 52(1) of the Charter). Thus, an SEP-holder’s right to injunctive relief must be balanced against both the effective application of Article 102 and the standard implementer’s individual right to conduct its business.49 The latter is particularly impacted in the case of SEP-based injunctions, as the implementer is prevented from selling any standard-compliant products, whereas in the case of non-SEPs, the defendant can often redesign its products to avoid enforcement of the injunction. Tellingly, in Scarlet Extended, the ECJ focused on the impact of an injunction on the defendant’s business, and did not even mention Article 47 of the Charter.50 Third, any Commission measure ordering the withdrawal of injunctive relief before national courts itself is obviously subject to judicial review by the EU Courts. In Otis, the Court of Justice recognized that, to the extent that the Commission brings its special powers to bear in national proceedings, the other party’s right of access to justice is safeguarded by the EU Courts’ full review of the Commission’s decisions.51 For the sake of completeness, denying injunctive relief to dominant SEPholders in certain cases would also not be incompatible with the EU Directive on the enforcement of IPRs, or with the EU’s obligations under the Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS). These instruments recognize that injunctive relief is discretionary, not automatic,52 and they obviously do not address the specific situation of an SEP-holder having given a FRAND commitment.

3.2.2 Refusal to license In Magill, IMS and Microsoft, the EU Courts developed criteria for the ‘exceptional circumstances’ under which a refusal to license intellectual property 49 See also Case C-12/08 Mono Car Styling SA, in liquidation v Dervis Odemis and Others [2009] ECR I-6653, para 49 (national rules imposing various conditions on the right of action granted to certain workers did not undermine the fundamental principle of effective judicial protection). 50 Scarlet Extended, cited above note 42, paras 41–46. 51 See Case C-199/11 European Union v Otis N.V. [2012] ECR I-000, para 63. 52 Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights, 2004 OJ L157/45. Article 9 of the Directive states that ‘Member States shall ensure that the judicial authorities may, at the request of the applicant: (a) issue against the alleged infringer an interlocutory injunction intended to prevent any imminent infringement of an intellectual property right […].’ Similarly, Article 44(1) of the TRIPS Agreement simply states: ‘The judicial authorities shall have the authority to order a party to desist from an infringement […].’ (emphasis added)

298  Competition, Regulation and Public Policies can be considered abusive. According to these criteria, there must be (1) a refusal to license (2) an input essential to carry out an activity on a downstream market, thus risking (3) a substantial elimination of competition on that market, and (4) stifling innovation to the detriment of consumers (5) without any objective justification. These criteria were developed in cases concerning ‘de novo’ refusals to license non-FRAND encumbered IP (ie, cases in which the IP holder had never licensed or promised to license the right in question), where the dominant undertaking had developed a de facto standard on its own. This situation is very different from that of an SEP-holder who has given a FRAND commitment as a quid pro quo for having its technology included in a standard and whose market power is derived principally from the standard’s adoption by the entire industry. It thus seems straightforward to argue that the giving of a FRAND-commitment – an essential safeguard to ensure that the standard-setting activity is compatible with Article 101 – is in itself the ‘exceptional circumstance’ under which an SEP-holder gives up the right to refuse to license its IP, in particular through the most obvious form of a refusal: the seeking of an injunction. In any case, in the vast majority of instances where an SEP-owner is seeking an injunction, even the narrow criteria of the case law developed for de novo refusals and de facto standards will be met:53 Refusal to license. As stated above, it is difficult to imagine a clearer form of a refusal to license a patent than seeking an injunction against the implementer based on that patent. The patent holder may see this as a negotiating tactic, but as long as an injunction application is pending (or as long as an injunction actually enforced), he cannot possibly argue that he is ‘willing’ to license when his actions show the contrary. Input that is essential to be active on a downstream market. As discussed above in section 3.1, by its very nature an SEP is essential to operate on a downstream market. Unlike in most other cases concerning refusals to license, this criterion 53 Moreover, even for such de novo refusals, the criteria listed above arguably are not exhaustive. In Case C-418/01 IMS Health [2004] ECR I-5039, para 38, the Court of Justice stated that the Magill criteria were ‘sufficient’ rather than ‘necessary’ for a finding an abuse. See also Commission Decision of 24 March 2004 in Microsoft, cited above note 9, para 555 (‘[T]here is no persuasiveness to an approach that would advocate the existence of an exhaustive checklist of exceptional circumstances and would have the Commission disregard a limine other circumstances of exceptional character that may deserve to be taken into account when assessing a refusal to supply.’); and the General Court’s 2007 judgment in Microsoft, cited above note 44, para 336 (‘In the light of the foregoing factors, the Court considers that it is appropriate, first of all, to decide whether the circumstances identified in Magill and IMS Health, paragraph 107 above, as described at paragraphs 332 and 333 above, are also present in this case. Only if it finds that one or more of those circumstances are absent will the Court proceed to assess the particular circumstances invoked by the Commission (see paragraph 317 above).’). Also relevant is Opinion of Advocate General Jacobs of 24 October 2004, Case C-53/03 SYFAIT v Glaxosmithkline, 2005 OJ C182/3, http://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1 396000924946&uri=CELEX:62003CC0053 para 68 (‘[T]he factors which go to demonstrate that an undertaking’s conduct in refusing to supply is either abusive or otherwise are highly dependent on the specific economic and regulatory context in which the case arises.’).

Sven B Völcker 

299  

should thus not raise any difficulty, at least not for ubiquitous and unavoidable standards such as wireless air interface standards. Substantial elimination of competition on downstream market. The General Court has held that, in refusal to deal cases, the relevant question is whether the refusal is liable to, or is likely to, eliminate effective competition.54 As the Commission states in the Guidance Paper on its enforcement priorities in applying Article 102 to exclusionary conduct, if an input is objectively necessary to compete effectively on a downstream market, ‘a dominant undertaking’s refusal to supply is generally liable to eliminate, immediately or over time, effective competition in the downstream market’.55 This presumption seems all the more appropriate in the case of assertions of SEPs, which by definition give the patent holder control over the entire downstream market. Prejudice to consumers. The ‘consumer harm’ prong of the established refusal to deal test is met where ‘the competitors that the dominant undertaking forecloses are, as a result of the refusal, prevented from bringing innovative goods or services to market’.56 This criterion is also satisfied where the SEP-holder seeks an injunction, as it denies consumers a product for which there is manifest demand. As the Commission noted in Google/MMI, even if the exclusion of products from the market is only temporary, in fast-moving technology markets it could cause serious harm.57

3.2.3 Injunctions as means of imposing ‘unfairly’ high (non-FRAND) royalties or other onerous terms on implementers As seen above, the seeking of an injunction plainly constitutes a refusal to license. In addition, injunctions brought by an SEP-holder should also be judged by their ultimate aim, ie putting pressure on the implementer to achieve terms more advantageous than those the implementer would have been willing to negotiate on an arm’s length basis, or indeed, more advantageous than the terms a court would have ordered if the SEP-holder had sought a judicial determination of the FRAND rate. The use of injunctions would in fact appear to be the perfect example of a mechanism of imposing unfair selling prices or other terms (such as extracting cross-licenses to differentiating IP). The Commission’s Horizontal Guidelines confirm that a FRAND royalty must be determined on an ex ante basis, ie on the basis of the patented technology’s value before the industry was locked into the 54 Microsoft, cited above note 44, paras 563–564. See also Case T-301/04, Clearstream Banking AG and Clearstream International SA v Commission [2009] ECR II-3155, para 148. 55 Guidance on its enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, 2009 OJ C45/7, para 85 (emphasis added). 56 Microsoft, cited above note 44, paras 643, 647, 648, 649, 652, 653 and 656. See also the Guidance Paper, cited previous footnote, para 87. 57 Google/Motorola, cited above note 15, para 107.

300  Competition, Regulation and Public Policies standard.58 By contrast, an SEP-holder who seeks an injunction many years after the standard is established by definition exploits the very hold-up power that results from the lock-in effect, and from the implementer’s dependence on the standard. However difficult an ex ante valuation may be to perform, it should be clear that the outcome of a negotiation conducted under the threat of an injunction will not in any way approximate the negotiated outcome that would have occurred under an ex ante valuation. An injunction is designed to produce a non-FRAND outcome, and is thus aimed precisely at securing the unfair terms that Article 102(a) does not allow a dominant SEP holder to impose on implementers of the standard. Substantially the same position has long been espoused in scholarly writing,59 and was recently adopted by the US Federal Trade Commission in its public interest statement before the US International Trade Commission. As the FTC explains, ‘a royalty negotiation that occurs under the threat of an exclusion order may be weighted heavily in favor of the patentee in a way that is in tension with the RAND commitment. High switching costs combined with the threat of an exclusion order could allow a patentee to obtain unreasonable licensing terms despite its RAND commitment, not because its invention is valuable, but because implementers are locked in to practicing the standard.’60 The fact that the use of injunctions is designed to achieve unfair conditions within the meaning of Article 102(a) is confirmed by arguments justifying injunctions as the only means of patent holders obtaining worldwide licenses to their entire SEP portfolio. The implication of such an argument is that damages actions before patent courts would yield significantly lower overall royalties, which as a matter of logic must be due to either: (i) patent protection not being available in some jurisdictions; or (ii) the SEP portfolio containing patents that would be found invalid or not infringed (ie, they would not truly be essential to any standard) if actually litigated. Seeking injunctions on a few patents and refusing to license them on FRAND rates unless the implementer also pays additional amounts for a worldwide license to other patents amounts to the imposition of unfair trading conditions contrary to Article 102(a) because the implementer is forced to pay for something that it does not want or need.61 See Horizontal Cooperation Guidelines, cited above note 8, paras 289–291. See, among others, Dolmans, ‘A Plea for Open Standards’, cited above note 43, at footnote 12; Mark Lemley and Carl Shapiro, ‘Patent Holdup and Royalty Stacking’, 85 Texas Law Review 2163 (2007); Suzanne Michel, ‘Bargaining For Rand Royalties In The Shadow Of Patent Remedies Law’, 77 Antitrust Law Journal 889 (2011). 60 Third Party United States Federal Trade Commission’s Statement on the Public Interest (6 June 2012), In re Certain Wireless Communications Devices, Portable Music & Data Processing Devices, Computers and Components Thereof, Inv. No. 337-TA-745. 61 See, eg, Case T-151/01 Der Grüne Punkt – Duales System Deutschland GmbH [2007] ECR II1607, in which the Commission and the General Court both considered that it was abusive to require a royalty payment for the use of a trademark when the licensee was not actually using the service denoted by the trademark. The Commission’s remedy required the dominant company to refrain from charging a ‘fee on the total amount of packaging bearing the green dot logo where it is shown that all or only some of that packaging has been taken back or recovered through another system’ (para 181). Requiring that the implementer license additional patents that it does not need may also amount to unlawful tying contrary to Article 102(d). 58 59

Sven B Völcker 

301  

3.2.4 Injunctions as discrimination Last but not least, the pursuit of injunctions to influence the outcome of negotiations between an SEP-holder and the implementer of a standard will in most cases also involve discrimination, as it will lead to the application of ‘dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage’. Typically, the SEP-holder will not seek to use injunctions systematically against all implementers of a standard. Rather, it will focus on the standard-implementers that are most vulnerable to an injunction, for example because they have significant sales in a jurisdiction where injunctions are particularly easy to obtain, or because they do not have a strong patent portfolio of their own that they could use to counter-sue the SEP-holder. The outcome of an injunction-influenced negotiation will vary from implementer to implementer, and will thus put at least some of them at a competitive disadvantage.

3.3 Objective justification for seeking injunctions on FRANDencumbered SEPs? Characterizing injunction applications for FRAND-encumbered SEPs as generally abusive does not preclude the possibility that they could exceptionally be compatible with Article 102 if they are ‘objectively justified’. Indeed, the Commission’s Google/MMI decision as well as its press releases accompanying the SOs against Samsung and Motorola Mobility all make reference to concerns about injunctions against ‘willing’ licensees (ie, licensees willing to pay a FRAND rate). This appears to suggest that injunctions against ‘unwilling’ licensees could be justified as a legitimate response by the IP-holder to protect its commercial interests in achieving ‘adequate’ remuneration for its IP and for its underlying investment. However, such an ill-defined justification would be difficult to square with the EU Courts’ case law and the Commission’s own policy guidance. The Union Courts have traditionally recognized a narrowly focused ‘protection of commercial interests’ justification for certain commercial conduct which can be characterized as ‘competition on the merits’ and which involves at least short-term consumer benefits.62 The ECJ’s Post Danmark judgment eschews any reference to 62 Case-T-24/93 Compagnie maritime belge transports SA and Others v Commission [1996] ECR II-1201, para 146; Case T-340/03 France Télécom SA v Commission [2007] ECR II-107, paras 185– 187; Case T-155/06 Tomra Systems SA and Others v Commission [2010] ECR II-4361, para 207; Case T-66/01 Imperial Chemical Industries Ltd v Commission [2010] ECR II-2631, para 295. See also AstraZeneca, cited above note 34, para 804 (‘It should be observed that the preparation by an undertaking, even in a dominant position, of a strategy whose object it is to minimise erosion of its sales and to enable it to deal with competition from generic products is legitimate and is part of the normal competitive process, provided that the conduct envisaged does not depart from practices

302  Competition, Regulation and Public Policies the ‘protection of commercial interests’ and instead stresses the consumer benefit requirement of any objective justification. In marked contrast, seeking to exclude a rival’s products from the market (and/or obtaining higher royalties) by means of an injunction benefits only the SEP-holder and causes consumer harm, at the very least in the short-term. It is precisely for this reason that the Commission’s Guidance Paper on exclusionary conduct recognizes investment protection arguments as an ‘objective justification’ in refusal to deal cases only in very limited circumstances.63 Indeed, under the Guidance Paper, a FRAND commitment given by the SEP-holder arguably precludes any investment-incentive defense from the outset. As the Commission states, concerns about undermining a dominant undertaking’s investment incentives are not in order where ‘regulation compatible with Community law already imposes an obligation to supply […] and it is clear, from the considerations underlying such regulation, that the necessary balancing of incentives has already been made by the public authority when imposing such an obligation to supply’.64 While standard-setting bodies will usually not be public authorities stricto sensu, the Commission’s wording nevertheless perfectly captures the FRAND ‘bargain’ required by the IPR rules of standard-setting organizations – FRAND terms being the quid pro quo for the substantial benefit accruing to the IP-holder when its technology is included in a standard. At any rate, it is incumbent upon the dominant undertaking to provide all the evidence necessary to demonstrate that the conduct concerned is objectively justified.65 In particular, ‘it falls on the dominant undertaking to demonstrate any negative impact which an obligation to supply is likely to have on its own level of innovation. If a dominant undertaking has previously supplied the input in question, this can be relevant for the assessment of any claim that the refusal to supply is justified on efficiency grounds.’66 For a host of reasons, it would appear very unlikely that an SEP-holder would be able to meet this evidentiary burden to justify an injunction. First, an SEP-holder would have to explain and substantiate its level of R&D investment in the technology it contributed in a bona fide way to the standard. Such a bona fide contribution would be in doubt where, for example, the SEPholder did not disclose its patent or patent applications in line with the SSO’s IPR policy before the standard was frozen. Independent studies show that, in past standardization processes, patent applications have surged in periods leading up to the finalization of standards, suggesting that opportunistic patenting is coming within the scope of competition on the merits, which is such as to benefit consumers.’ (emphasis added)). Guidance Paper, cited above note 55, para 89. Ibid, para 82. 65 Ibid, para 31. 66 Ibid, para 90. 63 64

Sven B Völcker 

303  

widespread.67 While determining an ‘appropriate’ return on R&D investment must generally also take into account the failure rate of such investment, it seems difficult to argue that work on major industry standards such as ETSI’s 3G and LTE standards constitutes a particularly risky activity, given the high degree of certainty that such standards will in fact be adopted and implemented by the industry. Second, any investment-justification defense for the use of injunctions to achieve an ‘appropriate’ level of royalties must take into account the substantial benefits of having one’s technology included in a standard, in particular for companies which themselves intend to manufacture standard-compliant products – typically the majority of those actively contributing their IP in the standardsetting process. It is in these companies’ interest to have their IP included in the standard irrespective of the higher royalties that they may hope to obtain by threatening injunctions for at least three reasons: (i) inclusion of the most powerful IP in the standard will accelerate the adoption and implementation of the standard and thus provide opportunities for all implementers; (ii) an IP-holder will want to favor the inclusion of its IP in the standard over any alternative technical solution that would be covered by another company’s IP even if it obtains only nominal royalties, simply because it will thus be able to avoid paying royalties to the other company when selling standard-compliant products; and (iii) obtaining an SEP portfolio for a given standard facilitates the negotiation of cross-licenses with other holders of SEP portfolios. The existence of powerful incentives to contribute technology to standards without any expectation of licensing revenue is illustrated by numerous successful standards that have developed under a royalty-free model, such as the Bluetooth standard initially developed by Ericsson and administered by the Bluetooth Special Interest Group.68 Third, even for companies that do not intend to practice the standard but nevertheless play a significant role in developing it, also by contributing their IP, it is not obvious that the ability to enjoin ‘unwilling’ implementers of the standard in the hope of obtaining higher royalties would necessarily foster innovation. Even ‘pure innovators’ will normally have an interest in standard specifications that will encourage rapid and universal adoption of the standard, even if they obtain relatively lower per-unit royalties. Fourth, the use of injunctions by certain SEP-holders to obtain higher royalties from certain implementers has an obvious impact on other SEP-holders that prefer an injunction-free standard, and thus their incentive to contribute IP to 67 Byeongwoo Kang and Rudi Bekkers, ‘Just-in-time inventions and the development of standards: How firms use opportunistic strategies to obtain standard-essential patents (SEPs)’, Working Paper 13.01, presented at 5th ZEW/MaCCI Conference on the Economics of Innovation and Patenting, Mannheim, 3-4 June 2013 (describing the phenomenon of a spike of patent applications just before or during the meetings of technical committees). 68 See http://www.bluetooth.com/Pages/Bluetooth-Home.aspx. Other examples of ‘default royaltyfree’ standards that have enjoyed widespread adoption include the DOCSIS and EuroDOCSIS standards used for the transmission of broadband over cable television networks, the Universal Serial Bus standards promulgated by the USB Implementers’ Forum, and the various web services standards issued by the Worldwide Web Consortium (W3C).

304  Competition, Regulation and Public Policies that standard, in that: (i) successful injunctions prevent the sale of devices and thus reduce royalties due to those IP-holders which negotiated royalties with that implementer without threat of injunctions; and (ii) higher royalties that one SEP-holder may ultimately obtain by leveraging the threat of injunctions reduce other SEP-holders’ royalties because they affect the implementer’s ability to pay royalties to such other SEP-holders and/or (to the extent that the implementer is able to pass on the increased royalties) make the implementer’s standardcompliant devices more expensive, thus reducing consumer demand for them. The highly ambivalent effects of injunctions obtained by one SEP-holder on overall investment incentives in a standard-setting environment should give competition authorities additional pause to accept any investment-incentive justification in the context of injunctions. Fifth, even if an SEP-holder were able to demonstrate that it has legitimate investment incentives worthy of protection, it would also have to show that there are no more proportionate means than injunctions to obtain FRAND royalties from companies perceived to be ‘unwilling’ to pay them – in particular, actions for FRAND-level damages in national courts. Essentially, the SEP-holder would have to show that the judicial system fails to offer patent holders effective protection through damages as opposed to injunctions. At least within the EU, it is difficult to see how such an argument could succeed. Articles 13 and 14 of the EU’s 2004 IP Enforcement Directive69 require Member States to ensure that the competent judicial authorities order an infringer to pay the right-holder damages ‘appropriate to the actual prejudice’ it has suffered. Article 14 of that Directive also provides for the ‘losing party pays principle’ to ensure that, unlike in some non-EU jurisdictions, the right-holder does not have to bear the legal costs of seeking redress before the courts of the Member States. The foregoing considerations show that, under Article 102, there is little room for any ill-defined ‘unwilling licensee’ justification for otherwise abusive refusals to license in the form of applications for injunctions in cases where the SEPholder is simply not satisfied with the royalty offered by the implementer. As with any objective justification under Article 102, the burden is on the dominant IP-holder to show not only why its own demand for royalties conforms with the FRAND principle but also why validation of its views of the FRAND rate by national courts in damages proceedings is not an appropriate alternative means of redress. *** In conclusion, the pursuit of injunctions by a dominant SEP-holder notwithstanding its FRAND commitment should be considered an abuse under Article 102 TFEU in most situations. Possible exceptions may exist, but the SEP-holder must be able to show a compelling justification. Simple disagreement as to whether the royalty 69

Cited above note 52.

Sven B Völcker 

305  

rate proposed by the SEP-holder is excessive is clearly insufficient, as judicial determination of the proper FRAND rate – without the threat of an injunction – will always be a more proportionate way for the SEP-holder to obtain adequate remuneration. A FRAND promise that allows for injunctions is a meaningless promise.

Cristina Caffarra and Pierre Régibeau*

Patent Explosion and Patent Wars: Hold-Up, Royalties and Misunderstandings over ‘Market Value’

1. Motivation and overview ‘Patent wars’ have now been raging for a few years and signs of appeasement are few. Following Microsoft’s assertion of its ‘ActiveSync’ patents, Apple’s litigation based partly on its ‘design’ patents and design rights, and the Rockstar consortium acquisition of Nortel patents Google acquired MMI to get hold of MMI’s mobile communication patent portfolio and continue ongoing litigation against Apple and Microsoft. This has led to a flurry of suits and countersuits and injunctions that have reached mainstream news – notably with the granting of an injunction against Samsung’s Galaxy products in the US, decisions by Judge Posner on the appropriateness of granting injunctive relieve to SEP owners and Judge Robart’s proposed methodology for the computation of FRAND royalties. By now, patent wars have engulfed a number of other players, including Samsung, Huawei and Nokia, and have spread to other jurisdictions such as South Korea and China. The recent patent wars are noteworthy because they are clearly part of a broader battle for supremacy between platforms in a converged mobile environment, with large asymmetries persisting between the parties involved: device manufacturers with a stash of patents deemed ‘standard essential’ because they have been involved early in the communication protocols; Microsoft with its traditional IP strength in software; Google, coming from the open source movement and therefore with virtually no patent portfolio until recently; Apple, very good at design and integration but less so at purely ‘technological’ innovation, and therefore with a very different patent portfolio. The convergence of the industry means we are really facing a battle for who gets the rents in combined devices, and this is what is making the problems so thorny – as there is a concern that parties may be using their IPRs as tools to exercise market power and raise rivals’ costs (or slow down their progress) in the market for smartphones and tablets. The hope that all will eventually reach patent settlements on their own and that ‘everything will be alright on the night’ appear to some overly optimistic. And * Respectively: Vice-President, CRA; Vice-President, CRA and Research Fellow, Imperial College. While CRA represents a number of parties involved in this debate, these are the personal views of the authors and do not in any way reflect the position of CRA clients

308  Competition, Regulation and Public Policies consumers might be harmed before the industry settles into some new, more peaceful equilibrium. The potentially harmful consequences of patent wars are magnified for industries that are heavily reliant on standards. Standards can be successful only if two conditions are met: firms that control important related technologies have to be induced to take part in the design of the standard; and access must be given to the standard on terms that allow for a competitive market for devices while insuring that the original innovations are properly rewarded. An all-out patent war makes it increasingly unlikely that those conditions can be fulfilled. On the one hand, firms that are typically very active in standard setting may have an incentive to use their ‘standard essential’ patents as weapons in the broader IP conflict, denying proper access to the standards to some of their rivals. On the other hand, firms with a lesser stake in SEPs have an incentive to claim that their rivals are indeed abusing the market power that the standard-setting process bestowed upon them, in the hope that regulators will then limit their rivals’ ability to rely on their SEPs to settle broader disputes. Because of the potential consumer harm, competition authorities have waded into this – albeit in a rather patchy manner. The main focus so far has been the potentially abusive role of SEP enforcement, with DG Comp recently opening investigations (eg, of the licensing practices of MMI and Samsung), spurred by a number of complaints, ‘to assess whether [they] abusively, and in contravention of a commitment they gave to the European Telecommunications Standards Institute (ETSI), used certain of their standard essential patent rights to distort competition in European mobile device markets, in breach of EU antitrust rules’. There are a number of complaints against other parties, and several investigations underway at the FTC and the DOJ. As they focus on SEPs, these investigations naturally run into the problem of meaningfully defining what the ‘commitments’ entered into by SEP holders actually are. In practice, this means putting some content behind the usual requirement that SEPs be licensed at Fair, Reasonable and Non-Discriminatory (FRAND) terms. But there is little consensus about what FRAND terms are. For example, while there is broad agreement that the royalties charged to a licensee should not reflect the additional market power that the patent’s inclusion into the standard confers, there is no accepted methodology to translate this principle into actual royalty levels. Because of this, both academics and competition authorities have been turning their attention to the process through which SEP holders and potential licensees might settle: if one does not know what the ‘right’ FRAND royalty should be, then it is important for the process through which actual royalties are set to be as transparent and balanced as possible. This explains the current focus on the use of injunctions in the investigations conducted on both sides of the Atlantic. As part of these investigations, some parties have also raised concerns about the fairly common practice of setting a royalty based on a percentage of the final market value of the whole product in which the relevant standard has been incorporated.

Cristina Caffarra and Pierre Régibeau 

309  

In their view, such a practice imposes an ‘unjustified tax’ on unrelated innovation developed by others. Consequently, SEP holders’ insistence on applying their ‘traditional’ royalty rates to producers of converged devices is seen as being ‘not FRAND’. The purpose of this chapter is to explore the economic foundations of this concern. We start by briefly sketching in Section 2 the reasons which have driven regulators to consider SEP-related issues on both sides of the Atlantic. We then discuss in Section 3 some economic insights on the question of sharing rents from complementary innovations. Based on these insights, we conclude that claims that ‘the market value of a device is not the appropriate base on which to calculate a FRAND royalty’ are misplaced, as are corresponding concerns about the choice of royalty base in agreements between SEP holders and licensees. Economic analysis is clear that original innovators need to appropriate a share of the value of follow-on innovation, in order for incentives to innovate to be preserved. Properly understood, the issue of whether a given rate is ‘FRAND’, and the issue of whether one should carve out functionalities before applying a given rate, are all about the share of value that original innovators can legitimately appropriate from follow-on innovations. The practical question is how that value can be pragmatically arrived at, and whether there may be feasible, reasonable ways of limiting rent extraction in a way that does not distort incentives for innovation on all sides. We evaluate some of the ideas being discussed (and their limitations) in Section 4.

2. holding up complementary innovation through IPR enforcement? 2.1 The hold-up problem Progress in the ICT sector typically takes the form of incremental innovations which build on the existing stock of products and features. Where complementarities are pervasive, and each component is protected by patents, the ‘next’ inventor making decisions on its own investment in patents or R&D may be discouraged if it faces someone with a large existing patent portfolio, from whom it may be difficult to license complementary technology. Without ex ante commitment to licensing, or the ability to negotiate ex post effectively with the patent holder, new investments can thus be held up. Standards also complicate matters. They are desirable because they ensure compatibility and interoperability, but the hold-up problem can be particularly severe where standards are concerned because standards involve perfect complementarity. FRAND commitments were designed to solve the hold-up problem, as a commitment against exploitation ex post; however, as noted earlier it is not clear what exactly FRAND means. Consequently, so the FRAND concept

310  Competition, Regulation and Public Policies does not solve the problem of limiting extraction (and therefore the potential hold-up). But why is this not simply an ex ante contracting problem? If standard-setting organisations write the right contracts and those contracts are then properly enforced by the courts, in principle we should be able to get around the hold-up problem. Why then have these problems not been resolved effectively, and why are regulators now thinking of using competition law to intervene? The reality is that the contracting problem is very difficult when there are many parties with technologies which are complements when it comes to setting a standard but which are incorporated in products, such as smart phones or tablets, that are actually competing hard in final markets. As different firms are likely to have different relative strengths in the technology markets and in the product markets, SSOdetermined rules for the licensing of SEPs will always be compromises that might not fully address the SSO-related hold up issue. The SSO rules themselves should therefore be seen as the result of a battle between parties trying to appropriate the rents created by the SSO’s activities. In other words, it should be no surprise if SSO rules are as much about who gets the rents from whom as about preventing SEP holders from exploiting the market power that they derive from the inclusion of their patents in a given standard. One should add that, in practice, SSOs also face a tradeoff between setting and enforcing rules that prevent the ex post exploitation of the additional market power granted by the standard-setting process and the need to get all firms with important patents to take part in this process. Given that SSOs have no power to regulate the behaviour of firms that decide not to participate, they would not in general be able to design contracts that enforce the first-best outcome. Faced with such shortcomings, regulators also appear to have developed a concern about the judiciary’s own ability to deal effectively with standard-related hold-up. Because of the SSOs’ own limitations, the task of the courts is not simply to ensure that some well-defined contract is enforced. Rather, they often find themselves in a position where they have to: determine what the obligations of SEP holders were or should have been; ensure that their own procedures (such as injunctions) do not allow one of the parties to gain the upper hand in negotiations towards an eventual settlement; and may eventually have to rule on what the elusive FRAND conditions for licensing actually are. At the very least, there is a sense that national courts do need some form of guidance in carrying out such a difficult set of tasks.

2.2 The FRAND/base problem In addition to injunctions (which are not discussed further here), one aspect which is attracting attention is the question of the ‘base’ on which the percentage royalty should be calculated – ie the ‘value’ to which the percentage should apply. This

Cristina Caffarra and Pierre Régibeau 

311  

is not a new problem: there is a long recognition that rate and base need to be ‘commensurate’ – ie the higher the base, the lower the percentage and vice versa – after all, what matters is the overall royalty payment which is made. We also know that there is great heterogeneity in commercial negotiations between parties about the structure of the royalty payment (often starting from a percentage royalty, and then progressing to lump sums, combinations of ongoing royalties and lump sums, compensatory payments, reductions in cash royalties based on reciprocal patent licences or licences for other products, and co-operation/technical support for a whole portfolio). Still, the issue which has attracted concern – with the decline of feature phones and the explosion of combination devices and of tablets – is that the value of these devices has increased significantly as a result of the inclusion of features which are technically unrelated to technologies which read on the SEPs in question. Does it follow that a royalty calculated on the overall market value of the product is a way of extracting undue rent from any other innovation that is unrelated to the standard itself? This concern is said to remain (in some form) even when the percentage opening position in discussions with potential licensees in practice is superseded in negotiations. Even though the final terms are the result of negotiation, the royalties realised per unit tend to be positively related to the net price of licensee products that make use of the patented technology. A royalty structure that provides the patentee with higher royalties per unit on higher-priced products – the concern goes – effectively allows him to realise a share of the added value created by the introduction of features that have nothing to do with the patented technology. Experts for the patent holders in litigation have responded that this argument ignores that the value of additional product features can be affected by the foundational technology, even if the technology that enables the additional feature does not directly depend on the foundational technology. These experts have explained that synergies can exist between the original patents and the gross profits that a licensee realizes from ‘non-infringing features’. These features can themselves be divided into two subgroups. On the one hand we have what we will call ‘dependent features’ which are those that, even though they do not infringe the SEPs could not be enabled without the technology covered by the SEPs. For example, a feature that relies on the availability of large quantities of data downloaded from the web would be of no interest without access to the patents that covers such data download. On the other hand we have what we call ‘symbiotic features’, which are features that could be used without access to SEPs but which become more attractive if access to the technologies covered by SEPs is secured. For example, viewing videos on a smart-phone is much more enjoyable if one can use the latest communication standard than if one has to rely on an earlier, less powerful, specification. One of the experts (who shall remain unnamed) provided the following example relating to the synergies between the cellular and Wi-Fi technologies covered by certain mobile patents and the various features of Apple’s iPhone:

312  Competition, Regulation and Public Policies [C]ellular and Wi-Fi capabilities are central to consumer demand for the iPhone. Without those capabilities, the iPhone would essentially be a music player/camera device. Apple could not charge the same price for such a device as it charges for the iPhone. In addition, it is unlikely that such a device could have achieved the same level of sales as the iPhone (even at a lower price).

Thus, while this problem is not new in the context of patent litigation, it is one of the aspects of the issues which are now considered by regulators. In the remainder of chapter we consider this issue in particular.

3. On the question of reasonable royalties and royalty base 3.1 Overview Expressed in economic terms, the concern is that imposing a variable royalty on the whole value of devices which incorporate multiple functionalities amounts to ‘taxing’ value-creating components of such devices that are not directly related to the patentee’s IP. The broader policy concern is that a royalty scheme that defines payments on such a broad base decreases incentives to invest in future improvements of the devices covered by the licensing agreement. Our discussion below explains that: – First, it is important to distinguish clearly between the total (expected) level of royalty payments made by the licensee, and the royalty base that merely determines how this total payment is made. Because the choice of royalty base is independent of the total amount received by the licensor, it is incorrect to conclude that using a broad royalty base gives rise in itself to an illegitimate ‘tax’ on the licensee’s income. – Secondly, the appropriate measure of a technology’s contribution to a complex device – such as a smartphone – is likely to be rather broad. Stated most generally, the contribution equals the difference between the licensee’s actual profits and those it would have realised if its products had not had access to the patented technology. On an ex post basis, this means considering the reduction in profits that the licensee would suffer if it were no longer permitted access to the patented technology. On an ex ante basis, this means considering the difference between the licensee’s actual profits and the profits it would have realised if the relevant standards had been based on the next-best alternative to the patented technology. Either way, a technology’s contribution would have to include not only the ‘standalone’ value of product features that depend directly on (and could not operate without) the patented technology, but also the additional value of features that are enhanced by the patented technology. Moreover, the economic contribution of the technology would also have to include additional value obtained from features that are independent of the

Cristina Caffarra and Pierre Régibeau 

313  

technology, to the extent that total sales of the device are increased by the presence of the technology. Indeed, even complementary products that are not physically connected to the device that embeds the infringed technology should also enter into the proper valuation of the technology’s contribution, to the extent that larger sales of the device promoted by the presence of the technology also generate extra sales of complementary goods and therefore additional profits for the licensee. We further argue that such an approach to the economic contribution of the infringed technology, which implies that the licensee can have a right to participate even in the licensee profits generated by features of the device that do not depend directly on the patented technology, is in the same spirit as the conclusions from the extensive economic literature on sequential innovations. The overwhelming conclusion of this literature is indeed that the first innovator should be given a share of the surplus created by followon innovators. – Thirdly, because the economic contribution of a technology is likely to vary according to product revenues, there are good economic reasons to use product revenues as the royalty base for determining how royalties should be calculated. Royalties calculated on this basis favour efficient risk-sharing and can help reduce asymmetric information. In that sense, a broad economic contribution directly implies payments metered on a broad royalty base. – Finally, we turn to the effects of a royalty payment scheme on incentives to innovate. It is inevitably the case that any payment scheme that expresses royalties as a percentage of downstream sales and/or profits will have some negative effect on incentives to innovate by downstream firms – even though these negative effects will be worse when royalties are based on profit than when they are based on sales.1 Importantly, the literature on sequential innovation suggests that, taking into account both the value of initial (upstream) innovation and the value of subsequent (downstream) innovation, innovation incentives for the economy as a whole are optimised when the initial (upstream) innovator is provided with a share of the incremental value associated with the subsequent (downstream) innovation. We also show that, in the case of a device that incorporates both features that are enhanced by the technology and other features that are not, the overall negative effects of variable royalties on downstream innovation are not necessarily smaller if the payment is computed on sales/profits attributable only to the enhanced feature rather than on sales/profits for the whole device. Thus, there cannot be a presumption that broad-based royalties are more harmful to downstream innovation than more narrowly focussed schemes. We discuss finally how regulators may be focusing on the issue of the ‘royalty base’ as a springboard for some broader thinking about the design of patent rights. 1 This is because a royalty based on profits also captures some of the return from cost-reducing efforts on the part of the licensee.

314  Competition, Regulation and Public Policies

3.2 Optimal compensation of upstream patent holders In a patent system that encourages innovation by providing successful innovators with rewards, it is desirable for the reward to be related to the economic value of the innovation. This raises two questions: how should the economic value of an innovation be measured; and what is the optimal share of this economic value that should go to the patent holder? These questions are addressed below. As part of this analysis, we discuss the issues raised when a downstream licensee increases the value of its products by adding ‘non-infringing features’ – ie features based on technology where there is no direct connection between this technology and the technology of the patent holder. As discussed above, such features might be either enabled or simply improved by the technology covered by SEPs.

3.2.1 Economic value of the infringed innovation From an economic point of view, the point of departure for any analysis of reasonable royalty payments must be the contribution of the licensed innovation to the licensee’s profits. The basic principle is straightforward: this contribution is equal to the licensee’s profits with access to the innovation, minus the licensee’s profits without access to the innovation. The difficulty is of course to determine what might have occurred without access, ie to get the correct counterfactual. In principle, the analysis of the economic value of an innovation should capture every possible channel through which the use of the technology might have affected the sales and profits of the infringer. While implementing this approach is never easy, this ‘overall profit’ approach is extremely useful as a framework in that it helps us identify the various types of effects that should be included in a meaningful assessment of the economic contribution of an innovation. This economic contribution of an innovation clearly involves any increase in the revenues obtained from selling the products that embody the innovation. Such extra revenue comes both from a price-increasing effect and a quantity-expanding effect: each individual might be willing to pay a higher price for the products with the innovation; and, for a given price, more sales of the products will be made. Methodologies that just look at the price difference between a product with the infringed innovation and one without are therefore incomplete, since they do not take into account the consideration that the infringer has likely exploited some of the advantage conferred by the innovation to expand its sales rather than increasing the price of its devices.

Cristina Caffarra and Pierre Régibeau 

315  

3.2.2 The economic value of an innovation when the licensee adds ‘unpatented features’ Let us now consider a situation where the infringing product consists of multiple components, sold as a whole, but where the patent in question reads only on a subset of the product components. As shorthand, let us refer to the other product components as ‘non-infringing features’ that are the result of downstream innovation. The introduction of non-infringing features and downstream innovation to the analysis has no effect on the general principle for how the economic value of an innovation should be measured. The same principle applies: the economic value of the upstream patented innovation continues to equal the difference between (a) the licensee’s profits when it has access to the patented technology and includes the unpatented features and (b) the licensee’s profits when it does not have access to the patented technology and pursues its next best alternative. To explain further, let V(U,D) equal the profits of the downstream infringing firm with access to the upstream and downstream innovations. Let V(U’,D) equal the downstream profits with access to the downstream innovation but without access to the upstream patented innovation; U’ is the best non-infringing alternative to the patented upstream technology. Let V(U,0) equal the downstream profits with access to the upstream technology but without the downstream innovation, and let V(U’,0) equal the downstream profits using the best non-infringing alternative to the patented upstream technology (but without access to the downstream innovation). Using this notation: • V(U,0) – V(U’,0) can be described as the standalone value of the patented upstream technology. It represents the increase in the value of the downstream infringing product created by access to the patented upstream technology, in the absence of the downstream innovation. • V(U,D) – V(U’,D) can be described as the incremental value of the patented upstream technology (assuming access to the downstream innovation). It represents the reduction in value if the downstream infringing product had access to the downstream innovation but had to remove the upstream patented technology. Adding and subtracting the standalone value to this expression for the incremental value, the incremental value of the patented upstream technology can be described further as: [V(U,0)-V(U’,0)]+[V(U,D)-V(U,0)]-[V(U’,D)-V(U’,0)] This re-expression highlights that if the downstream innovation adds more value when the upstream patented technology is present – ie if [V(U,D)V(U,0)]>[V(U’,D)-V(U’,0)] – then the incremental value of the patented upstream technology is greater than the standalone value of the technology and includes

316  Competition, Regulation and Public Policies some portion of the added value created by the downstream innovation. Put differently, if there are synergies between the downstream innovation and the upstream patented technology, these synergies can be regarded as part of the incremental value of the upstream patented technology.2 Note that in determining the portion of the value created by the downstream innovation that should be included in the incremental value of the patented upstream technology, it does not matter whether the downstream innovation is technically related in some way to the patented upstream technology. The only relevant consideration is whether the downstream innovation adds more value when the patented upstream technology is present. If the answer is ‘yes’, then the incremental value of the patented upstream technology is greater than its standalone value. A special case arises when, absent the infringed technology, the infringer would simply not have entered the market for the infringing device. In this case, the ‘extra sales’ imputable to the infringed technology are simply all sales, so that the economic value of the infringing technology is properly seen as the total profits obtained from the sales of the infringing devices.3 Note that this logic can also be applied in the case where the device that embeds the infringed innovation is usually sold with a number of other peripheral products. One can think of computer screens or printers that work with a PC, various external add-ons to smartphones (eg, a headset) or even the applications or content that can be displayed on infringing devices. Given that these peripheral products are clearly complementary to the device itself, the essential economic situation is really the same as in the case of multiple components embedded within a single device. At the very least, the infringed technology is the proper source of the additional profits that the infringer – if it sells those goods – makes from selling a larger number of these peripheral products because he/she sells a larger number of the infringing devices that trigger a demand for these products. Moreover, the increase in sales and/or prices of these devices due to a possible enhancement of their performance because of the infringed technology should also be part of a proper economic evaluation of the contribution of the infringed technology. It is also worth mentioning that this type of indirect contribution will be larger if the infringer sells its own proprietary system, as the lack of (perfect) compatibility with peripheral products made by others ensures that the infringer will capture the bulk of this correlated increase in peripheral product sales. The principles that we have just laid down in general terms are illustrated in the following example presented in figure 1. With full functionality, the smartphone 2 Downstream innovation also requires compensation, and therefore the introduction in the analysis of non-infringing features may have implications for how the economic value of an upstream innovation is shared. This subject is discussed further below. 3 In terms of the notation above, where the incremental value of the patented upstream technology equals V(U,D) – V(U’,D), if V(U’,D) = 0, the incremental value of the patented upstream technology simply equals V(U,D).

Cristina Caffarra and Pierre Régibeau 

317  

sells for 400 and generates linked sales of 300. If one were to eliminate the ‘dependent’ functionalities, ie those that crucially depend on the SEPs, the smartphone would retail for 300 and generate only 200 worth of linked sales. Finally, if one also considers that the performance of some other ‘symbiotic’ features would be lessened without access to the SEPs, the value of the phone drops a further 50, as does the value of linked sales. Overall, then, the total revenues from the sale of the phone and the linked sales would decrease by 150 + 150 = 300 if access to the SEPs could not be secured. This number represents the proper economic contribution of the SEPs. It is therefore this amount that needs to be divided up between licensor and licensee through negotiations. In this precise sense, then, this number is the proper base for the determination of royalty payments: the higher this base, the higher the royalty payments to SEP-holders should be.

Entire Market Value rule Overall, then, if there are synergies between the downstream innovation and the patented upstream technology, the economic contribution of the infringed technology will always include some elements (such as the increase in the total sales of the device) that relate to the value and sales of the whole device. In this sense, there is in fact a very robust economic logic for using what has become known in the debate before U.S. courts as the ‘Entire Market Value’ (‘EMV’) rule to determine the economic value of an infringed innovation.

318  Competition, Regulation and Public Policies The EMV rule essentially adopts, as the relevant base for royalty calculations, the market value of the infringing product in all cases where there is a reasonable presumption that the infringed innovation is ‘at the core’ of the infringing product, and is therefore a major determinant of that product’s overall market value. In this sense the EMV rule is closely linked to the above discussion of the economic contribution of an innovation. Where some applications of EMV appear to go wrong, and where the EMV rule understandably has attracted criticism, is in failing to draw a distinction between whether an innovation is ‘at the core’ of the infringing product and the incremental value of the innovation. Even if an innovation is ‘at the core’ of an infringing product, the proper measure of the economic contribution of the innovation remains the incremental effect of the innovation on the firm’s EMV – ie, the difference between the firm’s value with access to the technology in question, and its value if it does not use the relevant technology and pursues its next-best alternative. This principle of measuring economic contribution based on incremental value can be described as the Entire Marginal Market Value (‘EMMV’) rule.

3.3 The appropriate level of compensation for an IPR owner EEMV provides the correct basis for measuring the economic value of an innovation. The more difficult subject, both conceptually and practically, concerns going from an appropriate valuation of the infringed innovation to an appropriate level of compensation for the patent owner. One issue that arises in the context of SEPs (in a situation where the patents are truly essential) is that, because value is zero without access to any of the SEPs, the ex post incremental value of each patent equals the entire marginal market value of the downstream infringing product. This highlights a conundrum when there are complementary upstream patents – paying a significant share of incremental value to each upstream patent holder may exceed the total value of the infringing downstream product. In this situation, some rule for sharing the value created by the set of SEPs as a whole will be required. Allocating value across complementary upstream patents is a difficult problem that is beyond the scope of this chapter. Here we focus on the situation where a downstream infringing product requires access to a (single) upstream patented technology and demand for the downstream product is also enhanced by downstream innovation. We have already discussed how the incremental value of the upstream technology should be calculated in this situation: the incremental value of the upstream technology equals the difference between the value of the downstream product with access to both the upstream and downstream technologies, and the value of the downstream product without access to the upstream technology.

Cristina Caffarra and Pierre Régibeau 

319  

By the same logic, the incremental value of the downstream technology equals the difference between the value of the downstream product with access to both the upstream and downstream technologies, and the value of the downstream technology without access to the downstream innovation. Given these incremental values, and given the need to provide incentives for innovation at both the upstream and downstream levels, what can we say about the compensation that should be provided to upstream and downstream innovators? In the absence of any downstream innovative activity on the part of the infringer, or any upstream complementary patents, incentives to innovate would be maximised by allocating the whole EMMV to the upstream patent holder. In practice, however, the downstream licensee may be able to capture through bargaining some of the added value made possible by the patented innovation. But if the downstream device incorporates other innovations that are not covered by the infringed patent, one must ensure that these innovations are also properly rewarded. Building on the economic literature discussing sequential innovation, it is clear that, in order to encourage socially optimal innovation at the upstream level, the upstream patent holder should benefit from the increase in demand for the infringing product made possible by the downstream innovation – even if the downstream innovation is unrelated to the upstream patent in a technical sense. The problem of providing adequate incentives for upstream innovation in foundational technology and downstream innovation in product features is analogous to the problem analysed in the economic literature on sequential innovation where later firms innovate ‘on top of’ earlier innovations. The robust conclusion of this literature is that social welfare is maximised by allowing the enabling innovator to capture the full standalone value of its innovation (ie the value that the initial innovation would create in the absence of the follow-on innovation) plus a fraction of the further value from the follow-on innovation that is enabled by the initial innovation.4 While we do not provide here a review of this literature, a major insight is that it is socially efficient to increase the reward of the first innovator by reducing (by at least some amount) the private reward of second innovators. The intuition for this result can be seen most clearly in the case where the first innovation has no commercial value by itself but makes the development of commercially valuable follow-on innovations possible. If the initial innovator did not get part of the private reward generated by the follow-on innovation, there would never be any innovation.

4 See Jerry Green and Suzanne Scotchmer, ‘On the Division of Profits in Sequential Innovation’, 26 Rand Journal of Economics 20 (1995); Carmen Matutes, Pierre Regibeau and Katharine Rockett, ‘Optimal Patent Design and the Diffusion of Innovations,’ 27 Rand Journal of Economics 60 (1996); Ted O’Donoghue, ‘Patentability Requirement for Sequential Innovation’, 29 Rand Journal of Economics 654 (1998); Vincenzo Denicolò, ‘Two-stage Patent Races and Patent Policy’, 33 Rand Journal of Economics 488 (2000); Robert Hunt, ‘Patentability, Industry Structure and Innovation’, 52 Journal of Industrial Economics 401 (2004); Katharine Rockett, ‘Property Rights and Inventions’, in Bronwyn Hall and Nathan Rosenberg, eds., Handbook of the Economics of Innovation, Vol. I, NorthHolland, 2010, Chapter 7.

320  Competition, Regulation and Public Policies

3.4 The form of payment and the royalty base The previous section showed that, if there are synergies between downstream innovation and upstream patented technology, the incremental value of the upstream patented technology includes these synergies. A separate question is how these synergies should be divided between upstream and downstream innovators. Building on the literature on sequential innovation, the previous section also showed that the upstream patent holder should realise some portion of any synergies between the upstream and downstream innovations in order to ensure optimal innovation incentives at both the upstream and downstream levels. We now assume that the total amount of compensation to the upstream patent holder has been determined, and that T is the total (expected) reward that the infringed patent holder ought to receive. We then consider whether the form of this payment matters, in particular with respect to innovation incentives at the upstream and downstream level. While the royalty payment can of course be made according to a large variety of formulae, we will restrict ourselves to simple and commonly used schemes: lump sum royalty payments, per-unit royalties and royalties expressed as a percentage of some royalty base (eg, the value of downstream sales). In considering this question, note first that one would expect upstream patent holders and downstream licensees to settle on the form of payment that maximises their joint surplus since, in a bargaining framework, this would ensure that both parties end up better off than if they had settled on any other form of payment. Thus, one would expect both parties – the upstream patent holder and the downstream licensee – to treat all aspects of the form of payment as subject to negotiation. This includes the size of lump sum payment, the royalty rate (if any) and the base to which this rate is applied. It would be a mistake to think of some aspects of the payment form, such as the royalty base, as being ‘imposed’ by the licensor on an unwilling licensee. If the joint surplus could be increased by changing the form of the payment, it would be in the mutual interests of the parties to do so. Applied to this case, this ‘Coase Theorem’ observation implies that if royalty payments tend to be calculated by multiplying a royalty rate times the value of downstream sales, there should be a presumption that this is economically efficient. We should also add that negotiations between the licensor and the licensee tend to occur after the licensor has made most of the investment that is likely to affect the total expected joint surplus from the relationship, while it seems natural that the licensee has not yet sunk as large a share of the investments that will help determine the total surplus obtained over the course of the agreement. This implies that, when agreeing on the terms of the licensing contract – and in particular when agreeing on the form of payment – the parties will pay significant attention to the potential effect of such terms on future innovation by the licensee, and will mostly disregard any potential effect on the licensor’s own incentives to innovate. The concern that the form of licensing payment agreed upon might

Cristina Caffarra and Pierre Régibeau 

321  

unduly restrict further innovation by the licensee appears therefore rather odd. Note further that, even if the upstream patent holder ‘calls the shots’ and is not willing to negotiate over the form of the license agreement, imposing a royalty base that would restrict future innovation by the licensee would not be in the unilateral interest of the licensor. Even with complete bargaining power, the licensor has an incentive to maximise the size of the surplus that it can appropriate. Building on this backdrop, we now consider how the form of a royalty payment might affect joint surplus. It is first useful to highlight the case where the features of the downstream product are fixed and there is no further downstream innovation (‘static effects’). We then consider the possibility of further downstream innovation.

3.4.1 Static effects A static framework (where all features of the infringing device have already been developed) is still useful in order to isolate two main mechanisms through which the form of payment might affect the total surplus shared by the two firms – so that some forms of payment might actually be privately preferred to others. The first mechanism is the potential effect of the royalty payment on the marginal cost of production of the infringing devices. With lump sum payments, or with a royalty-free cross-licensing agreement, the licensing contract does not affect the marginal cost of production of the downstream firm, so the total amount that the licensee must pay has no direct effect on the price of the relevant devices. On the other hand, any scheme such that the total amount paid increases with the number of units of the devices sold increases the marginal cost of production of the licensee and leads to higher consumer prices. This is true whether we consider a per-unit royalty or a royalty expressed as a percentage of a base (such as total sales) that increases with output (over the relevant range). The effect of an increase in marginal cost on the joint surplus available to the licensor and licensee depends on whether or not the licensor competes with the licensee in the relevant downstream markets. If the licensor and licensee do not compete with each other, then the joint surplus is maximised by making the licensee as competitive as possible, ie by choosing a payment scheme that does not increase its marginal cost.5 The parties would therefore both be better off with lump sum payments or royalty-free cross licensing. If, on the other hand, the licensor and licensee are rivals in the downstream markets, then their joint profits might be enhanced by raising the costs of the licensee in order to decrease the intensity of downstream competition. If so, then we would expect the parties to agree on some form of output-related royalty base. The second mechanism relates to risk aversion. Realistically, how much the infringed innovation will actually contribute to the profits of the licensee is 5 We are dealing with the single relationship between the licensee and the licensor. The situation where the licensor licenses a number of firms that compete with each other downstream would differ.

322  Competition, Regulation and Public Policies uncertain. The parties might agree about the expected value of the EMMV, but its actual realisation cannot be known for sure when the licensing contract is signed. If the parties are risk-averse, then they would prefer not to assume the full risk associated with this uncertainty. In particular, the licensee might be reluctant to settle for a single upfront payment and be left exposed to the risk that the EMMV will actually turn out to be much lower than expected. In such a context, the total surplus to be shared between licensor and licensee is maximised by allocating risk between parties so as to minimise their joint ‘disutility’ from risk-aversion. Typically, this is achieved by agreeing a form of payment that ensures that the licensee pays more if EMMV turns out to be unexpectedly high, and pays less if it turns out to be unexpectedly low. This is best achieved by choosing a royalty base that is closely linked to EMMV, and applying a percentage royalty to that base. A closely related mechanism arises because of asymmetric information between the licensor and the licensee. If the licensee has a better idea of how useful his technology is likely to be, then a licensor who is confident in his technology should be willing to ‘put his money where her mouth is’ and agree a form of payment such that her total compensation will be greater when his technology proves effective than if it does not. Again, this involves choosing a base that is closely related to EMMV and applying a percentage royalty to that base. This suggests that there are good reasons why the parties might want to use a royalty base that is closely related to the EMMV. This begs the question: why not use the EMMV itself? From our discussion of the EMMV concept, it should be clear that actually computing the EMMV would be a rather challenging exercise. Not only does it require the specification of an adequate counterfactual, it would require information on the effects of the licensed technology on possibly a large variety of products, including some products that do not directly incorporate the licensed technology. Pragmatically, then, the best that one can hope for is to find an easily computed and easily verified royalty base that one can reasonably expect to be linked fairly closely to the notion of EMMV. One candidate for a royalty base related to EMMV could be a measure of the licensee’s profits on his/her relevant product line. The principal problems with using profits as a royalty base are measurement and verification. Profits equal revenues minus costs. While revenues might be fairly easy to measure and monitor, costs are anything but. Not only would complex issues of allocation of common costs have to be resolved, it also seems reasonable to assume that the licensee would have much better information about the cost involved and could therefore manipulate the measure to the detriment of the licensor.

3.4.2 Form of royalty payment and economic incentives with further downstream innovation Downstream innovation and the addition of new product features are likely to keep evolving over the lifetime of a licensing agreement. In particular, new

Cristina Caffarra and Pierre Régibeau 

323  

features enabling new functionalities are likely to be added and the performance of existing features might be enhanced. Both adding new features and enhancing existing ones require investment. For simplicity, let us assume that the necessary investment would be undertaken by the licensee. The relevant question then is: would some types of royalty schemes – and in particular some types of royalty bases – be more favourable to continuing investment by the licensed party? Before we address this question directly, remember that, to the extent that access to the patented upstream technology has positive effects on the increase in downstream revenues and profits resulting from new features, these synergies should be regarded as part of the incremental value of the patented upstream technology – even if the new features do not rely in any technical sense on the upstream patent and even if the downstream innovations in question take place many years after the development of the patented upstream technology. Moreover, as the future development of the device’s functionality is likely to be very uncertain, agreeing on a percentage royalty computed on a base that includes the revenues generated by such future investments helps to achieve some risksharing benefits. The main potential problem with a percentage royalty expressed on a base that contains future innovation by the licensee is that such agreements will tend to have some negative effect on the licensee’s incentive to carry out such future improvements. The intuition is straightforward: the licensee is more likely to invest in a project if it can expect to capture 100% of its return than it is if it has to turn over X% of the return to the licensor. If royalties were paid to the upstream patent holder on a lump-sum basis, this negative effect on incremental downstream investment incentives would disappear. In this sense, there is a tradeoff between the innovation-reducing effect of running royalties and their risk-sharing benefits. Is it therefore a problem to include, in the royalty base, sources of value not directly related to the infringed innovation (ie, not part of EMMV)? Concretely, is the use of a scheme based on the sales of a device problematic given that some of the future improvements on the device are neither enabled nor enhanced by the infringed technology? Regulators seem to be moving towards a view that innovation incentives could be sharpened if the royalty base excluded revenues related to downstream innovations unrelated to the upstream patented technology. We note again that one would expect the chosen royalty payment scheme to reflect the parties’ opinion as to which of these two effects is strongest. A running royalty on a broad royalty base should therefore be seen as a sign that the parties believe that their joint surplus is likely to be higher under such a scheme even if it might have some chilling effect on further downstream innovation. To better understand how a broad royalty base could result in more downstream innovation than a narrower royalty base (ie a royalty base that excluded downstream revenues related to features where there was no synergy between the value of these features and the upstream patented technology), let us begin with some notation. As before, let U represent the upstream patented technology. Let D1 be the downstream innovation which has synergies with the upstream patented

324  Competition, Regulation and Public Policies technology; let D2 be the unrelated downstream innovation. For purposes of this discussion, let us assume away measurement problems – although in the real world, ease of measurement can be a significant issue and a reason to adopt a broad royalty base. The value of the downstream product assuming access to the upstream patented technology and access to both types of downstream innovation is V(U,D1,D2). To say that D2 is unrelated to the patented upstream technology means that the value created by D2 does not depend on whether the downstream firm has access to the patented technology: V(U,D1,D2)-V(U,D1,0)=V(U’,D1,D2)-V(U’,D1,0) By contrast, the value created by D1 is affected by whether the downstream firm has access to the patented technology: V(U,D1,D2)-V(U,0,D2)>V(U’,D1,D2)-V(U’,0,D2) Suppose that one were able to exclude the incremental value of D2 from the royalty base, and suppose that this produced an adjusted royalty base that was 50% of the unadjusted royalty base. Under these assumptions, if the agreed royalty on a base equal to the total present and future sales of the device were X%, then the agreed royalty on a narrower royalty base that excludes the effects of ‘unrelated innovations’ would be 2X%. So, with a broad royalty base, incentives to invest on developing both enabled and non-enabled features would be hampered by the X% share of the revenues from investment implied by the royalty. This would decrease the incentive to invest moderately and uniformly across various types of features. If the royalty base were instead calculated by subtracting the value attributable to innovations unrelated to the upstream patented technology, then (a) the royalty scheme would have negative effects only on downstream innovations where there were synergies between the value created by the innovations and the patented upstream technology; and (b) the negative effect of downstream innovation of this type would be greater than if a broad royalty base were used (in which case the negative effects on downstream innovation incentives would be spread across the two categories of downstream innovation). As the following examples show, it is not possible to determine which type of royalty scheme/royalty base hampers innovation least without making detailed assumptions about the sensitivity of investments to expected income. Assume that, in the absence of any running royalty, there would be an overall improvement in the value of the device equal to ∆V, with ‘enabled features’ (D1) accounting for ∆V/2 and ‘non-enabled features’ (D2) accounting for the other ∆V /2. Now assume that a ‘tax’ of X% is small enough to ensure investment in all innovation projects, but a tax of 2X% kills off all projects. In such a case, innovation would not be affected if a broad base is chosen but it would be halved if a narrower base were chosen since the royalty rate would have to be doubled in order to give the same appropriate reward to the patent holder. Clearly, then, left to their own device, the parties would choose a broader base.

Cristina Caffarra and Pierre Régibeau 

325  

Now assume that an X% royalty reduces the proportion of projects successfully completed from 1 to α < 1 and a 2X% royalty reduces the proportion of projects completed from 1 to β < α < 1. With a broad royalty base, the rate would be X% leading to a total level of innovation equal to α∆V. With a narrower base, the rate would be 0% on non-enabled features so that the corresponding level of further innovation for these features would be ∆V/2. For the enabled features, the rate would be 2X% and the resulting level of innovation would therefore be β∆V/2. The total level of innovation is therefore equal to (1+β)/2. So the level of innovation is greater with a narrower base if (1+β)/2>α or β>2α-1. So if a royalty rate of x involved a 10% reduction in innovation (α = 0.1), then opting for the narrower base would increase overall innovation if β > 0.8, ie if the decrease in innovation is less than proportional to the increase in the royalty rate. Therefore, there cannot be an unqualified presumption that computing the payment (not the total value) of royalties on a base that includes ‘unrelated’ features is more nefarious for innovation than opting for a base that includes only the features that are enabled or enhanced by the licensed technology. Overall, and especially given that the broader base is easier to define and measure, it is not at all clear that a broader royalty base would be bad for further innovation.

3.4.3 Inclusion of heterogeneous devices in the royalty base One of the features of the regulators’ current concerns appears to be a sense that it is not appropriate (or does not make sense) to have the same royalty rate applying across a range of rather different devices. The economic contribution of a given set of technologies will vary from one device to another. This would depend not only on the proportion of the device’s features that are enabled or enhanced by the technologies but also on the set of complementary products that might accompany the device. In principle, then, it could be argued that both the total contribution of the licensor and his contribution ‘per device sold’ should vary across devices. If the royalty payment scheme specifies the same rate applied to the same type of base (eg, net sales) for all devices, then clearly some devices make a larger contribution to the total transfer made to the licensor while others make a smaller contribution to this transfer. But it would be very hard in practice to discriminate across products and determine precise economic valuations over a wide product range. It therefore makes sense to adopt a single rate for devices that belong broadly to the same family, while possibly allowing for different rates for sufficiently different devices (eg smartphones and game consoles). Whether the benefits of such modulation outweigh the additional transaction costs that such an approach would necessarily introduce is something that the parties seem best placed to determine during their negotiations. As for the choice of royalty base, again it seems that both parties would have the same interest: choosing the payment scheme that maximises their joint surplus.

326  Competition, Regulation and Public Policies

3.5 A (tentative) summary Certain SEP owners have traditionally started negotiations by proposing to license their portfolio of patents for a given rate (2.25%, 2.4%, etc.) of the net sales of the devices deemed to infringe their intellectual property. Certain prospective licensees have made vocal complaints that imposing a variable royalty on the value of whole devices amounts to ‘taxing’ value-creating components of such devices that are not directly related to the relevant IP. Translated into policy terms, this may point to a concern that a royalty scheme that defines payments on such a broad base would decrease incentives to invest in future improvements of the devices covered by the licensing agreement. The nature of the debate is often confused. This chapter has sought to clarify that: Firstly, one must be very careful when defining what is meant by a ‘tax on unrelated features’. The fact that the royalties are paid as a function of the whole value of the device does not mean that the licensor unjustly profits from surplus created by others, including the licensee. In negotiations, the total expected value of the royalty transfers between the two parties will be naturally based on the economic contribution that the licensed technology makes to the devices involved. The fact that the total transfer might then be calculated and paid as a percentage of a base that includes some value that is unrelated to the infringed technology does not affect the total level of the transfer and therefore cannot be seen in itself as an additional ‘tax’ of the licensee’s income. Secondly, the level of compensation for the licensor must be linked to the economic value that its technology creates. This economic value will typically include not only the standalone value of the features enabled by the technology but also the increase in the value of features enhanced by the technology (synergies) and the extra profit on the whole device arising because of the additional sales which are induced by the increased value of the product incorporating the infringed technology. Overall, then, even when the value of a number of features of the end device is completely independent from the infringed technology, the economic contribution of this technology will include some of the revenues coming from the (extra) sales of these independent features. Moreover, the economic value created by the technology also extends to the profit margin on all additional sales of complementary products if their sales are boosted by the increased sales of the device. Thus, the proper measure of economic value logically extends to some of the value associated with ‘independent features’, and might also extend to the value of products that are not even physically linked to the device that embeds the infringed technology. The measure of value that is relevant to determine the overall level of compensation, reached either through bargaining or through other means, is therefore likely to be quite broad-based indeed. Furthermore, because of uncertainty and asymmetric information, there are good economic reasons why the base used for the payment of the agreed total

Cristina Caffarra and Pierre Régibeau 

327  

(expected) transfer should be linked to the measure of the economic contribution of the technology. One would therefore expect the royalty base itself to be rather broad, reflecting the breadth of the contribution to economic value. Turning to the special case of devices that incorporate multiple functionalities, we then ask whether a rate of royalty applied to the value of the whole product would have a greater chilling effect on future innovation than a proportionally increased royalty rate defined on the value contributed only by the features that are in fact enabled by the relevant technology. In other words, keeping the total amount of royalty transfers from licensee to licensor constant, is further innovation more limited if the royalty payment is determined on a narrow or on a broad base? We show that innovation is larger with a narrow base if the reduction in innovation increases less than proportionally with the rate of royalty required to keep the total transfer unchanged. There is therefore no general presumption that choosing a broad royalty base would decrease innovation compared to the alternative of a broader base. Finally, both the level of the total royalty transfer and the form of payments (including the base) are determined as part of a bargaining process between licensor and licensee. In this process, both parties have a common interest in maximising the size of the total surplus that they get to share. In particular, the licensor has no interest in unduly limiting future innovation by the licensee if this simply adds to the (expected) size of the pie to be shared. In fact, when it comes to choosing the base for royalty payments, the parties essentially have common incentives: they both want to maximise future innovation; they both benefit from alleviating asymmetric information issues and from sharing uncertainty optimally; and they both gain from choosing simple schemes that are easy to implement. Of course, a licensee can still benefit from complaining about the form of payment in the hope that the clear logical and economic distinction between such form and the legitimate total level of transfer might be missed – so that an argument against the broad base would actually lead to a lowering of the overall royalty payment. This would be a logical and economic fallacy.

4. Food for thought: royalty base as part as a broader (re)design of patent protection? We have so far discussed the role of the royalty base in the sole context of (FRAND) licensing. We have therefore taken the current state of the patent protection system as given. However, it is also interesting to speculate as to how regulators may be thinking of using the royalty base as part of some broader redesign of patent rights. Academic economists have of course long complained that a ‘one size fits all’ patent system does not make sense, as the economic context in which patent

328  Competition, Regulation and Public Policies rights are exercised differs significantly across various sectors of activities. It is for example widely recognised that very significant patent protection might be needed more in industries like pharmaceuticals, where a significant invention can be covered by relatively few patents, than in ICT-type industries where the development of any useful product relies on a large number of innovations covered by an even greater number of patents. In such complex industries, extensive patent protection strengthens ‘patent thickets’ and might be damaging to all. More controversially, one can also argue that patent protection should be adapted to the length of the innovation cycle in the sector of application. There are indeed two good reasons why one might argue that patent life should be shorter in industries with short product/innovation cycles: First, in such fast-moving industries, the rewards from being first to market can already be quite considerable. If this ‘first mover advantage’ is significant, then it might suffice to provide adequate incentives to innovate. There might be no need to provide inventors with additional profits through patent protection, or at least through long-lived patent protection. Secondly, while it is clear that early innovators should receive some of the surplus accruing to later innovators building on their initial contribution, one might feel that such a reward should decrease as the technological distance between the first innovation and its successor increases. In fast-moving industries, considerable technological distances can grow quickly. The traditional objection to the design of ‘sector-specific’ patent rights is that, at the time of the grant, it is not possible to identify with any certainty the sectors in which the innovation covered by the patent will eventually be applied. One possibility could be to grant patents with uniform maximum lifetimes of twenty years but to specify that the effective lifetime will depend on the eventual fields of use of the patent. A given patent could then last ten years in one of its applications and twenty in another. The argument would be that, since both IP Law and IP-related competition law already make heavy use of the ‘field of use’ concept, such a system would not raise any new conceptual difficulties. Moreover, as fields of use would be determined ex post, ie when a product relying on the patent enters some market(s), competition authorities may be able to determine which markets the product belongs to, and hence which patent protection length it should enjoy. But we should be very careful about thinking this is where the solution lies. While the ‘field of use’ approach might be conceptually straightforward, it still raises very significant practical difficulties. One could make a possible argument for implementing sector-specific patent length in an indirect fashion, for instance by ensuring that royalty bases become progressively ‘thinner’ over time. Concretely, this could take the form of accepting royalties based on the price of the full devices for a few years but then ‘freezing’ the base at the value of a device with the functionalities available at the end of such a period. Again this raises practical issues as to how the continuing value of this ‘frozen’ device would be computed in later years. Moreover, real-life contracts achieved through

Cristina Caffarra and Pierre Régibeau 

329  

commercial negotiations already often contain clauses which, for example, limit the royalty payment over time or use a tapered profile. The suggestion may therefore be redundant for all practical purposes. Still, the question seems at least worthy of debate and further thought.

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?

Introduction In the wake of a number of judgments of the Court of the Justice of the European Union and the General Court of the European Union,1 this chapter discusses the interaction between transparency policy and EU competition law, focusing mainly on issues relating to Articles 101 and 102 TFEU. As it stands today, EU law provides for access to the European Commission’s file in Article 101 and 102  TFEU procedures to three categories of persons. Firstly, the defendants in these procedures benefit from a right of access,2 which is intended to enable the effective exercise of their rights of defence.3 In principle, the defendants’ right of access does not extend to business secrets, confidential information or the Commission’s internal documents or those of Member States’ competition authorities.4 Secondly, complainants benefit from a right of access to the file, although it is more limited than that enjoyed by defendants.5 Thirdly, third parties who can demonstrate a sufficient interest in the outcome of the procedure (‘interested third parties’) enjoy some right of access to the file only when they have been heard by the Commission in the context of the administrative procedure.6 * Judge of the Court of Justice of the European Union. At the time of writing, Judge of the General Court. The views expressed are entirely personal. 1 To cite only two of the most recent judgments, see Case C‑477/10 P Commission v Agrofert Holding [2012] ECR I-000; Case C‑404/10 P Commission v Éditions Odile Jacob [2012] ECR I-000. 2 See Article 27(2) of Council Regulation 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty Article, 2003 OJ L1/1; Article 15(1) of Commission Regulation 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, 2004 OJ L123/18 (‘Implementing Regulation’); Point 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, 2004 OJ C325/7 (‘Access to File Notice’). 3 See Point 7 of the Access to File Notice, cited previous footnote. 4 See Article 15(2) of the Implementing Regulation, cited above note 2. 5 See Article 8(1) of the Implementing Regulation and Points 30 and 31 of the Access to File Notice. 6 See Article 13 of the Implementing Regulation; DG Competition, Best Practices on the conduct of proceedings concerning Articles 101 and 102 TFEU, available at http://ec.europa.eu/competition/ consultations/2010_best_practices/best_practice_articles.pdf, at §§ 90 and 91; Case T‑198/03 Bank Austria Creditanstalt v Commission [2006] ECR II‑1429, paras 29 and 31.

366  Competition, Regulation and Public Policies However, EU law does not provide for any right of access to the file to any other persons, ie the general public.7 Without a right of access to the file and because evidence is essential to third parties wishing to claim damages for loss caused by infringements of competition law provisions, such third parties have tested numerous other routes to gain access to material proving the existence of such infringements.8 There are four main paths. Firstly, they have resorted to Article 15 TFEU and 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 (‘Transparency Regulation’),9 which provide for a general right of access to documents held by EU institutions, as well as national rules providing for a right of access to documents held by national institutions (including competition authorities).10 Secondly, they have requested disclosure of the required evidence in the context of judicial proceedings before national courts. In relation to such proceedings, national courts may order disclosure inter partes, ie from the defendants in an action brought by a damage claimant, and disclosure from third parties, including from their national competition authority. National courts may also ask the Commission to communicate information in its possession.11 Furthermore, those courts may try to obtain the disclosure of evidence located in another Member State.12 Thirdly, third parties have attempted to gain access to evidence before courts of third countries, in particular the United States.13 Fourthly, third parties have also sought leave to intervene before the 7 See, in that respect and for a discussion of the extent of the right of access to the file during the administrative procedure, Gaëlle Bontinck and Tom Snels, ‘Business Secrets and Confidential Information in Proceedings pursuant to Articles 101 and 102 TFEU’, (2011/3) European Journal of Consumer Law 547, 554–559. 8 See Gianni De Stefano, ‘Access of Damage Claimants to Evidence Arising out of EU Cartel Investigations: A Fast-evolving Scenario’, 3 Global Competition Litigation Review 95, 96–100 (2012). 9 OJ L145/43. 10 For an example, see the French Act of 1978 (loi n° 78-753 du 17 juillet 1978 relative à la liberté d’accès aux documents administratifs et à la réutilisation des informations publiques), last modified in 2011 (loi n° 2011-525 du 17 mai 2011 de simplification et d’amélioration de la qualité du droit). 11 See Article 15(1) of Regulation No 1/2003. 12 See Article 1 of Council Regulation 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, 2001 OJ L174/1; and to a lesser extent, Article 31 of Council Regulation 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, 2001 OJ L12/1. See also Laurence Idot, ‘Access to Evidence and Files of Competition Authorities’, in Jürgen Basedow, Stéphanie Francq and Laurence Idot (eds), International Antitrust Litigation: Conflict of Laws and Coordination (Hart Publishing, 2012) pp 259 et seq, at pp 265–267; De Stefano, ‘Access of Damage Claimants’, cited above note 8, at p 99. De Stefano explains that Regulation 1206/2001 has been used in German damages actions and refers in that respect to the German report on damages actions to DG Competition, http://ec.europa.eu/competition/antitrust/actionsdamages/study.html. 13 For example, in the Vitamins case, private plaintiffs sought discovery of corporate statements submitted by a leniency applicant to the European Commission in the context of a claim for treble damages before a U.S. federal court. See Re Vitamins Antitrust Litigation, Report of Special Master (99–197) (TFH), 2002 U.S. Dist. LEXIS 26490 January 23, 2002 D.D.C., denying motion to reconsider at Re Vitamins Antitrust Litigation (99–197) (TFH), 2002 U.S. Dist. LEXIS 25815, December 18, 2002 D.D.C. See also Ingrid Vandenborre, ‘The confidentiality of EU Commission cartel records in civil litigation: the ball is in the EU Court’, 32 European Competition Law Review 116, 116–122 (2011).

Küllike Jürimäe 

367  

General Court in actions for annulment lodged against Commission decisions in a competition law infringement with a view to getting access to the evidence relied on by the Commission.14 It follows that, as a matter of fact, numerous different authorities (administrative, judicial and legislative) may be called upon to perform a very delicate balancing exercise. While public and private enforcement both contribute to the effectiveness of EU competition law,15 their methods and procedures may involve conflicting interests. In particular, transparency, which is a corollary to the right to claim damages for loss caused by infringements of competition law provisions, sits uneasily with the Commission’s willingness to encourage undertakings to cooperate with it by providing, for the purpose of its investigations, information they consider as sensitive or confidential. In this context, this chapter will consider two main issues. On the one hand, it will attempt to determine who should ultimately perform the balancing exercise mentioned above. It may already be asserted, at this stage of the analysis, that the task of balancing the different interests at stake in the EU should not be left to foreign courts. The question is therefore whether that task should be the responsibility of EU Courts, of Member States’ courts or of the legislator (EU or national).16 On the other hand, this contribution will discuss how the balance should be struck with the help of relevant case law concerning the interaction between access to information/evidence and public enforcement of EU competition law. The question, in that respect, is whether specific protection should be afforded to material relating to Article 101 and 102 TFEU procedures, in particular to evidence provided by the undertakings under investigation. Those issues will be handled in three successive steps. First, this article will analyse how the EU Courts have so far struck this balance. Second, the same type of analysis will be carried out with respect to national courts and legislative authorities. Third, on the basis of those analyses, the appropriateness and necessity of an intervention by the EU legislator will be assessed.

14 See, eg, orders of 25 October 2011 in Case T‑28/11 KLM v Commission, not reported; Case T‑38/11, Cathay Pacific Airways v Commission, not reported; Case T‑40/11, Lan Airlines v Commission, not reported; Case T‑46/11, Lufthansa v Commission, not reported; Case T‑62/11, Air France-KLM v Commission, not reported; and Case T‑63/11, Air France v Commission, not reported. 15 See Case C‑453/99 Courage v Crehan [2001] ECR I‑6297, para 27. See also Eric Barbier de la Serre, ‘Beyond effectiveness and uniformity: The Rise of Consistency’, Concurrences No 2-2012, pp 26 et seq, at p 33; Assimakis Komninos, ‘Relationship between Public and Private Enforcement: quod Dei Deo, quod Caesaris Caesari’ in Philip Lowe and Mel Marquis (eds), European Competition Law Annual 2011: Integrating Public and Private Enforcement of Competition Law: Implications for Courts and Agencies (Hart Publishing, 2014). 16 Because decisions of administrative bodies are ultimately reviewed by courts, one need not analyse whether the task of balancing the different interests at stake should be the responsibility of administrative authorities. See Barbier de la Serre, ‘Beyond effectiveness and uniformity’ cited previous footnote, at p 33.

368  Competition, Regulation and Public Policies

1. EU case law concerning the interaction between transparency and public enforcement of competition law: how has the balance been struck? The EU Courts are seized of issues concerning the interaction between transparency and public enforcement of EU competition law through three different types of actions. The first of these are direct actions lodged, on the basis of the Transparency Regulation, against Commission decisions refusing access to evidence relating to Article 101 and 102 TFEU procedures. Second, there are indirect actions (ie references for preliminary rulings) lodged by Member State courts which have to deal with an issue related to access to such evidence (in the framework either of damages proceedings or of proceedings relating to a refusal of access to material relating to Article 101 and 102 TFEU procedures pursuant to the national transparency legislation). The final type consists of requests for leave to intervene, in cases where the EU Courts have been asked to annul Commission competition law decisions, with a view to obtaining documents relied on by the Commission in the course of its investigation. Direct actions concerning the application of the Transparency Regulation to documentation relating to Article 101 and 102 TFEU procedures represent the majority of the cases in which the EU Courts have had to rule on the interplay between transparency and EU competition law. Hence, the first subsection below (1.1.) will start with a brief analysis of the case law concerning the basic principles underlying the application of the Transparency Regulation in general, ie not specifically in the field of competition law. Pursuant to that case law, free access to documents should be the principle with only limited exceptions, although, in certain cases, the Court of Justice has greatly facilitated the reliance by the EU institutions on those exceptions. The second subsection below (1.2.) will dwell on the case law concerning the interaction between transparency and State aid law as well as merger law. In those areas of EU law, the EU courts have so far decided to provide special protection of the interests of the undertakings under investigation, thereby derogating from the principle of free access to documents. In a third subsection (1.3.), the interaction between transparency and EU competition law, limited to Articles 101 and 102 TFEU, will be analysed. The analysis will reveal that uncertainty remains as to whether the interests of the undertakings under investigation pursuant to those provisions should be protected. At this point, it must be emphasized that the requests for leave to intervene mentioned above will not be dealt with in this chapter. This is because it is settled case law that the right of a party, whether a main party or an intervener, to be given access to the file or to procedural documents may be used solely for the purpose of exercising the rights conferred in the context of the same action, to the exclusion of any other purpose.17 It follows that information accessed during the course of 17

See Case T‑168/01 Glaxo Wellcome v Commission, order of 5 August 2003, not reported, para 28.

Küllike Jürimäe 

369  

proceedings before the EU Courts cannot, in principle, be used in subsequent litigation, including an action for damages before national courts.18

1.1. Case law concerning the Transparency Regulation: from widest access to general presumptions Regulation No 1049/2001 was adopted in furtherance of Article 255 of the EC Treaty (later replaced by Article 15 TFEU).19 Pursuant to the latter provision, EU institutions must conduct their work as openly as possible,20 which implies that citizens (meaning EU citizens as well as natural or legal persons residing or established in the EU) must have a right of access to the documents of those institutions. Accordingly, the basic principle underlying Regulation No 1049/2001, as set out in Article 1 and Recital 6 of the Preamble to that regulation, is to ensure the widest possible access to documents. That principle is subject to the exceptions set out in Article 4 of Regulation No 1049/2001. Besides exceptions concerning the protection of the public interest (as regards, eg, public security, defence and military matters and international relations) and of the privacy and the integrity of the individual, Article 4 of that Regulation contains exceptions on grounds of private interests.21 In particular, Article 4(2) of Regulation No 1049/2001 provides that ‘the institutions shall refuse access to a document where disclosure would undermine the protection of commercial interests of a natural or legal person, including intellectual property, court proceedings and legal advice, the purpose of inspections, investigations and audits, unless there is an overriding public interest in disclosure’. Furthermore, Article 4(3) of Regulation No 1049/2001 protects the decision-making process of the EU institutions, unless there is an overriding public interest in disclosure. Article 4 does not contain any specific exception providing protection to undertakings cooperating in competition law investigations. The first judgments delivered by the EU Courts with regard to the Transparency Regulation stuck closely to the letter and spirit of that regulation, ie free access to documents with limited exceptions. However, the Courts rapidly introduced greater flexibility with regard to the duties imposed on the EU institutions pursuant to that Regulation. As will be seen below (subsection 1.2.), the use of such a greater flexibility has to date been upheld by the EU Courts in cases involving State aid and merger laws. 18 See Bontinck and Snels, ‘Business Secrets and Confidential Information’ cited above note 7, at pp 571 and 572. 19 With regard to Regulation 1049/2001, see, eg, Ian Harden, ‘The Revision of Regulation 1049/2001 on Public Access to Documents’, 15 European Public Law 239 (2009); Marc Maes, ‘L’accès du Public aux documents des Institutions de la Communauté européenne : vers une révision du règlement (CE) 1049/2001’, (2007/2) Revue de Droit de l’Union Européenne 411. 20 Article 15(1) TFEU. 21 See Bontinck and Snels, ‘Business Secrets and Confidential Information’ cited above note 7 at p 560.

370  Competition, Regulation and Public Policies

1.1.1. The principle: concrete and individual examination of each of the requested documents In the very first cases concerning the Transparency Regulation,22 the General Court went back to the principles developed in the case law elaborated in relation to the 1993 code of conduct on public access to Council and Commission documents.23 That case law underlined that the public should be granted the widest possible access to documents.24 In furtherance of that case law,25 the General Court stated that the exceptions to the principle of access to documents had to be interpreted strictly.26 It also made it clear that the examination which an EU institution must undertake in order to apply one of the exceptions contained in Article 4 of Regulation No 1049/2001 must be carried out in a concrete and individual manner, as opposed to an abstract and general examination.27 In that respect, it held that an assessment of documents by reference to categories rather than on the basis of the actual information contained in those documents is insufficient. This is so because in its examination the institution must assess specifically whether an exception invoked actually applies to all the information contained in those documents.28 By the same token, the institution must assess the possibility of granting the applicant partial access to the documents at issue, as envisaged by Article 4(6) of Regulation No 1049/2001.

1.1.2. The exception: the notion of ‘general presumption’ However, within a relatively short period, the Court of Justice admitted the possibility for the EU institutions to base the decisions whereby they refuse to grant access to documents on the basis of one or several of the exceptions set down in Article 4 of Regulation No 1049/2001 on general presumptions which apply to categories of documents. This means that the EU institutions are spared from supplying explanations as to how access to each requested document might specifically and effectively undermine the interest protected by an exception laid down in that article. In 2008, in Turco,29 the Court of Justice acknowledged, as a matter of principle, that the Council of the European Union could base its decisions refusing access to See footnotes 26 and 28 below. 1993 OJ L340/41. 24 See, eg, Case T‑14/98 Hautala v Council [1999] ECR II‑2489, para 81. 25 See, eg, Case T‑111/00 British American Tobacco International (Investments) v Commission [2001] ECR II‑2997, para 40; Case T‑123/99 JT’s Corporation v Commission [2000] ECR II‑3269, para 64. 26 Case T‑2/03 Verein für Konsumenteninformation v Commission [2005] ECR II‑1121, para 106. 27 Ibid, para 71. 28 See ibid, para 73; Joined Cases T‑391/03 and T‑70/04 Franchet and Byk v Commission [2006] ECR II‑2023, para 117. 29 See Joined Cases C‑39/05 P and C‑52/05 P Sweden and Turco v Council [2008] ECR I‑4723, paras 50, 66 and 67. 22 23

Küllike Jürimäe 

371  

a document on such general presumptions but denied that possibility in the case at hand. In 2010, the Court of Justice concretely allowed the EU institutions to rely on general presumptions. It did so for the first time in Commission v Technische Glaswerke Ilmenau,30 a case concerning EU State aid law, which will be further analysed in subsection 1.2.1. Subsequently, in API,31 it granted specific protection to pleadings lodged by one of the EU institutions in the context of proceedings before the EU Courts. In that respect, it allowed a general presumption that the disclosure of those pleadings would undermine the protection of the court proceedings, for the purposes of the second indent of Article 4(2) of Regulation No 1049/2001, while those proceedings were pending.

1.2. Access to documents in competition law cases: acknowledgement of the sensitivity of competition law investigations In the competition law field, while the General Court repeatedly refused to afford any special status to requested competition law documents, the Court of Justice has maintained the possibility for institutions to rely on the notion of general presumption as developed in Turco32 with a view to refusing access to those documents on the basis of categories. Arguably, by so doing, the Court of Justice demonstrated its readiness to take account of the sensitive nature of competition proceedings and to admit that the way the exceptions laid down in Article 4 of Regulation No 1049/2001 had been applied up to that point may not have been adapted to that type of proceedings.33

1.2.1. Transparency and State aid law As regards EU State aid law, the General Court’s judgment in Technische Glaswerke Ilmenau v Commission,34 a case concerning the Commission’s refusal to give access to documents relating to an ongoing State aid investigation, is to be understood as asserting, first, that the Transparency Regulation does apply to State aid cases and, second, that no adjustments are necessary when applying that Regulation to that field of law, in order to take account of its specificities. In particular, the General Court held that the obligation for an institution to undertake a concrete, individual assessment of the content of the documents covered in the application for access is an approach to be adopted as a matter of Case C‑139/07 P Commission v Technische Glaswerke Ilmenau [2010] ECR I‑5885, para 61. Joined Cases C‑514/07 P, C‑528/07 P and C‑532/07 P Sweden v API [2010] ECR I‑8533, para 94. 32 See above note 29. 33 See Laurence Idot, ‘Le règlement n° 1049/2001 doit-il s’appliquer aux ‘procédures concurrence’? À propos des affaires Technische Glaswerke Illmenau, Odile Jacob et Argofert’, Europe No 10, octobre 2010, étude 11. 34 Case T‑237/02 Technische Glaswerke Ilmenau v Commission [2006] ECR II‑5131, set aside on appeal: Case C‑139/07 P, cited above note 30. 30 31

372  Competition, Regulation and Public Policies principle, which applies to all the exceptions in paragraphs 1 to 3 of Article 4 of Regulation No 1049/2001. This principle applies irrespective of the field to which the documents sought relate, and concerns, in particular, cartels or the control of public subsidies.35 The General Court added that such an examination might not be necessary in particular circumstances.36 Nonetheless, it emphasized that the Commission had justified the refusal of access by reference to the exception based on the protection of the purpose of inspections and investigations. In that respect, the Commission had maintained that the disclosure of documents relating to current investigations might undermine the conduct of the examination of the complaint by compromising the dialogue between the Commission, the Member States and the undertakings concerned.37 The General Court held that so general an assessment failed to demonstrate that there were special circumstances of the individual case leading to the conclusion that it was not necessary to undertake a concrete, individual examination of the documents concerned.38 On appeal, the Court of Justice, in Commission v Technische Glaswerke Ilmenau,39 overturned the General Court’s judgment and acknowledged the existence of a general presumption that disclosure of documents in the administrative file in principle undermines the protection of the objectives of investigation. It is to be noted that although the case concerned an ongoing State aid procedure, the Court of Justice did not explicitly rely on that fact. Hence, the judgment left some uncertainty as to whether the notion of general presumption could apply in cases where the State aid procedure had already been closed. It is also important to stress that the conclusion reached by the Court of Justice is based mainly on two lines of reasoning. First, in the procedures for reviewing State aid, interested parties other than the Member State concerned do not have a right to consult the documents in the Commission’s administrative file.40 Second, in contrast with cases such as Turco,41 in which the EU institutions acted in their capacity as legislative authorities and with regard to which the Court of Justice held that wider access to documents should be authorised, documents relating to procedures for reviewing State aid fall within the framework of the administrative functions of those institutions. In view of that argumentation, one might wonder whether the Court of Justice’s reasoning was designed to apply solely to State aid procedures or whether it would be extended to other areas of competition law.

Ibid, para 85 of the judgment of the General Court. Ibid, para 86. Ibid, para 88. 38 Ibid, para 89. 39 Cited above note 30. 40 Ibid, paras 58 and 61. 41 See above note 29 and accompanying text. 35 36 37

Küllike Jürimäe 

373  

1.2.2. Transparency and merger law The Court of Justice confirmed the existence of such a general presumption in the field of EU merger law in two recent cases – Commission v Agrofert42 and Commission v Éditions Odile Jacob43 – and overturned the General Court’s judgments.44 Both cases concerned a Commission decision refusing to disclose documents relating to merger control proceedings to third parties. The Commission had relied on four series of exceptions laid down in Article 4(2) and (3) of Regulation No 1049/2001: the exception relating to the protection of the purpose of inspections, investigations and audits; the exception relating to the protection of commercial interests; the exception relating to the protection of the institution’s decision-making processes; and the exception relating to the protection of legal advice. In both cases, the General Court had annulled the Commission decisions relying on the established case law45 and confirmed that the Commission ought to have undertaken a specific, individual assessment of the content of the documents covered by the application for access with a view to ensuring that the disclosure of those documents would have undermined the interests protected by the exceptions mentioned in the previous paragraph.46 In Agrofert v Commission, the General Court also admitted, with regard to the exception concerning the protection of legal advice, the possibility for the Commission to base its decisions on general presumptions, but considered the conditions for so doing were not met in the case at issue.47 In contrast, the Court of Justice clearly stated in both cases that there did exist a general presumption that the disclosure of documents exchanged between the Commission and undertakings in the course of merger proceedings undermined in principle both the protection of the purpose of investigations and the protection of the commercial interests of the undertakings involved in those proceedings.48 It added that this presumption covered not only proceedings already closed but also pending proceedings.49 Moreover, it stated that such a presumption also applied to the Commission’s internal documents when the merger control procedure was not definitively closed.50 See Agrofert Holding, cited above note 1, para 64. See Éditions Odile Jacob, cited above note 1, para 123. 44 See Case T‑111/07 Agrofert Holding v Commission, judgment of 7 July 2010, not reported, paras 125–132; Case T‑237/05 Éditions Odile Jacob v Commission [2010] ECR II‑2245, paras 41–47. 45 See, eg, Verein für Konsumenteninformation, cited above note 26; Franchet and Byk, cited above note 28. 46 See Éditions Odile Jacob, cited above note 44, para 44; Agrofert Holding, cited above note 44, paras 89, 101 and 110. 47 See Agrofert Holding, cited above note 44, paras 125–129. 48 See Éditions Odile Jacob, cited above note 1, paras 118 and 123; Agrofert Holding, cited above note 1, paras 57 and 64. 49 See Éditions Odile Jacob, cited above note 1, para 124; Agrofert Holding, cited above note 1, para 66. 50 With regard to internal documents, see Éditions Odile Jacob, cited above note 1, para 130. In that case, although the Commission had already adopted a decision on the merger it was reviewing, the merger control procedure was not considered to be definitively closed since that decision had been challenged before the EU Courts. Consequently, the Commission could be called upon, depending 42 43

374  Competition, Regulation and Public Policies In reaching such a conclusion, the Court of Justice expressed its intention to take account of the specificities of merger control proceedings and underlined that the Transparency Regulation could not be applied to those proceedings without restriction. In that respect, three elements of its reasoning are noteworthy. First, in justifying the application of a general presumption, the Court referred to the objective of merger control proceedings, which consists of ascertaining whether or not a merger gives the notifying parties a market power which may significantly impede competition and insisted on the fact that, with this objective in mind, the Commission gathered sensitive commercial information.51 Second, the Court of Justice explained that the EU legislator had, in the Merger Regulation, sought to ensure the balance between the obligation on the undertakings concerned to send the Commission possibly sensitive commercial information (which enables assessment of the compatibility of the proposed transaction with the common market) and the guarantee of increased protection for the information so provided (by virtue of the requirement of professional secrecy and business secrecy).52 This balance would be jeopardised by allowing generalised access on the basis of the Transparency Regulation to the documents exchanged in the context of merger proceedings. Third, it mentioned the strict rules in the relevant legal provisions regarding the treatment of information obtained or established in the context of such proceedings.53 Accordingly, the Court of Justice’s judgments demonstrate a preparedness to grant specific protection to undertakings involved in merger proceedings. The following subsection will inquire into whether such willingness exists with regard to access to documents in cases concerning Articles 101 and 102 TFEU.

1.3. Access to documents in competition law cases: remaining uncertainty As already mentioned, to date the General Court has adopted an approach in merger and State aid cases that favours a strict application of the right of access to documents, as envisaged in the Transparency Regulation, over the protection of both the Commission’s investigations and the interests of the undertakings under investigation. As will be shown in subsection 1.3.1, this is also true of the interaction between transparency and Article 101 and 102 TFEU procedures. However, whereas, with regard to the relationship between transparency and EU State aid and merger rules, the Court of Justice has delivered final judgment and on the result of the legal proceedings, to recommence its investigation with a view to the eventual adoption of a new decision on the merger in question. A contrario, see Agrofert Holding, cited above note 1, paras 75–79. 51 52 53

Éditions Odile Jacob, cited above note 1, para 115; Agrofert Holding, cited above note 1, para 56. Éditions Odile Jacob, cited above note 1, para 121; Agrofert Holding, cited above note 1, para 62. Éditions Odile Jacob, cited above note 1, para 118; Agrofert Holding, cited above note 1, para 59.

Küllike Jürimäe 

375  

adopted a position favouring the protection of the Commission’s investigations and decision making procedure, as of this writing it has not yet done so with regard to access to documents and Article 101 and 102 TFEU procedures. On the contrary, it has arguably heightened uncertainty as to how the balance should be struck with regard to the different interests at stake. It must be stressed, before examining the relevant case law, that when balancing the right of access of third parties to materials gathered in Article 101 and 102 TFEU proceedings against the protection of the Commission’s investigation and decision-making procedure and interests of the undertakings under investigation, the EU Courts actually have to consider how public and private enforcement of antitrust law will be affected. This can only be described as a Catch-22 situation, since public and private enforcement both contribute to the effective enforcement of competition law. On the one hand, access to evidence is essential for aggrieved parties should they wish to succeed in their damages claims, but such access undermines the cooperation of undertakings in investigations, which is itself a prerequisite to public enforcement. On the other hand, the Commission’s protection of the secrecy and confidentiality of the material in its possession is key to securing the cooperation of undertakings and, therefore, to public enforcement; but a regime of strong protection works to the detriment of private enforcement. In that context, it must be further noted that the question of access to cartel evidence is particularly sensitive, especially insofar as material provided to the Commission by leniency applicants is concerned. As will be seen below, most of the cases dealt with by the EU Courts concern, at least to some extent, evidence provided by leniency applicants in a cartel investigation.

1.3.1. The General Court’s approach: no special protection With regard to the application of the Transparency Regulation to documents relating to Article 101 and 102 TFEU proceedings, three of the General Court’s judgments are noteworthy. Those judgments undoubtedly demonstrate that the General Court, as with State aids54 and mergers,55 has so far adopted an approach that favours a strict application of the right of access to documents over the protection of both the Commission’s investigations and decision making procedure and the commercial interests of the undertakings under investigation. First, one must make reference to Verein für Konsumenteninformation v Commission,56 which is one of the first cases in which the General Court had to apply Regulation No 1049/2001. The case concerned a cartel known as the ‘Lombard Club’, in which Austrian banks had fixed interest rates for certain investments and loans. The Verein für Konsumenteninformation (‘VKI’), a consumer association, was involved in judicial proceedings in Austria, claiming 54 55 56

See above section 1.2.1. See above section 1.2.2. See above note 26.

376  Competition, Regulation and Public Policies damages on behalf of its members from one of these banks because of an incorrect adjustment of the interest rates which were applied to some loans. With a view to gathering evidence for the purpose of those proceedings, VKI requested authorisation from the Commission to consult the administrative file relating to the Lombard Club decision. The Commission rejected VKI’s request. To that end, it divided the documents in the Lombard Club file into 11 separate categories and explained that those categories were covered by one or more of the following exceptions provided for by the Transparency Regulation: the exception relating to the protection of the purpose of investigations; the exception relating to the protection of commercial interests; the exception relating to the protection of court proceedings; and the exception relating to the protection of privacy and integrity of the individual. The General Court annulled the Commission decision. It held that the examination which an EU institution must undertake in order to apply one of the exceptions contained in Article 4 of Regulation No 1049/2001 must be carried out in a concrete and individual manner, ie in respect of each document, as opposed to an abstract and general examination.57 It added that an assessment of documents by reference to categories rather than on the basis of the actual information contained in those documents is insufficient, since the examination required of an institution must enable it to specifically assess whether an exception invoked actually applies to all the information contained in those documents.58 By this judgment, the General Court demonstrated its desire to adhere strictly to the terms of the Transparency Regulation and paid no particular attention to the specificities of EU competition law. In that regard, it must be observed that the judgment does not contain any mention whatsoever of a possible need to protect the commercial interests of the undertakings under investigation or the duty of the Commission in ensuring the effectiveness of the public enforcement of competition law. Nor does it undertake any explicit balancing exercise between those interests and the right of access to cartel evidence. The only element for which it expressed any sympathy was the burden of work involved in a concrete, individual examination of the documents contained in a file such as that in the Lombard Club case.59 Nevertheless, it refused to take that factor into consideration because the Commission had failed to explain why the alternatives to such an examination (ie, alternatives less restrictive than refusing outright to grant access, as the Commission had done) also represented an unreasonable amount of work.60 Second, CDC61 is worth mentioning. The case concerned the cartel in the hydrogen peroxide market, in which the Commission found nine undertakings had participated. CDC had unsuccessfully requested access to the table of contents of the Commission’s case file with a view to using that information to See Verein für Konsumenteninformation, cited above note 26, paras 70, 71 and 84. Ibid, para 73. 59 Ibid, para 122. 60 Ibid. 61 Case T‑437/08 CDC Hydrogene Peroxide v Commission [2011] ECR II-8251. 57 58

Küllike Jürimäe 

377  

claim damages on behalf of undertakings affected by that cartel. The Commission rejected the request on the basis of the exceptions relating to: protection of the purpose of the investigation activities; protection of the commercial interests of the undertakings which took part in the cartel; and protection of the Commission’s decision‑making process. The General Court annulled the Commission decision. Two elements in the Court’s reasoning deserve particular attention. First, with regard to the exception relating to the protection of commercial interests, the General Court held that the interest of a company which took part in a cartel in avoiding damages actions cannot be regarded as a commercial interest within the meaning of the Transparency Directive and, in any event, does not constitute an interest deserving of protection.62 The General Court recalled, in that respect, that any individual has the right to claim damages for loss caused by conduct which is liable to restrict or distort competition.63 Second, it dismissed the Commission’s interpretation of the exception relating to the protection of the purpose of the investigation activities. In the Commission’s view, that exception may be relied on, in a general way, to refuse disclosure of any document likely to undermine the Commission’s cartel policy and, in particular, its leniency programme. The General Court held that ‘acceptance of the interpretation proposed by the Commission would amount to permitting the latter to avoid the application of Regulation No 1049/2001, without any limit in time, to any document in a competition case merely by reference to a possible future adverse impact on its leniency programme’.64 The Court stressed that in the present case the Commission had refused to disclose a document which contained no information likely to damage the interests of the companies which applied for leniency.65 It also made clear that reliance on that exception was limited to cases in which the investigation was ongoing.66 The judgment in CDC is also worth examining for one other reason. Contrary to its approach in Verein für Konsumenteninformation v Commission, the General Court considered the different interests at stake when a third party tries to gain access to cartel evidence and explicitly undertook a balancing exercise in that regard. As a result of this balancing exercise, the General Court might be viewed as having made a policy choice, namely that of favouring the interests of the damage claimants over those of undertakings that have participated in a cartel but later take advantage of the Commission’s leniency programme. Third, in EnBW,67 the General Court annulled a decision of the Commission to refuse to disclose the documents in its possession with regard to the gas insulated switchgear cartel. The documents had been requested by EnBW Energie BadenWürttemberg AG, which considered itself to be a victim of that cartel. Ibid, para 49. Ibid. Ibid, para 70. 65 Ibid. 66 Ibid, paras 60 and 62. 67 Case T‑344/08 EnBW Energie Baden-Württemberg v Commission [2012] ECR II-000, set aside on appeal: Case C-365/12 P, judgment of the Court of Justice of 27 February 2014, not yet reported. 62 63 64

378  Competition, Regulation and Public Policies The General Court’s reasoning is remarkable in two respects. First, it reiterated the statements made in CDC,68 thereby stressing its intention to strike the balance in favour of damage claimants. Second, it considered it inappropriate to follow the solution developed by the Court of Justice in Commission v Technische Glaswerke Ilmenau69 and therefore dismissed the possibility for the Commission to base its decisions rejecting requests for access to documents on general presumptions applying to categories of documents. To that end, the General Court pointed out that there is a manifest difference between Technische Glaswerke Ilmenau and the case at hand.70 It referred to the fact that, in contrast to the circumstances of the request by EnBW, in Technische Glaswerke Ilmenau the Commission had not yet reached a final decision. It also noted that Regulation No 1/2003 could not be compared to Regulation No 659/1999,71 which concerns State aid. In so doing, the General Court expressed its reluctance to make any specific adjustments when applying the Transparency Regulation in cartel cases.

1.3.2. The Court of Justice’s approach: remaining uncertainty While, as outlined above, the General Court deliberately undertook a balancing exercise, the Court of Justice has recently increased uncertainty with regard to how it considered that the balance should be struck and whether, in Article 101 and 102 TFEU cases, it would follow the trend initiated in its State aid and merger rulings. In contrast to the cases mentioned in the previous sections, Pfleiderer72 did not concern the application of Regulation No 1049/2001. In this well-known case, the Amtsgericht Bonn, a Bonn local court, referred a question for a preliminary ruling to the Court of Justice. This was in the context of proceedings between Pfleiderer AG and the Bundeskartellamt concerning an application for access to the file relating to the cartel in the decor paper sector. Pfleiderer, a customer of the fined undertakings, applied for full access to the Bundeskartellamt’s file (including leniency documents) with a view to preparing its action for damages. The Bundeskartellamt partly rejected the application, prompting Pfleiderer to bring an action before the Bonn local court. The Amtsgericht Bonn asked the Court of Justice whether the provisions of EU competition law were to be interpreted as meaning 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 to a competition authority of a Member State. See EnBW, judgment of the General Court, cited previous footnote, paras 125 and 148. Cited above note 30. See EnBW, judgment of the General Court, cited above note 67, paras 57, 59 and 60. 71 Council Regulation 659/1999 of 22 March 1999 laying down detailed rules for the application of Article [108 TFEU], 1999 OJ L83/1, now amended by Council Regulation 734/2013 of 22 July 2013, 2013 OJ L204/15. 72 Case C‑360/09 Pfleiderer AG v Bundeskartellamt [2011] ECR I-5161. 68 69 70

Küllike Jürimäe 

379  

The Court of Justice confirmed the absence of binding common EU rules on the right of access to documents relating to a leniency procedure.73 Nevertheless, it replied more precisely that the provisions of EU cartel law must not be interpreted as precluding a person who has been adversely affected by EU competition law infringement and is seeking to obtain damages from being granted access to documents relating to a leniency procedure.74 Therefore it is for the Member State courts, on the basis of national laws, to determine whether such access must be permitted or refused by weighing the interests in favour of disclosure of the information and in favour of the protection of the information provided voluntarily by the applicant for leniency.75 The Court made it evident that those interests were, on the one hand, the need to ensure the effectiveness of leniency programmes and, on the other hand, the right to claim damages for loss caused by a competition law infringement.76 It is interesting to note that the Court of Justice refrained from following Advocate General Mazák’s opinion, in which the Advocate General proposed to indicate what leniency material could and could not be disclosed.77 Briefly, in substance, the Court of Justice sent the case back requesting the national judge to carry out a balancing exercise. This judgment has already attracted a great deal of attention from legal commentators.78 Some have interpreted the judgment as a refusal to decide: unable to choose between two equal means of ensuring the effectiveness of the application of Article 101 TFEU, the Court of Justice left that choice to national judges.79 Others noted that the Court of Justice entrusted the national courts with a mission that seemed genuinely impossible.80 They went on to express the view that the judgment had to be interpreted either as meaning that the Court of Justice had accepted that its ruling would provoke a long series of preliminary references, or that the Court of Justice had considered that it was not within its remit to correct the chaotic consequences of incomplete EU legislation and therefore wanted to force the EU legislator to face its responsibilities.81 Yet others have interpreted Ibid, para 23. Ibid, para 32. 75 Ibid, para 30 and operative part of the judgment. 76 Ibid, paras 26 and 28. 77 Opinion of Advocate General Mazák of 16 December 2010 in Pfleiderer, cited above note 72, at paras 43–47. The Advocate General proposed that access to voluntary self-incriminating statements made by a leniency applicant should not, in principle, be granted but that, apart from those statements, injured parties should have access to all other pre-existing documents submitted by a leniency applicant in the course of a leniency procedure. 78 See, eg, Robin Van der Hout and Miriam Firmenich, ‘Access to documents containing confidential business information – The application of Regulation (EC) 1049/2001 in cartel cases and the need for reform’, Zeitschift für Europarechtliche Studien 647, 653–655 (2011/4); Sven Völcker, case note, ‘Case C‑360/09, Pfleiderer AG v. Bundeskartellamt, Judgment of the Court of Justice (Grand Chamber) of 14 June 2011’, 49 Common Market Law Review 695 (2012); Gaëtane Goddin, ‘The Pfleiderer Judgment on Transparency; The National Sequel of the Access to Document Saga’, 3 Journal of European Competition Law and Practice 40 (2012). 79 See Laurence Idot, ‘Les leçons du droit international privé’, Concurrences no. 2-2012, pp 41 et seq, at p 46. 80 See Barbier de la Serre, ‘Beyond effectiveness and uniformity’, cited above note 15, at p 33. 81 Ibid, at p 33–34. 73 74

380  Competition, Regulation and Public Policies the judgment as offering the Commission and national competition authorities support in their approach to resist or limit access to corporate statements submitted by leniency applicants.82 Whatever interpretations may be made on the judgment, there is no doubt that by entrusting the balancing exercise to Member States’ courts, the judgment has created a lot of uncertainty as to how requests for access to cartel files and, in particular, to leniency material will be treated. Section 2 below, which deals with the manner in which the interaction between transparency and Article 101 and 102 TFEU procedures has so far been handled by Member States, will confirm this assertion. Before moving on to the examination of how the Member States have dealt with this issue, it must be mentioned that, as of this writing, an appeal against EnBW is currently pending before the Court of Justice.83 The Court of Justice may well seize that opportunity to actually undertake the balancing exercise it entrusted to Member States’ courts in Pfleiderer.

2. Member States’ approach concerning the interaction between transparency and public enforcement of competition law: how has the balance been struck? In Pfleiderer, the Court of Justice pointed to the lack of binding EU rules on the right of access for persons adversely affected by a cartel to documents relating to leniency procedures. It ruled that without such binding rules it was for Member States to enact and apply national rules on the right of access.84 Although Pfleiderer referred only to the absence of rules on the right of access to leniency material, the same is in fact true for all material related to Article 101 and 102 TFEU proceedings. Thus far in the EU, there has been no harmonisation of rules relating to that question. Subsection 2.1. will demonstrate that although national laws tend to limit the right of access to competition law files, thereby granting specific protection both to national competition authorities and the undertakings under investigation, they differ from Member State to Member State and are further subject to the interpretations of national competition authorities and national judges. Hence, See Komninos, ‘Relationship between Public and Private Enforcement’, cited above note 15. See Commission v EnBW Energie Baden-Württemberg, cited above note 67. On 27 February 2014, following the completion of this chapter, the Court of Justice decided to extend the general lines of its transparency jurisprudence in the areas of State aid and merger control to the context of concerted practices. It thus held, contrary to what had been suggested by the General Court, that the Commission is in principle entitled to refuse access to documents in its case file without first carrying out a document-by-document examination, unless this general presumption against disclosure is rebutted by the party seeking access by showing specifically how access to the documents is necessary to bring an action before a national court. 84 See Pfleiderer, cited above note 72. 82 83

Küllike Jürimäe 

381  

forum shopping has been encouraged.85 Subsection 2.2. will show that Pfleiderer has made the situation even more complicated.

2.1. Divergence of national laws A quick comparison of the relevant laws of most of the Member States shows that the majority of those laws provide for some sort of limitation on the right of access to competition law files held by national competition authorities. However, beyond this general trend, the details of the legal solutions vary tremendously. Among the strictest Member States is Austria. Austrian law is very protective of the activities of its competition authority and of undertakings under investigation. Article 39(2) of the Austrian Kartellgesetz provides for access to the file to third parties only if all parties involved give their express consent.86 Where any party involved refuses to grant access, the national court is left with no discretion and cannot order the disclosure of the documents contained in the case file.87 This rule is the same for all documents contained in the file and does not apply exclusively to leniency material. In the Netherlands, the relevant legal provisions are also restrictive towards third parties seeking access to material related to Article 101 and 102 TFEU proceedings with a view to a follow-on action, but slightly less so than in Austria. A damage claimant may, prior to or during litigation, request to inspect and make a copy of documents associated with its legal relations to another person.88 Nonetheless, the claimant must identify the documents with a reasonable degree of precision and may not engage in a so-called ‘fishing expedition’.89 In addition, Dutch law provides for a privilege of non-disclosure for document holders bound by a duty of confidentiality. It also provides for a right not to disclose a document when there are serious reasons for the refusal (eg, confidentiality of the information) or if it may reasonably be assumed that the proper administration of justice is safeguarded even if the information requested is not produced, bearing in mind that the exercise of that right may be reviewed by a court.90 Italy, Poland and Spain are less restrictive. The legal provisions applicable in those three Member States provide third parties with some right of access to the See, eg, Idot, ‘Access to Evidence’, cited above note 12, at p 264. See Christina Hummer, ‘The Austrian Cartel Court brings a preliminary ruling before the European Court of Justice on the question of access to cartel files by third parties adversely affected by a cartel (Printing chemical producers)’, 12 October 2011, e-Competitions no. 40181. 87 See Völcker, case note, cited above note 78, at p 712. 88 Article 843a of the Dutch Civil Code of Procedure. 89 Anthony Maton et al., ‘The effectiveness of National Fora in Europe for the Practice of Antitrust Litigation’, 2 Journal of European Competition Law and Practice 489, 505 (2011). 90 Article 843a of the Dutch Civil Code of Procedure. See also George Cumming and Mirjam Freudenthal, Civil Procedure in EU Competition Cases before the English and Dutch Courts, Kluwer Law International, 2010, pp 203–204. 85 86

382  Competition, Regulation and Public Policies files of their national competition authorities91 but deny such a right to leniency material.92 More liberal Member States are Belgium, France, Greece and Sweden. France93 and Greece94 prohibit access to third parties in the course of the administrative procedure but accept some right of access in the course of judicial proceedings. Furthermore, they do not provide for any specific rules with regard to leniency material. It must be noted that according to a recent ruling French courts can order the disclosure of leniency-related documents.95 Belgium96 and Sweden97 provide for some right of access both in the course of the administrative procedure and in the course of judicial proceedings. They do not provide for any specific legal rules concerning leniency material. Germany and the United Kingdom appear to be the most liberal Member States. In the United Kingdom,98 third parties do not enjoy any right of access during the administrative procedure but may benefit from extensive disclosure, including leniency material, in the context of judicial proceedings. In Germany, there is a statutory presumption in favour of access by damage claimants to documents in the competition authorities’ files, including leniency material.99 Although access to such leniency material has been informally restricted since 2006 by an administrative internal document,100 in February 2009 the Amtsgericht Bonn ordered the Bundeskartellamt to disclose the case file, including documents provided by leniency applicants. The applicant in that case was none other than Pfleiderer! Subsequently, in the wake of Pfleiderer, the German legislator tabled 91 In Italy, see Article 13 of Presidential Decree 217/98 of 30 April 1998, GURI, 9 July 1998, no. 158. In Poland, see Article 69(1) of the 2007 Act on the Protection of Competition and Consumers, Dz. U. No 50, poz. 331 z. In Spain, see Article 31 of Implementing Regulation 261/2008 of 22 February 2008. 92 In Italy, see Article 10 bis of the communication on immunity from fines and on the reduction of their amount, deliberation of 6 May 2010, no. 21092. In Poland, see Article 70 of the 2007 Act on the Protection of Competition and Consumers (cited previous footnote 91). In Spain, see Article 51 of Implementing Regulation 261/2008 of 22 February 2008. 93 See Law no. 2011-525 of 17 May 2011; and, eg, Christophe Lemaire and Simon Naudin, ‘Disclosure before the commercial judge is not in breach of article L. 463-6 if these documents were already known to the other parties (Outremer Telecom and Orange Caraïbes)’, Concurrences, n° 1-2012, 198. 94 See Article 19(6) of the Rules of Procedure of the Greek Competition Authority, Article 16 of Act 1599/1986, and Article 5 of the Code of Administrative Procedure. 95 See Cour de Cassation, judgment of 19 January 2010, Semavem, no. 08-19.761. 96 See Coordinated Act of 15 September 2006 on the protection of economic competition, MB, 29 September 2006, p 50613; and Article 877 of the Judicial Code. 97 See Swedish Competition Act (2008:579) and Chapter 38, Article 8, of the Swedish Code of Procedure. 98 See, eg, Maton et al., ‘The effectiveness of National Fora’, cited above note 89, at pp 497–499; and National Grid Plc v ABB Ltd & Others [2012] EWHC 869 (Ch). 99 See Völcker, cited above note 78, citing (at p 712) Section 406e of the German Code of Criminal Procedure in conjunction with Section 46(1) and (3) of the German Law on Administrative Offences. 100 Bekanntmachung Nr. 9/2006 über den Erlass und die Reduktion von Geldbußen in Kartellsachen, or Bonusregelung, 7.03.2006, point 22, available at http://www.bundeskartellamt.de/wDeutsch/ download/pdf/Merkblaetter/Merkblaetter_deutsch/06_Bonusregelung.pdf.

Küllike Jürimäe 

383  

a proposal restricting damage claimants’ access to leniency material.101 However, that proposal was later withdrawn.102 Having illustrated this diversity of national legal provisions, the discussion now turns to examine the way in which national judges have applied the balancing exercise prescribed in Pfleiderer.

2.2. Application of Pfleiderer by national judges An analysis of the first cases rendered by national courts after Pfleiderer demonstrates, first, that the solutions proposed differ and, second, that those courts faced difficulties as to how to undertake the balancing exercise.

2.2.1. Diverging solutions Three judgments delivered by national courts following Pfleiderer deserve attention. They demonstrate that the balancing exercise prescribed by that judgment has produced different outcomes depending on the national court. First, one must mention the Amtsgericht Bonn’s judgment after Pfleiderer. In a judgment of 18 January 2012, the Amtsgericht stated that the Bundeskartellamt had to release non-confidential versions of decisions and any seized documents but had the right to prevent access to leniency applications and accompanying documents. This was the same Bonn court that had ordered the disclosure of leniency materials in February 2009.103 Second, a regional Czech court judgment in early 2012104 dismissed the actions for annulment brought by the participants in the colour picture tubes cartel against the decision of the Czech Competition Authority (‘CCA’) of 19 November 2010. The applicants claimed inter alia that the CCA had violated their rights of defence by not granting full access to leniency materials. The regional court held that Pfleiderer had not provided clear guidance as regards access to leniency materials and that, in those circumstances, it considered that such materials, given their sensitive nature, deserved special protection during the administrative procedure. Therefore, that material had to be protected until the Statement of Objections was issued. Following this, the undertakings under investigation should have limited access to the material in order to defend themselves. 101 See Draft Bill regarding the 8th Amendment to the German Act against Restraints on Competition, which proposes Restrictions for Claimants to Access Leniency Applications, GCLR 2012, 5(1), R5. 102 See Thomas Kapp, ‘Das Akteneinsichtsrecht kartellgeschädigter Unternehmen: Bonn locuta, causa finita?’, WuW, 05/2012, pp 474 et seq, at p 482. 103 On the decision of the Amtsgericht, see National Grid, cited above note 98, para 21. 104 Regional Court of Brno, judgment of 23 February 2012. See also ‘Czech Republic: anticompetitive agreements – cartel’, 33(6) European Competition Law Review N81-82 (2012).

384  Competition, Regulation and Public Policies Third, the judgment of the High Court of England and Wales in the National Grid case is also noteworthy.105 This case was a follow-on private action brought after the Commission had imposed fines on participants in the gas insulated switchgear cartel. In support of a damages claim, National Grid Plc sought disclosure of relevant parts of the Commission decision and also of documents submitted to the Commission during its investigation, which might contain leniency materials. The judgment is of interest in four respects. First, the High Court stated that, compared to the Commission, national judges are better placed to determine whether material submitted under the EU leniency programme should be disclosed. In particular, it stated: ‘the Commission is naturally far less well placed than the national court to assess the relevance and importance of the disclosure being sought to the litigation before the court’.106 Second, while weighing the interest in disclosure against the need to protect an effective leniency programme, the High Court focused on the principle of legitimate expectations and stated that, when submitting information to the Commission under the EU leniency programme, leniency applicants had no legitimate expectations that leniency documents would not be disclosed, since the grant of immunity or leniency cannot protect an undertaking from the civil law consequences of its infringement.107 Third, the High Court considered that it had to determine whether disclosure was proportionate. In that respect, it took into account the degree of relevance of the information sought for the purpose of the claim108 and the availability of that information from other sources.109 Fourth, the High Court made it clear that the judgment of the Amtsgericht Bonn in Pfleiderer was of little assistance, since the balancing exercise is to be conducted on a case-by-case basis.110 In view of its assessment, the High Court ordered that certain parts of the documents had to be disclosed to National Grid Plc. It follows from National Grid that British courts do not exclude that leniency materials may, under certain circumstances, have to be disclosed.

2.2.2. Uncertainty as to how to carry out the balancing exercise Besides the divergent solutions advocated by national judges, it must also be emphasized that those judges have encountered difficulties in undertaking the balancing exercise prescribed in Pfleiderer. In that respect, one must mention, first, the reaction of Judge Roth111 on reading Pfleiderer. He explained that he did not know how to carry out the prescribed balancing exercise in practice and that the judgment led him to ask the Commission See National Grid, cited above note 98. Ibid, para 26. Ibid, para 34. 108 Ibid, paras 45–55. 109 Ibid, paras 42–44. 110 Ibid, para 60. 111 See National Grid, cited above note 98. 105 106 107

Küllike Jürimäe 

385  

to submit its views, by way of a written intervention, as to how it thought such an exercise should be carried out.112 Second, and more importantly, it should be noted that a preliminary ruling case, Donau Chemie, is pending, as of this writing, before the Court of Justice.113 The reference was lodged by the Oberlandesgericht Wien. As mentioned earlier, Austrian law is very protective towards its competition authority and the interests of undertakings under investigation. It provides for access to the file to third parties only if all parties involved give their express consent. This means that a balancing exercise is, in principle, excluded when one or more parties oppose disclosure of the file (including leniency material). The question faced by the Oberlandesgericht Wien was whether Pfleiderer meant that it had to conduct the balancing exercise even when national law seemed to prohibit such an exercise. One legal commentator has already pointed to two elements with regard to the answer that may be given by the Court of Justice.114 First, he argues that one element of Pfleiderer seemed to indicate that the national court could not be forced to carry out the balancing exercise. In that respect, he drew attention to the statement of the Court of Justice pursuant to which there are no provisions of EU law precluding access to documents, which could, in his view, be interpreted as requiring the court to engage in a balancing exercise only where national law leaves them the freedom to do so. Second, he referred to Courage and Crehan,115 which, in his view, instructed national courts to disregard national procedural rules that made it practically impossible or excessively difficult for victims of competition law infringements to obtain compensation. He deduces from that judgment that it is theoretically possible that, even where a national legislature has decided to deny claimants access to the competition authority’s file, a national court could nevertheless allow it. In view of the uncertainty surrounding how to handle the tensions between access to case files and public enforcement of competition law at the level of the EU judge, of the national judge and of the national legislators, the appropriateness of an intervention by the EU legislator will now be examined.

3. Should the EU legislator intervene to strike the balance? The present section will, first, discuss the reasons militating in favour of the adoption of an EU legislative instrument (3.1.). Second, it will analyse the relevant proposals the EU legislator has already prepared and/or tabled (3.2.). Third, in 112 See Lewis Crofts, ‘English Court to Consult EC on Disclosure in Switchgear Cartel Damage Claim; Mulls ECJ Reference’, MLex, 15 June 2011. 113 See Case C‑536/11 Bundeswettbewerbsbehörde v Donau Chemie AG and Others, judgment of the Court of Justice of 6 June 2013, not yet reported. 114 See Völcker, case note, cited above note 78, at pp 712–713. 115 See Courage and Crehan, cited above note 15.

386  Competition, Regulation and Public Policies view of the latter proposals, it will explore the best available options for a future intervention of the EU legislator (3.3.).

3.1. Reasons militating in favour of an intervention The reasons militating in favour of an intervention of the EU legislator flow mainly from sections 2 and 3 above and will therefore only be recapped briefly in the present section, starting with the reasons that may be identified at Member State level and continuing with those that may be identified at EU level.

3.1.1. At Member State level It follows from section 2 that, at Member State level, an intervention of the EU legislator is warranted both to give guidance to the national judge with regard to the balancing exercise prescribed in Pfleiderer and to harmonize, at least to a certain degree, national rules on access by the general public to documents related to Article 101 and 102 TFEU procedures. Furthermore, an intervention of the EU legislator would considerably reduce the risks of forum shopping identified above. One may add that, besides looking for the most attractive jurisdiction in the Member States, damage claimants have also attempted to rely on United States laws to obtain evidence they needed to lodge successful judicial proceedings in Member States’ courts. Section 1782(a) of Title 28 of the U.S. Code empowers U.S. district courts to compel persons within its jurisdiction to produce evidence for ‘use in a proceeding in a foreign or international tribunal’.116 Enacting EU legislation could also diminish the incentive to try to employ that possibility. Finally, as a matter of principle, as one commentator has noted,117 one may consider that the issue of the interaction between transparency and public antitrust enforcement may, in certain circumstances, ‘have EU-wide effects’. To that extent, it is a ‘question of general EU interest that cannot be left to national courts’.

3.1.2. At EU level At EU level, an intervention of the EU legislator may be opportune, first, because no clear direction has so far been given by the EU Courts as to how to handle the interface between access to competition files and public enforcement of competition law. 116 See Vandenborre, ‘The confidentiality of EU Commission cartel records’ cited above note 13, at p 122; Eric Barbier de la Serre, ‘L’assistance judiciaire américaine au soutien d’une plainte en droit communautaire de la concurrence’, (2005/1) Revue de Droit des Affaires Internationales 35. 117 See Barbier de la Serre, ‘Beyond Effectiveness and Uniformity’, cited above note 15, at p 33.

Küllike Jürimäe 

387  

Second, as mentioned earlier, most of the relevant cases the EU Courts have so far dealt with were brought on the basis of Regulation No 1049/2001. As that regulation was enacted with a view to allowing citizens the widest possible access to documents of the EU institutions, it may not be the most appropriate tool to regulate transparency issues in competition law matters.118 This may be deduced by analogy from the rulings of the Court of Justice in Commission v Technische Glaswerke Ilmenau,119 Commission v Agrofert Holding120 and Commission v Éditions Odile Jacob,121 which have confirmed, in the State aid and merger fields, that some adjustments were necessary when applying the Transparency Regulation in those domains of law. It may therefore prove useful to exclude competition law from the scope of Regulation No 1049/2001 and to try and draft a piece of legislation concerning the right of access to competition material that is better suited to the particularities of that area of law.122

3.2. What has been done so far by the EU legislator? In order to contribute to the discussion on the way in which the EU legislator should intervene to regulate the interaction between transparency and public enforcement of competition law, one must present a review of the unsuccessful attempts made so far to legislate on the matter. In particular, efforts were made to amend Regulation No 1049/2001 and to regulate private enforcement.

3.2.1. Unsuccessful attempt to amend Regulation No 1049/2001 On 30 April 2008, the Commission proposed an amendment to Regulation No 1049/2001.123 This tackled the question of the interaction between transparency and, generally, the interests at stake when an investigation is pending. Article 2(6) of the proposal read as follows: Without prejudice to specific rights of access for interested parties established by [EU] law, documents forming part of the administrative file of an investigation or of proceedings concerning an act of individual scope shall not be accessible to the public until the investigation has been closed or the act has become definitive. Documents 118 See Idot, ‘Access to Evidence’, cited above note 12, at p 273; Idot, ‘Le règlement n° 1049/2001’, cited above note 33; Onno Brouwer and Daniel Mes, ‘The Tide of Transparency has reached the Antitrust Shores: the Use of Regulation 1049/2001 in EU Antitrust (Related) Proceedings’, in Carl Baudenbacher, Claus Gulmann and Koen Lenaerts (eds), Liber Amicorum en l’honneur de Bo Vesterdorf (Bruylant, 2007) pp 451 et seq, at p 472. 119 See above note 30. 120 See above note 1. 121 See above note 1. 122 See, eg, Idot, ‘Le règlement No 1049/2001’, cited above note 33. 123 Proposal for a Regulation of the European Parliament and of the Council regarding public access to European Parliament, Council and Commission documents, COM(2008)229 final.

388  Competition, Regulation and Public Policies containing information gathered or obtained from natural or legal persons by an institution in the framework of such investigations shall not be accessible to the public.

In substance, it follows from point 3.2. of the Explanatory Memorandum to the proposal that this new provision, if it were enacted, would have two practical consequences. On the one hand, it would completely exclude access by the general public to documents contained in the administrative file of the investigation until the investigation has been closed or until the relevant decision can no longer be challenged by an action for annulment. The idea is that, during the investigation, only the specific rules governing the field in which that investigation takes place apply, eg the regulations concerning competition and trade defence. On the other hand, information obtained by the relevant EU institution from a natural or legal person would be protected even after the relevant decision became definitive. In practice, this would mean, for example, that documents submitted to the Commission by leniency applicants would never be accessible.124 However, on 11 March 2009, the European Parliament amended the abovementioned proposal125 and set aside the new Article 2(6) that the Commission had proposed to insert within Regulation No 1049/2001. Nonetheless, as of early 2011, the European Parliament still had not taken a vote and adopted its official position at first reading. Accordingly, on 21 March 2011, almost three years after the legislative procedure came to a standstill, the Commission tabled a new amendment to Regulation No 1049/2001.126 This proposal is limited to the changes warranted by the entry into force of the Lisbon Treaty. Point 5 of the Explanatory Memorandum to this new proposal mentions that the discussions in the European Parliament and in the Council have shown strongly diverging views about amending Regulation No 1049/2001. Furthermore, point 8 of that Explanatory Memorandum states that the revised proposal is without prejudice to the ongoing procedure on the basis of the Commission’s proposal of April 2008. By October 2012, no revision of Regulation No 1049/2001, either on the basis of the 2008 or the 2011 proposal, had been adopted. This reflects the difficulty of reconciling transparency and the protection of competition law investigations.

124 See Eric Barbier de la Serre and Stéfanie Quiles, ‘Conjugaison du private enforcement et des programmes de clémence : que le législateur de l’Union tranche’, Revue Lamy de la Concurrence, no. 29, octobre-décembre 2011, pp 59 et seq, at p 65. 125 Proposal for a Regulation of the European Parliament and of the Council regarding public access to European Parliament, Council and Commission documents (recast) (COM(2008)0229 – C60184/2008 – 2008/0090(COD)), 2008 OJ C87E/362. 126 Proposal for a Regulation of the European Parliament and of the Council amending Regulation (EC) No 1049/2001 regarding public access to European Parliament, Council and Commission documents, COM(2011) 137 final.

Küllike Jürimäe 

389  

3.2.2. Unsuccessful attempt to regulate private antitrust enforcement In 2009, following publication of a White Paper (2 April 2008)127 and a Green Paper (19 December 2005)128 both on damages actions for breach of [EU] antitrust rules, the Commission prepared a proposal for a Council Directive on rules governing damages actions for infringements of Articles [101] and [102] of the Treaty.129 The proposal was never officially tabled. The approach adopted by that proposal may be regarded as a horizontal one. In point 1.2. of the Explanatory Memorandum to the proposal, the Commission highlighted that the practical exercise of the right to compensation was often rendered difficult or impossible and that, in practice, victims of infringements of EU competition rules only rarely obtained reparation of the harm suffered, thereby foregoing several billion euros each year in compensation. With a view to ensuring proper access to effective redress mechanisms, the proposal set out to remove all major obstacles standing in the way of victims’ right to obtain compensation. Accordingly, it gathered individual proposals designed to tackle issues as diverse as, for instance, the right to compensation, the right to collective redress, access to evidence and the recognition within the whole of the EU of the effect of decisions made by national competition authorities or national courts. With regard to access to evidence, it follows from recital 13 and Article 7 of the 2009 proposed directive that disclosure of evidence would have been subject to a decision of a court and under its strict control. A disclosure order would have been granted where a damage claimant had shown plausible grounds to suspect that it had suffered harm through an infringement of Article 101 or 102 TFEU.130 In addition, the party requesting disclosure would have had to: show that the evidence was relevant to substantiate its claim; specify the pieces of evidence it wished to obtain (or specify narrow categories of this evidence); and show that it was unable to obtain that evidence itself.131 However, courts would have been able to refuse to grant the order if the disclosure would be disproportionate. In assessing proportionality, the court would have had to consider, in particular, the possibility that the evidence to be disclosed might have contained confidential information.132 In that respect, the 2009 proposal would have required Member States to provide for effective measures so that national courts could protect confidential information to the greatest extent possible while also ensuring that relevant evidence containing such information was available in the action for damages.133 127 White Paper on damages actions for breach of [EU] antitrust rules, COM(2008)  165  final, 2 April 2008. 128 Green Paper on damages actions for breach of [EU] antitrust rules, COM(2005) 672 final, 19 December 2005. 129 See http://allegati.unina.it/postlaurea/perf/Exclusiva_Juridico.pdf. 130 Article 7(1) of the proposal for a Council Directive on rules governing damages actions for infringements of Article [101] and [102] of the Treaty. 131 Ibid, Article 7(2). 132 Ibid, Article 7(3). 133 Ibid, Article 7(3).

390  Competition, Regulation and Public Policies Moreover, Article 8 of the 2009 proposal provided that corporate statements made in the context of leniency programmes as well as settlement submissions would never be subject to disclosure. It also would have precluded national courts from ordering disclosure where a competition authority had shown that disclosure would undermine an ongoing investigation. Hence, the proposal is protective of the interests of the undertakings that cooperate in the investigation as well of those of the competition authorities. However, it must be stressed that the context in which this proposal was put forward was not very favourable, which probably explains why it was never officially tabled. The content of that proposal was largely inspired by that of the White Paper. Yet the overall reaction to the White Paper (which had been preceded by the Green Paper outlining the main obstacles to an effective EU system of damages claims) was unenthusiastic,134 as it created numerous concerns notably within certain Member States. One of the major concerns of Member States was that the proposal, as it stood in the White Paper, might lead to the development of a U.S.-style litigation culture. The proposal also raised certain legal issues such as the question of its legal basis.135

3.3. What should be done? Against this background, the question that naturally comes to mind is whether the time is now ripe for renewed action on the part of the EU legislator and, if so, in what form.

3.3.1. Is the time ripe for a new proposal from the EU legislator? It appears, as transpires from sections 2.1. and 2.2. above, that Pfleiderer has created the necessary momentum for renewed intervention of the EU legislator. This momentum is confirmed, first, by a resolution that the heads of the European Competition Authorities, meeting in the framework of the European Competition Network, adopted on 23 May 2012.136 In that resolution, they voice their concerns following the Pfleiderer judgment. In clear and emphatic terms they express the view that the protection of leniency materials is fundamental for the effectiveness 134 See, eg, Jeroen Kortmann and Christof Swaak, ‘The EC White Paper on Antitrust Damage Actions: Why the Member States are (Right to be) Less Than Enthusiastic’, 30 European Competition Law Review 340, 340–341 (2009). 135 Ibid, at pp 349–351. See also Veronica Pinotti and Dana Stepina, ‘Antitrust Class Actions in the European Union’, 2 Journal of European Competition Law and Practice 24, 31 (2011). 136 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, available at http://ec.europa. eu/competition/ecn/leniency_material_protection_en.pdf. See also Christophe Lemaire and Pascal Cardonnel, ‘Le Réseau européen de la concurrence adopte une “résolution” visant à protéger les documents issus des procédures de clémence en cas d’action en réparation, Concurrences, No 3-2012, pp 184–185.

Küllike Jürimäe 

391  

of anti-cartel enforcement; but they also welcome the development of private enforcement of competition rules, in particular through damages actions, as a complementary tool. The ripeness of a new legislative proposal is confirmed by the publication on 19 June 2012 by the European Parliament of a study entitled ‘Collective Redress in Antitrust’.137 Despite its title, the study, which was commissioned by the Committee on Economic and Monetary Affairs, exhaustively examines all issues related to the interaction between public and private enforcement and is not merely focussed on the possibility of lodging collective actions. It recommends the adoption, on the basis of Article 103 TFEU, of a regulation which would govern collective redress in the area of competition law as well as any related issues. The study suggests, inter alia, that leniency applicants should be protected from damages actions. It must be emphasized that the study does not reflect the European Parliament’s official position. To this end, the Commissioner and the Director-General for Competition recently announced that a new legislative proposal would be tabled by the end of 2012.138 They both explained, in substance, that it was necessary to clarify the relationship between private actions and public enforcement by the Commission and national competition authorities, notably as regards the protection of leniency programmes. This is confirmed by the Commission Work Programme for 2012,139 pursuant to which a Proposal for a Directive on rules governing actions for damages for infringements of competition law provisions is due to be submitted in December 2012.

3.3.2. What form might an intervention of the EU legislator take? Although in the past the EU legislator tried to regulate the interaction between transparency and public enforcement issues by amending Regulation No 1049/2001, this does not appear to be the best available option. First, that Regulation is a general instrument concerning access to documents. One may therefore argue it is not well suited to take account of the specific nature of access to competition law documents, as described, for example by the heads of the European Competition Authorities in their resolution mentioned above. Second, Regulation No 1049/2001 only applies at EU level; it does not regulate transparency issues that arise at national level. 137 European Parliament, Collective Redress in Antitrust (IP/AE/ECON/ST/2011-19, PE 475.120), available at http://www.europarl.europa.eu/committees/en/econ/studies.html. See also Christophe Lemaire, ‘La Commission des affaires économiques et monétaires du Parlement européen publie une étude sur les recours collectifs’, Concurrences, No 3-2012, p 185. 138 See Alexander Italianer, ‘Public and private enforcement of competition law’, speech, Brussels, 17 February 2012; Joaquín Almunia, ‘Antitrust Enforcement: Challenges Old and New’, speech, St. Gallen, 8 June 2012. 139 Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, Commission Work Programme for 2012, Delivering European Renewal, COM(2011) 777 final. See in particular the annex entitled ‘Planned Commission initiatives until end of 2012’.

392  Competition, Regulation and Public Policies It follows that, from a purely theoretical point of view, the best option to regulate the interaction between transparency and public enforcement issues seems to be, as suggested by the above-mentioned study published by the European Parliament, to develop a lex specialis in the area of competition law which would apply both at EU and at national level, and whose overall aim would be to strike an acceptable balance between public and private enforcement. Such a lex specialis would thus cover all legal issues relevant to such a balance, including the right to compensation, access to documents, collective redress, and leniency. Although it is not the aim of this chapter to make suggestions as to the substance of a future legislative act, it is worth noting that, with regard specifically to the interrelation between transparency, private enforcement and public enforcement of antitrust law, one of the main issues would be that of the status of leniency materials. In that respect, the mere fact that leniency would be addressed in a legislative instrument would to some extent confer legal status on leniency programmes as opposed to their current soft law character. Such a change of legislative status is essential to make sure that those programmes are not at risk when faced with general transparency rules which currently have the status of law.140 Furthermore, the study referred to above suggests reducing the civil liability of the first leniency applicant either by excluding the obligation to be jointly liable for the whole amount of the damage or by granting immunity from civil responsibility, unless the other cartelists become insolvent and the first leniency applicant is the only undertaking that has the capacity to compensate the cartel’s victims. In so doing, it deviates from the proposal for a Council Directive on rules governing damages actions for infringements of Articles [101] and [102] of the Treaty mentioned above, which provided for the exclusion of any disclosure of corporate statements made by leniency applicants.141 The solution put forward in the study may be an interesting avenue to explore in view of a future lex specialis, as the complete exclusion of disclosure of corporate statements may be extremely detrimental to the right of aggrieved parties in their attempt to obtain compensation.

Conclusion The present chapter has shown that an intervention of the EU legislator to regulate the interaction between transparency, private enforcement and public enforcement is long overdue. Apart from all the arguments mentioned above in support of such an intervention, one must remember that it is not the task of a judge to make policy choices. The judge’s role in this context should be limited Barbier de la Serre, ‘Beyond effectiveness and uniformity’, cited above note 15, at p 34. However, the study comes back to the solution that was proposed in the White Paper mentioned above. 140 141

Küllike Jürimäe 

393  

to balancing the different policies. An intervention of the EU legislator would have the advantage of setting out the main principles as regards the interrelation between transparency, private enforcement and public enforcement of competition law while still leaving it to the judge to undertake the required balancing exercise on a case-by-case basis; an exercise for which it is particularly well suited. It now remains to be seen whether the different actors involved in the legislative process will manage to reach an agreement about the numerous legal issues that should be tackled within the future legislative act.

Epilogue This chapter was completed in October 2012. Since then, the question of the interrelation between transparency and public enforcement of competition law in the EU has witnessed significant developments.142 This has necessitated some updates. On 6 June 2013, the Court of Justice delivered its judgment in Donau Chemie. It held that EU law precludes a provision of national law which forbids third-party access to documents forming part of the cartel file without the consent of all the parties to the proceedings and which excludes any possibility for national courts of weighing up the interests involved. To reach that conclusion, the reasoning developed by the Court of Justice is straightforward. Pursuant to Pfleiderer, national courts must carry out a balancing exercise to protect the effective application of, inter alia, Article 101 TFEU and the rights that provision confers on individuals.143 In particular, a rule which in principle denies access to any document forming part of competition proceedings is liable to make it impossible or, at the very least, excessively difficult to protect the right to compensation conferred on parties adversely affected by an infringement of Article 101 TFEU.144 Consequently, national law must not be developed in such a way as to preclude any possibility for the national courts to conduct the weighing of the respective interests on a case‑by‑case basis.145 The Austrian rule at issue effectively precluded national courts ruling on a request for access to the file from weighing the interests protected by EU law.146 A further interesting aspect of the judgment is that the Court of Justice expressed its opinion as to what the outcome of the balancing exercise should be. It insisted on the fact that a refusal of access to documents in the context of an application 142 See also Case T-380/08 Netherlands v Commission, judgment of the General Court of 13 September 2013, not yet reported. In this case, contrary to what it had done in EnBW, the General Court extended the notion of a ‘general presumption’ against disclosure to cover EU cartel law. I will not go into detail about that case. 143 See Donau Chemie AG, cited above note 113, paras 30–31. 144 Ibid, para 32. 145 Ibid, para 35. 146 Ibid, paras 36–37.

394  Competition, Regulation and Public Policies for leniency would entail a double benefit for leniency applicants. Not only would such applicants benefit from immunity from fines but they would also be – at least to a certain extent – protected from actions for damages.147 The Court stated that such a refusal of access should ‘be based on overriding reasons relating to the protection of the interest relied on and applicable to each document to which access is refused’.148 It added that ‘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’.149 Undoubtedly, the aim of going beyond what had been asked by the Oberlandesgericht Wien was to put an end to the legal uncertainty engendered by Pfleiderer. It is still unclear whether that objective has been achieved. As a matter of fact, the possibility of disclosure of leniency documents might deter cartel participants from taking advantage of a leniency programme. Does this constitute an actual or merely theoretical risk of undermining the effectiveness of the national leniency programme? Accordingly, it could be difficult for national judges to determine what the Court of Justice meant when referring to an actual risk of undermining the public interest. It is also worth underlining that while Advocate General Jääskinen in his opinion of 7 February 2013 considered that the Austrian rule at issue infringed EU law, his view as to how the relevant interests should be weighed differed substantially from what the Court of Justice stated in its judgment. In particular, contrary to the Court of Justice, which requested a document-by-document balancing exercise, the Advocate General insisted on the importance of affording specific protection to leniency programmes. In that respect, he held that access by third parties to voluntary self-incriminating statements made by a leniency applicant should not in principle be granted.150 He also made clear that, in his view, a rule which not only provides absolute protection for the participants in a leniency programme, but also requires the interests of other participants to a restrictive practice to be balanced against the interests of the alleged victims, would be more appropriate.151 To sum up, both Pfleiderer and Donau Chemie demonstrate the need for legislative intervention. In that respect, it must be noted that it is partly in reaction to the former judgment that the Commission adopted a proposal for a directive on private enforcement152 on 11 June 2013.153 Differing from the Court of Justice in Donau Chemie, the Commission deems it necessary to grant special Ibid, para 47. Ibid 149 Ibid, para 48. 150 Opinion of AG Jääskinen in Donau Chemie, cited above note 113, para 55. 151 Ibid, para 64. 152 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. See in particular point 1.2 of the Explanatory Memorandum accompanying the proposed directive. 153 For initial impressions of the Commission’s proposals, see Philip Lowe and Mel Marquis (eds), European Competition Law Annual 2011: Integrating Public and Private Enforcement of Competition Law – Implications for Courts and Agencies (Hart Publishing, 2014). 147 148

Küllike Jürimäe 

395  

protection to some documents provided by leniency applicants. More precisely, the Commission proposes to strictly prohibit national courts from disclosing corporate statements.154 It should be pointed out that the Commission included in its proposal a specific provision aimed at protecting leniency applicants from undue exposure to damages claims155 by limiting immunity recipients’ liability. Pursuant to that provision, immunity recipients are liable to cartel victims other than their direct or indirect purchasers or providers only when such victims are unable to obtain full compensation from the other undertakings that were involved in the cartel.156 In addition, where 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 would be determined in the light of its relative responsibility for that harm.157 Only time will tell whether this proposal will be more successful than those of the past.

Ibid, article 6(1)(a). Ibid, recital 28. 156 Ibid, article 11(2). 157 Ibid, article 11(4). 154 155

Cons. Rosa Perna*

The Role of Courts in Reconciling Competition, NonCompetition and Constitutional Imperatives: The Italian Experience

1. Antitrust law in the EU and Italian legal systems 1.1 The European Treaties The protection of free competition and of the related public and private interests is part of the genetic code of European Union law. This has been true since the establishment of the European Economic Community.1 The basic idea of European antitrust law is that competition between enterprises helps enhance progress and general welfare, and promotes the efficiency of the economic system, whereas a lack of control on the market might cause a progressive reduction in economic pluralism. The protection of the consumer has long been seen as an additional and derived effect of the protection of competition and free market. In the Treaty of Rome, competition was the object of Community policy, and a standard to be reached. With the Treaty of Maastricht, it became an informing principle of Community antitrust law.2 The objectives and values of the Community, and now the Union, have been increasingly consolidated. Competitive markets are today part of the bedrock of European economic life.

* Administrative judge, Rome. Member of the European Association of Administrative Judges (EAAJ). 1 These figurative expressions are borrowed from Roberto Caponigro, ‘Interessi e regole di tutela negli ambiti nazionale e comunitario’, a presentation delivered at a seminar entitled ‘L’Europa del diritto: i giudici e gli ordinamenti’, Lecce, 27–28 April 2012. 2 See Maria De Benedetto, ‘Il principio della concorrenza nell’ordinamento italiano’, Rivista della Scuola Superiore dell’economia e delle finanze, 12/2004, pp 103 et seq. (thorough and diachronic analysis of the principle of competition in the Italian law system).

398  Competition, Regulation and Public Policies

1.2 The Italian legal system 1.2.1 The freedom of economic initiative On the other hand, in 1947 the Italian Constitution under Article 41 explicitly recognized the freedom of economic initiative of individuals;3 the first paragraph of that provision guaranteed the right of enterprise as an individual right, whereas the third paragraph offered constitutional guarantees to public powers in order to subordinate the right of enterprise or its exercise for reasons of public interest.4 In Article 41, therefore, there was no organizing idea of the market but the mere need to protect the market itself from the possible realization of collectivist ideologies.5 The concept of competition as such was thus initially absent from the Italian Constitution. In academic literature6 it has been pointed out that the freedom of economic initiative does not concern the principle of competition, as the first one is ‘vertical’ insofar as it indicates spheres of action of the individual protected as against the State. By contrast, competition implies a horizontal relationship among entrepreneurs, ie among persons who exercise the freedom of economic initiative. The relationship between Article 41 of the Constitution and the Treaty provisions on competition has also been described as the connection between a subjective and an objective phenomenon, considering that the constitutional rule establishes a subjective situation of individual freedom, whereas European law expresses a model of economic and legal relationships. More precisely, Article 41 guarantees the protection of the right of enterprise but does not entrust itself with the choice of the system for the process of liberalization and the model of market as a basis for economic relationships. European law, on the other hand, takes the opening of the market and the freedom of competition as necessary premises for those economic interactions.7

1.2.2 The principle of competition Law No 287/1990 later introduced provisions into the Italian legal system, which correspond to the then-Community antitrust rules. Oddly enough, Article 1(1) states Article 41 of the Constitution provides: 1- Private economic enterprise is free. 2- It may not be carried out against the common good or in such a manner that could damage safety, liberty and human dignity. 3- The law shall provide for appropriate programmes and controls so that public-sector and private-sector economic activity may be oriented and co-ordinated for social purposes.’ 4 See De Benedetto ‘Il principio’, cited above note 2 (citing M.S. Giannini, Diritto pubblico dell’economia, Il Mulino, 1992, p 175). 5 Ibid. 6 Ibid. 7 See Fabio Cintioli, ‘L’art. 41 della Costituzione tra il paradosso della libertà di concorrenza e il ‘diritto della crisi’’, Diritto e società, 3–4/2009, pp 373 et seq. 3

Cons. Rosa Perna 

399  

that the provisions of the Law itself are adopted to give effect to Article 41 of the Constitution, which as explained above provides for the protection of the right of economic initiative. Article 1(4) adds that the interpretation of the Law is to be made in accordance with the principles of Community (now EU) antitrust law. Considering that Article 41 did not really articulate a principle of competition, and that Law 287/1990 nevertheless incorporated by reference Community principles for its own interpretation, several authors have envisaged the risk of conflict between the two legal systems, ie the constitutional and the antitrust provisions. Various theoretical solutions for their reconciliation have therefore been proposed. Nevertheless, the issue lost its relevance with the adoption of Constitutional Law No 3/2001, which reformed Title V of the Constitution. Article 117(2)(e) thus explicitly acknowledged the principle of competition, inscribing ‘competition protection’ among the matters reserved to the exclusive legislation legislative power of the Italian State (as opposed to the Regions). Moreover, according to Article 117(1) as modified, legislative powers are to be exercised – both by the State and by the Regions – in compliance not only with the Italian Constitution but also with the constraints deriving from European Union legislation and from international obligations. The constitutional affirmation of the principle of competition has given a clear and distinct constitutional basis to the Italian antitrust law; in addition, all constraints of European derivation in the matter must be respected by national and regional legislator. As a consequence, all antitrust laws and regulations that the national judge is required to apply, of both internal and European origin, flow directly from, or are recognized by, the fundamental law of the State. This must be kept in mind when considering possible conflicts of laws and principles. As a consequence of the introduction of the Community/EU principle of competition, ‘the market’ has been incorporated within the Constitution as a positive value for the promotion of economic and organizational efficiency.8 The Constitutional Court has subsequently intervened to rule on the new distribution of competences,9 and in so doing has extensively qualified the notion of the protection of competition so as to encompass competitive relations on the market and not to exclude promotion of competition by the State, while leaving Regions free to adopt more favourable pro-competitive measures.

1.2.3 Economic liberalizations and market regulation Recent national legislation in the matter of economic liberalizations must be taken into account as an expression of a general tendency, on the one hand toward a reduction of public measures that restrict access to markets, and on the other 8 See Giuliano Amato, ‘Il mercato nella Costituzione’, in Quaderni costituzionali, 1992, cited in De Benedetto (above note 2). 9 Constitutional Court, decision No 14 of 18 December 2003.

400  Competition, Regulation and Public Policies hand toward a ‘softening’ of the administrative discretion in the adoption of those measure still in force.10 In the first case, there is a mitigation of certain quantitative or numerical parameters imposed in a certain economic sector; to this end, Decree No 201 of 6 December 2011 (the so-called Decreto salva-Italia), and in particular Article 34(3) thereof, by means of a general provision abolishes restrictions concerning geographical areas or minimum distances among enterprises. The provision was explicitly adopted in order to guarantee free competition under conditions of equal opportunity, the correct functioning of the market and a uniform access for consumers to goods and services in the national territory. In the second case, a market is opened to competition, but this intervention requires that the access of new operators be made not only possible but also profitable by the adoption of pro-competitive measures. As a consequence, the process of liberalization has led to the emergence of a new administrative function, ie market regulation, which is entrusted to neutral, independent administrative authorities and aimed at promoting conditions of competition on the liberalized markets. However, there is no conflict or overlapping with the scope and aim of action of the Italian Antitrust Authority. The administrative authorities pursue certain specific public interests such as the granting of certain standards in public services, the supervision of tariff rates, the transparency of markets, the sound and prudent operation of intermediaries, the freedom and secrecy of communications, and so on.

2. Judicial review of acts of the Italian Antitrust Authority 2.1 Introduction In the Italian legal system where, as in other European countries, the prevailing opinion does not endow independent authorities, including the Antitrust Authority, with a judicial character. The activities of such institutions are thus subject to the control of the courts.11 The involvement of judges in the matter of competition law has in fact occurred more in relation with the measures of the Authorities than with the disputes between individuals. 10 For a concise but complete analysis, see Roberto Giovagnoli, ‘Liberalizzazioni, semplificazioni ed effettività della tutela’, presented on the occasion of the Premio Sorrentino 2012 award seminar, Rome, 12 June 2012. 11 See Enrico Galanti, ‘Discrezionalità delle Autorità amministrative indipendenti e controllo giudiziale’, Quaderni di Ricerca Giuridica della Consulenza Legale, no. 64, June 2009, http://www. bancaditalia.it/pubblicazioni/quarigi/qrg64/qrg_64/Quaderno_64.pdf.

Cons. Rosa Perna 

401  

To carry out such control, the judge has had to ‘specialize’ very rapidly in the subject, trying not to be conditioned, on one hand, by a sort of diffidence towards a brand new authority and, on the other hand, by a certain deference in reviewing acts of a high level of technicality. In this framework, the public interest has been placed in the protection of competition itself, without claims for directing the activity of private citizens.12

2.2 The European perspective on judicial review Of course, any analysis of judicial control of the acts of an antitrust authority cannot be reasonably confined within the boundaries of a single legal system. The analysis must necessarily reckon with the EU law of competition, which has seen the European Commission and the Court of Justice, by means of a constant interaction, performing a leading role in the growth of the national antitrust authorities and of the relative jurisdictions within the Member States. The national judges, then, have become part of a system which encompasses the jurisdictions of all the member States and which uses the case law of the EU Courts as a frame of reference. This process of harmonization has led to a wider phenomenon of close interaction among the various national administrative laws.13 In a sort of circular causation, the comparison among different models pertaining to Member States contributes to the creation of an EU rule, which in turn affects the interpretation of domestic rules, in a process of reciprocal harmonizing integration. The EU Courts control the ‘legality’ of the acts of the Commission, and the scrutiny of the Courts is extended to the merits of the Commission’s decision as far as fines are concerned.14 The Treaty on the Functioning of the European Union does not contain any rules regarding the kind of judicial review that can be performed by the General Court and the Court of Justice with regard to the Commission’s competition decisions. However, the EU case law has opted for a type of judicial review which, without limiting itself to a formal verification of the procedural rules, comes to a full verification of the facts and of the criteria for the application of the antitrust rules (ie rules prohibiting agreements restricting free competition and abuses of a dominant position); whereas, in the case of complex economic evaluations, the Courts must restrict themselves to considering the evaluation performed by the Commission, verifying the respect of the proceeding rules and of the motivation, the material correctness of the facts and the lack of an abuse of power or of a gross error in evaluation – but without substituting Margherita Ramajoli, Attività amministrativa e disciplina antitrust, Milano, 1998, at p 356. On the issue of judicial review of the antitrust measures we cannot but refer to the ample and rich report by Roberto Chieppa, ‘Jurisdictional Control over the Decisions of the Antitrust Authorities’, in Enrico Adiano Raffaelli, ed., Antitrust between EC law and National Law, Giuffrè-Bruylant, 2005, pp 103 et seq. 14 See Article 261 TFEU and Article 31 of Regulation No 1/2003. 12 13

402  Competition, Regulation and Public Policies their own evaluation.15 The Court of Justice expressly denies the possibility of a ‘substitutive’ control of the judge on complex technical evaluations carried out by the Commission.16

2.3 Judicial review in Italy EU case law leaves to the national law of each Member State the choice of the competent judges and the procedural conditions of appeals aimed at protecting the individuals’ rights based on EU legal precepts, as long as these conditions: are not less favourable than those concerning analogous internal appeals (principle of equivalence); and do not make impossible or excessively difficult the protection of the rights conferred by EU law (principle of effectiveness).17 The Italian legislator has assigned the task of reviewing the decisions of the Antitrust Authority to the Administrative Courts, which have exclusive jurisdiction pursuant to Article 133(1)(1) of Legislative Decree No 104 of 2 July 2010,). At first instance, actions for annulment must be brought before the Administrative Tribunal of Rome, or ‘TAR’. (Ibid., Article 135(1)(b).) By contrast, actions under contract law and actions for damages18 are attributed to the ordinary courts. Two different jurisdictions, administrative and ordinary, are therefore respectively competent for public and private enforcement. In the field of competition, legal and economic knowledge is so intermingled that the judge is required to be legally correct and economically proper at the same time. In fact, considering that the administrative power antitrust law is basically marked by the so-called ‘technical discretion’,19 the judge has to interpret the rules even in fields where he lacks specific competence. Needless to say, the lack of technical knowledge on the part of the judge may also occur in other fields, namely in those left to the (regulatory) competence 15 The EU case law on the subject is broadly recalled in Marta Moretti, ‘Valutazioni economiche complesse in materia antitrust e self restraint dei giudici dell’U.E.’, Diritto dell’Unione Europea, 2/2009, pp 315 et seq. 16 Case C-225/91 Matra SA v Commission [1993] ECR I-3203; Case C-157/96 The Queen v Ministry of Agriculture, Fisheries and Food, Commissioners of Customs & Excise, ex parte National Farmers’ Union [1998] ECR I-2211. 17 See Chieppa, ‘Jurisdictional Control’, cited above note 13; Case C-120/97, Upjohn Ltd v The Licensing Authority established by the Medicines Act 1998 and Others [1999] ECR I-223, discussed in Roberto Caranta, ‘Tutela giurisdizionale effettiva delle situazioni giuridiche soggettive di origine comunitaria and incisività del sindacato del giudice nazionale (Kontrolldichte)’, in Rivista italiana diritto pubblico comunitario, 2/1999, pp 503 et seq. 18 Actions for damages vis-à-vis the Authority are within the exclusive jurisdiction of the administrative judge. 19 The technical discretion is connoted by the application of rules of a non-exact science having a certain degree of disputability; it therefore differs from the ‘technical verification’, where an exact science is applied in order to reach a certain outcome. It also differs from ‘administrative discretion’, where a public authority acts according to a public purpose, for which it is responsible under the law, by means of selection, acquisition, comparison and evaluation of public and private interests implied in its action.

Cons. Rosa Perna 

403  

of (other) independent administrative authorities;20 in the case of antitrust law, however, a further difficulty arises owing to the use, at a legislative and administrative level, of many ‘indeterminate juridical concepts’, such as those implied in the definitions of ‘abuse of dominant position’, ‘relevant market’ and ‘restrictions of competition’, since EU and national antitrust rules give meaning in law to concepts belonging to economics. As for the limits of the national court’s review of the acts of the antitrust authority, it has been pointed out that the administrative court can check the facts considered in the antitrust proceedings as well as the evaluation process by which the authority has applied the law. However, where the legitimacy of action and the correct use of the underlying technical rules have been ascertained, the review cannot go so far as to substitute the court’s evaluation for the one already carried out by the administration, who remains the only entitry that exercises antitrust powers.21 Over time, the national courts have developed a stronger, more incisive review of the antitrust authority’s decisions. This review has been deemed to consist of a comprehensive re-examination of the technical evaluations made by the authority, as well as of the economic principles and the indeterminate legal concepts it has applied;22 and it is to be conducted having recourse to rules and technical knowledge belonging to the same disciplines applied by the administration.23 In this way, the court can both reassess the technical choice of the administration and apply the correct interpretation of the relevant indeterminate legal concepts to the factual controversy.24 At the same time, the previous distinction between ‘strong’ and ‘weak’ judicial review has been abandoned.

2.4 Judicial review and the protection of individual legal situations The Italian legislator’s choice to give exclusive jurisdiction to the administrative court to review administrative antitrust decisions has drawn a lot of academic criticism. In the Italian system, the administrative court has jurisdiction not only for the protection of legitimate interests, for which he is ordinarily competent (Article 103 of the Constitution), but also for the protection of individual rights. Legitimate interests, or legally protected interests, can be defined as the advantages granted to an individual who is subject to the administration’s public power. The ‘owner’ of the interest can thus influence the proper exercise See Galanti, ‘Discrezionalità delle Autorità’, cited above note 11. Among many others, see Cons. Stato, VI, 12 February 2007, No 550; Cons. Stato, VI, 10 March 2006, No 1271; TAR Lazio, Roma, I, 24 August 2010, No 31278; TAR Lazio, Roma, I, 29 December 2007, No 14157; TAR Lazio, Roma, I, 30 March 2007, No 2798; TAR Lazio, Roma, I, 13 March 2006, No 1898. 22 Cons. Stato, VI, 8 February 2007, No 515. 23 Cons. Stato, VI, 23 April 2002, No 2199. 24 Cons. Stato, VI, 2 March 2004, No 926. 20 21

404  Competition, Regulation and Public Policies of administrative powers. Generally speaking, the exclusive jurisdiction of the administrative court aims at providing certainty as the legal means of redress in matters where it is hard to define the boundary between rights and legitimate interests.25 However, in the field of competition law, the application of legal provisions and/or administrative acts is deemed wanting in any degrading effect with regard to the legal positions of infringers. As a consequence, such individual positions represent rights, often of a constitutional rank, whose recognition should be attributed to the ordinary judge, the latter being the natural judge of the rights in question.26 On the other hand, it has been pointed out that the compatibility of legitimate interests with formally restricted structures of power depends on an evaluation implying technical discretion. Legitimate interests may therefore be compatible with the notion of fundamental rights (such as the freedom of economic initiative) when the exercise of power is bound to indeterminate legal concepts, as it is in antitrust proceedings.27 As has been noted, at times the Italian Antitrust Authority exercises real discretionary powers. This is the case, for example, when it adopts decisions in the field of concentrations taking account of relevant interests in the national economy (Article 25 of Law No 287/1990).28 Furthermore, a complainant has a legitimate interest if it has a substantial interest that is opposed to anticompetitive conduct and if he contests the failure of the Authority to execute its repressive powers. In EU law, the distinction between rights and legitimate interests is unknown. However, the national qualification of subjective juridical situations is a matter to which the EU law system is indifferent, provided the principle of effectiveness is respected. The exclusive jurisdiction of the administrative judge thus does not entail any ‘EU contradiction’,29 considering the similarities between the EU action 25 See Ignazio Marino, ‘Autorità garante della concorrenza e del mercato e giustizia amministrativa’, in Diritto dell’economica, 3/1992, pp 578 et seq. 26 See Chieppa, ‘Juridictional Control’, cited above 13 (citing G. Ghidini and V. Falce, ‘Giurisdizione antitrust: l’anomalia italiana’, Mercato Conccorenza Regole, 2/1999, pp 317 et seq.). 27 Ibid, citing F. Merusi, ‘Giustizia amministrativa and autorità amministrative indipendenti’, Diritto amministrativo, 2/2003, pp 181 et seq.). On indeterminate juridical concepts, see Caponigro, ‘Interessi e regole di tutela’, cited above note 1, citing Salvatore Veneziano, ‘Il controllo giurisdizionale sui concetti giuridici a contenuto indeterminato e sulla discrezionalità tecnica in Italia’ (2005), www. giustizia-amministrativa.it. Veneziano holds that the category of indeterminate juridical concepts relates to a particular legislative technique – prone to the need of the legal system’s flexibility – in which a legal provision, in order to identify a fact as producing legal effects, does not describe the fact itself in a precise and exhaustive manner but allows for an integrative operation by the interpreter by using undetermined concepts that will be completed and specified with extra-juridical elements or criteria that are inserted into the legal paradigm. The problem of the extension and scope of the indeterminate concepts is that such legal concepts are first interpreted and applied by the administrative authority and only later, upon appeal against the administrative act, by the administrative judge. 28 See Chieppa, ‘Jurisdictional Control’, cited above note 13. 29 See Ghidini and Falce, cited above note 26, at p 326.

Cons. Rosa Perna 

405  

for annulment under Article 263 TFEU and the Italian action for annulment of administrative acts.30 The qualification of individual positions from an EU law perspective in terms of legitimate interest and the assignment of their recognition to a court different from the ordinary judge, ie the administrative judge, is thus proper and justifiable whereas, in accordance with the principles of full and effective jurisdictional protection of those positions, a contraction of their protection by the courts would not permissible.31

3. Reconciling competition, non-competition and constitutional imperatives 3.1 The issue at a glance At this point of the study some reflections are called for. In the field of competition law, the application of legal provisions and/or administrative acts may be challenged if it degrades the legal position of those subjected to antitrust measures: such individual positions represent rights, often of a constitutional rank, expressing values and protecting interests that may not always coincide, or can even collide, with those underlying the antitrust rules. From another prospective, as it has been inferred, a position of legally protected interest may not be incompatible with the notion of fundamental rights (such as the freedom of economic initiative) when the exercise of power is – as in the antitrust proceedings – bound to use indeterminate juridical concepts. The interpretation of such general clauses may represent a tool given to the judge in order to resolve conflicts between two rules of law by moulding the economic concepts implied in the legal text. Article 41 of the Italian Constitution explicitly recognizes the freedom of economic initiative of individuals but at the same time enables public powers to subordinate the right of enterprise or its exercise where necessary for the pursuit of a public interest. This twofold provision expresses an inner ideological tension of the Constituent Assembly which, in terms of practical application, can lead to situations of conflict between an individual’s right of enterprise and the exercise of public powers in defence of the market, an interest transcends the interest of individuals. See Caponigro, ‘Interessi e regole di tutela’, cited above note 1. Ad. Plenaria, Cons. St., 29 July 2011, n. 15 permits an atypical action for ‘ascertainment’. This action is not provided for by the Italian Code of administrative proceedings, but if sustained by an actual and present interest, any time the codified actions do not satisfy properly the need for the protection of recognized rights, other types of actions may be necessary in accordance with the constitutional and European precepts recalled by Article 1 of the Code itself. 30 31

406  Competition, Regulation and Public Policies The constitutional affirmation of the principle of competition by means of Law No 3/2001 and Article 117(2)(e) has given a clear and distinct constitutional basis to the antitrust law in Italy. The Italian antitrust provisions are therefore the direct expression and application of a constitutional principle; for this reason, their legitimacy can be justified and must be scrutinized according to constitutional principles. In case of conflict with other national laws, the hierarchy of norms should apply, steering the interpreter towards a solution respectful of constitutional values. Furthermore, by virtue of the constitutional precept laid down in Article 117(1), all constraints deriving from EU law and international obligations must be respected by the national and regional legislators. It follows that European antitrust law has risen to a constitutional rank and binds the exercise of legislative power at both national and regional level. As the Constitutional Court has stated,32 in direct actions for constitutional review of newly enacted national and regional legislation, the provisions of EU law serve as interstitial rules through which the conformity of that legislation with the Constitution must be tested. In other words, it is possible in practice to apply the limits laid down in the first paragraph of Article 117 of the Constitution, with the result that a national or regional provision that is incompatible with EU law will be declared unconstitutional because it infringes Article 117(1). It infringes, in particular, the obligation on the part of the Italian legislator to comply with the duties deriving from EU law. All in all, antitrust laws and regulations that the national court is required to apply have a direct origin in, or direct recognition from, the fundamental law of the State. Moreover, recent economic liberalizations have opened up markets to more competition; this has required the adoption of pro-competitive measures in order to make it not only possible but also profitable for new operators to gain access to those markets. A new administrative function, the regulation of markets, has thus emerged. The aim of this regulation is to promote conditions of competition on the liberalized market but it also seeks to secure specific public goals assigned to it. Although, in theory, there should be neither any contrast nor any overlap in the aims and scope of action between antitrust and regulation, in practice some unclear situations of conflict may occur.

3.2 The principle of competition as defined by the Italian Constitutional Court Over time, the Constitutional Court,33 often while scrutinizing the constitutionality of regional laws suspected of encroaching upon the scope and competence of the Case C-169/08 Presidente del Consiglio dei Ministri v Regione Sardegna [2009] ECR I-821. Observations made in this section follow the organic and complete survey on the constitutional judgments after the 2001 reform conducted by Anna Argentati, ‘La giurisprudenza della Corte Costituzionale in materia di ‘tutela della concorrenza’ a dieci anni dalla riforma del Titolo V della Costituzione’ (October 2011), http://www.agcm.it/component/joomdoc/doc_download/3003-sr5.html. 32 33

Cons. Rosa Perna 

407  

legislation of the State under Article 117 of the Constitution, has further defined the principle of competition, stating that the protection of competition has to encompass both a static (or conservative) dimension, prearranged to maintain existing competitive markets which may be to distortive strategies of enterprises or public subjects; and a dynamic (or proactive) profile, aimed at liberalizing markets and promoting the establishment of competitive market structures (decision No 14 of 13 January 2004). According to the Constitutional Court, the protection of competition is only one of the instruments of national economic policies. But the concept of competition must reflect the notion operating at EU level, which emerges from regulatory interventions, antitrust regulation and measures designed to promote an open and competitive market. With regard to regulatory interventions, decision No 339 of 16 December 2009 – recalling an earlier decision (No 336/2005) – legitimated State legislation (d.l. No 112/2008)34 regulating the energy and communications sectors, the Court having recognized the importance of strategic infrastructures and of an efficient network for the economic development of the country, and for competition. The realization of programs for access to critical energy infrastructure was thus necessary to develop a free competition on the market. A similar approach was taken in decision No 246 of 24 July 2009 and in decision No 142 of 23 April 2010 in the field of water management, where it was found that a State planning activity conducive to integrated water management was permissible for the purpose of overcoming fragmentation in the management of water resources. In the cases mentioned above, the highlighted importance of values such as integration and planning in strategic fields may have found a counterpart in the individual right of enterprise, whose protection is nevertheless guaranteed by a constitutional precept as well (Article 41 of the Constitution, as discussed earlier). The dialectic comparison among competition and individual rights of economic initiative has been dealt with in more resolutely, and formally reconciled, in the sectors of professional tariffs and taxi services. The relevant pronouncements were made in cases brought by certain Regions concerning the constitutional legitimacy – on grounds of alleged encroachment on regional legislative competences – of some of the provisions of Decree-Law No 223 of 4 July 2006 (the so-called ‘Decreto Bersani’). In decisions No 443 and No 452, both of 21 December 2007, the Court declared the questions to be unfounded. The challenged provisions were aimed at promoting more competitive market structures and promoting, at the same time, the freedom of choice of consumers. In particular, Article 2 of the Decree abrogated the legislative and regulatory provisions of fixed or minimum tariffs for liberal professions; Article 6 allowed the allocation of new taxi licences and even the combination of several of them in the hands of the same person. 34 Decree Law n. 112/08 had been contested by certain Regions on the basis that it could violate concurrent regional legislative competences in the fields of energy and government of the territory.

408  Competition, Regulation and Public Policies Shop opening and closing times have also been scrutinized by the Constitutional Court. In decision No 150 of 21 April 2011, the Court first observed that the regulation of shops’ opening times was a matter of ‘commerce’, and thus belonged to the competence of the Regions according to Article 117(4) of the Constitution. Nevertheless, the Court considered it necessary to evaluate whether the relevant regulations were capable of harming the protection of competition. In the case at hand, which concerned Abruzzo’s Regional Law No 17 of 12 May 2007), the Court stated that the ‘regulation on Sunday and holiday opening of retailers’ not only pursued the same objective envisaged by Legislative Decree No 114/98 (which had first liberalized the market) – but that it had opened the market and eliminated barriers and constraints to the free pursuit of economic activity. It had even broadened the scope of the liberalization, increasing the number of days in which ‘Sunday and holiday opening’ was allowed. In doing so it enhanced the free choice of consumers and enterprises as well. The Region had thus exercised its competence in the matter of commerce by way of a regulation that not only did not impinge upon the prerogatives of the State but also produced pro-competitive effects, albeit in a marginal and indirect way. Of major relevance with regard to competition was the Court’s scrutiny of regional legislation imposing higher charges, both economic and administrative, than those provided for by the ‘Code of Electronic Communications’ upon private operators (ie, Law n. 259/2003 implementing EU directives in the matter of electronic communications). More specifically, a regional tax (Tuscany’s Regional Law No 54 of 6 April 2000) had imposed, in breach of Article 93 of the Code, charges for verifications and controls of installations of mobile phone telephony in the regional territory. With decision n. 272 of 22 July 2010, while declaring the illegitimacy of the contested provisions under Article 117 of the Constitution, the Court nevertheless stated that Article 93 expresses a fundamental principle in the legal system of communications, as it prohibited the imposition of (other) charges on operators in order to grant a uniform and non-discriminatory treatment to all of them, the aim of the provision also being the protection of competition in the form of equality of treatment (see also decision No 336/2005). The distribution of medicine is the sector in which it has been most difficult for competition-based reasoning to penetrate into the Court’s approach. This is explained by the importance attributed by the Court to the right to health. In decision No 27 of 4 February 2003, the Court considered the legitimacy of restraints on opening periods of pharmacies established by a regional law, specifically, Lombardy Regional Law n. 21 of 3 April 2000. The Law stated that the legislative choice aims at pursuing public health, which implies limits on competition among pharmacies that are of instrumental nature. Viewing the restraints on opening periods as having the same motivation as that of the numerical rationing of pharmacies, and having ascertained that numerical rationing aims at a better performance of public service, the Court concluded that restraints on

Cons. Rosa Perna 

409  

opening periods ensure the territorial and time continuity of the service to citizens and guarantee a certain body of users to retailers. The Court did not evaluate the adequacy and proportionality of the measure but expressed a judgement as to the inner consistency of the legal set of rules in force, not without adding that a remoulding of the legislative wording would fall outside the scope of the Court’s inquiry.35 Decision No 76 of 28 March 2008 concerned Article 32 of the Constitution, which protects the right to health. In this case the Court upheld the constitutionality of a State law in the pharmaceutical sector (Royal decree No 1265 of 27 July 1934). The law had conditioned the opening of a pharmaceutical seat in derogation to the demographic criterion on ‘the verification of sole objective requirements linked to topographic conditions and practicability’ (see also decision No 295/2009). After recalling the considerations already made in previous decisions, the Court stated that the right to health legitimates such planning in order to guarantee the broadest and most rational coverage of the whole territory in the interest of citizens’ health. It is worth noting that in these decisions from the pharmaceutical sector the Court did not inquire whether the legal constraints scrutinized were necessary and proportionate, and whether the general interest would not have been better served by less restrictive measures, such as the introduction of a minimum, and not a maximum, limit on the pharmaceutical seats permitted in a given territory.36 Competition was thus perceived more as a risk than as an opportunity for the general public. To wrap up this survey on some of the constitutional pronouncements of the last decade, we can conclude (see in particular decision No 235 of 22 July 2011) that the protection of competition has been enriched legislative provisions that: a) target commercial conduct that negatively influences the competitive structure of markets (antitrust measures); b) aim at opening a market or at consolidating its opening through the elimination of barriers to access, reduction or elimination of restrictions on the free pursuit of entrepreneurial activities and competition (protection of competition in the market); c) pursue the objective of structuring competition to realize the widest opening of the market to all economic operators (protection of competition for the market). All in all, the protection of competition encompasses not only the application of antitrust laws, provisions on liberalization of markets and competitive procedures for public procurement, but also all initiatives aimed at supporting the competitiveness of the system and the development of the market, considered as an essential infrastructure for the achievement of real competition.37 35 For this case and other related case law, see Nicola Salerno, ‘Le farmacie nel diritto dell’economia’, Il diritto dell’economia, 1/2011, pp 99 et seq. 36 See Argentati, ‘La giurisprudenza della Corte Costituzionale’, cited above note 33. 37 Ibid.

410  Competition, Regulation and Public Policies

3.3 The national judge between constitutional precepts and EU law constraints 3.3.1 National judges and national laws Even in the field of competition, where the EU Treaty and the EU Courts determine most of the substance of the applicable law, national judges apply, first and foremost, national laws. Although it is common to speak of a paradigm consisting of a coherent and uniform application of the antitrust legal framework across the EU, it is important to keep in mind that the starting point for a judge is typically the national law framework,38 at least when checking whether the formal requirements of an appeal are met. In addition, it must be considered that national procedural rules, both for the antitrust authorities and for the Courts, and the organizational set-up of the authorities as well as the judicial review system, are established by the national legislator. It is of course commonplace that judges have to consider and respect the requirements of EU law, even when applying national procedural law. That may well mean accepting an appeal brought by an operator which according to national law would not have had legal standing but which nevertheless by European standards, if the relevant criteria are met, has to be considered as an ‘undertaking’ affected by the decision of a national antitrust authority. At the core of Italian antitrust, Law No 287/1990 has introduced provisions that correspond to the EU antitrust rules. They aim at preventing the restriction of competition within the domestic market or in a relevant part of it. According to the wording of Article 1, the provisions of the Law give expression to Article 41 of the Constitution, which as explained earlier protects the freedom of economic initiative. Article 1 also makes clear that the interpretation of the Law must be in accordance with the principles of EU antitrust law. Here a first task for the judge comes into sight, as he is implicitly called on to articulate the content of the freedom of economic initiative and to link it with the substance of EU antitrust law, which provides the basis of his interpretation. He must do this in particular when he uses his judicial discretion to review the use of technical discretion in the antitrust acts contested before the court. When dealing with national rules, the judge is subject (only) to the law (Article 101 of the Constitution). He does not have the power to pronounce a national law to be illegal or unconstitutional. Even if a legislative provision is contrary to the principles set out in the Italian Constitution, that provision can only be repealed, or declared unconstitutional, by the Constitutional Court. 38 See Hans Peter Lehofer, ‘National Judges and EU Institutions’, presentation made at a seminar on ‘Implementing the Revised Regulatory Framework in Electronic Communications. The Role of National Judiciary’, Brussels, 28 November 2011.

Cons. Rosa Perna 

411  

Consequently, Italian courts do not have a power to ‘disapply’ a law where they have a doubt about a law’s compatibility with Constitution. If a judge has such doubts, he must submit a request for a preliminary ruling to the Constitutional Court. In doing so he will indicate that the Italian rule is alleged to be inconsistent with one or more provisions or principles laid down in the Italian Constitution. If the Constitutional Court declares that the rule is unconstitutional, the rule is repealed from the legal system. This judgment of unconstitutionality does not apply only inter partes; it has erga omnes and is binding on all Italian courts.

3.3.2 The European dimension As national judges, we touch on the European dimension in different ways. First of all, there is the straightforward task of applying EU law and national laws implementing EU directives. In the course of this exercise it is possible to put questions to the European Court of Justice if issues arise that are not acte clair and if the Court of Justice has not yet resolved them.39 In other cases, when a court considers that a provision of Italian law doesn’t comply with EU law,40 and if it has no doubts about the interpretation of the European provisions, the court cannot declare the Italian law to be inconsistent with European law but it must disregard the national provision by virtue of the precedence of EU law.41 This is, of course, the principle of the primacy of EU law,42 which implies the power and the duty to disapply the relevant national provisions.43 Such a judgment has effect only for the specific case at hand; it will not have erga omnes effect. For other purposes, therefore, the national provision is not repealed but remains valid and effective, although it could potentially be disapplied in subsequent cases.

3.3.3 Antitrust law and regulation As already mentioned, although they do not differ in their aims and scope of action, in practice, some situations of conflict may arise between antitrust and regulatory provisions. In this regard it is relevant to note that recent provisions (Article 35(1) Ibid. For a concise and rich survey of the relations among national, EU and ECHR rules, see Rosanna De Nictolis, ‘An overview of administrative justice in Italy’, presentation made at the Meeting with the President of the European Court of Human Rights, Rome, 2 May 2012. 41 Case 6/64 Flaminio Costa v ENEL [1964] ECR 585 (Eng Spec edn). 42 Case 106/77 Amministrazione delle Finanze dello Stato v Simmenthal SpA [1978] ECR 629. 43 Ibid, paras 21 and 24 (‘[E]every national court must, in a case within its jurisdiction, apply Community law in its entirety and protect rights which the latter confers on individuals and must accordingly set aside any provision of national law which may conflict with it, whether prior or subsequent to the Community rule. […] [A] national court which is called upon, within the limits of its jurisdiction, to apply provisions of Community law is under a duty to give full effect to those provisions, if necessary refusing of its own motion to apply any conflicting provision of national legislation, even if adopted subsequently, and it is not necessary for the court to request or await the prior setting aside of such provision by legislative or other constitutional means.’). 39 40

412  Competition, Regulation and Public Policies of Decree Law No 201 of 6 December 2011) have attributed to the Italian Antitrust Authority a special capacity in judicial proceedings to lodge an appeal against general, regulatory or particular administrative acts of other public administrations if such measures violate rules aimed at protecting competition and the market. 1.  Concurrent rules. Certain pronouncements of the Council of State in Plenary Assembly (decisions No 11–13 and 15–15 of 11 May 2012) stress the importance, in the field of electronic communications, of special consumer protection provisions laid down in the sector-specific legislation, and their priority relative to the general provisions laid down in the Code of Consumer. The Council of State noted that the sectoral rules aiming at consumer protection are complete in their content and establish the competence of the Communications Authority (Agcom) to sanction violations. For these reasons, the Court concludes that the system of electronic communications has no lack of protection for the consumer, and there is no need for supplementary interventions by the Antitrust Authority. Moreover, Directive No 2005/29/CE (transposed into the Italian Consumer Code) establishes that special consumer protection provisions laid down in other regulations prevail over its own provisions. The Council of State explained that this principle of ‘specialty’ was applicable not only in cases of actual conflict between the two systems but also where the special provisions add further elements beyond those contained in the general provisions. It follows that the ex ante, sector-specific regulation did not allow ex post interventions by the Antitrust Authority. The exclusive action of Agcom would ensure the good functioning of the Administration. We can conclude that, in this case, the antinomy and the possibility of contrast between regulations were overcome on the basis of the principle of ‘specialty’: priority was given to the corpus juris of the sector-specific regulation and of the concurrent principle of the ‘good functioning of the Administration’, a principle enshrined in Article 97 of the Constitution. Consumer protection has thus developed, after having long been considered an additional and derivative effect of the protection of competition, into an autonomous object of attention for the legislator and, consequently, for the national courts. 2.  Conflicting rules. On the other hand, electronic communications also offer a pertinent example of a matter subject to a regulatory power and characterized by a high degree of technicality. In this sector, a court may thus be confronted with the need to reconcile different and conflicting rules, not only when an antitrust decision is contested but also when a regulatory act is challenged. While performing this task, the judge has a peculiar ‘tool’ at his disposal, ie, his discretionary power to shape the economic concepts implied in legal texts. The judge can thus reassess the technical choice of the authority and can apply the correct interpretation of the relevant indeterminate juridical concepts to the factual controversy.

Cons. Rosa Perna 

413  

Thus, with regard to price controls, we can recall case n. 1239/09, H3G v Agcom.44 In this case the plaintiff contested decision n. 667/08/Cons of 26 November 2008, concerning wholesale termination services for fixed-line and mobile phone voice telephony, owing to the fact that it had imposed a scheduled reduction of tariffs for the operator in question as from 1 July 2010 in order to reach total symmetry among operators within a reasonable time. The first instance court rejected the claim, upholding the legitimacy of the contested decision. According to the court, Agcom had correctly used the costs borne by an efficient operator as a decisive benchmark.45 In cases such as this, the main issue for the court is what exactly a cost-oriented price is; a second question, however, is whether the model of cost adopted by the Authority can produce an anticompetitive effect on the relevant market. With regard to the first issue, the BU-LRIC (bottom up long-run incremental cost) model, which has been used by the Italian regulatory authorities, is based on hypothetical efficient operator’s costs and – as stated in the relevant Commission Recommendation – by means of a so-called symmetric price control, it leads to the same price for each operator. This means that, at the beginning of the regulatory cycle, the biggest, ex-incumbent operators have the lowest termination rates while the smallest operators, who are also the last entrants on the market, have the highest termination rates. These rates are supposed to steadily converge until they become symmetric after a certain programmed number of years. Coming to the second issue, from a general economic perspective46 the costoriented price, in a rough simplification, is the equilibrium price on the competitive market at which the price is equal to the average cost and marginal cost. However, not all regulatory-economics argues in favour of cost-oriented price regulation. Moreover, careful mathematical methods have even deduced that cost-oriented price regulation does not ensure long-term competitiveness, whereas incentive regulation seems necessary.47 Against that background it is easy to conclude that, in a sector like telecommunications, where the fixed (constant) costs are high and the marginal costs tend toward zero, and the competitive market’s conditions, ie a negligible market share, is not met, the unit cost – which is the per-minute cost of the mobile termination rates – necessarily depends on call volume and traffic. In such a situation, each operator’s cost-oriented price is different, and it decreases with the increase of the market share. Consequently, when evaluating the consistency of a regulatory act establishing symmetric tariffs with the principles of the competition law, a judge might hold 44 For a broad review of the Italian case law in the field of electronic communications, see Rosa Perna, ‘Interim measures and retroactive effect of remedies’, presented at a seminar entitled ‘Implementing the revised regulatory framework in electronic communications’, Brussels, 28 November 2011. 45 TAR Lazio, III ter, 11 February 2011, No 1336, upheld on appeal: Cons. Stato, 23 May 2011, No 3106. 46 For a deeper analysis, see András Kovács, ‘The assessment of the merits, the appropriate expertise and the deference to NRAs’ , presented at the seminar ‘Implementing the revised regulatory framework in electronic communications’, Brussels, 28 November 2011. 47 Ibid.

414  Competition, Regulation and Public Policies that converging to the symmetric prices would be a consistent and logical process only if it could be assumed, in an ex ante examination, that the development of the operator’s market shares might trend towards equalization, so that by the time the termination rates become symmetrical the operator’s market share and turnover will be the same as those of its competitors. This was indeed the focal point of the argument put forward by H3G, the smallest Italian operator, against the symmetric charges.

4. Final remarks It is time we drew our conclusions. First of all, the introduction into national law of EU antitrust principles puts the pre-existing principle of the freedom of economic initiative in a somewhat new light. Although it not designed to have any connection with competition law, Article 41 of the Italian Constitution has nevertheless been able to develop as part of a more modern conception of the ‘economic constitution’.48 Although the market was initially regarded by the Constituent Assembly in negative terms,49 thanks to the introduction of the EU competition law principles, the market has over time assumed its place within the formal Constitution and today has a positive meaning since competition promotes economic and organizational efficiency. The constitutional reform of 2001, whereby the amended Constitution granted wide legislative competence to the Regions, has on the one hand shown the dangers of a geographically fragmented economic regulation. On the other hand, the reform has highlighted the need to reshape all state, regional and local regulation in a pro-competitive direction. With regard to this latter purpose, even a possible cooperation between the Constitutional Court and the Antitrust Authority has been envisaged.50 In particularly, it has been suggested that the Authority should be enabled to raise, directly before the Court, questions as to the lawfulness of legislation that violates constitutional principles of competition. It thus seems clear that the process of interpretation and application of laws and regulations circling round the principle of competition has taken on great importance. Moreover, it must be considered that the principle of competition is a rule characterized by a higher degree of abstraction and flexibility than a legal provision. This implies that, in practical cases, the principle is cannot be derogated See De Benedetto, ‘Il principio della concorrenza nell’ordinamento italiano’, cited above note 2. See Amato, ‘Il mercato nella Costituzione’, cited above note 8 (referenced by De Benedetto, ibid). 50 Marcello Clarich, ‘Servizi pubblici e diritto europeo della concorrenza: l’esperienza italiana e tedesca a confronto’, Rivista trimestrale di diritto pubblico, 2003, pp 113 et seq., as reported by De Benedetto, cited above note 2. 48 49

Cons. Rosa Perna 

415  

from; it can only be applied or, when confronted by other overriding principles and values, discarded. Considering the abstract possibility of the need for a judicial balance among principles of the same constitutional rank, we can conclude that the effective affirmation of the principle of competition will thus depend also on the importance that the courts are inclined to recognize to other values in conflict or in contrast with that principle.

Douglas H Ginsburg and Daniel E Haar*

Resolving Conflicts between Competition and Other Values: The Roles of Courts and Other Institutions in the US and the EU

Introduction The laws of competition often lead to results that conflict – or at least appear to conflict – with the requirements of other laws, regulations, or policies in the same jurisdiction. For example, a law may specifically permit behavior proscribed by the antitrust laws, such as allowing competitors to set rates collectively, as did a 1948 law for the benefit of the railroad industry in the United States.1 Even if another law does not directly contradict the laws of competition in that way, the blunt instrument of competition enforcement may interfere with a more finelyhoned regulatory scheme tailored to, for example, complicated securities practices.2 ‘Conflict’ therefore can be an expansive concept; so conceived, it comes into play whenever a court’s imposition of liability for a violation of competition law would directly frustrate some other law or public policy. Indeed, even though antitrust law and a particular regulatory regime may in general yield similar results because both were designed to further a similar end, the overlap of multiple legal regimes with differing standards for liability may create costly legal uncertainty in borderline cases such as Credit Suisse. In recent years courts in the United States have become more attuned to this second type of conflict.

* Douglas H Ginsburg is Senior Circuit Judge, US Court of Appeals for the District of Columbia Circuit and Professor of Law, George Mason University Law School (at the time of writing, New York University Law School). Daniel E Haar is a attorney in the Antitrust Division, US Department of Justice. The views stated herein are solely the authors’ own and do not express any policies or positions of the institutions with which they are affiliated. We thank Jose Antonio Batista de Moura Ziebarth for helpful comments on an earlier draft. 1 The Reed-Bulwinkle Act of 1948, Ch 491, 62 Stat. 472 (1948) (codified as amended at 49 USC § 10706 (2006)). This legislation expressly empowered the now-abolished Interstate Commerce Commission (ICC) to grant approval to agreements establishing rate bureaus for the purpose of setting rates collectively. 2 See, eg, Credit Suisse Securities (USA) v Billing, 551 US 264 (2007).

418  Competition, Regulation and Public Policies The appropriate way in which to resolve any particular conflict depends upon the answer to two related questions. The first is: 1. Is the conflict one of means alone, or do the laws truly serve different and hence irreconcilable ends? When two conflicting laws further the same value, such as economic efficiency (whether understood as consumer welfare or as total welfare), the conflict is one of means alone. In such a case, it is possible that a single institution with sufficient access to the relevant facts and expertise in economics could determine the appropriate balance between them. When, on the other hand, the values underlying two laws diverge, there is little reason to think any institution could determine the optimal balance between them; the best that may be done is to decide which of the two must give way. If there were an agreed-upon single metric of social welfare, then the task of reconciling the laws would be easier. Most competition experts probably believe it is more desirable to maximize economic efficiency than to favor the economic interest of some subset of producers or consumers in the economy. In a democratic society, however, we must expect, though perhaps not happily, that more narrow values may and do find their way into legislation. In fact, the values a society holds are far from stable over time. For example, the economic downturns and depressions in the late 19th and early 20th Centuries led to widespread public discontent with market capitalism and its associated value of competition, and that discontent entered the legal regime of the United States through many statutes enacted in the New Deal era and beyond. Later in the 20th Century, the turn in the academy toward economic analysis of law and the deregulatory movement to which it gave rise led to a general convergence upon economic efficiency as the proper goal of economic regulatory legislation; as a result, the way courts and agencies reconciled competition laws with more recently enacted statutes had to change. As we discuss below, the locus of much decision-making in the United States has shifted over this period from the legislature to the courts; at the same time, the European Commission’s Directorate-General for Competition (DG Comp) and the European Courts developed their own distinct tools for reconciling competition laws with anticompetitive regulatory schemes. There can be no guarantee on either side that the current emphasis upon economic efficiency will last: with the recent financial crisis, and the popular skepticism about free markets it has engendered, we may be entering another period of disagreement about the fundamental values that should underlie legislation.3 If that happens, the legislature may become more active, again resolving conflicts in favor of values other than competition. 3 See James Cooper and William Kovacic, ‘US Convergence with International Competition Norms: Antitrust Law and Public Restraints on Competition’, 90 Boston University Law Review 1555, 1564–65 (2010) (noting that the ‘aftermath of the financial crisis has not only directly affected government involvement in the economy, but in general has fed impulses to rely less on competition, and more on government dictate, to organize the economy’).

Douglas H Ginsburg and Daniel E Haar  419   The second question pertinent to our inquiry is the following: 2. Which institutions are best placed to resolve a conflict between competition and other laws, regulations, or priorities? We consider three possible answers to this question: (1) the legislature; (2) a competition agency;4 and (3) a court. The institution best situated to solve a particular conflict depends upon the nature of that conflict.5 A competition enforcement agency will likely have the experience and the technical knowhow as well as the appropriate set of institutional norms with which to reconcile competition law with another legal regime that is likewise intended to serve economic efficiency, such as, perhaps, a patent or a securities law. A competition agency would be ill-suited, however, to reconcile competition law with a law meant to serve a wholly different goal, such as the protection of labor unions. A union, which may unite employees of competing firms in order to raise producers’ cost of labor, is structurally equivalent to a cartel fixing the price of an input. But a competition agency could hardly be expected to resolve a resulting conflict in a principled or unbiased fashion. Moreover, furthering two irreconcilable values, like serving two masters, could only bring confusion to the mission of the agency and lower its morale. 4 A competition agency is typically a unit of the executive branch of government, as are the Antitrust Division of the US Department of Justice (DoJ) and the Directorate-General for Competition in the European Commission. We do not consider the resolution of purely intra-executive branch conflicts between competition and other values because such conflicts are straightforwardly resolved within the hierarchy of the executive branch. For example, if the Antitrust Division were to raise concerns about the antitrust implications of a merger between two defense contractors, which merger the Department of Defense favors, any disagreement not resolved by the heads of the DoJ and the DoD would be resolved by the President or his delegate. Cf Competitive Impact Statement, United States v Raytheon Co, No. 1:97-CV-02397 (DDC Oct 22, 1997) (noting cooperation between DoJ and DoD in approving purchaser of divested assets or a trustee of assets). For an extended discussion of the interplay between the DoD and the antitrust enforcement agencies in defense industry mergers in the US, see Mark Schwartz, ‘The Not-So-New Antitrust Environment for Consolidation in the Defense Industry: The Martin Marietta-Lockheed Merger’, 1996 Columbia Business Law Review 329 (1996). For the remainder of this chapter we focus our analysis upon the competition agency as the most relevant actor within the executive branch, although we recognize other executive branch departments and officials may also play important roles in resolving conflicts between antitrust and other laws or policies. Some competition agencies, such as the Federal Trade Commission (FTC) in the US, are not subject to executive branch oversight. When the source of a conflict between the antitrust laws and another law or policy is a difference of opinion between an executive agency and an independent agency, the conflict cannot be resolved within the executive branch of government and therefore may be resolved by a court, see, eg, FTC v Alliant Techsystems Inc., 808 F. Supp. 9, 12, 24 (DDC 1992) (preliminarily enjoining merger between two suppliers of ammunition where testimony of Army witnesses ‘indicated a professional and individual preference for the merger’ and the Army itself had ‘no objection to the merger’), or by the legislature, see, eg, 49 USC § 10706(a)(2)(A) (exempting from the antitrust laws any agreement between rail carriers if approved by the Surface Transportation Board). 5 Although the primary focus of this chapter is on the role of courts in resolving such a conflict, we consider other institutions that may be involved in this process in order to illuminate more fully the role of courts and to evaluate whether a court is always the best-placed institution to resolve the conflict.

420  Competition, Regulation and Public Policies The balancing of irreconcilable goals is best left to the legislature. This is not to say the legislature always, or even usually, does the ‘right’ thing – however measured. Indeed, it is a fact of life that legislatures pass laws with goals other than, and frequently inimical to, economic efficiency. During the New Deal era, for example, the United States Congress passed many laws that broadly6 or narrowly7 but quite intentionally facilitated the reduction of output and, consequently, increased the prices of goods, which slowed the nation’s recovery from the depths of the Great Depression.8 If there is a large enough constituency supporting such a law, not only is the legislature the more legitimate institution to strike that political balance; a competition agency would be unsuited to the task, which is not matched to its strengths and its culture. A court, similarly, lacks the tools, as well as the democratic legitimacy, with which to balance fundamentally opposing values. Therefore, if the legislature is to pass a law that reduces economic efficiency, then it would be best for it to state that goal expressly, or at least to say expressly that the competition laws do not apply where the new law does apply lest the courts try to balance irreconcilable values and end up attributing their own preferences to the unexpressed will of the legislature. Where, then, do courts enter the picture? Courts occupy a unique position: according to the dominant jurisprudential philosophies in America and, we believe, in Europe, a court must give effect to the general will of the legislature as applied to the specific case before it. The court is therefore never the first mover: since its role is not to initiate law-making but rather to interpret the law, even if that requires it to fill a gap, the court must always look to the prior decision (explicit or implicit) of another institution for guidance about how to reconcile antitrust with an overlapping regulatory scheme. The court’s job is straightforward when the legislature has said the competition laws will not apply to a particular activity or a particular class of organizations, but the legislature does not always do so; nor can a legislature be expected to foresee, and thus to deal expressly with, every conflict between a new law and the existing corpus juris. 6 See, eg, Fair Labor Standards Act of 1938, Pub. L. 75–718, ch. 676, 52 Stat. 1060 (codified as amended at 29 USC §§ 210, et seq.) (establishing a federal minimum wage, requirement of overtime pay, etc.); National Industrial Recovery Act [NIRA] of 1933, ch. 90, 48 Stat. 195 (previously codified at 15 USC §§ 701–712), invalidated as unconstitutional, ALA Schechter Poultry Corp v United States, 295 US 495 (1935), Title I of which empowered trade associations to pass codes of ‘fair competition’ and exempted such agreements from the antitrust laws. 7 See, eg, Motor Carriers Act of 1935, 49 Stat. 543, ch. 498 (codified at 49 USC 31502) (limiting entry into trucking industry); Agricultural Adjustment Act of 1933, Pub. L. 73–10, 48 Stat. 31 (subsidizing farmers for removing land from production), invalidated as unconstitutional, United States v Butler, 297 US 1 (1936). 8 See Letter from John Maynard Keynes to President Franklin D. Roosevelt (Dec. 30, 1933), in ‘From Keynes to Roosevelt: Our Recovery Plan Assayed’, N.Y. Times, Dec. 31, 1933, at XX2, http:// newdeal.feri.org/misc/keynes2.htm (‘N.I.R.A. ... probably impedes Recovery’; ‘stimulation of output by increasing aggregate purchasing power is the right way to get prices up; and not the other way round’); Milton Friedman and Anna Schwartz, A Monetary History of the United States, 1867–1960, Princeton University Press, 1963, at 494–495 (NIRA raised prices of goods and lowered profits, leading to a decrease in private investment and a slowing of the economic recovery).

Douglas H Ginsburg and Daniel E Haar  421   When two laws would yield conflicting results but neither expressly repeals or overrides the other, a court must use some method of statutory interpretation in order to discern what the will of the legislature requires in the circumstances before it. A court may also determine that the application of competition law to a particular practice would interfere with the work of a regulatory agency even if simultaneous compliance with both competition law and sectoral regulation would not necessarily entail contradictory obligations; the court will then interpret the legislature as having intended to give the regulators space in which to operate unobstructed by antitrust enforcement. The more courts are willing to treat such a tension as a conflict, and therefore to find an undertaking or a practice exempt from competition law,9 the more active the role courts play in defining the limits of competition law and the proper balance between it and other bodies of law and regulation. In addition to applying the express or implied will of the legislature, a court must apply its fundamental law – whether it be a national constitution or, in the case of the EU, the Treaty on the Functioning of the European Union (TFEU) and other ‘primary law’. When the application of competition law would conflict with a fundamental law, of course, the latter law must prevail. As illustrated by the Supreme Court’s decision in Superior Court Trial Lawyers Ass’n v FTC,10 however, determining the proper reach of the US Federal Constitution, and therefore whether it precludes application of the antitrust laws in a particular situation, is no easy task. In that case, the court of appeals had determined that the First Amendment guarantee of free speech protected from antitrust liability a labor strike by government-funded lawyers who represented indigent defendants in criminal cases; although the strike was intended most immediately to increase the lawyers’ wages, the court held that because the striking lawyers intended thereby to improve the quality of representation provided to the indigent defendants, they were exempt from the antitrust laws if they did not have market power, which question had not been resolved by the FTC.11 The Supreme Court reversed that judgment and concluded the ‘expressive component’ of a strike or an economic boycott is not substantial enough to warrant special consideration under the antitrust laws.12 In Europe, competition law itself is part of the fundamental law (ie, the TFEU) and hence more likely to prevail in the event of a conflict with another law. It is therefore in general unnecessary for a European court to ask, as had the court of appeals in Superior Court Trial Lawyers, whether the fundamental law precludes the application of competition law. The fundamental competition law necessarily trumps an ordinary law or a regulatory regime, as shown in the case of Deutsche Telekom AG v Commission; the European Court of Justice held Deutsche Telekom liable for the exclusionary effect of its wholesale prices even though those prices See below note 26 and accompanying text. 856 F.2d 226 (D.C. Cir. 1988), rev’d, 493 US 411 (1990). 11 Ibid. at 250. 12 493 US at 431–432. 9

10

422  Competition, Regulation and Public Policies had been specifically approved by the German telecommunications regulatory commission.13

1. Reconciling Conflicts in the United States The methods by which the United States and Europe reconcile a conflict between competition and other laws differ more as a result of history than any difference of philosophy. Because in the United States the principal antitrust law – the Sherman Antitrust Act – was passed in 1890, most potentially conflicting regulatory legislation came in a later era. Applying the principle that a later legislative enactment prevails over an earlier one, a US court faced with a potential conflict will generally ask whether the later statute expressly or implicitly repealed the earlier statute.14 If the later statute does not repeal the earlier one, then the courts must turn to other methods by which to resolve the conflict. Courts in the US reconcile a potentially conflicting application of antitrust and other laws in one of four ways, depending upon the precise wording of the other law. They may (1) interpret an express statutory exemption to preclude application of the antitrust laws; (2) interpret a statute to require a case-by-case balancing of competition and other values; (3) interpret a conflict between antitrust and a later law to preclude application of the antitrust law; or (4) interpret the vaguely worded antitrust laws narrowly so as to avoid a conflict with a later-enacted law.

1.1 Express Legislative Repeal of the Antitrust Laws Many laws passed in the early to mid-20th Century expressly provided for repeal of the antitrust laws. For example, § 6 of the Clayton Act of 1914 exempts labor unions and agricultural cooperatives by declaring: ‘The labor of a human being is not a commodity or article of commerce. Nothing contained in the antitrust laws shall be construed to forbid the existence and operation of labor, agricultural, or horticultural organizations, instituted for the purposes of mutual help.’15 See below note 44 and accompanying text. See Posadas v National City Bank of New York, 296 US 497, 503 (1936) (‘Where provisions in the two acts are in irreconcilable conflict, the later act to the extent of the conflict constitutes an implied repeal of the earlier one’). Some cases describe the effect of repeal, whether express or implied, as conferring an ‘exemption’ or an ‘immunity.’ See, eg, Connell Constr. Co. v Plumbers, 421 US 616, 621–622 (1975) (‘basic sources of organized labor’s exemption from federal antitrust laws are §§ 6 and 20 of the Clayton Act’); and cf. Professional Real Estate Investors, Inc. v Columbia Pictures Industries, Inc., 508 US 49, 56 (1993) (‘Those who petition government for redress are generally immune from antitrust liability.’). 15 15 USC § 17. Similarly, the Webb–Pomerene Act of 1918 states ‘[n]othing contained in the Sherman Act shall be construed as declaring to be illegal an association entered into for the sole purpose of engaging in export trade and actually engaged solely in such export trade [etc.]’ so long as the agreement does not harm domestic exporters or harm competition within the United States. 15 USC § 62. Title I, § 5 of the NIRA (1933) provided ‘any code [of fair competition] ... [or] any action complying with the provisions 13 14

Douglas H Ginsburg and Daniel E Haar  423   Unsurprisingly, such enactments coincided with the widespread rise of values inimical to the notion of free competition that undergirds market capitalism. Both because such exemptions are less common in more recent legislation and because a court’s job in applying such an exemption is fairly straightforward, we do not dwell further upon this situation.

1.2 Judicial Balancing of Competition and Other Values Some laws have required courts themselves to balance the value of competition against a broader conception of the public interest, even though courts are not wellequipped to perform that task. Under the Bank Merger Act of 1966, for example, after a court finds the effect of a proposed merger ‘may be substantially to lessen competition,’ the banks may defend their merger on the ground that ‘the anticompetitive effects of the proposed transaction are clearly outweighed in the public interest by the probable effect of the transaction in meeting the convenience and needs of the community to be served.’16 Similarly, when the Federal Trade Commission, pursuant to § 13(b) of the FTC Act,17 asks a court temporarily to enjoin a merger in order to preserve the status quo pending a trial on the merits, the court must determine whether granting the injunction will serve the public interest – a capacious concept that may incorporate values other than competition, such as full employment18 or national security19 – in addition to whether the government is likely to succeed at trial. As noted above, courts are not well-equipped to perform such a balance. Due perhaps to the rise of economic analysis of law in the 1970s and the associated movement toward deregulation, however, the Congress since then has been more likely to thereof ... shall be exempt front the provisions of the antitrust laws.’ Likewise, the Reed-Bulwinkle Act of 1948 exempted ‘[p]arties to any agreement approved by the [Interstate Commerce] Commission ... from the operation of the antitrust laws.’ 62 Stat. 473 (codified as amended at 49 USC § 10706(b)(2)). 16 12 USC § 1828(c)(5)(B), (c)(7)(B). See also United States v First City Nat. Bank of Houston, 386 US 361 (1967) (holding trial court must decide de novo, even after regulatory approval, whether an anticompetitive merger serves the public interest). The Bank Holding Company Act, 12 USC § 1843(j) (2)(A), contains a similar provision: ‘[Federal Reserve] Board shall consider whether performance of the [acquired nonbanking] activity by a bank holding company or a subsidiary of such company can reasonably be expected to produce benefits to the public, such as greater convenience, increased competition, or gains in efficiency, that outweigh possible adverse effects, such as undue concentration of resources, decreased or unfair competition, conflicts of interests, unsound banking practices, or risk to the stability of the United States banking or financial system.’ 17 15 USC 53(b) (granting district courts jurisdiction to issue preliminary injunction ‘[u]pon a proper showing that, weighing the equities and considering the Commission’s likelihood of ultimate success, such action would be in the public interest’). Section 4 of the Sherman Act (15 USC § 4) and § 15 of the Clayton Act (15 USC § 25) grant district courts similar authority in a suit brought by the US Department of Justice. 18 See FTC v Weyerhauser, 665 F.2d 1072, 1083 (D.C. Cir. 1981) (‘improvement in the area employment rate’ weighed against enjoining likely anticompetitive merger even though it ‘would not qualify as [a] defense[] to the charged violation of Section 7 of the Clayton Act’). 19 See FTC v PPG Industries, Inc., 798 F.2d 1500, 1508 (D.C. Cir. 1986) (‘avoiding foreign ownership of military suppliers’ might weigh against enjoining likely anticompetitive merger).

424  Competition, Regulation and Public Policies pass economic legislation consistent with principles of efficiency and less likely to provide for express exemption from, or balancing against, the antitrust laws.

1.3 Implied Repeal of Antitrust Law Without express guidance from the legislature, federal courts have developed a doctrine of implied repeal to deal with situations where application of the competition laws would conflict, or at least be in tension, with a more recently enacted law. This doctrine, especially as the Supreme Court has extended it in the last few years, is itself in some tension with the Supreme Court’s oft-repeated teaching generally that ‘repeals by implication are not favored,’20 and specifically that ‘[r]epeal [of the antitrust laws] is to be regarded as implied only if necessary to make [a law such as] the Securities Exchange Act work, and even then only to the minimum extent necessary.’21 Pursuant to that ‘cardinal principle,’22 US courts historically would find a conflict between two laws only when application of one would lead to an outcome incompatible with what application of the other would have required. For example, in Silver v New York Stock Exchange, the Supreme Court asked whether the Securities Exchange Act of 1934 had implicitly repealed the antitrust laws insofar as § 1 of the Sherman Act would have prohibited (as a group boycott) the members of a securities exchange from adopting a rule barring direct telephone lines to non-member brokers.23 The Court held that the antitrust laws did apply in that situation because the exchange had not given the plaintiff non-member notice and an opportunity to be heard, but that did not finally resolve the substantive relation of the antitrust laws to the Securities Exchange Act.24 In another case, the Court applied the same doctrine and held the Securities Exchange Act did, however, repeal the antitrust laws to the limited extent that SEC regulations implementing that Act prior to 1975 allowed a stock exchange to fix the commissions charged by member brokerage firms, a horizontal agreement that otherwise would have been proscribed by § 1 of the Sherman Act.25 In more recent cases, courts have lowered the bar somewhat for concluding the antitrust laws have been implicitly repealed, doing so where application of the antitrust laws would simply interfere in some way with the functioning of a regulatory scheme. Accordingly, the Supreme Court has distinguished Silver, holding in Credit Suisse that the Securities Exchange Act repealed the antitrust laws insofar as they would have applied to the way syndicates of securities underwriters in initial public offerings agreed upon commission rates and upon the joint marketing of securities. The Court based this determination upon Silver v New York Stock Exchange, 373 US 341, 357 (1963). Ibid. 22 Ibid. 23 Ibid. 24 Ibid. at 369-370. 25 Gordon v New York Stock Exchange, Inc., 422 US 659, 685–686 (1975). 20 21

Douglas H Ginsburg and Daniel E Haar  425   (1) the existence of regulatory authority under the securities law to supervise the activities in question; (2) evidence that the responsible regulatory entities exercise that authority; ... (3) a resulting risk that the securities and antitrust laws, if both applicable, would produce conflicting guidance, requirements, duties, privileges, or standards of conduct ... [and because] (4) ... the possible conflict affected practices that lie squarely within an area of financial market activity that the securities law seeks to regulate.26

Even though prior cases such as Silver had suggested the Court would find an implied repeal only in situations of clear ‘repugnancy’ or ‘incompatib[ility]’ between antitrust and another body of law,27 the Court dramatically extended the notion of incompatibility in Credit Suisse to reach a situation in which two laws imposed liability for similar conduct and therefore application of the antitrust law – in the Court’s view – provided little incremental benefit.28 This option of holding that a later law implicitly repeals the antitrust laws in a specific context gives US courts significant power to define, and thus to shrink, the scope of application for the antitrust laws, as shown in Figure 1 below. Figure 1. Credit Suisse: Securities Laws and Regulations Repeal the Antitrust Laws

Antitrust Laws

Boundary of antitrust as decided in Credit Suisse

Securities laws and Regs

Conduct as issues: application of antitrust laws repealed by securities laws/regs

As compared with express repeal by the Congress, which often shields blatantly anticompetitive behavior from application of antitrust laws, the implied repeal doctrine, as developed by the courts, limits the scope of the antitrust laws only to the degree necessary to avoid interfering with a regulation that similarly targets anticompetitive behavior in a more specific manner and is tailored to particular Credit Suisse, cited above note 2, at 275. Ibid. 28 Ibid. at 284 (‘In sum, an antitrust action in this context is accompanied by a substantial risk of injury to the securities markets and by a diminished need for antitrust enforcement to address anticompetitive conduct.’). 26 27

426  Competition, Regulation and Public Policies business practices.29 Express statutory repeals are typically applied to a broad class of actors or activities, such as labor unions, whereas implied repeals are typically limited by the court to a narrow class of commercial behavior, such as, in Credit Suisse, the setting of commission rates by a syndicate of underwriters that markets securities in an initial public offering.30

1.4 Narrow Interpretation of the Antitrust Laws Fourth, and finally, a court may determine that the vaguely worded antitrust laws were not intended to apply to certain activities regulated by other institutions. For example, the Supreme Court held over a century ago that the framers of the Sherman Act ‘never contemplated’ that it would apply to ‘reasonable and legal conditions imposed upon the assignee or licensee of a patent by the owner thereof, restricting the terms upon which the article may be used and the price to be demanded therefor.’31 In United States v Line Material Co., the Supreme Court likewise stated ‘the possession of a valid patent ... does not give the patentee any exemption from the provisions of the Sherman Act beyond the limits of the patent monopoly,’32 which is a roundabout way of saying the patent laws do displace the antitrust laws to the extent of the rights granted by a patent.33 29 In Credit Suisse, for example, the Supreme Court emphasized, in holding that the actions of the underwriters were impliedly exempt from the antitrust laws, that the SEC had already forbidden many of the challenged practices at issue, albeit in a more detailed and nuanced fashion than the antitrust laws would have done. Ibid. at 279. 30 See ibid. at 268. 31 E. Bement & Sons v National Harrow Co., 186 US 70, 92 (1902). 32 333 US 287, 308 (1948). 33 Where ‘the limits of the patent monopoly’ lie, however, is a vexed question. For years, the federal courts of appeals had tightly restricted the application of the antitrust laws as applied to so-called ‘reverse payments,’ ie, payments made pursuant to an agreement to settle patent litigation in which the party claiming infringement (typically a branded pharmaceutical manufacturer) agrees to pay the allegedly infringing party (typically the manufacturer of a generic version of the branded drug). As the Court of Appeals for the Eleventh Circuit reiterated in 2012, ‘absent sham litigation or fraud in obtaining the patent, a reverse payment settlement is immune from antitrust attack so long as its anticompetitive effects fall within the scope of the exclusionary potential of the patent.’ Federal Trade Commission v Watson Pharmaceuticals, Inc., 677 F.3d 1298 (11th Cir. 2012), rev’d Sub. nom FTC v Activis, 133 S Ct 2223 (2013); see also In re Tamoxifen Citrate Antitrust Litig., 466 F.3d 187, 213 (2d Cir. 2006) (similar); In re Ciprofloxacin Hydrochloride Antitrust Litig., 544 F.3d 1323 (Fed Cir. 2008). However, the Third Circuit disturbed this relatively pro-patent consensus of the courts of appeals when it allowed plaintiffs to proceed to trial on a claim under § 1 of the Sherman Act against the parties to a reverse payment agreement. In re K-Dur Antitrust Litigation, 686 F.3d 197 (3d Cir. 2012). The D.C. and Sixth Circuits had previously allowed antitrust suits to proceed against parties to reverse payment agreements. See Andrx Pharms., Inc. v Biovail Corp. Int’l, 256 F.3d 799 (D.C. Cir. 2001); In re Cardizem CD Antitrust Litig., 332 F.3d 896 (6th Cir. 2003), but in those cases the branded drug manufacturer had paid the manufacturer of the generic simply to delay the start of the generic’s 180-day exclusivity period under the HatchWaxman Act without resolving the patent and antitrust litigation between the two firms. Accordingly, there was no clear circuit split before the decision of the Third Circuit in K-Dur. The Third Circuit pointed to the ‘almost unrebuttable presumption of patent validity’ that flowed from the other circuits’ ‘scope of the patent’ standard, which presumption it deemed unjustified in light of various empirical analyses showing generic drug manufacturers often prevailed against branded drug manufacturers in suits for infringement. K-Dur, 686 F.3d at 214. Following the completion of this chapter, the Supreme Court

Douglas H Ginsburg and Daniel E Haar  427   In a quite prominent case involving this fourth method of reconciling conflicts, Verizon Communications Inc. v Law Offices of Curtis V Trinko, the Supreme Court decided the antitrust laws do not impose upon an incumbent telephone company a duty to lease to a competing carrier access to parts of its network where the Federal Communications Commission has already imposed such a requirement pursuant to the Telecommunications Act of 1996.34 Rather than holding that the telecommunications laws and regulations had pro tanto repealed the antitrust laws,35 as the securities laws and regulations were held to do in Credit Suisse, the Supreme Court in Trinko couched its decision as merely interpreting the scope of the earlier-adopted Sherman Act. The Court noted that, except in very limited circumstances, a company has no duty under the antitrust laws to deal with its competitors;36 therefore, it characterized the question before it as whether to broaden the reach of the antitrust laws, not whether to narrow that reach, which would have been the question raised by the doctrine of implied repeal. The Court explained that imposing a new duty under the antitrust laws when a telecommunications regulation already performed much the same function could not be expected to yield more social benefit than social cost.37 In that way, the Court in Trinko reached much the same end as it had in Credit Suisse – the giving way of the antitrust laws in favor of a later-enacted regulatory law – but by a different route. For a graphic representation of the Court’s analysis in Trinko, see Figure 2. Figure 2. Trinko: A Narrowing Interpretation of the Antitrust Laws

Antitrust Laws

Boundary of antitrust as decided in Trinko

Telcomm. Laws/Regs Telecomm. laws and Regs

Conduct at issue: antitrust laws do not apply

resolved this conflict among the circuit courts, overturning the Eleventh Circuit’s assertion that reverse payment settlements are immune from antitrust attack. See FTC v Actavis, Inc., 133 S Ct 2223 (2013). 540 US 398 (2004). Indeed, the Telecommunications Act of 1996 foreclosed the doctrine of implied repeal by expressly providing that ‘nothing in this Act or the amendments made by this Act shall be construed to modify, impair, or supersede the applicability of any of the antitrust laws.’ 110 Stat. 143, 47 USC § 152, note. 36 540 US at 408. 37 Ibid. at 412 (‘Against the slight benefits of antitrust intervention here, we must weight a realistic assessment of its costs.’). 34 35

428  Competition, Regulation and Public Policies Trinko is like Credit Suisse in another way: it significantly expanded one of the four judicial methods of resolving a conflict with the antitrust laws. Prior to Trinko, the Supreme Court had used the fourth method in a rather modest fashion. As we have seen, for example, the Court held the Sherman Act of 1890 was not intended to qualify the previously enacted patent laws. In Trinko, by contrast, the Supreme Court held that the Sherman Act was not intended to reach practices that were first regulated by the Telecommunications Act of 1996, passed more than a century later. State action immunity, which first arose in the case of Parker v Brown, 317 US 341 (1943), is another doctrine by which courts narrow the reach of the antitrust laws. If a ‘clearly articulated’ state statute permits conduct by a private actor that would otherwise violate the antitrust laws, and that conduct is ‘actively supervised’ by the state, then it does not come within the condemnation of the antitrust laws. For example, in California Retail Liquor Dealers,38 the Supreme Court held wholesalers who set resale prices for wine pursuant to a state authorizing statute were not immune from liability under the Sherman Act (as it was then interpreted to prohibit all resale price maintenance (RPM)39) because the state did not actively supervise those prices. Although states have no authority to override a federal law, which the Constitution declares the supreme law of the land, the Supreme Court reached much the same result through this ‘state action’ doctrine by interpreting narrowly the intended scope of the federal antitrust laws so as not to reach conduct authorized by state law and supervised by the state. That is, they concluded the Congresses that passed the Sherman and Clayton Acts in 1890 and 1914 did not intend that the antitrust laws be applied to the states acting in their capacity as sovereigns or, by extension, to private parties whose anticompetitive behavior is explicitly permitted and actively overseen by a state.40 Two commentators have argued the logic of state-action immunity should extend beyond instances of active supervision by the states and, accordingly, that federal courts should not construe the antitrust laws to reach conduct that can be adequately redressed under state law. They argue the antitrust laws should Cal. Retail Liquor Dealers Ass’n v Midcal Aluminum, Inc., 445 US 97 (1980). See Dr. Miles Med. Co. v John D. Park & Sons Co., 220 US 373 (1911) (holding minimum RPM a per se violation of § 1 of Sherman Act); Albrecht v Herald Co., 390 US 145 (1968) (holding maximum RPM per se unlawful under § 1). But see Leegin Creative Leather Products, Inc. v PSKS, Inc., 551 US 877 (2007) (abandoning the per se rule in favor of the rule of reason for claims that a particular instance of minimum RPM violates the Sherman Act); State Oil Co. v Khan, 522 US 3 (1997) (same regarding claims for maximum RPM). 40 For a critique of state action immunity as unnecessarily shielding anticompetitive conduct from federal antitrust scrutiny, see Cooper and Kovacic, cited above note 3, at 1585 (‘As currently construed by the courts, the state action ... doctrine[] sweep[s] too far, protecting anticompetitive conduct that harms consumers and [does not] advance[] ... the value[] of federalism’). But see ibid. at 1589 (concluding, due to the entrenched aversion of federal courts to intruding upon state policy-making, the best way to overcome the pernicious effects of the doctrine would be for the Congress to ‘to amend the antitrust laws to note explicitly that it intended to supplant anticompetitive state regulation’ rather than for the Supreme Court to overrule Parker v Brown). Cooper and Kovacic would ‘distinguish state programs that serve the “public interest” from those that are merely designed to enrich favored producers,’ and narrow the application of state action immunity to practices of the former type. Ibid. at 1590. 38 39

Douglas H Ginsburg and Daniel E Haar  429   not deal with cases of so-called ‘patent holdup,’ in which a patentee uses its leverage to demand a higher royalty after an industry has adopted a standard for which its patent is essential, because state tort and contract remedies, as well as federal patent law remedies, are adequate for such cases.41 At least to date, the courts have not taken up that invitation. In Rambus Inc. v FTC,42 the court did narrow the range of viable antitrust claims involving patent holdups to situations where, absent fraud against the standard setting organization, a patent would not have been made essential; the court did so, however, without reference to the adequacy of state law contract remedies to handle situations of holdup short of fraud. Indeed, we know of no case thus far in which a court has limited the scope of the federal antitrust laws because there is an adequate state (or patent) law alternative, although that is a jurisprudential possibility opened up by the Parker and the Trinko decisions. *** Each of the four approaches we have surveyed may have its virtues, but some do seem preferable to others. A legislature can save a court a lot of effort, and avoid the risk the court will err, by providing an express exemption from the antitrust laws when it passes a law inimical to economic efficiency. The legislature puts a court in a much more awkward and uncertain position when, as in the Bank Merger Act, it directs the court to balance the harm to competition against other aspects of the ‘public interest.’ Even if an adept economist could fit diverse public interest concerns within the single metric of economic efficiency, a court is not equipped to perform the detailed cost-benefit analysis such a balancing requires. Conversely, a court may be better suited to determine, by following the four-part analysis of Credit Suisse, whether a later law implicitly repealed the antitrust laws within the commercial setting it governs, than it is to discern the intended scope of the Sherman Act of 1890 when that law overlaps with a contemporary regulation, as the Court had to do in the Trinko case.

2. Reconciling Conflicts in Europe Like the United States, Europe has dealt with conflicts in a way shaped by its history, and the picture therefore looks very different. Competition law in the European Union is a product of the EU Treaties. Before the original Treaty of Rome came into effect in 1958, several European countries had nationalized major industries. Whereas competition law developed in the United States from the late 41 Bruce Kobayashi and Joshua Wright, ‘Federalism, Substantive Preemption, and Limits on Antitrust: An Application to Patent Holdup’, 5 Journal of Competition Law and Economics 469, 479 (2009). 42 522 F.3d 456 (D.C. Cir. 2008).

430  Competition, Regulation and Public Policies 19th Century, competition law in Europe developed at a time when skepticism about economic regulation, along with the paramount need to create a single market, could influence the European Court of Justice to give competition laws a broad construction. Most important, the Court of Justice and the General Court are guided by the principle of the ‘primacy’ of EU law, according to which the EU Treaties and EU legislation are essentially supreme over the laws of the Member States.43 Furthermore, the EU competition rules are Treaty law, as opposed to statutory law. Therefore, whether sectoral regulation is adopted at the Member State level (in which case it may be a means of transposing EU Directives) or whether EU sectoral rules are directly applicable without further need for national regulation, EU competition law will in principle override such regulation whenever there is a conflict. In contrast to the situation in the United States, the EU Courts have interpreted competition laws quite broadly: ‘[I]t is only if anti-competitive conduct is required of undertakings by national legislation, or if the latter creates a legal framework which itself eliminates any possibility of competitive activity on their part, that [Articles 101 and 102 TFEU] do not apply.’44 In the quoted case, the ECJ held that a price squeeze violated Article 82 EC (now Article 102 TFEU) despite the approval of the telecommunications company’s wholesale and retail rates by the German national regulatory authority. This is in marked contrast to the US Supreme Court’s decision in Pacific Bell Telephone Co. v linkLine, which held that a price squeeze did not violate § 2 of the Sherman Act when both the retail prices at issue were subject to the supervision of the Federal Communications Commission.45 The ECJ’s decision illustrates how little deference is shown by the EU Courts to a sectoral regulator’s actions if they are deemed to conflict with the competition rules of the Treaty.46 43 See Case 6/64 Flaminio Costa v E.N.E.L. [1964] ECR 585 (Eng. Spec. edn). See also Giorgio Monti, ‘Managing the Intersection of Utilities Regulation and EC Competition Law’, 4 Competition Law Review 123 (2008). 44 Case C-280/08 Deutsche Telekom AG v Commission [2010] ECR I-9555. This doctrine provides an even narrower exemption from the competition laws than does the analogous immunity for state action in the United States because a private firm is exempted in Europe only if the law of a Member State leaves it no choice but to engage in anticompetitive conduct, whereas in the United States a practice is exempt so long as it is ‘clearly articulated and affirmatively expressed as state policy’ and ‘actively supervised by the State itself,’ Cal. Retail Liquor Dealers Ass’n, cited above note 38, at 105 (internal quotation marks and citation omitted). For a generally favorable comparison of the European doctrine as contrasted with state action immunity in the United States, see Mel Marquis, ‘Abuse of Administrative Power to Restrict Competition in China: Four Reflections, Two Ideas and a Thought’, in Michael Faure and Xinzhu Zhang, eds., The Chinese Anti-Monopoly Law: New Developments and Empirical Evidence, Edward Elgar, 2013, pp 73 et seq., for example at 80–84 (paper also available on SSRN: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2090138). 45 555 US 438 (2009). 46 As with conflicts between EU competition law and national regulation, the EU Courts similarly construe competition law broadly even when that law is in tension with a community-wide sectoral regulation; they have refused to exempt the practices of regulated industries in the absence of a block exemption issued by the Commission. See Case 45/85 Verband der Sachversicherer e.V v Commission [1987] ECJ 405 (confirming that Articles 101 and 102, or at that time Articles 81 and 82 EC, applied to insurance companies).

Douglas H Ginsburg and Daniel E Haar  431   Although the EU competition rules generally prevail over non-fundamental laws in the event of a conflict, there is an equally fundamental exception: Article 101(3) of the TFEU itself provides that the prohibitions of Article 101(1) may ‘be declared inapplicable’ to an agreement with an anticompetitive effect if it (1) ‘contributes to improving the production or distribution of goods or to promoting technical or economic progress,’ (2) ‘allow[s] consumers a fair share of the resulting benefit,’ (3) does not ‘impose on the undertakings concerned restrictions which are not indispensable to the attainment of these objectives,’ and (4) does not ‘afford such undertakings the possibility of eliminating competition in respect of a substantial part of the products in question.’ A determination under Article 101(3) of whether an agreement is exempt from Article 101(1) is typically done either informally by companies themselves who ‘self-assess’ the compatibility of their agreements with Article 101 or by the European Commission in the form of a block exemption, or it is done ex post in the context of, for example, litigation before a national court.47 Formally, at least, Article 101(3) exempts an agreement from Article 101(1) only if application of the latter provision would create what we have defined above as a ‘conflict of means,’ not in the event of a ‘conflict of ends,’ ie, when a competing goal would frustrate economic efficiency. Viewed in that light, Article 101(3) looks like a codification of the rule of reason familiar to US antitrust jurisprudence.48 Nevertheless, the European Commission has in some Article 101(3) cases given weight to values other than competition and thus determined the overall social utility of an agreement by balancing harms to competition against ill-defined benefits unrelated to competition; as a result, competition laws gave way in conflicts over ends and not just means.49 These cases are problematic for reasons described earlier: neither competition agencies nor courts are well-equipped to balance harm to competition against a benefit to, for example, the environment, even if an economist might be able to express such diverse values according to the single metric of economic efficiency. In the last decade or so, however, the European Commission has seemed less inclined to use Article 101(3) to import non-competition values into its analyses.50 Instead, the Commission has generally put Article 101(3) to good use as a basis for issuing ex ante ‘block exemptions’ for certain pro-competitive business practices. See Council Regulation 1/2003, Articles 1, 5 and 6, 2003 OJ L1/1. See, eg, United States v Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001) (rule of reason requires ‘plaintiffs [to] show that [the defendant’s] conduct was, on balance, anticompetitive’) (emphasis supplied). 49 In Commission Decision of 4 July 1984 – Synthetic Fibres, 1984 OJ L207/17, the European Commission found an agreement to reduce total capacity for the manufacture of synthetic textile fibers fell under the exemption now in Article 101(3) because the agreement ‘raise[d] the profitability and restore[d] the competitiveness of each party.’ Of course, cartel agreements also increase the member firms’ profitability, and it is difficult to see why ‘overcapacity’ should present an exception to the general rule against cartel agreements. See also Commission Decision of 17 September 2001 – DSD, 2001 OJ L319/1 (balancing benefits to the environment from a reduction in packaging waste against the anticompetitive effects of an agreement among manufacturers and distributors of packaging). 50 See Commission Decision of 2 April 2003 – French Beef, 2003 OJ L209/12 (holding French beef farmers’ agreement setting a minimum price was not justified as a reaction to the ‘exceptional circumstances’ brought about by a reduction in beef consumption due to fear of mad cow disease). 47 48

432  Competition, Regulation and Public Policies In doing so, the Commission has generally followed the conclusion by DG Comp that a particular type of agreement is likely to be either benign to begin with or that, if it does restrict competition, it is nevertheless on balance pro-competitive from the perspective of Article 101(3). Some of these block exemptions are sector-specific: • Insurance companies had argued, starting at least as early as the 1980s,51 that what are now Articles 101 and 102 could not be applied to them directly, that is, without implementing legislation, because of specific circumstances in their industry. Eventually, in 2003, the Commission adopted a block exemption allowing (1) agreements among insurance companies for the purposes of sharing data on costs and risks, (2) the creation of re-insurance pools, and (3) several other competitively benign practices.52 • Transportation was exempted under Council Regulation 141/62, but that exemption was largely withdrawn in 2002. Now a handful of narrower block exemptions apply to specific practices in the transportation sector, such as agreements on various technical matters53 and certain agreements among small and medium sized firms.54 Many block exemptions, however, are not directed at specific sectors of the economy, but rather go to a particular practice or type of practice regardless of the industry in which it may occur.55 In theory, block exemptions allow the Commission substantially to reduce legal uncertainty by exempting ex ante various beneficial business practices from potential liability under the competition laws. However, the EU Courts interpret block exemptions ‘narrowly’ or ‘strictly,’56 which tends to limit the gains in legal certainty that the exemption might otherwise provide. Moreover, the party arguing for application of the exemption bears the burden of persuasion.57 In light of the EU Courts’ practice of giving a narrow construction to block exemptions and a broad construction to the competition laws even when they are in tension with another law, we see that competition law may have a substantially broader application in Europe than it does in the United States.

See Commission Decision of 30 March 1984 – Nuovo CEGAM, 1984 OJ L99/24. Commission Regulation 358/2003, 2003 OJ L53/8. 53 Regulation 1017/68, Article 3, 1968 OJ L175/1. 54 Ibid., Article 4 (permitting agreements by small and medium sized transportation firms, as defined by carrying capacity, regarding various joint activities including financing and the purchase of equipment). 55 See, eg, Commission Regulation 316/2014 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of technology transfer agreements, 2014 OJ L93/17 (replacing Commission Regulation 772/2004, 2004 OJ L123/11), which exempts certain classes of patent-related and knowhow-related agreements from strict application of the competition laws. 56 See Case C-70/93 BMW v ALD Auto-Leasing [1995] ECR I-3439. 57 Council Regulation 1/2003, cited above note 47, Article 2. 51 52

Douglas H Ginsburg and Daniel E Haar  433  

3. Comparative Evaluation As we have seen, the reach of the competition laws in the United States and in Europe is subject to rather different institutional constraints. In the United States, antitrust law is ordinary law that has no inherent primacy over other federal statutes, such as those that create and empower sectoral regulators, whereas in the EU, competition law is fundamental and hence may not be repealed via the normal legislative process. Consequently, the US Congress, because it can repeal or otherwise narrow the antitrust laws at will, plays a much larger role in defining the scope of antitrust law than does the European legislator (which today typically means the European Parliament and the Council acting as co-legislators). Similarly, due to the judicially developed state action doctrine, the legislatures of individual states in the United States have more scope to limit the application of antitrust laws compared to the legislatures of Member States in the European Union. Notwithstanding the contrary impression one might get from looking solely at Credit Suisse and Trinko, therefore, it may be in the United States rather than in Europe where the judiciary plays the more modest role in reconciling competition and other values. When a US court decides a case involving the scope of the antitrust laws, it does so as a matter of legislative interpretation; its decision can be overturned by the Congress.58 By contrast, because the competition laws in Europe are ‘constitutional’ in nature, a court decision interpreting those laws is not realistically subject to revision by the European legislator. If the EU Courts were to interpret the competition laws in such a way as to stifle competition or innovation – which an untoward decision can surely do, as we saw in the United States for many decades – the EU could be stuck with that interpretation for the indefinite future.59 The best solution in Europe, however, may be not for the EU Courts to emulate the US courts by taking upon themselves the task of reconciling competition and other values; the legitimacy of judicial policy-making is more questionable with respect to matters that are not subject to legislative correction. Rather, it may be best to look to DG Comp and the competition authorities of 58 This process works in both directions. The Congress may overturn a broad judicial interpretation of the antitrust laws, as it did in enacting the labor exemption in §§ 6 and 20 of the Clayton Act; that statute overturned decisions such as Loewe v Lawler, 208 US 274 (1908), in which the Supreme Court had held a labor union that had organized a boycott a of non-union employer’s products was subject to antitrust liability as a combination in restraint of trade. Conversely, the Congress may also overturn a judicially created exemption from the antitrust laws, as it did in the Curt Flood Act of 1998, 15 USC § 26b, which limited the antitrust exemption for major league baseball recognized by the Court in Federal Baseball Club of Baltimore, Inc. v National League of Professional Baseball Clubs, 259 US 200 (1922) and reaffirmed in Flood v Kuhn, 407 US 258 (1972). 59 See Douglas H. Ginsburg and Eric M. Fraser, ‘The Role of Economic Analysis in Competition Law’, in Ian McEwin, ed., Intellectual Property, Competition Law and Economics in Asia, Hart Publishing, 2011, chapter 3, p 35 (‘From 1890 into the 1970s competition law in the United States was not economically coherent’); Robert H. Bork and Ward S. Bowman, Jr., ‘The Crisis in Antitrust Law’, 65 Columbia Law Review 363, 364 (1965) (‘Anti-free-market forces now have the upper hand [in antitrust jurisprudence] and are steadily broadening and consolidating their victory’).

434  Competition, Regulation and Public Policies Member States to reconcile conflicts with the competition laws as they arise. By issuing block exemptions, the Commission can develop sensible interpretations of the competition laws that give clear guidance to private parties and give due consideration to other regulatory schemes that also seek to advance consumer welfare. In the United States, therefore, the courts are often the best placed institution to reconcile the conflicting demands of antitrust and other laws, whereas in Europe it is DG Comp (or a national competition authority) that is so placed. Because the United States and the European Union, accordingly, turn to rather different institutions with different virtues to reconcile conflicts (of means or of ends), there are different advantages to the methods adopted by each jurisdiction. We shall compare those institutions along three dimensions – certainty, expertise, and flexibility.

3.1 Reduction of uncertainty The European practice of using block exemptions to resolve conflicts seems at first glance to provide more certainty for private actors than does American practice, which relies upon case-by-case resolution.60 By announcing in advance that certain practices are exempt from the competition laws, the European Commission provides more guidance on the applicability of those laws. By contrast, in the United States the question of whether a practice is exempt will not be resolved until the issue is pressed in litigation, and even then the court’s decision may and probably will resolve the uncertainty across a more narrow range than would a block exemption. However, because the EU Courts construe block exemptions narrowly and the competition laws broadly, the scope for enforcement of competition law may be wider in Europe than it is in the U.S, as illustrated by the Deutsche Telekom case. Therefore, insofar as there is uncertainty inherent in Articles 101 and 102 themselves, the future judicial development of the competition laws is concomitantly uncertain and the method by which Europe resolves conflicts therefore may provide less certainty than might at first appear.

3.2 The role for expertise In Europe, DG Comp plays a leading role in drafting block exemptions for promulgation by the Commission. Courts, by contrast, have a smaller role in defining the boundaries between competition policy and other bodies of law. If 60 Of course, Article 101(3) TFEU also allows for granting exemptions ex post on a case-by-case basis, while the US antitrust enforcement agencies issue ex ante ‘guidelines’ to facilitate business planning. For present purposes, however, the relevant comparison is between the EU’s use of ex ante block exemptions and the US system of resolving actual conflicts ex post.

Douglas H Ginsburg and Daniel E Haar  435   a competition agency can have enough information and foresight to draft block exemptions that are neither too broad nor too narrow, then the European approach would seem to be more sensible than looking to the courts, as is done in the US, to decide whether a practice is implicitly exempt from the antitrust laws. Perhaps owing to the influence that the economic analysis of law has had upon the US judiciary, the Supreme Court seems confident of its ability to engage in an implied repeal analysis, as in Credit Suisse, and in the creative tailoring of the antitrust laws, as in Trinko. Such analyses require complex judgments regarding, among other things, (1) the incremental benefit of applying competition law to conduct already subject to sectoral regulation, and (2) the risks of false positives inherent in enforcing competition law. Without doubt, there are some American judges with the knowledge and training needed to perform that type of analysis, but we think it unclear whether the judiciary as a whole has the appropriate expertise to justify giving it this expansive role. For that reason, the US system runs the risk of generating lower court decisions of varying analytic quality, which may give only uncertain guidance to undertakings potentially subject to antitrust liability.61 But because any first instance decision that the antitrust laws have been repealed by a subsequent statute (or otherwise narrowed by conflicting regulation) is likely to be reviewed by a court of appeals and perhaps by the Supreme Court, the relevant question may be only whether appellate judges are equipped for this task.

3.3 Flexibility On the other hand, by relying more heavily upon the courts, the United States requires that conflicts be resolved ex post, on a case-by-case basis, which means courts base their decisions upon a factual record of real business conduct and its known effects upon competition.62 This approach minimizes the risk of the court or agency striking the wrong balance by acting before it has accumulated sufficient information to make an informed determination, and perhaps thereby reaching an inefficient decision. Of course, in Europe the Council and the European Commission can also take advantage of hindsight by modifying block exemptions in the light of their accumulated experiences, and they have done so on occasion.63

61 See Michael Baye and Joshua Wright, ‘Is Antitrust Too Complicated for Generalist Judges? The Impact of Economic Complexity and Judicial Training on Appeals’, 54 Journal of Law and Economics 1 (2011) (adducing empirical evidence that having basic economics training improves quality of a judge’s opinions in ordinary antitrust cases but not in complex antitrust cases). 62 See, eg, Microsoft, cited above note 48, at 93 (declining to apply per se rule against tying ‘because of the pervasively innovative character of platform software markets, tying in such markets may produce efficiencies that courts have not previously encountered and thus the Supreme Court had not factored into the per se rule as originally conceived’). 63 See Council Regulation 1/2003, cited above note 47, Article 36, repealing the exemption from competition law previously given to the transportation industry.

436  Competition, Regulation and Public Policies

4. Conclusion Because the United States and Europe have developed different methods of resolving conflicts between competition and other values, reflecting their different constitutional structures, it is difficult to draw firm comparative conclusions from their experiences. Each approach has its virtues as well as its flaws. By relying heavily upon the courts, the United States runs the risk that lower court judges, unskilled in economic analysis, will provide only uncertain guidance to businesses or will err when deciding that the antitrust laws should give way to another regulatory regime in a particular circumstance. Appellate review and Congressional oversight, however, may ensure that the costs of any such error are not necessarily substantial. Moreover, the United States has the added advantage of flexibility inherent in case-by-case adjudication to resolve conflicts. In Europe, by contrast, the cost of a judicial error may be much greater because the ‘constitutionalization’ of competition law in the Treaty precludes legislative oversight of the courts’ competition decisions, thereby increasing the likelihood that any judicial error will be long-lived. The Commission has mitigated that risk somewhat by using its limited legislative powers to resolve conflicts ex ante in block exemptions. That practice has the added advantage of giving greater certainty to the business community than does case-by-case judicial decision making. Finally, both the United States and the EU have managed to incorporate expertise into their systems in different ways. By delegating to DG Comp the decision of whether to exempt a practice or entity from competition law, the European Commission can rely directly upon the agency’s economics expertise and experience. The United States has given expert competition agencies little authority to resolve such conflicts, but it has nonetheless managed for several decades to make antitrust jurisprudence increasingly coherent by relying upon appellate judges to deploy at least a basic understanding of economics.

Ian S Forrester

Competition Law and Public Policy Considerations

These annual gatherings in the hills overlooking Florence have materially contributed to the development of the law, by encouraging participants to think about fundamental principles. Our topic is how to reconcile common sense, public interest considerations and the ostensibly bleak prohibitory language of the European Union’s competition rules. This chapter will review the ways in which enforcement has softened (or not) to take account of such not easily measurable considerations. It often happens that a practice which involves some restrictive feature is defended on the grounds of public interest. Closer examination may reveal that national law permits the practice; that the practice is part of the normal business model of the accused and others like it; that the sector is of national importance but that its practices are not sanctioned or compelled by national law; or that the practice is part of an extensively regulated industry. Sometimes the measure is a kind of price control, rate-setting or other guild-sponsored imposition. Sometimes the measure is justified by reference to safety, health, the environment, economic crisis or higher sporting or cultural objectives. Having examined a number of successful rationalisations in European competition law, and a greater number of rejected rationalisations, I have to record that it is not easy to find a consistent pattern. Equally it is not easy to be sure whether the outcome was really altered by the public policy, or whether the factor was merely mentioned as something vaguely helpful. The protection of the environment has been invoked in a number of competition cases, but it is difficult to say whether there is a policy of relaxing normally-applicable competition law principles to achieve green goals. The one public policy which seems to be a genuine and consistent driver of competition law is market integration. My tentative conclusion is therefore not that we need more examples of the relaxing of the competition law rules in favour of causes deemed worthy, like market integration, response to the crisis, or environmental protection, but rather that we need to recalibrate competition law principles to take account of modern doctrines. And that recalibration means not just in published guidance but in arguments submitted to a court which has a judicial standard of legality rather than correctness. US antitrust law, although applied in a very different procedural context, is significantly more ‘friendly’ to business interests, though more fierce in punishing cartels and cartellists. * Queen’s Counsel at the Scots Bar; Honorary Professor, University of Glasgow; White & Case, Brussels. Warm thanks are expressed to Strati Sakellariou, Dikigoros of the Athens Bar, Peter Hodal, Advokat of the Bratislava Bar, and Sandra Keegan of the Louisiana Bar, all of whom have so cheerfully shared their knowledge, ideas and insights. The opinions expressed are wholly personal.

438  Competition, Regulation and Public Policies This chapter considers some ancient judicial precedents about restrictive contractual terms, crisis cartels, regulatory measures which restrict competition, collective bargaining, and a miscellany of supposedly relevant public policies. It concludes with a reflection on whether the market integration feature of competition policy needs a fresh look. The public policy of market integration has had a very major impact on competition policy, on the rules governing intellectual property, on the enforcement of competition law and on individual decisions. European competition law is unique in the world in attributing such significance to market integration. It would be proper to give serious reflection to whether this huge policy consideration should be recalibrated in light of modern notions of economic effects and benefits for consumers. I have suggested that the unique emphasis of European competition law on market integration had become a civil religion, and further suggested that the religion would benefit from some sceptical analysis. It would do no harm to reflect on whether we are truly better off with it.

1. Public policy and restrictive contracts: not a new problem In the era before the Sherman Act,1 courts confronted the tension between rigid, consistent, inflexible clarity and flexible, unpredictable, arbitrary unpredictability. In the English courts in the reign of the first Queen Elizabeth, litigations arose about contracts in restraint of trade. The problem commonly encountered was constraints on the right of a worker to earn his (always a ‘he’ as far as I know) living from his trade. These constraints were sometimes entered into by family members, potential competitors or others lacking evident authority to bind the worker not to compete in a particular town or village. In Colgate v Bacheler,2 it was held that This condition is against law, to prohibit or restrain any to use a lawful trade … which is against the benefit of the commonwealth, for being freemen it is free for them to exercise their trade in any place.

The old absolute rule was set forth as follows: Any deed by which a person binds himself not to employ his talents, his industry, or his capital in any useful undertaking in the kingdom, would be void, because no good reason can be imagined for any person’s imposing such a restraint on himself.3

In the famous case of Nordenfelt v Maxim Nordenfelt Guns and Ammunition,4 the inventor and entrepreneur sold out his business to an acquirer, and accepted what 1 The Sherman Act of 1890, probably the best claimant to the title of the world’s first antitrust law. But it would be wrong to assume that the Sherman Act was the first legal measure to address restrictive practices. 2 Cro Eliz 872, 78 Eng Rep 1097 (1596) 3 Homer v Ashford (1825) 3 Bing. 322. 4 Nordenfelt v Maxim Nordenfelt Guns & Ammunition Company (1894) AC 535.

Ian S Forrester  439   we would now call a non-compete obligation, restraining himself from engaging ‘either directly or indirectly, in the trade or business of a manufacturer of guns, gun mountings or carriages … or ammunition’ anywhere on earth for 25 years. The relevant market was worldwide, and the potential customers were ‘Governments and potentates, great and small, civilized and savage, who for purposes offensive or defensive desire … Nordenfelt guns with suitable ammunition’. Thus Mr. Nordenfelt was a precursor of Dr. Reuter, who challenged the length of his noncompete restriction after selling his business to BASF.5 The plaintiff/inventor/ patentee (by now bankrupt and arguing his own case – fluently and powerfully – before the House of Lords) relied on the absolute notion of restraint of trade, especially as set forth in Homer v Ashford, cited above. He relied on a number of older cases which struck down as unenforceable restraints on the freedom of individuals to do business. Lord McNaghten (a celebrated name in legal history) described the tension between competition and the validity of private dealings as follows: In the age of Queen Elizabeth all restraints of trade, whatever they were, general or partial, were thought to be contrary to public policy, and therefore void (Colgate v Bacheler6). In time, however, it was found that a rule so rigid and far-reaching must seriously interfere with transactions of every-day occurrence. Traders could hardly venture to let their shops out of their own hands; the purchaser of a business was at the mercy of the seller; every apprentice was a possible rival. So the rule was relaxed.

As the Lord Chancellor put it: Courts will rightly refuse to enforce any compact by which an individual binds himself not to use his time and talents in prosecuting a particular profession or trade, when its enforcement would obviously or probably be attended with these injurious consequences. But it must not be forgotten that the community has a material interest in maintaining the rules of fair dealing between man and man. It suffers far greater injury from the infraction of these rules than from contracts in restraint of trade. I think it is now generally conceded that it is to the advantage of the public to allow a trader who has established a lucrative business to dispose of it to a successor by whom it may be efficiently carried on. That object could not be accomplished if, upon the score of public policy, the law reserved to the seller an absolute and indefeasible right to start a rival concern the day after he sold. Accordingly it has been determined judicially, that in cases where the purchaser, for his own protection, obtains an obligation restraining the seller from competing with him, within bounds which having regard to the nature of the business are reasonable and are limited in respect of space, the obligation is not obnoxious to public policy, and is therefore capable of being enforced. (emphasis added)

5 6

Commission Decision 76/743/EEC, 1976 OJ L254/40. Cro Eliz 872, 78 Eng Rep 1097 (1596).

440  Competition, Regulation and Public Policies The Lord Chancellor said that the function of the courts: when a case like the present is brought before them, is, in my opinion, not necessarily to accept what was held to have been the rule of policy a hundred or a hundred and fifty years ago, but to ascertain, with as near an approach to accuracy as circumstances permit, what is the rule of policy for the then present time. When that rule has been ascertained, it becomes their duty to refuse to give effect to a private contract which violates the rule and would, if judicially enforced, prove injurious to the community. When the series of cases, from the earliest to the present time, are carefully considered, I think they will be found to record the history of a protracted struggle between the principle of common honesty in private transactions, on the one hand, and the stern rule which forbade all restraints of trade on the other.

So there was a defeat for the individual who invoked the basic, absolute principle and tried to escape his self-imposed duty not to compete. A worldwide ban on post-sale competitive activity was, on balance, upheld. The courts’ discussion of whether to follow a rigid but narrow rule or a ‘rule of reason’ approach was intriguingly modern. The courts were aware of the need to avoid formalist rigidity, but equally they attached high importance to the binding effect of business decisions. The application of European competition law as codified in the Treaty of Rome, as amended, (the TFEU, we can be sure, will not be with us in ten years’ time, when some new euphonious Treaty has been devised) is primarily the responsibility of the European Commission, which has traditionally been accorded much discretion by the European Courts. Although the Commission is the ‘Guardian of the Treaty’, it is led by 27 Commissioners appointed by the Member States. These are leading political figures, nearly always former Ministers or even Prime Ministers. Their function is, properly, political. Each is exposed to political influences which are not necessarily coincidental with the objectives of competition law. The exposure to political influences combined with the responsibility of applying the competition law inevitably creates some tension. It used to be that competition law in Europe was more ‘political’ and less rigid than in the United States; and even today there are probably more occasions in Europe than in the US where one policy is balanced against another. This tension is evident in the relationship between competition law and industrial policy, competition policy and market integration, competition law and health care, and competition law and public policy on the regulation of professional bodies and sporting ones. That does not mean that enforcement is ‘politicised’ (though it might be), but it does mean that political goals and public interest choices will never be far away from a big case. As noted in my reflections for the 1997 workshop in this series,7 competition policy has been invoked for many different reasons and in many different contexts as a cure for many diseases. In its 2011 Report on Competition Policy the Commission stated that it used competition policy and enforcement as an 7 Ian S Forrester, in Claus-Dieter Ehlermann and Laraine Laudati (eds), European Competition Law Annual 1997: The Objectives of Competition Policy (Hart Publishing 1998) p 359.

Ian S Forrester  441   instrument in the resolution of the financial and sovereign debt crises and as a contributing factor to the wider policy objectives of the Europe 2020 strategy: growth, jobs and competitiveness.8

2. Competition law and Industrial policy: crisis cartels The tension between industrial policy and competition law has been very evident in cases of industrial overcapacity. No Treaty provision specifically provides for the possibility of a suspension of the application of competition rules in times of economic crisis or in the case of industry facing structural overproduction. However, in instances where an industry is plagued with overcapacity and similar structural problems, industry members have claimed that there has been a market failure and that horizontal cooperation to limit production is justified. Indeed, they argue, is it not sensible that an industry combination can resolve the problem better, in a more European manner, than competition law? Why demand bankruptcies and disruptions when negotiated reductions can achieve the same result? Thirty years ago, in its 12th Report on Competition Policy, the Commission stated that it may be able to condone agreements in restraint of competition which relate to a sector as a whole, provided they are aimed solely at achieving a coordinated reduction of overcapacity and do not otherwise restrict free decision-making by the firms involved.9

The Commission noted however that ‘the necessary structural reorganization must not be achieved by unsuitable means such as price-fixing or quota agreements’.10 It added that it could authorise sectoral agreements only where the four conditions of Article 101(3) TFEU (Article 85(3) EEC, as greybeards call it) are met, notably: (i) a detailed and binding programme of closures which ensures that overcapacity is irreversibly dismantled and that, while the programme is in operation, no new capacity is created; (ii) consumers not to be deprived of the freedom of choice between competitors; (iii) any information exchange to be solely with a view to supervise capacity reductions; and (iv) limited duration. This approach was reflected in various Commission decisions concerning sectors which had been struggling with production over-capacity. Agreements were notified. Viscount Davignon and other Commissioners helped to broker conclusions. Competition law was in those days at its most political. There were a few cases in which competition authorities considered such arguments in defence only after the competition investigation had been initiated; in that 8 Report on Competition Policy 2011, COM ( 2012) 0253, page 1. Also available at http://eur-lex. europa.eu/LexUriServ/LexUriServ.do?uri=COM:2012:0253:FIN:EN:PDF. 9 XIIth Report on Competition Policy – 1982, paragraph 39. 10 Ibid.

442  Competition, Regulation and Public Policies context, the existence of structural market problems was considered mostly as an attenuating circumstance in the calculation of fines. It seems unlikely that such deals would obtain Commission blessing today, but perhaps that assumes such problems will not arise again. It is also relevant to note that the European Coal and Steel Community was all about industrial coordination and capacity-sharing, so such arrangements have a venerable provenance. We may yet be surprised. I shall now offer some examples in specific cases.

Synthetic fibres and petrochemicals In Synthetic Fibres,11 the Commission concluded that market forces had failed to bring about the reduction in overcapacity necessary to achieve an effective competitive structure and exempted the agreement under what is now Article 101(3) TFEU. The agreement allowed the parties to achieve higher capacity utilisation rates, better individual specialisation and to restructure in a way that facilitated the organisation of accompanying social measures for retraining and redeployment of workers made redundant. The Commission also found that competition would not be eliminated as there were imports from third countries and substitutable products. The Commission granted a temporary exemption, subject to the condition that the parties refrain from exchanging information about their individual output and deliveries. 12 Other examples of exemption under Article 101(3) TFEU include bilateral restructuring agreements where some producers agreed to stop making certain products and to specialise in others. Most of these cases concerned situations of overcapacity in the petrochemical industry, with the Commission finding that market forces were too slow to bring about restructuring changes on an individual basis such that reciprocal specialisation could be attained only through an arrangement between the parties. Such bilateral agreements (eg, BPCL/ICI and ENI/Montedison) reduced the parties’ portfolio and avoided overlaps following the restructuring, so the Commission found that an exemption under Article 101(3) TFEU could be granted.13 The Commission considered that the agreements allowed the parties to rationalise and reduce excess capacity more rapidly and Commission Decision of 4 July 1984 – Synthetic Fibres, 1984 OJ 1984 L207/17. See Case C-176/99 Arbed SA v Commission [2003] ECR I-10687, the ‘steel beams case’, in which producers of steel were granted a – legal – ‘quota system’ following the Commission’s declaration of a manifest crisis within the meaning of Article 58 ECSC. In 1980, after having attempted to manage the crisis by way of unilateral voluntary commitments given by undertakings as regards the amount of steel put on the market and minimum prices (the Simonet Plan) or by fixing guide and minimum prices (the Davignon Plan, the Eurofer I agreement), the Commission declared that there was a manifest crisis within the meaning of Article 58 of the ECSC Treaty and imposed mandatory production quotas for, inter alia, beams. That Community system came to an end on 30 June 1988. 13 See Commission Decision of 19 July 1984 – BPCL/ICI, 1984 OJ L212/1 and Commission Decision of 5 May 1988 – Bayer/BP Chemicals, 1988 OJ L150/35, for the restructuring of the petrochemical industry in the UK; Commission Decision of 4 December 1986 – ENI/Montedison, 1987 OJ L5/13, concerning the restructuring of the Italian petrochemical industry. 11

12

Ian S Forrester  443   radically than would have been possible otherwise. An exemption under Article 101(3) TFEU was granted.14

Stichting Baksteen15 A rare example from the 1990s of an authorised crisis cartel was Dutch Brickmakers, which concerned the grant of an exemption to an industry capacity reduction programme between Stichting Baksteen (‘the Brick Foundation’) and sixteen brick producers. The agreement, as originally notified, included the fixing of production quotas and the allocation of all the country’s productive capacity. The Commission considered this approach to be ‘inappropriate and could not be justified solely on the ground of a crisis in the brick industry’.16 The original plan was withdrawn and replaced by a new, less rigid, agreement aimed at reducing capacity in order to restore balance of supply and demand. The decision begins with a technical analysis of why capacity reductions are necessary. The production of bricks is capital-intensive so profitability depends on a high capacity utilization rate. Capacity utilization had fallen by ten per cent, prices had fallen by thirty per cent in real terms, consumption by twenty per cent, and stocks had risen by twelve per cent over the desirable level. The parties therefore agreed to cut capacity by about 200 million bricks, by closing down seven production units owned by four different producers, for at least thirty years. These producers would refrain from selling, for at least thirty years, the dismantled plants to producers within a radius of 500 kilometres of the Dutch frontier. The costs of these closures, including social costs, were to be financed, in ‘solidarity’, by the sixteen firms which signed the agreement, through a compensation fund managed by the Brick Foundation. The implementation of a social plan for the sector, negotiated with the trade unions, would also be monitored by the Foundation. All parties to the agreement were prohibited from introducing new capacity. Article 101(1) was evidently applicable, since the agreement, although ultimately pro-competitive, restricted the parties’ means of production, investments, and competitive strategies. It could also affect intra-Community trade. Although the market for bricks is fairly regional (the weight of bricks, and their low intrinsic value, make transport over long distances unprofitable), the Commission considered that trade with the bordering regions of Belgium, Germany, and the UK could be influenced. However, the Commission felt that the agreement qualified for an exemption under Article 103(3) for the following reasons: (i) the agreement was aimed at increasing the profitability of the Dutch brick industry, and its return to normal competitiveness; (ii) in addition, because of the co-ordination of the closures, the 14 15 16

See citations in the previous footnote. Commission Decision of 29 April 1994 – Stichting Baksteen, 1994 OJ L131/15. Ibid, para 6.

444  Competition, Regulation and Public Policies restructuring would be carried out in acceptable social conditions; (iii) consumers were given a fair share of the resulting benefit; (iv) in the long term, the industry would be restored to health and offering competitive supplies; (v) in the short term consumers would continue to have a choice of supplier and security of supply. The Commission accepted that the effect on prices would be offset by diminishing stockpiling costs. The measures were considered indispensable to the attainment of reduction in capacity. The parties would not have been willing or able, in the absence of coordinated action, to make capacity cuts: certain producers had only one kiln and were thus technically incapable of making any capacity cuts; other producers would not have decided to reduce capacity if there was no certainty that competitors would follow their example and provide financial support. The reader is slightly surprised that the outcome was positive.

FNCBV17 The Dutch Brickmakers case may have engendered, ten years later, optimism as to the prospects for an agricultural crisis cartel. But it was not to be. In December 2006 the then Court of First Instance dismissed an action brought by four French federations of farmers and one French federation of slaughterers. The Court upheld a widely publicized Commission Decision18 which imposed substantial fines on the federations for concluding a minimum price agreement and agreeing to suspend or at least limit beef imports into France from other Member States. The circumstances were exceptional. The agreement was concluded in an atmosphere of crisis and tension fuelled by widespread demonstrations by farmers, with some tacit tolerance and even support from the French government. (France has an intriguing history of vulnerability to rioting farmers.) Farmers’ organisations relied on the following factors: (i) drop in the consumption of beef as a result of the ‘mad cow’ crisis; (ii) intervention measures taken by the Community and national authorities aimed at restoring balance in the beef market; (iii) loss of consumer confidence, linked to the fear of ‘mad cow’ disease; and (iv) the situation of farmers who, despite Community adjustment measures applied by France, were faced with slaughterhouse entry prices for cows which were falling again, while consumer prices remained stable. The Court found that it constituted a restriction ‘by object’ and rejected attempts by the parties to invoke exceptional circumstances. Equally, the Court refused to accept that ‘the freedom of trade unions to protect their collective interests justified conduct which infringed Article 81(1) of the EC Treaty’.19 17 Joined Cases T-217/03 and T-245/03 FNCBV and Others v Commission [2006] ECR II-4987, upheld on appeal: Joined Cases C-101/07 and 110/07 Coop de France bétail et viande and Fédération nationale des syndicats d’exploitants agricoles (FNSEA) and Others v Commission [2008] ECR I-10193. 18 Commission Decision of 2 April 2003 – French Beef, 2003 OJ L209/12. 19 FNCBV, paras 81–94, inducing the regulated character of the market, the ministerial intervention, and the background of the crisis.

Ian S Forrester  445   The applicants went on to argue that the agreement should benefit from the exemption in Regulation 26/6220 for activities related to production and trade in agricultural products which are necessary for realizing the objectives of the common agricultural policy. The Court rejected this argument, as the agreement did not stabilize the market or aim to ensure a fair standard of living for the agricultural community, two objectives of the policy. Alternatively, the applicants asked for a substantial reduction of the fines imposed by the Commission (€12 million (FNSEA); €1.44 million (FNB); €600,000 (JA); €144 million (FNPL); €720,00 (FNICGV); €480,000 (FNCBV)), and indeed, the General Court took into account the economic context of the agreement, but only for purposes of setting the amount of the fine. Here the Court increased the 60% reduction granted by the Commission to a 70% reduction due to the exceptional circumstances of the mad cow crisis and the fact that the agreement was exclusively between federations dealing with two levels of the production chain of a basic agricultural product, 21 a slightly curious rationale for an unsurprising concession.

Irish Beef 22 More recently, in December 2008, Irish Beef concerned an agreement between Irish beef processors to set up the Beef Industry Development Society (BIDS) in order to reduce production overcapacity on the Irish beef market. BIDS produced a rationalisation plan which provided, inter alia, for a reduction in capacity of the beef processing industry by 25%. Under the plan, BIDS would offer a standard contract to processors withdrawing from the market (the goers) to whom a compensation payment would be made from a fund generated by a levy on members of BIDS remaining in the sector (the stayers). The Irish Competition Authority challenged the plan, and a reference to the ECJ followed. The Dutch brickmakers and the Italian petrochemical producers fared a lot better. The ECJ viewed the BIDS arrangements as being intended to enable several undertakings to implement a common policy whereby some processors would be encouraged to withdraw from the market in order to reduce the overcapacity problem which was affecting the beef processors’ profitability by preventing them from achieving economies of scale.23 But each operator should independently determine its own commercial behaviour on the common market, rather than through coordination with others.24 20 Council Regulation 26/62 of 4 April 1962 applying certain rules of competition to production of and trade in agricultural products, 1962 OJ 30/993, Special Edition 1959-1962, page 129. 21 FNCBV, paras 360–361. 22 Case C-209/07 The Competition Authority v Beef Industry Development Society Ltd, Barry Brothers (Carrigmore) Meats [2008] ECR I-8637. 23 Ibid, para 33 24 Ibid, para 34.

446  Competition, Regulation and Public Policies In so ruling, the ECJ rejected the argument made by BIDS that the arrangements could not constitute a restriction of competition ‘by object’ as their intention was not to restrict competition but to rationalise and make more competitive the beef industry by reducing production over-capacity: [E]ven supposing it to be established that the parties to an agreement acted without any subjective intention of restricting competition, but with the object of remedying the effects of a crisis in their sector, such considerations are irrelevant … Indeed, an agreement may be regarded as having a restrictive object even if it does not have the restriction of competition as its sole aim but also pursues other legitimate objectives ….25

The arguments in favour of the structural reasonableness of the BIDS deal could be considered, if at all, only in the context of an exemption under Article 101(3) TFEU. However, we learn from GSK Spain26 that the merits of an unwelcome request for exemption must be fairly considered. Such agreements will not be considered as per se illegal, since it is theoretically possible that they satisfy the criteria of Article 101(3) TFEU. But winning the argument will not be easy.27

Conclusion on industrial policy and competition law Blessing a crisis cartel has not been known in Europe for about 20 years and seems unlikely now. It seems that industrial restructuring agreements would most probably be considered unlawful.28 Although there are precedents from the 1980s and 1990s, and although the rationales of those decisions are not time-sensitive, Ibid, para 21 (emphasis added). Case T-168/01 GlaxoSmithKline Services Unlimited v Commission [2006] ECR II-2969. Joined Cases 56/64 and 58/64 Établissements Consten SARL and Grundig-Verkaufs-GmbH v Commission [1966] ECR 299, paras 342, 343 and 347; Case T-17/93 Matra Hachette SA v Commission [1994] ECR II-595, para 85; Case T-168/01 GlaxoSmithKline Services Unlimited v Commission [2006] ECR II-2969, para 233. 28 See the Contribution on Crisis Cartels submitted by the European Union to Session III of the OECD’s Global Forum on Competition, 17–18 February 2011, DAF/COMP/GF/WD(2011)20, 27 January 2011. According to the Commission a ‘crisis cartel’ (Contribution to the Global Forum, paragraphs 5 and 38) may only be tolerable in the rare event that competitive market forces alone would not remove excess – structural (and not cyclical) – overcapacity. Such structural overcapacity only exists ‘where over a prolonged period all the undertakings concerned have been experiencing a significant reduction in their rates of capacity utilisation and a drop in output accompanied by substantial operating losses and where the information available does not indicate that any lasting improvement can be expected in this situation in the medium-term.’ (XIIth Report on Competition Policy (1982), point 38). True, there may be situations in which market forces alone will not be able to remedy problems of overcapacity. This rare event is, according to the Commission, marked by the high costs of giving up overcapacity and stable, transparent and symmetric market structures. Thus, the undertakings involved would be reluctant to give up overcapacity but will rather wait and suffer economic losses for a while until competitors exit the market (Contribution to the Global Forum, paragraphs 8–11, and 42). This is referred to as a ‘war of attrition’ which could lead to a severe waste of economic resources (Contribution to the Global Forum, paragraphs 8 and 11). Hence, this could substantially harm the industry’s competitiveness and ultimately even affect consumers (Contribution to the Global Forum, paragraph 11). Thus the Commission does not entirely close the door on an exemption, but confirms that the door is very stiff. 25 26 27

Ian S Forrester  447   it is not likely that future parties will be able to invoke them in the present era. Again, we may be surprised, as times are hard.

3. Cautious deference to regulation Competition law often has to consider specific policy objectives pursued by national legislation or professional organisations. Policy objectives, when expressed in specific regulations, may restrict competition at the expense of promoting a public policy goal. The European Courts have often had to reconcile public policy grounds as set by the Member States (or as borne in mind by Member States when acting) against their potential anticompetitive effect. Some of these disputes have touched the very interesting topic of private freedom of initiative, and even private concertation, within a context of public regulation. Where the regulator acts as the industry recommends, or in other circumstances acts so as to narrow the scope of independent action for individual businesses, the usual rules govern the private operators. There are a number of disparate authorities, and it is not easy to fit them into a consistent analytical framework. I propose to divide the authorities into three categories: (i) national legislation, notably as to price, where the Member State argues that it does not actively favour anticompetitive arrangements or reinforce their effects, (ii) regulation of professional organisations, notably in sport and (iii) specific social policy objectives. In Pascal Van Eycke,29 the Court had to review the Belgian legislation governing the interest which may be paid on savings deposits. The interest rates advertised by the banks were reduced after the adoption of the legislation. This legislation had to be viewed in its legal and economic context, as the interest on savings had been exempt from taxation, and there was a significant increase in interest rates which led also to a significant increase of interest on loans. As the selfregulatory attempts to address these problems had failed, the Belgian Ministry of Finance imposed a regulation that fixed the maximum level of the basic interest rate with respect to tax exemptions. Mr. Van Eycke suspected that the rates he was paying were not set purely by market forces. He did not win, but he moved the law forward a little. The Court held that: Articles 85 and 86 of the Treaty are per se concerned only with the conduct of undertakings and not with national legislation. Articles 85 and 86, in conjunction with Article 5, require the Member States not to introduce or maintain in force measures, even of legislative nature, which may render ineffective the competition rules applicable to undertakings. Such would be the case, if a Member State were to require or favour the adoption of agreement, decisions or concerted practice contrary to Article 85 or to 29

Case 267/86 Pascal Van Eycke v ASPA NV [1988] ECR 4769.

448  Competition, Regulation and Public Policies reinforce their effects, or to deprive its own legislation for taking decisions affecting the economic sphere.30

Therefore, the Court did not consider the national legislation to be problematic. In the later case of Meng,31 the Court was forced to clarify its position. What if the Member State adopts legislation clearly prohibiting normal competition, but there is no direct evidence that there are or were any agreements between undertakings which were favoured or required or whose effects were reinforced, and the legislation does not delegate powers to private traders? German administrative bodies adopted rules forbidding insurance brokers from providing rebates to customers corresponding to part or all of the commission which the brokers received from insurers on issuance of a policy; the rules applied in the fields of health and liability insurance. Another way of describing the practice would be as a prohibition of price competition between brokers. While there was no evidence of agreements in these sectors, there had apparently been agreements in other insurance sectors. Meng, a broker prosecuted for infringing these rules (that is, for being unfairly pricecompetitive), challenged the rules, among others, for their incompatibility with the competition rules of the Treaty.32 The stakes in the case were high. The Commission tried to extend the Van Eycke test by claiming that Member States may not ‘adopt a public measure which restricts competition taken with the specific purpose of permitting undertakings to avoid application of Articles 85 and 86 without some public interest being claimed in support’. Repeating the Van Eycke formula, the Court stated that there were no agreements to which to link the regulations, and no delegation. The fact that there might have been agreements in neighbouring sectors of the insurance industry was irrelevant; the required agreement had to be in the same sector.33 The Court also took a cautious line in Ohra Schadeverzekeringen NV,34 which involved a prosecution for infringements of a Dutch law prohibiting insurance companies from granting rebates to those whom they insured. The Court stated that the Dutch rules neither imposed nor encouraged an illegal understanding between those in the insurance industry. The rules in question did not therefore reinforce an anticompetitive understanding and, moreover, had not been preceded by any private sector agreement or understanding. In addition, private-sector operators were given no individual responsibility by the legislation for its Ibid, para 16. Case C-2/91 Wolf W. Meng [1993] ECR I–5751. 32 See further Ian Forrester and Christopher Norall, ‘Annual Survey of Competition Law’, 13 Yearbook of European Law 427, 480–481 (1993). 33 See also Hans Gilliams ‘Competition Law and Public Interest: Do We Need to Change the Law for the (Liberal) Professions’, in Claus-Dieter Ehlermann and Isabela Atanasiu (eds), European Competition Law Annual 2004: The Relationship between Competition Law and (Liberal) Professions (Hart Publishing, 2005) p 11: ‘The regulation that the public authorities had imposed was preceded by a (presumably illicit) agreement between undertakings, and the Commission had pointed out to the ECJ that the public regulation in essence had reproduced the substance of that agreement. The Court nevertheless refused to confirm the applicability of Article 81 EC to the regulation.’ 34 Case C-245/91 Ohra Schadeverzekeringen NV [1993] ECR I-5851. 30 31

Ian S Forrester  449   implementation. The legislation itself imposed the prohibition for which Ohra was being prosecuted. A similarly abstemious approach was followed in Reiff.35 The facts were different in that they involved the setting of road freight tariffs by boards composed of experts selected by the federal Ministry of Transport from a list of nominees recommended by industry members and trade associations. The relevant law provides that the members of the boards would serve in an honorary capacity and would not be bound by orders or instructions (from the companies or associations who paid their salaries, for example). The Minister could attend meetings or send a representative. An advisory committee could issue an opinion on each tariff rate proposal, and the Minister must either approve the rate or impose a different one. Had the Court felt bolder, it could have observed that if they occurred in a private manner, the meetings of the members of the tariff boards would manifestly have infringed the rules; the Court could also have enquired whether in reality the rates proposed were routinely endorsed by the Minister or whether in reality the board’s recommendation was no more than a recommendation. A private cartel which enjoys the chance of having its determinations made legally compulsory is indeed in a most privileged position. The term ‘unholy alliance’ has often been given to the willing co-operation between the private sector and the public sector to impose private sector decisions by public law compulsion. The judgment did not give carte blanche to unholy alliances. It noted that in the present case the members were independent experts given the job of fixing tariffs in the general interest, subject always to the supervision of public bodies which could overrule them. The Court cautiously held that in these particular circumstances, the EC Treaty did not prohibit such an arrangement.36 The fact that three cases were decided in parallel suggests that the Court knew well what it was doing. Any référendaires who longed for excitement were disappointed. Likewise, in Arduino,37 the legality of a scale setting the minimum and maximum fees chargeable by lawyers in Italy for their services survived hostile analysis. The Italian Ministry of Justice adopted the scale on the basis of a draft prepared by the governing body of the Italian Bar. Before approving a fee-scale, the Ministry was obliged, under a statutory procedure, to consult the Italian Council of State and the Inter-ministerial Committee on Prices. Arduino was a disappointed client. The Court was asked to give its ruling on whether Articles 10 and 81 EC (now Article 4(3) TEU and Article 101 TFEU) precluded Member States from adopting a measure, in accordance with statutory procedures, approving a tariff fixing minimum and maximum fees. The ECJ held that, although the EC competition rules focused on the conduct of undertakings, Article 81 (read in conjunction with Article 10), prohibited Member States from adopting rules rendering the application of EC competition law ineffective. Articles 10 and 81 were infringed: 35 Case C-185/91 Bundesanstalt für den Güterfernverkehr v Gebrüder Reiff GmbH & Co. KG [1993] ECR I-5801. 36 Forrester and Norall, ‘Survey’, cited above note 32, at 482. 37 Case C-35/99 Manuele Arduino and Compagnia Assicuratrice RAS SpA [2002] ECR I-1529.

450  Competition, Regulation and Public Policies where a Member State requires or favours the adoption of agreements, decisions or concerted practices contrary to Article 81 or reinforces their effects, or where it divests its own rules of the character of legislation by delegating to private economic operators responsibility for taking decisions affecting the economic sphere.38 In that regard, the fact that a Member State requires a professional organisation to produce a draft tariff for services does not automatically divest the tariff finally adopted of the character of legislation.39

The Court noted that the draft required ministerial approval before it could enter into force, and that the Italian courts might depart from the maximum and minimum limits. The Italian State could therefore not be said to have delegated to private economic operators responsibility for decisions affecting the economic sphere (which would have the effect of depriving the provisions of the character of legislation). Thus, the fee schedules were held to be legal, a remarkable outcome: a trade association imposed a tariff on its members in the public interest, but did so with legislative backing, and the measure was lawful.40 The Commission was, predictably, disappointed and sought to limit the scope of the doctrine. The same question on Italian lawyers’ scale fees arose in the Cipolla case.41 The question was whether the Court could be persuaded to be somewhat bolder than in Arduino. The Commission called for a reconsideration of Arduino, amongst others, in light of the opinions of Advocates General Jacobs42 and Léger43 who suggested a three-fold test: (1) the public authorities of the Member State concerned exercise effective control over the content of the agreement; (2) the State measure pursues a legitimate aim in the public interest, and (3) the State measure is proportionate to the aim which it pursues. The proportionality test put forward by Advocates General Jacobs and Léger would have left a door open to challenge the details of specific tariffs.44 But the judgment in Cipolla closely follows that in Arduino, with the Court focusing on the procedural guarantees that ensure that the tariffs are in the general interest, and not just in the interest of the bar association that fixes the draft tariff. Not a great surprise.45

4. Deference to professional organisations and their regulations Public policy considerations also play a role in self-regulation of professional organisations which inevitably limit members’ freedom of activity and modes of Ibid, para 34. Ibid, para 36. 40 Ian Forrester and Jacquelyn MacLennan, ‘Annual Survey of EC Competition Law’, 22 Yearbook of European Law 2001–2002 499, 551 (2003). 41 Joined Cases C-94/04 and C-202/04 Cipolla and Macrino [2006]. 42 Opinion in Joined Cases C-180/98 to C-184/98 Pavel Pavlov and Others v Stichting Pensioenfonds Medische Specialisten [2000] ECR I-6451. 43 Opinion in Arduino, cited above note 38. 44 Hans Vedder, ‘Of Jurisdiction and Justification: Why Competition is Good for Non-Economic Goals, But May Need to be Restricted’, 6 Competition Law Review 51, 67–68 (2009). 45 Ibid, at 68. 38 39

Ian S Forrester  451   behaviour. The Court seems to give what could be called deference if the restraint of the competition is justified by objective grounds relative to the proper conduct of the regulated activity. The leading case on deference to public policy-based anticompetitive regulations of professional organisations is Wouters.46 A Dutch lawyer, Mr. Wouters, applied to the Dutch Bar association for authorisation to enter into partnership with an accounting firm. The Bar rules prohibited lawyers from entering multidisciplinary partnerships with accountants. Mr. Wouters initiated proceedings before the Dutch courts, and the case was referred for a preliminary ruling to the ECJ. The Dutch Bar association was to be regarded as an association of undertakings within the meaning of Article 101(1) TFEU. The Court was also willing to accept that the Dutch ban on multidisciplinary partnerships might limit production and technical development within the meaning of Article 101(1) TFEU, and was likely to affect trade between EC Member States: indeed, it applied to foreign lawyers and accountancy firms wishing to form partnerships with practitioners in several Member States. The Court concluded that the regulation was liable to have an effect on competition. This was not a controversial conclusion, since multi-disciplinary partnerships were under consideration, Arthur Andersen had entered into association with large law firms, and Bars (and professions) were divided as to how to provide for such challenges. Some forbade this form of economic activity, a few did not. However, the Court then went to examine whether the resulting restrictions were inherent for the pursuit of those objectives, noting that: not every agreement between undertakings or every decision of an association of undertakings which restricts the freedom of action of the parties or of one of them necessarily falls within the prohibition laid down in Article 85(1) of the Treaty [Article 101(1) TFEU]. For the purposes of application of that provision to a particular case, account must first of all be taken of the overall context in which the decision of the association of undertakings was taken or produces its effects. More particularly, account must be taken of its objectives, which are here connected with the need to make rules relating to organisation, qualifications, professional ethics, supervision and liability, in order to ensure that the ultimate consumers of legal services and the sound administration of justice are provided with the necessary guarantees in relation to integrity and experience.... It has then to be considered whether the consequential effects restrictive of competition are inherent in the pursuit of those objectives.47

The Court upheld the right of a Member State to establish rules governing the exercise of a professional activity in its territory, and accepted the Bar’s argument that its rules were needed to promote the independent giving of legal advice and avoid conflicts of interest.48 The ECJ attributed much weight to the freedom 46 Case C-309/99 JCJ Wouters, JW Savelbergh and Price Waterhouse Belastingadviseurs BV v Algemene Raad van de Nederlandse Orde van Advocaten [2002] ECR I-1577. 47 Ibid, para 97. 48 Ibid, para 99: ‘in the absence of specific Community rules in the field, each Member State is in principle free to regulate the exercise of the legal profession in its territory (Case 107/83 Klopp [1984] ECR 2971, paragraph 17, and Reisebüro, paragraph 37). For that reason, the rules applicable to that profession may differ greatly from one Member State to another.’

452  Competition, Regulation and Public Policies of the Bar to regulate its own affairs and, in so doing, to restrict the freedom of its members to engage in certain activities. It held that the anticompetitive restrictions resulting from the regulation do not go beyond what is necessary to ensure the proper practice of the legal profession. The Court did not pursue a lowest common denominator approach, and clearly did not try to second-guess the Bar’s rules. Accordingly, the EC competition rules did not prohibit the taking of a carefully considered measure by the Bar,49 which disallows a quite large category of potential economic activity. The collapse of Arthur Andersen in the wake of the collapse of its client Enron has subsequently cooled the enthusiasm for professional experimentation (critics might say miscegenation). Wouters was followed by a more specific case, Meca-Medina,50 where the two European Courts reached their parallel conclusions by very different routes. Mr. Meca-Medina and Mr. Majcen were two professional athletes who competed in long-distance swimming contests. They tested positive for Nandrolone and in consequence were suspended by the International Swimming Federation under the Olympic Movement’s Anti-Doping Code. The two athletes complained to the Commission, alleging that the International Olympic Committee’s rules on doping control were not compatible with the Community rules on competition and freedom to provide services. Following the Commission’s rejection of their complaint, they brought an action before the CFI to have the decision set aside. The CFI dismissed their action and held that the rules on doping did not fall within the scope of Community competition law. By contrast, on appeal, the Court of Justice held that sport is subject to Community law insofar as it constitutes an economic activity, and that the punitive nature of the rules at issue and the magnitude of the penalties applicable if they were breached were capable of producing adverse effects on competition. Rules of that kind could indeed prove excessive, owing both to the way in which the circumstances in which penalties for doping might be imposed were distinguished from those in which they were not, and to the severity of those penalties.51 To escape the prohibition on distortion of competition laid down by the Treaty, the restrictions imposed by those rules must be limited to what is necessary to ensure the proper conduct of competition.52 However: As regards the overall context in which the rules at issue were adopted, the Commission could rightly take the view that the general objective of the rules was, as none of the Forrester and MacLennan, ‘Survey’, cited above note 41, at 549 Case C-519/04 P David Meca-Medina and Igor Majcen v Commission [2006] ECR I-6991. 51 Ibid, para 42: ‘The compatibility of rules with the Community rules on competition cannot be assessed in the abstract. Not every agreement between undertakings or every decision of an association of undertakings which restricts the freedom of action of the parties or of one of them necessarily falls within the prohibition laid down in Article [101 TFEU]. For the purposes of application of that provision to a particular case, account must first of all be taken of the overall context in which the decision of the association of undertakings was taken or produces its effects and, more specifically, of its objectives. It has then to be considered whether the consequential effects restrictive of competition are inherent in the pursuit of those objectives (Wouters and Others, paragraph 97) and are proportionate to them.’ (emphasis added) 52 Ian Forrester, Jacquelyn MacLennan and Assimakis Komninos, ‘Annual Survey of EC Competition Law’, 27 Yearbook of European Law 2005–2006 521, 540 (2008). 49 50

Ian S Forrester  453   parties disputes, to combat doping in order for competitive sport to be conducted fairly and that it included the need to safeguard equal chances for athletes, athletes’ health, the integrity and objectivity of competitive sport and ethical values in sport.53

It did not appear that these restrictions went beyond what was necessary to ensure that sporting events took place and function properly. Thus the anti-doping measures under challenge were legitimate and proportionate regulations which were within the scope of the competence of the body adopting them. Thus constitutionally the measures did not escape scrutiny under the competition rules, but they were to be respected as a good faith effort to address an important problem. However, matters were treated differently in another case concerning sport, Piau,54 on the regulation of football players’ agents. Mr. Piau challenged the FIFA Players’ Agents Regulation as being anticompetitive. The Court noted that ‘the regulation was adopted by FIFA of its own authority and not on the basis of rule-making powers conferred on it by public authorities in connection with a recognised task in the general interest’55 unlike the case of Wouters. Piau suggests a different treatment of cases where the regulation is adopted by private bodies with conferred rule-making power and private bodies applying their internal rules. There seems to be an evident contrast with Meca-Medina. The Court nonetheless went on to conclude that the rules on agents were, on their merits, a proportionate and appropriate way to prevent abuses and to protect players. The CFI did not apply the Wouters test under Article 101(1) but under Article 101(3) TFEU, and concluded that restrictions stemming from the compulsory nature of the licence might benefit from an exemption. The existence of the regulation was justified by the past behaviour of the players’ agents that harmed players and clubs, financially and professionally and ‘FIFA’s objective to raise professional and ethical standards for the occupation of players’ agent in order to protect players who have a short career.’56 So although the rule-maker had not been publicly endowed with a public function, what it did on its merits deserved endorsement, although the conditions for exemption under Article 101(3) TFEU refer to economic benefits and not public policy benefits. In Wouters and in Meca-Medina, the ECJ held that not every agreement or decision restricting the freedom of action of undertakings (or of one of them) necessarily falls within the prohibition. In order to decide whether the Article 101(1) TFEU prohibition applies, national courts and the national competition authorities should consider public interest issues, in spite of the restriction of certain parties’ freedom of action. Provided the effect on competition results from the pursuit of those objectives, the prohibition may not apply. Put another way, the ECJ confirmed that the competitive rules of the Treaty are not governed exclusively by competition law concerns in sporting matters. Other legitimate Meca-Medina, cited above note 50, at para 43. Case T-193/02 Laurent Piau v Commission [2005] ECR II-209, appeal dismissed, Case C-171/05 [2006] ECR I-37*. 55 Ibid, para 74. 56 Ibid, para 102. 53 54

454  Competition, Regulation and Public Policies considerations should be respected. If the measures concerned are necessary for a worthwhile purpose (the promotion of the integrity of the legal profession was one such), they may be objectively justified and fall outside the prohibition of the competition rules.57 The CFI seemed to conduct a similar analysis in Piau. Although it had to evaluate if all of the conditions for exemption under Article 101(3) TFEU are met (which only refer to economic benefits and not public policy benefits) it also referred to policy reasons behind the regulation.

Collective bargaining A good example of cartellised behaviour with strong economic consequences is wage negotiations between employers and trade unions. The ECJ appears to have created a specific category of public policy exceptions from competition law for ‘collective bargaining agreements’. In Albany,58 the ECJ held that agreements concluded in the context of collective negotiations between management and labour, by virtue of their nature and purpose, fall outside the scope of Article 101(1) TFEU, where the purpose of the agreement is to improve conditions of work and employment.59 The Court said in particular, it is beyond question that certain restrictions of competition are inherent in collective agreements between organisations representing employers and workers. However, the social policy objectives pursued by such agreements would be seriously undermined if management and labour were subject to Article 101 TFEU when seeking jointly to adopt measures to improve conditions of work and employment. It therefore follows from an interpretation of the provisions of the Treaty as a whole … that agreements concluded in the context of collective negotiations between management and labour in pursuit of such objectives must, by virtue of their nature and purpose, be regarded as falling outside the scope of Article 101(1) TFEU.60

The Court seems to be willing to give a free pass to some social policy objectives of Member States and exclude assessment of agreements concluded for the execution of these objectives under competition law. The open question is how broad is the category of social policy objectives to which the Court will grant automatic deference without analysing the anticompetitive nature of the agreements in these areas. The Court’s decision in the celebrated Bosman61 case is not easy to reconcile with Albany. Mr. Bosman was the hero and victim of a challenge to the rule Forrester and MacLennan, ‘Survey’, cited above note 41, at 550. Case C-67/96 Albany International BV v Stichting Bedrijfspensioenfonds Textielindustrie [1999] ECR I-5751. 59 Heike Schweitzer, ‘Competition Law and Public Policy: Reconsidering an Uneasy Relationship’, EUI Working Paper 2007/30, at p. 4. 60 Albany, cited above note 59, paras 59–60. 61 Case C-415/93 Union royale belge des sociétés de football association ASBL v Jean-Marc Bosman [1995] ECR I-4921. 57 58

Ian S Forrester  455   whereby when a new club hired a young professional footballer, even if he was out of contract and was free to accept employment, the new club should pay a ‘training indemnity’ to the old club in recognition of its contribution to the development of the player in his early days. The rule was intended to reward, and actually did reward, grassroots efforts to develop talent and also to share revenues between wealthier and poorer clubs. The rule was said to be indistinctly applicable to domestic and international transfers. Mr. Bosman’s transfer from Liège to Dunkirk fell through because the Belgian football authorities did not supply promptly to the French football authorities the paperwork to allow him to play professionally in France. The question was whether the practice was to be examined under the rules of competition or of free movement of workers. The former presented the difficulty that was articulated in Albany. The latter presented the difficulty that Article 48 EC (now Article 45 TFEU) applied to public sector restrictions, not private sector ones. Article 48 EC had always been applied to governmental discrimination against foreigners. It had never been applied to private arrangements which regulated the structure of an employment market. The exceptions provided in Article 48 address public policy, public security and public health. It was difficult to see how these might be relevant to the problems of private employment, because Article 48, while relevant to government action, was ill-adapted to deal with the details of employment relationships. It was assumed that Article 85(1) would be more relevant in that football associations agreed between each other upon rules to govern the transfer of out-of-contract players. However, it was unclear what would be the consequences if collective bargaining agreements were to fall within the scope of Article 85(1). Should concertation between employers, or between unions, or between the two, concerning terms of employment be regarded as vulnerable under the competition rules? The case created huge media interest. There were numerous television crews swarming around the Court on the day of the oral argument, and a flood of extravagant reports about footballer slaves awaiting their freedom, or football clubs awaiting extinction. The football authorities contended that the transfer fee system applies to every professional player; if the formality of a transfer certificate (domestic or international, as the case may be) applied to all transfers, the obstacle to freedom would not be obvious (Mr. Bosman claimed that international transfers were more difficult than domestic transfers; UEFA denied this): those rules [regarding transfer fees] are likely to restrict the freedom of movement of players who wish to pursue their activity in another Member State by preventing or deterring them from leaving the clubs to which they belong even after the expiry of their contracts of employment with those clubs.62 Since they provide that a professional footballer may not pursue his activity with a new club established in another Member State unless it has paid his former club a transfer fee 62

Ibid, para 99.

456  Competition, Regulation and Public Policies agreed upon between the two clubs or determined in accordance with the regulations of the sporting associations, the said rules constitute an obstacle to freedom of movement for workers.63

The consequence of the judgment was that Article 48 applied to a private sector measure which imposed financial obstacles to taking up a profession or a job whether the job was taken in one’s own country or another country. This was a quite radical position, a surprise for most observers. although the rules … apply also to transfers between clubs belonging to different national associations within the same Member State and are similar to those governing transfers between clubs belonging to the same national association, they still directly affect players’ access to the employment market in other Member States and are thus capable of impeding freedom of movement for workers.64 Consequently, the transfer rules constitute an obstacle to freedom of movement for workers prohibited in principle by Article 48 of the Treaty.65

Thus, Article 48 could apply to the transfer rules. Did this mean they were absolutely prohibited? Or could they be justified, and if so, by reference to what concepts? The Court said an exception to Article 48 could be available if those rules pursued a legitimate aim compatible with the Treaty and were justified by pressing reasons of public interest. But even if that were so, application of those rules would still have to [respect the principle of proportionality by not being too burdensome]. 66

If the Court had stopped there, transfer rules could have been legal if they were economically justifiable, in the public interest, and proportionate. It would have been for national courts to decide in particular circumstances whether given rules or given applications of the rules were legitimate. But the Court did not stop there. It went on to discuss the detailed factual justifications offered for transfer fees, and condemned those justifications. In view of the considerable social importance of sporting activities and in particular football in the Community, the aims of maintaining a balance between clubs by preserving a certain degree of equality and uncertainty as to results and of encouraging the recruitment and training of young players must be accepted as legitimate.67

This must have been welcome reassurance. But The application of the transfer rules is not an adequate means of maintaining financial and competitive balance in the world of football. Those rules neither preclude the richest clubs from securing the services of the best players nor prevent the availability of financial resources from being a decisive factor in competitive sport, thus considerably altering the balance between clubs.68 Ibid, para 100. Ibid, para 103. 65 Ibid, para 104. 66 Ibid, para 104. 67 Ibid, para 106. 68 Ibid, para 107. 63 64

Ian S Forrester  457   Assuming this to be true, the virtue of transfer fees was not that they remove the benefits of wealth from wealthy clubs, but that smaller clubs were given financial benefits. Thus transfer payments from wealthy clubs to smaller clubs were, according to the football authorities, a valuable mechanism for redistribution of income within the sport. It must be accepted that the prospect of receiving transfer, development or training fees is indeed likely to encourage football clubs to seek new talent and train young players.69

This was a recognition of what could be a pressing reasons of public interest. But: because it is impossible to predict the sporting future of young players with any certainty and because only a limited number of such players go on to play professionally, those fees are by nature contingent and uncertain and are in any event unrelated to the actual cost borne by clubs of training both future professional players and those who will never play professionally. The prospect of receiving such fees cannot, therefore, be either a decisive factor in encouraging recruitment and training of young players or an adequate means of financing such activities, particularly in the case of smaller clubs.70

This paragraph seems unconvincing, though I am ready to confess the narrowmindedness of a disappointed advocate. Why should uncertainty of outcome be a bad thing for an economic organization, especially in the field of sport? However, if training young players may be a lawful justification for the transfer rules, why should it be unlawful because of the difficulty of matching the fee generated by one player’s transfer with the costs of training that player? Of course a club’s earnings in this manner were contingent and uncertain. But many clubs (like Brechin Rovers, Brønby and others) earned vital income in this manner when their young stars moved on to bigger clubs. The Court thus concluded: Article 48 of the Treaty precludes the application of rules laid down by sporting associations, under which a professional footballer who is a national of one Member State may not, on the expiry of his contract with a club, be employed by a club of another Member State unless the latter club has paid to the former club a transfer, training or development fee.71

The Court said that there was no need to consider the legality of the system under the competition rules for as long as the transfer system was illegal under a different part of the Treaty. The debate continued!

5. ‘It is not for you to decide’ The EU Courts’ jurisprudence and the Commission’s decision-making practice have also generated another group of public policy cases where private practices Ibid, para 108. Ibid, para 109. 71 Ibid, para 114. 69 70

458  Competition, Regulation and Public Policies are rejected under the competition rules, even though public policy arguments are invoked.

‘Safety would be at risk’ Hilti AG v Commission72 concerned the disposition of the appeal against the 6,000,000 ECU fine imposed in December 1987 on the largest producer of nail guns, nails, and cartridge strips in Europe. The proceedings before the Commission seem to have been rather bad-tempered, and the intensity of the debate before the Court seems also to have been particularly vigorous. The question was whether Hilti could properly forbid the use of rival nails on its guns, on the grounds, inter alia, of safety. The Court upheld the Commission’s finding that nail guns, nails, and cartridge strips constituted three different markets. Hilti had argued that there was only one market for fixing systems of all brands made by all manufacturers. The Court rejected this argument, stating that in the absence of other legislation all producers were free under Community law to make products designed for use with other producer’s devices, unless by so doing they violated a patent or other intellectual or industrial property right. If Hilti were correct, nail gun producers could exclude the use with their equipment of products other than those bearing their own trademark. Hilti’s arguments about the alleged incompatibility or inferiority of its competitors’ nails, which Hilti claimed were unsuitable for Hilti nail guns, were rejected. A company in a dominant position could not, on its own initiative, take measures to eliminate from the market products it considered dangerous or at last inferior in quality to its own. The Court noted that Hilti had not taken any legal steps to establish that its competitors’ products were dangerous when used with Hilti products. A company’s interpretation of national rules on a producer’s responsibility for its own products could not prevail over the Community’s competition rules. This is the line which has broadly been followed thereafter. It is not likely that an enforcer of the competition rules will refrain from applying the otherwise applicable rule by reference to private concerns about safety; but cases will arise where products manifestly present regulatory challenges and where it would be reckless to ignore these.

‘The publicly-conferred privilege is antiquated and discriminatory’ It regularly happens that privileges conferred by a public authority have an effect on competition and are challenged for that reason. One may contrast the 72

Case T-30/89 Hilti AG v Commission [1991] ECR II-1439.

Ian S Forrester  459   grant of exclusive access to a cemetery (Bodson v Pompes Funèbres)73 with the grant to a docking company of a compulsory exclusivity reinforced by a nationality clause. In Merci Conventionali Porto di Genova SpA v Siderurgica Gabriella SpA,74 an Italian steel company and the company which had been given the exclusive right to operate in the port of Genoa were in litigation over the charges for port services, which the customer regarded as excessively costly and inadequate. Article 110 of the Codice Navigazione provided that unloading operations in the port of Genoa must be reserved to docking companies whose workers were Italian nationals. The Genoa court referred a number of questions concerning the propriety under Community law of the state of affairs which prevailed in the port, and which apparently had the effect of obliging the unfortunate shipper to pay a lot of money for undesired services. The European Court observed that a port undertaking which had the exclusive right to perform port operations for a third party was indeed an undertaking granted exclusive rights by a Member State within the meaning of Article 90(1) EEC (now Article 106(1) TFEU). As a result, Member States could not enact or maintain in force any measure which was contrary to the rules of the Treaty. Article 48 EC prohibited a Member State from adopting provisions reserving to its own nationals the right to work in a national enterprise. As regards the exclusive rights given to certain enterprises at the port, the Court recalled that an enterprise which enjoyed a legal monopoly in a substantial part of the Common Market could be dominant, and had no difficulty in concluding that the market for unloading services at the port of Genoa was sufficiently large to constitute a substantial part of the Common Market. Merely creating a dominant position by granting exclusive rights was not in itself incompatible with Article 90. However, the result would be different when the enterprise endowed with these rights was led to exploit its dominant position in an abusive manner by the mere exercise of the rights conferred upon it, or where such rights were liable to create a situation where the enterprise would be tempted to abuse its position. The Court then observed that the national regulations in question were liable to encourage or to permit the payment for services not requested, the charging of disproportionately high prices, or the refusal to use modern technology in favour of slower and more costly techniques, or to grant price reductions unless at the same time charging other users more. The Court therefore held that the Member State, by adopting a rule such as was before the Court, had created a situation contrary to Article 90. Postal delivery services are another area of contention, where cost-efficiency and public service seem to be opposed.

Case 30/87 Corinne Bodson v Pompes Funebres [1998] ECR 2479. Case C-179/90 Merci Conventionali Porto di Genova SpA v Siderurgica Gabriella SpA [1991] ECR I-5889. 73 74

460  Competition, Regulation and Public Policies

‘The technology leads to unfair competition’ Italy v Commission75 was procedurally unique in that one Member State was appealing against a competition decision condemning the practices of a public entity in another Member State and even more intriguing in that the government of the other Member Stare intervened in support of the Commission. In 1982, the Commission had condemned British Telecom under Article 86 EEC, finding that it had abused its dominant position by restricting UK message-forwarding agencies from relaying telex and telephone messages originating outside the UK to other countries. British Telecom said that it applied the restrictions only because of international agreements to which it was party along with other national telecommunication authorities. The fact that British Telecom did not appeal against the Commission’s decision and that the British Government intervened in support of the Commission against Italy’s appeal suggests strongly that British Telecom was glad to be free of an inconvenient agreement. In those circumstances, it must have been evident that Italy’s prospects of success were not good. Italy argued that a public entity’s activities could be challenged under Article 90 EEC or, conceivably, Article 169 EEC, but not under Article 86 EEC. The Court replied that the management of a telecommunications network was a business activity subject to Article 86. The practices complained of were not compelled by the Government or by statute. The Court also rejected the suggestion that a national monopoly can maintain schemes intended to preserve its monopoly, since otherwise there could be a breach of Article 222 EEC (Article 345 TFEU), which states that national rules on property ownership shall be unaffected by the Treaty. Then the Italian Government asserted that the message-forwarding agencies in question were abusing the public network … by using special equipment, which with the aid of computer techniques enable a large number of messages to be forwarded in a very short time.76

The Court answered first that this had not been proved, and then added, more significantly: In the second place the employment of new technology which accelerates the transmission of messages constitutes technical progress in conformity with public interest and cannot be regarded per se as an abuse.77

In dealing with the interpretation of Article 90(2), the Court tartly pursued this point in quite capitalist terms: It should be observed that, whilst the speed of message-transmission made possible by technological advances undoubtedly leads to some decrease in revenue for BT, the presence in the United Kingdom of private forwarding agencies attracts to the British public network, as the applicant itself oversees, a certain volume of international 75 76 77

Case 41/83 Italy v Commission [1985] ECR 873. Ibid, para 24. Ibid, para 26.

Ian S Forrester  461   messages and the revenue which goes with it. The Italian Republic has totally failed to demonstrate that the results of the activities of those agencies in the United Kingdom were, taken as a whole, unfavourable to BT, or that the Commission’s censure of the schemes at issue put the performance of the particular tasks entrusted to BT in jeopardy from the economic point of view.78

Finally, the Court stated that the international agreements which bound British Telecom did not, in fact, proscribe the practices which British Telecom was reluctantly prohibiting on the basis of the agreements. The result of the appeal was not, in all the circumstances, very surprising. However, the judgment was important as the national operator maintained direct control of individual subscribers’ use of the network, and strongly discouraged the intervention of intermediaries who offered a better deal to subscribers. The case is authority for the proposition that avoiding the use of better technology is a bad reason for prohibiting the use of superior technology. The public policy seemed unpersuasive and the Court condemned it.

‘The need for numbers’ Decca Navigator System79 presented a complex and intriguing decision under both Article 86 EEC and Article 85 EEC in a field (maritime radio navigation systems) bordering on a public service. The facts, in addition to being dense and complicated, certainly encouraged the Commission to present the case as a populist defence of users against monopoly power which had been applied in a way which seems to have been almost callous. Racal Decca developed and operated a system known as the Decca Navigator System (DNS) based on transmission of signals from groups of land-based stations which are received by special receivers on ships or boats at sea. Racal Decca provided the signals in the UK and Denmark. It was also the principal manufacturer of receivers, which it hired out or, after 1983, sold, the proceeds being used to finance the operation of the system. With the arrival of competing suppliers (one a subsidiary of Philips) in the early 1980s, Racal Decca resorted to a combination of litigation, negotiation based on a dubious claim of copyright in information relating to DNS, and altering the frequency of signals (thereby apparently causing some navigation difficulties to boats using even Racal Decca’s own equipment). It succeeded in concluding agreements with two of the competing suppliers which essentially assigned the market for pleasure boats to them, reserving commercial shipping to Racal Decca; these were notified to the Commission, which also received complaints. The Commission found that Racal Decca was dominant in the separate markets for transmission of signals (despite arguments that there could not be a market for 78 79

Ibid, para 33. Commission Decision of 21 December 1988 – Decca Navigator System, 1989 OJ L43/27.

462  Competition, Regulation and Public Policies a service for which there was no distinct payment) and supply of the receivers. The abuses were the agreements with the two suppliers, which excluded them from the commercial market, and the changes in signals, which obstructed or coerced those competitors who were unwilling to enter into such agreements. However, the Commission seems to have made heavy weather in dealing with some of Racal Decca’s defences. Racal Decca said its service was costly to operate, and it needed a guaranteed source of revenue to finance it. The Commission replied righteously that: no undertaking has the right to ensure the continuation of its business by means which infringe ... competition law. Therefore, the criticized behaviour remains abusive even if there were no other alternatives to those of ceasing to supply and abandoning the market for DNS transmissions.80

It went on to find that alternative solutions were available, namely, either having the transmission chains taken over by the State and running them for the State on a service basis (as had been done in most countries), or, when the technology became available, equipping the receivers with renewable electronic devices which would have to be purchased from Racal Decca. However, one may feel that the idea that Article 86 might have the effect of compelling dominant enterprises to sell their business to the State is a little startling. Again, rather unpersuasive public service arguments were indignantly rejected.

The civil religion of parallel trade For a young competition lawyer of my generation in Europe, it was axiomatic that competition law favoured parallel trade, and that contractual hindrances to parallel trade were the easiest way to incur a fine. At a time when enforcement was rather passive and not very investigative, the existence of a contractual ban was simple to verify. Defences criticising free-riding or arguing that the ban had not been enforced were always unsuccessful. Fines rose to levels deemed high in those days (the fines in the Pioneer cases81 in 1979 were a watershed at some ten million dollars), and all sorts of products were targeted (clothes, alcohols, construction materials, tennis balls, cars). If enforcement policy were a tree, by far its biggest branch would be parallel trade. Academic criticism in the 1980s and 1990s of what I have called the ‘civil religion’ of parallel trade focussed more on the absence of effects and the textual absolute nature of the infringements than on whether consumers were really better off as a result. One can debate whether the policy was a diversion from better targets like cartels or state aids, or for its time made good sense. Today the Common Market is more economically diverse than ever, and price discrimination may be wise commercial policy. Indeed, without price discrimination, lower80 81

Ibid, para 113. Commission Decision of 14 December 1979 – Pioneer Hi-Fi Equipment, 1980 OJ L60/21.

Ian S Forrester  463   priced markets might never be served at all. It is probably desirable that the classic and unique feature of EU public policy as to competition should be reappraised. I suggest that the most consistent theme of public policy running through forty-five years of enforcement is market integration. The policy adopted by the Commission as an enforcement priority in Grundig v Consten82 and endorsed by the Court remained a very high priority for thirty-five years. Contractually banning parallel trade was close to an absolute offence so grave that not fuelling parallel trade was said to be an offence in Bayer/Adalat.83 Following the judgments in that case, and following the embracing of the relevance of effects and consumer welfare, the preoccupation with helping parallel traders has diminished. The contending policies (common sense versus the civil religion) were examined in two pharmaceutical cases referred from Greece, where the volumes exported had reached astonishing levels. (Over 80% of some medicines delivered to patients and pharmacies in the UK had been put on the market in Greece at the very low Greek price and immediately exported to UK wholesalers.) Ever larger volumes were ordered by Greek wholesalers who wanted to supply wholesalers in Germany or the UK, where prices were much higher. The core problem was how to decide the volumes of product which should be delivered to exporting wholesalers. It was agreed that there were no contractual prohibitions. It seemed strange to imagine that producers of medicines must supply the North Sea countries at the price in one of the weakest Mediterranean economies. Yet there is a rich volume of precedent blessing the activity of parallel trade. The judges addressed the gap that the Bayer/Adalat judgment had left open: could unilateral volume reductions by a dominant player constitute an abuse? Which policy should prevail? In Lelos,84 the ECJ broadly confirmed the legitimacy of dominant pharmaceutical companies taking steps to protect their commercial interests even if this involves limiting parallel trade. While the ECJ rejected some of the technical and economic arguments of the pharmaceutical industry as to the role of parallel trade and the worthiness of its protection under EC competition law, the judgment confirms that dominant pharmaceutical companies can refuse to supply wholesalers with ‘significant quantities of products that are essentially destined for parallel export’. The factual background to the case was as follows. In 2000 and 2001, complaints were lodged with the Hellenic Competition Commission (HCC) against GlaxoSmithKline (GSK) by a number of wholesalers, alleging that by limiting supplies of three drugs from its Greek subsidiary, GSK was abusing its dominant position. The wholesalers in question had been addressing ever-larger orders of these three medicines to GSK, mainly for export, to exploit the price differentials in prescription medicines between EU Member States. In 2000, GSK took the Consten and Grundig, cited above note 27. Case T-41/96 Bayer AG v Commission [2000] ECR II-3383, upheld on appeal: Joined Cases C-2/01 P and C-3/01 P Bundesverband der Arzneimittef-Importeure eV and Commission v Bayer AG [2004] ECR I-23. 84 Joined Cases C-468/06 to C-478/06 Sot. Lélos kai Sia EE v GlaxoSmithKline [2008] ECR I-7139. 82 83

464  Competition, Regulation and Public Policies decision to suspend supplies of these medicines to wholesalers for a few weeks to ensure that pharmacy supplies were restored. Subsequently, it decided to supply wholesalers with quantities corresponding to Greek annual consumption plus a safety margin (amounting up to 25% of annual Greek consumption). In 2003, the HCC referred the case to the ECJ,85 asking whether GSK’s refusal to supply, in unlimited quantities, all the orders placed by wholesalers, could constitute an abuse of dominant position. The Court had two astonishingly divergent opinions from Advocate General Jacobs and Advocate General Colomer, so the outcome was by no means certain. In its judgment, the ECJ began by confirming that it may be an abuse of a dominant position for a dominant pharmaceutical company to prevent all parallel trade. This was not a controversial proposition as it was agreed that GSK had never prohibited all exports, but on the other hand if limitless volumes have to be delivered to Greek wholesalers at the Greek price, great financial and commercial disruption would be caused. The ECJ considered that parallel trade may in principle open up an alternative source of supply to buyers, and may bring some benefits to the final consumer. But a dominant pharmaceutical company must be able to defend its own commercial interests by adopting reasonable and proportionate measures: the Community rules on competition are also incapable of being interpreted in such a way that, in order to defend its own commercial interests, the only choice left for a pharmaceuticals company would be not to place its medicines on the market at all in a Member State where the price of those products are set at a relatively low level.86

The ECJ continued: Thus, although a pharmaceuticals company in a dominant position, in a Member State where prices are relatively low, cannot be allowed to cease to honour the ordinary orders of an existing customer for the sole reason that that customer, in addition to supplying the market in that Member State, exports part of the quantities ordered to other Member States with higher prices, it is none the less permissible for that company to counter in a reasonable and proportionate way the threat to its own commercial interests potentially posed by the activities of an undertaking which wishes to be supplied in the first Member State with significant quantities of products that are essentially destined for parallel export.87

Thus a dominant pharmaceutical company can protect its own commercial interests when confronted with orders that are out of the ordinary, in terms of quantity in light of both the previous business relations with the wholesaler concerned and the requirements of the market in the Member State concerned. It is for the national court to decide whether the orders placed by the wholesalers were ordinary and whether GSK’s decision to limit supplies was proportionate to the need to protect its own commercial interests. 85 Case C-53/03 Synetairismos Farmakopoion Aitolias & Akarnanias (Syfait) and Others v GlaxoSmithKline plc and GlaxoSmithKline AEVE [2005] ECR I-4609. 86 Lélos, cited above note 82, para 68. 87 Ibid, para 71.

Ian S Forrester  465   The ECJ thus confirmed that dominant pharmaceutical companies are entitled to adopt measures responding to – but not prohibiting or eliminating – parallel trade, and that such measures are not contrary to the EC competition rules. This was not an easy or obvious battle. I cannot be optimistic that these disputes will end with the Lelos case; hence the recommendation that the civil religion be not immune from doctrinal reflection.

6. Other public policy considerations: the environment and others The Treaty’s competition rules apply to all industrial sectors. The TFEU provides only for a limited application of the competition rules in the agricultural sector, and there have been a few cases on food products, notably French spirits. However, the secondary EU legislation makes competition law applicable to agriculture with certain exemptions to Article 101 TFEU in order to ensure a proper functioning of the Common Agricultural Policy and its national organisations of agricultural markets.88 The TFEU also provides for exemptions for nuclear energy and military equipment.89 There are also certain exceptions for certain agreements on rail, road and inland waterway transport aiming at technical improvements or achieving technical cooperation. There used to be a special block exemption for liner conference agreements, which in effect allowed price-fixing, market-sharing and revenuepooling cartels, as liner conferences would bring stability to the market.90 The scope of this exemption was narrow, however, and in a series of decisions the Commission condemned market-sharing and price-fixing arrangements which were held to fall outside the block exemption.91 The block exemption for liner shipping was repealed in 2006 (with effect from 2008). There are currently no specific EU rules or guidance on how environmental benefits associated with a horizontal agreement between competitors may take the agreement outside the prohibition on anticompetitive agreements,92 but such 88 Council Regulation 1184/2006 of 24 July 2006 applying certain rules of competition to the production of, and trade in, agricultural products, 2006 OJ L214/7. 89 EU Member States may not adopt regulations that would deprive EU competition rules of their effectiveness. Thus, Member States (legislative, executive, regulatory authorities) may not adopt regulations that would encourage or force undertakings to violate EU competition law and if they do, regulations should be disapplied. A Member State, for instance, may not approve any price-fixing arrangements between companies contrary to the EU competition rules. 90 Council Regulation 4056/86 of 22 December 1986 laying down detailed rules for the application of Articles 85 and 86 of the Treaty to maritime transport, 1986 OJ L378/4. 91 See, eg, Case T-86/95 Compagnie Générale Maritime and Others v Commission [2002] ECR II1011; Case T-395/94 Atlantic Container Line AB v Commission [2002] ECR II-875. 92 The Commission provided some guidance on the assessment of environmental agreements in its 2001 Horizontal Guidelines (Commission Notice – Guidelines on the applicability of Article 81 to horizontal co-operation agreements, 2001 OJ C3/2). However, the revised Horizontal Guidelines published in 2011 (Communication from the Commission – Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements,

466  Competition, Regulation and Public Policies arguments are often made. In practice, the Commission mentioned environmental benefits in its exemption decisions, to the extent that those benefits could be subsumed under the general conditions of Article 101(3): to benefit from the exemption under that provision, the agreement must ‘contribute to improving the production or distribution of goods or advance technological and economic progress’ and allow ‘consumers a fair share of the resulting benefit’. Thus, for example, in KSB/Goulds/Lowara/ITT,93 which concerned the joint research and development of a more environment-friendly pump that would lead to energy savings, the Commission referred to the environmental benefits as an ‘improvement in operating characteristics’ and stressed the fact that consumers would buy them at the same price as conventional pumps. In Assurpol,94 the Commission exempted a co-operation agreement between insurance and reinsurance firms aiming at improving the knowledge of risks, at creating financial capacity and at developing technical expertise in insuring environmental damage risks, reasoning that due to these benefits the agreement could be seen as contributing to ‘technical and economic progress’. It is difficult to know if the ‘green’ credentials of the deal really affected the outcome. In another case, the Commission indicated its readiness to take a favourable view of a voluntary commitment sponsored by a trade association that aimed at reducing energy consumption by televisions and video recorders in standby mode.95 One has to doubt whether environmental factors were in these cases more than background context and decorative arguments. In DSD,96 the Commission exempted certain exclusivity clauses in a country-wide system for the collection and recovery of sales packaging that met the requirements of German and Community packaging and environmental legislation. While it admitted that the system in question was consistent with the objectives of German and EU environmental legislation, this did not stop it from considering the exclusivity clauses anticompetitive and from exempting them only under specific obligations. The exemption was granted supposedly because the agreement generated environmental benefits that were considered to contribute to technical and economic progress. There is an interesting exception to my thesis that the environment does not really shape competition policy. Indeed, it was a coordinated discontinuation of production (echoes of crisis cartels!). In CECED,97 the Commission came close to treating environmental interests as a core argument for granting an exemption to a restrictive agreement, approving an agreement to stop production with a view to improving the environmental performance of products. The participants in the agreement, nearly all the European producers and importers of domestic washing 2011 OJ C11/1), no longer include a section on environmental agreements. The Commission’s approach is that such agreements should be analysed in the framework of standardisation agreements. Commission Decision of 12 December 1990 – KSB/Goulds/Lowara/ITT, 1991 OJ L19/25, para 27. Commission Decision of 14 January 1992 – Assurpol, 1992 OJ L37/16, para 38. 95 EACEM, 1998 OJ C12/2 (Article 19(3) Notice), in particular paras 11 and 12. 96 Commission Decision of 17 September 2001 – DSD, 2001 OJ L319/1, paras 142–146. 97 Commission Decision of 24 January 1999 – CECED, 2000 OJ L187/47. 93

94

Ian S Forrester  467   machines, agreed to stop producing or importing into the EU the least energyefficient machines in order to reduce the energy consumption of such appliances and thereby reduce pollutant emissions from power generation. The Commission exempted the agreement reasoning that [a]lthough participants restrict their freedom to manufacture and market certain types of washing machine, thereby restricting competition within the meaning of [Article 101(1) TFEU], the agreement fulfils the conditions for exemption under [Article 101(3) TFEU]: it will bring advantages and considerable savings for consumers, in particular by reducing pollutant emissions from electricity generation. The Commission decision to exempt the agreement takes account of this positive contribution to the EU’s environmental objectives, for the benefit of present and future generations.98

Other Commission decisions show that the Commission is sceptical about justifying hard-core competition restraints such as price-fixing or output limitation, notwithstanding possible environmental benefits.99 One striking example is VOTOB,100 where the Commission condemned a Dutch industry association representing six companies offering bulk liquid tank storage facilities for agreeing on a uniform environmental surcharge to customers. VOTOB had entered a covenant with the Dutch government accepting important reductions in the firms’ emissions of certain pollutants. To finance their efforts, the firms jointly decided to ask their clients to pay an environmental surcharge, and agreed on the surcharge level. The Commission reasoned that the agreement on the fixed fee harmonised the costs and thus excluded competition on an important price component and was not necessary to achieve the environmental benefits. It is difficult to reconcile the VOTOB and CECED approaches. One of the difficulties in analysing the miscellany of arguments and rationales invoked in competition cases is knowing whether the invoked factors were taken seriously and might have made a difference to the outcome. As we can see in connection with the environment, it is unclear whether possible benefits for the environment have had the effect of changing the otherwise applicable rules of competition, or their application. By contrast, there can be no such hesitation with respect to market integration. The whole point of these cases was indeed to dispense with the otherwise necessary consideration of the economic merits of the measure at stake.

7. Conclusions Many public policy arguments are made in the course of competition cases in Brussels and in Luxembourg. There are many cases where public policy See XXXth Report on Competition Policy (2000), paras 96–97. See Commission Decision of 21 December 1988 – Ansac, 1990 OJ L152/54, para 23; see also 2001 Horizontal Guidelines, cited above, para 188. 100 See XXIInd Report on Competition Policy (1992), paras 177 et seq. 98 99

468  Competition, Regulation and Public Policies is invoked as a ground for a tolerant application of the competition rules; the attempt is usually unsuccessful. Most appear to have failed, and even when they have been successfully invoked, the decisions often give the impression that the public policy consideration has been decorative rather than substantive. As to crisis cartels and coordinated capacity reductions, modern practice probably indicates that they are rarely capable of being defended but the door is not completely closed. In many cases, arguments of public health and safety have been rejected. In considering price-setting by public authorities, the ECJ has been notably deferential to state discretion, even if private interests seemed very close behind. Governmental policy in setting prices has been accorded tolerant noninterference. In connection with the environment, I have difficulty in deciding whether we are seeing competition policy being applied intelligently or whether environmental policy is being skilfully balanced against competition policy. One big exception to the inconsistent fortunes of policy arguments in competition law is market integration, which has been a potent and usually consistent force which has overridden normal competition law analysis. It is time to refresh that doctrine, reflecting on whether it needs adaptation for this millennium. The merits of the civil religion of market integration as a driver of competition policy deserve to be reconsidered.

Philip Lowe*

Conclusions

1. In the context of the latest major economic recession, competition authorities do not appear to have given in to any pressure to relax enforcement of antitrust and merger rules in order to help companies stay in the market and preserve jobs. In Europe, some evidence suggests relatively low rates of exit of large firms with much higher rates for smaller ones. While the recession tends to reveal more starkly the competitive disadvantages of weaker firms, the tendency is for the scarce resources of private and public budgets to be used for short-term measures to increase ‘short-term profitability’ and without provision for investment in the structural reforms and innovation which will improve long-term competitiveness. State aid control can usefully guide investment into these areas. The application of the ‘as efficient competitor’ test,1 accompanied by a convincing theory of harm to consumers, can be a bulwark against protecting competitors rather than competition. In the banking sector, concerns about systemic risk and long-term financial stability dominate over the need for bank restructuring, and the weakness of individual banks and low levels of interbank lending exacerbate the recession. 2. The focus of competition law on preventing or sanctioning a weakening of competition related only to market power makes it less relevant and visible at a political level at a time when the principal concern of public policies is to stimulate economic recovery, including measures to promote competition and innovation, for example by encouraging new entrants. An excessively narrow definition of the remit of competition authorities, limited to the control of market power, could lead to a reduction in their influence, whereas their role in advocating pro-competitive solutions to meet market failures in other areas of public policy could be crucial to identifying ways out of economic and financial crisis. 3. Competition policy within any jurisdiction also needs to take into account the external trade environment, in particular because a narrow assessment of an agreement or merger may ignore their potential negative impact on security of supply or import dependence. 4. The aim of competition policy, including but not limited to law enforcement, is generally accepted as being: to make markets work for the benefit of consumers. * Non-Executive Director, UK Competition & Markets Authority. At the time of writing, DirectorGeneral, DG Energie, European Commission. 1 The ‘as efficient competitor’ test was endorsed in the Commission’s Guidance Paper on exclusionary abuses and has likewise been endorsed by the ECJ. See, eg, Case C-208/08 P Deutsche Telekom AG v Commission [2010] ECR I-9555, para 177.

470  Competition, Regulation and Public Policies However, it shares this objective with consumer protection agencies as well as sectoral regulators in fields such as transport, telecommunications, energy and others. Sanctions or remedies imposed in an individual competition case or market investigation can have a positive game-changing effect on the way markets work. However, regulation, and the ex ante or ex post action of regulators, may sometimes produce a more timely and/or effective solution to a competition problem than competition law can do. Equally, a competition authority may concentrate on solving what is effectively a short-term problem on the market rather than concentrating on longer term improvements in market functioning which a regulator may have a greater focus on. On the other hand, regulators may frequently satisfy immediate political and consumer concerns by imposing short-term controls on price and output, which can harm competition in the long term. The general message here is that there needs to be more constructive complicity between competition authorities and regulators in identifying best solutions, with humility on both sides. It does not follow from this that competition law enforcement and regulatory powers need to be brought together into one authority. In some jurisdictions, this has led to some confusion in aims and analysis. 5. This confusion is not helped by a lack of clear definition of the activities of certain regulatory bodies. Public policy aims, such as higher environmental standards, are for the benefit of society as a whole under the broad heading of social welfare, and they should not be attributed solely to the objective of consumer welfare. For example, sometimes the strict definition of consumer welfare can be problematic because it may not capture broader concerns that consumers regard as important, or because it may not account for consumers not yet born. In recent years, governments have also tended in recent years to assign to sectoral regulators tasks which extend their role into areas of public policy such as safety. This adds to the difficulty of establishing clear and precise enforcement priorities and clearly allocating tasks between different agencies. 6. The work of sectoral regulators overlaps directly with competition authorities when regulators deal with market power. In the United States, the Trinko case2 calls in some circumstances for a pre-emption of antitrust law where regulation already deals with the conduct in question. However in the European Union, the competition rules can be and have been applied successfully to strengthen enforcement with respect to abuses of dominance such as in the case of margin squeeze. 7. In addition, the responsibilities of sectoral regulators extend to issues such as interoperability and interconnection, which can create and/or improve the functioning of markets. When addressing such issues, regulators can contribute significantly, as already intimated, to improving the overall competitive environment through remedies which may be difficult to achieve by means of individual competition law decisions. 2

Verizon Communications v Law Offices of Curtis V Trinko, LLP, 540 US 398 (2004).

Philip Lowe 

471  

8. The European Commission is also in a position to use law and regulation related to European market integration to improve competitive conditions, for example with respect to market entry. 9. The extent to which competition law decisions – in particular, under Article 101(3) TFEU – need to take account of, integrate or subsume, other public policy objectives still gives rise to some debate. In general, however, there are strong arguments in terms of clarity and operationality which suggest it is better to pursue non-competition public policy objectives through other distinct regulatory instruments rather than allowing private parties themselves to invoke such public policies in defence of a priori anticompetitive agreements. In any event, competition authorities and judges tend inevitably to take account of the impact of all public policies on the competitive environment surrounding an individual case. 10. Progress towards the achievement of the liberalisation objectives set out in sectoral regulation remains partial in some areas. Some significant advances have been made in air transport, telecoms and energy, and there have been direct benefits for consumers. However, much remains to be done, whether through application of the competition and state aid rules, or through sectoral legislation. 11. A key area of interaction between competition law enforcement and other public policies is innovation. The protection of intellectual property rights is essential to provide incentives for innovation. However the traditional, somewhat negative approach of antitrust enforcers towards instruments such as patent pools may not adequately reflect the fact that cross-licencing and other related practices may sometimes help build contestability vis-à-vis existing dominant players who have so-called essential patents. Refinements to the European Commission’s approach to patent pools and other technology licensing issues are on the way (see below). 12. As a general rule, high technology markets are dynamic, fast-moving and innovative. They should not normally give rise to competition cases. But companies can abuse their market power by, for example, using their intellectual property rights to act as a gatekeeper and frustrate follow-on innovation, or by exercising undue influence on the setting of standards, or failing to respect FRAND principles. 13. In a context of the increasing instrumentalisation by market players of IP litigation as an instrument to promote their commercial interests, antitrust enforcement must clearly distinguish between competitive ‘noise’ and genuine issues of competition concern. This is essential if they are to assign the correct level of priority to IP-related issues in their overall enforcement effort. At the same time, their assessment needs to take full account of conflicts and problems which can more appropriately be dealt with by patent authorities and standard-setting organisations. In principle, the activities of all these sets of bodies – patent authorities, standard-setting organisations and competition authorities – can be mutually enforcing, in particular insofar as the prospect of

472  Competition, Regulation and Public Policies antitrust enforcement can act as an ultimate deterrent against anticompetitive exploitation of IP rights. 14. Recent EU experience suggests that the Guidelines on technology transfer agreements provide a basis for the Commission to discourage certain practices and concentrate its case-related enforcement work on key precedent-setting cases with real economic impact. However, there had been a lack of empirical evidence on the precise impact of the Guidelines on the market and on commercial practices. A recent review process has shed some light on these practices and has yielded a revised block exemption and revised Guidelines.3 15. Courts have the ultimate responsibility, in interpreting either constitutional law or measures adopted by the legislator, for dealing with the interplay of regulations relating to competition and with those implementing other public policies. In the US, the Trinko case and a number of sectoral exemptions clearly circumscribe the field of application of antitrust legislation. In contrast, in the EU, both the European Courts and the national courts have a more complex task which implicates the primacy of EU law, the concurrent application of competition and sectoral rules, and the appropriate weight to be given to legislation and action under the heading of the different policy objectives set out in the EU Treaties. In this respect, the following measures offer the potential to strengthen the effectiveness of the application of EU law: – the application of Articles 101 and 102 TFEU in conjunction with other provisions of the Treaty; – the application of the state aid rules, whether directly or on a decentralised basis; – the duty of the Member States to assist in the application of the competition rules; and – the potential for national courts to disapply national laws that contradict primary or secondary EU law. 16. If competition and related public policies are to be endowed with legitimacy, the European Courts and the national courts must be in a position to assess whether competition law and regulation have the potential, when applied in specific cases, to meet the objectives set out for them by the legislator. They can only do this if they carry out a thorough review of the substance of a case; in some instances, this requires courts to apply more rigour than they have sometimes thus far been prepared to do. 17. Movement of the courts in this direction would place increased responsibility on competition and sectoral authorities at EU level to tell the courts a convincing economic story about the impacts of a merger, an agreement or other conduct on competition. Both competition authorities and courts need to take explicit account of the inevitable time lag between policy developments and court judgments on such developments, and the evolution in the case law 3 The new rules on licensing will take effect on 1 May 2014. For details, see http://ec.europa.eu/ competition/antitrust/legislation/transfer.html.

Philip Lowe 

473  

which that implies. At EU level, the Commission has a special responsibility, as it were, to provide the European Courts with a coherent interpretation of the interplay between competition and sectoral legislation. The principle of transparency, including access to documents, has recently drawn the close attention of the European Courts. Access to documents for citizens, under appropriate conditions, is an essential instrument to ensure that the Commission’s administration of public policies is transparent and effective. At the same time, giving access to leniency statements in cartel cases can chill the process of evidence collection and can thus frustrate current and future investigations. There remains a strong case for developing and refining legislation at EU level in order to prevent commercial litigators from abusing rights of the access granted to all citizens. In June of 2013, the Commission proposed new measures that may have an impact on the degree and conditions of ‘discoverability’ of leniency statements.4 However, some issues linked to the Transparency Directive will likely continue to evolve irrespective of the shape the EU’s private enforcement initiative ultimately takes.

4 For discussion of the Commission’s proposals, see Philip Lowe and Mel Marquis (eds), European Competition Law Annual 2011: Integrating Public and Private Enforcement – Implications for Courts and Agencies (Hart Publishing, 2014).

References

I. Publication, speeches and presentations A Ackerman, Bruce and Richard Stewart, ‘Reforming Environmental Law’, 37 Stanford Law Review 1333 (1985) Aghion, Philippe, Julian Boulanger and Elie Cohen, ‘Rethinking Industrial Policy’, Bruegel Policy Brief No. 2011/04. Aghion, Philippe et al., ‘Competition and Innovation – An Inverted U Relationship’, 120 Quarterly Journal of Economics 701 (2005) Almunia, Joaquín, ‘Antitrust enforcement: Challenges old and new’, speech, St. Gallen, 8 June 2012 Almunia, Joaquín, ‘EU competition policy and innovation’, SPEECH/13/697 of 13 September 2013¸ Florence Amato, Giuliano, Antitrust and the Bounds of Power, Hart Publishing, 1997 Anderson, Robert D., ‘The Interface Between Competition Policy and Intellectual Property in the Context of the International Trading System’, 1 Journal of International Economic Law 655 (1998) Anderson, Robert D., ‘Systemic Implications of Deeper TransAtlantic Convergence in Competition/Antitrust Policy’, in Simon Evenett and Robert Stern, eds., Systemic Implications of Transatlantic Regulatory Cooperation and Competition, World Scientific Publishing, 2011, chapter 7 Anderson, Robert D. and Simon Evenett, ‘Incorporating Competition Elements into Regional Trade Agreements: Characterization and Empirical Analysis’, 2006, http://www.evenett.com/research/workingpapers/CompPrincInRTAs. pdf Anderson, Robert D. and Alberto Heimler, ‘Abuse of Dominant Position: Enforcement Issues and Approaches for Developing Countries’, in Vinod Dhall, ed., Competition Law Today: Concepts, Issues and the Law in Practice, Oxford University Press, 2007, chapter 2 Anderson, Robert D. and Alberto Heimler, ‘What has Competition Done for Europe? An Inter-Disciplinary Answer’, Aussenwirtschaft (the Swiss Review of International Economic Relations), 62. Jahrgang (2007), Heft IV, pp. 419 et seq.

476  References Anderson, Robert D., Abraham Hollander, Joseph Monteiro and W. T. Stanbury, ‘Competition Policy and Regulatory Reform in Canada, 1986–1997’, 13 Review of Industrial Organization 177 (1998) Anderson, Robert D. and Peter Holmes, ‘Competition Policy and the Future of the Multilateral Trading System’, 5 Journal of International Economic Law 531 (2002) Anderson, Robert D. and Frédéric Jenny, ‘Competition Policy, Economic Development and the Role of a Possible Multilateral Framework on Competition Policy: Insights from the WTO Working Group on Trade and Competition Policy’, in Erlinda Medalla, ed., Competition Policy in East Asia, Routledge, 2005, chapter 4 Anderson, Robert D., William E. Kovacic and Anna Caroline Müller, ‘Ensuring integrity and competition in public procurement markets: a dual challenge for good governance’, in Sue Arrowsmith and Robert D. Anderson, eds., The WTO Regime on Government Procurement: Challenge and Reform, Cambridge University Press, 2011, chapter 22 Andrews, Philip and Paul Gorecki, ‘The Panda Waste case: Conflicts between National Competition Law and Regulation’, 6 European Competition Journal 543 (2010) Arnason, Ragnar, ‘A Review of International Experience with ITQs’, CEMARE Report No. 58 (2002) Arrow, Kenneth, ‘Economic Welfare and the Allocation of Resources for Invention’ in Various Authors (National Bureau of Economic Research), The Rate and Direction of Inventive Activity: Economic and Social Factors, 1962, pp. 609 et seq., http://www.nber.org/chapters/c2144 B Baccaro, Vincenzo, ‘Collecte et recyclage des batteries usagées en Italie: droits exclusives, exigences environnementales et droit de la concurrence’, (2001/1) Competition Policy Newsletter 39 Baldwin, Robert, Martin Cave and Martin Lodge Understanding Regulation: Theory, Strategy and Practice, 2nd ed., Oxford University Press, 2012 Barazza, Stefano, ‘Post Danmark: the CJEU calls for an effect-based assessment of pricing policies’, 5 Journal of European Competition Law and Practice 466 (2012) Barbier de la Serre, Eric, ‘L’assistance judiciaire américaine au soutien d’une plainte en droit communautaire de la concurrence’, (2005/1) Revue de Droit des Affaires Internationales 35 (2005) Barbier de la Serre, Eric, ‘Beyond effectiveness and uniformity: The Rise of Consistency’, Concurrences n° 2-2012, pp. 26 et seq.

References  

477  

Basedow, Jürgen, Stéphanie Francq and Laurence Idot, eds., International Antitrust Litigation: Conflict of Laws and Coordination, Hart Publishing, 2012. Baumol, William and Gregory Sidak, Toward competition in local telephony, MIT Press, 1994 Baye, Michael and Joshua Wright, ‘Is Antitrust Too Complicated for Generalist Judges? The Impact of Economic Complexity and Judicial Training on Appeals’, 54 Journal of Law and Economics 1 (2011) Bekkers, Rudi and Joel West, ‘Standards, Patents and Mobile Phones: Lessons from ETSI’s Handling of UMTS’, 7 Journal of IT Standards and Standardization Research 13 (2009) Beato, Paulina and Jean-Jacques Laffont, ‘Competition in public utilities in developing countries’, Inter-American Developing Bank Washington, D.C., Sustainable Developing Department, Technical Papers Series, 2002, http:// cdi.mecon.gov.ar/biblio/docelec/MU2007.pdf Bergqvist, Christian, ‘Final curtain or another around on Post Danmark?’, 34 European Competition Law Review 287 (2013) Bessen, James and Eric Maskin, ‘Sequential Innovation, Patents and Imitation’, 40 RAND Journal of Economics 611 (2009) Bhattacharjea, Aditya, ‘Export Cartels: A Developing Country Perspective,’ Working Paper No. 120, Centre for Development Economics, Delhi School of Economics, 2004, http://www.cdedse.org/pdf/work120.pdf Bork, Robert, The Antitrust Paradox, Basic Books, 1978 Bork, Robert and Ward Bowman, Jr., ‘The Crisis in Antitrust Law’, 65 Columbia Law Review 363 (1965) Bontinck, Gaëlle and Tom Snels, ‘Business Secrets and Confidential Information in Proceedings pursuant to Articles 101 and 102 TFEU’, (2011/3) European Journal of Consumer Law 547 (2011) Bouckaert, Jan, Theon van Dijk and Frank Verboven, ‘Access regulation, competition, and broadband penetration: An international study’, 43 Telecommunications Policy 661 (2010) Bourgeois, Jacques and Denis Waelbroeck, Ten years of effects-based approach in EU competition law: State of play and perspectives, Brussels, Bruylant, 2012 Bourreau, Marc, Carlo Cambini and Steffen Hoernig, ‘Ex ante regulation and coinvestment in the transition to next generation access’, 36 Telecommunications Policy 399 (2012) Brouwer, Onno and Daniel Mes, ‘The Tide of Transparency has reached the Antitrust Shores: the Use of Regulation 1049/2001 in EU Antitrust (Related) Proceedings’, in Carl Baudenbacher, Claus Gulmann and Koen Lenaerts, eds., Liber Amicorum en l’honneur de Bo Vesterdorf, Bruylant, 2007, pp. 451 et seq.

478  References Buschle, Dirk, ‘Unbundling of State-Owned Transmission System Operators – Effective Remedy or Eyewash?’, 1 European Networks Law And Regulation Quarterly 49 (2013) C Calabresi, Guido and Douglas Melamed, ‘Property Rules, Liability Rules, and Inalienability: One View of the Cathedral’, 85 Harvard Law Review 1089 (1972) Cardoso, Ricardo, Sandra Kijewski, Oliver Koch, Patrick Lindberg and Károly Nagy, ‘The Commission’s GDF and E.ON Gas decisions concerning longterm capacity bookings: Use of own infrastructure as possible abuse under Article 102 TFEU’, (2010/3) Competition Policy Newsletter 8 Caranta, Roberto, “Tutela giurisdizionale effettiva delle situazioni giuridiche soggettive di origine comunitaria and incisività del sindacato del giudice nazionale (Kontrolldichte)”, in Rivista italiana diritto pubblico comunitario, 2/1999, pp. 503 et seq. Cass, Ronald and Keith Hylton, Laws of Creation – Property Rights in The World of Ideas, Harvard University Press, 2013 Cave, Martin, ‘Encouraging infrastructure competition via a ladder of investment in telecommunications’ 30 Telecommunications Policy 223 (2006) Chang, Ha-Joon, 23 Things They Don’t Tell You About Capitalism, Bloomsbury Press, 2010 Chieppa, Roberto, “Jurisdictional Control over the Decisions of the Antitrust Authorities”, in Enrico Adiano Raffaelli, ed., Antitrust between EC law and National Law, Giuffrè-Bruylant, 2005, pp. 103 et seq. Choumelova, Dessislava and Juan Delgado, ‘Monitoring competition in the telecommunications sector: European Commission Sector Enquiries’, in Pierre Buiges and Patrick Rey, eds., The Economics of Antitrust and Regulation in Telecommunications, Edward Elgar, 2004, pp. 269 et seq. Cintioli, Fabio, ‘L’art. 41 della Costituzione tra il paradosso della libertà di concorrenza e il “diritto della crisi”’, Diritto e società, 3-4/2009, pp. 373 et seq. Connor, John, ‘Price Effects of International Cartels in Markets for Primary Products’, paper presented at the Symposium on Trade in Primary Product Markets and Competition Policy, World Trade Organization, Geneva, 22 September 2011 Cooper, James and William Kovacic, ‘U.S. Convergence with International Competition Norms: Antitrust Law and Public Restraints on Competition’, 90 Boston University Law Review 1555 (2010)

References  

479  

Cumming, George and Mirjam Freudenthal, Civil Procedure in EU Competition Cases before the English and Dutch Courts, Kluwer Law International, 2010 D Davies, Gareth, ‘Democracy and Legitimacy in the Shadow of Purposive Competence’, European Law Journal, DOI: 10.1111/eulj.12079 (2013) De Stefano, Gianni, ‘Access of Damage Claimants to Evidence Arising out of EU Cartel Investigations: A Fast-evolving Scenario’, 3 Global Competition Litigation Review 95 (2012) Denicolò, Vincenzo, ‘Two-stage patent races and patent policy’, 33 Rand Journal of Economics 488 (2000) Dolmans, Maurits, ‘A Tale of Two Tragedies – A Plea for Open Standards, and Some Comments on the RAND Report, Open Standards & Antitrust’, Concurrences no. 1, 2010 Ducrey, Patrik, ‘The Agreement between Switzerland and the EU Concerning Cooperation in the Application of their Competition Laws’, 4 Journal of European Competition Law and Practice 437 (2013) E Edlin, Aaron and Joseph Farrell, ‘Freedom to Trade and the Competitive Process’, NBER Working Paper No. 16818 (2011) Ehlermann, Claus-Dieter Ehlermann, ‘Reflections on a European Cartel Office’, 32 Common Market Law Review 471 (1995) Ehlermann, Claus-Dieter and Laraine Laudati, eds., European Competition Law Annual 1997: The Objectives of Competition Policy, Hart Publishing 1998 Ehlermann, Claus-Dieter and Louisa Gosling, eds., European Competition Law Annual 1998: Regulating Communications Markets, Hart Publishing, 2000 Ehlermann, Claus-Dieter and Isabela Atanasiu, eds., European Competition Law Annual 2004: the Relationship between Competition Law and (Liberal) Professions, Hart Publishing, 2005 Ehlermann, Claus-Dieter and Isabela Atanasiu, eds., European Competition Law Annual 2005: The Interaction between Competition Law and Intellectual Property Law, Hart Publishing, 2007 Ehlermann, Claus-Dieter and Mel Marquis, eds., European Competition Law Annual 2008: Antitrust Settlements, Hart Publishing, 2010 Evenett, Simon J., Margaret Levenstein and Valerie Suslow, ‘International Cartel Enforcement: Lessons from the 1990s’, 24 The World Economy 1221 (2001).

480  References F Fisher, Franklin, ‘Economic Analysis and “Bright Line” Tests’, 4 Journal of Competition Law and Economics 129 (2007) Forrester, Ian and Christopher Norall, ‘Annual Survey of Competition Law’, 13 Yearbook of European Law 427 (1993) Forrester, Ian and Jacquelyn MacLennan, ‘Annual Survey of EC Competition Law’, 22 Yearbook of European Law 2001–2002 499 (2003) Forrester, Ian, Jacquelyn MacLennan and Assimakis Komninos, ‘Annual Survey of EC Competition Law’ 27 Yearbook European Law 2005–2006 521 (2008) Fox, Eleanor, ‘The WTO’s First Antitrust Case – Mexican Telecom: A Sleeping Victory for Trade and Competition’, 9 Journal of International Economic Law 271 (2006) Fox, Eleanor and Lawrence Sullivan. ‘Antitrust – Retrospective and Prospective: Where Are We Coming From? Where Are We Going?’, 62 New York University Law Review 936 (1987) François, Joseph and Miriam Manchin, ‘Institutions, infrastructure, and trade’, Policy research working paper No. 4152, World Bank, 2007 Friedman, Milton and Anna Schwartz, A Monetary History of the United States, 1867–1960, Princeton University Press, 1963 Fröberg, Martin, ‘Developing Countries and International Trade: The Occurrence and Prevention of Export Cartels’, undated, University of Maastricht, http:// arno.unimaas.nl/show.cgi?fid=15750 G Gans, Joshua and Stephen King, ‘Access Holidays and the Timing of Infrastructure Investment’, 80 Economic Record 89 (2004) Garzaniti, Laurent and Matthew O’Regan, Telecommunications, Broadcasting and the Internet: EU Competition Law and Regulation, 3rd edition, Sweet & Maxwell, 2010 Geradin, Damien, ‘EC Competition Law and Environmental Protection’, 2 Yearbook of European Environmental Law 117 (2002) Geradin, Damien, ‘Limiting the Scope of Article 82 of the EC Treaty: What can the EU learn from the U.S. Supreme Court’s Judgment in Trinko in the wake of Microsoft, IMS, and Deutsche Telekom?’, 41 Common Market Law Review 1552 (2004) Geradin, Damien, ‘Refusal to Supply and Margin Squeeze: A Discussion of Why the ‘Telefonica Exceptions’ are Wrong’, TILEC Discussion Paper 51 (2011)

References  

481  

Geradin, Damien, Anne Layne-Farrar and Nicolas Petit, EU Competition Law and Economics, Oxford University Press, 2012 Geradin, Damien and Robert O’Donoghue, ‘The Concurrent Application of Competition Law and Regulation: the Case of Margin Squeeze Abuses in the Telecommunications Sector’, 1 Journal of Competition Law and Economics 355 (2005) Geradin, Damien and J. Gregory Sidak, ‘European and American Approaches to Antitrust Remedies and the Institutional Design of Regulation in Telecommunications’, in Martin Cave, Sumit Majumdar and Ingo Vogelsang, eds., Handbook of Telecommunications Economics v. II, North-Holland, 2005, pp. 518 et seq. Gerber, David, ‘Constitutionalising the Economy: German Neo-liberalism, Competition Law and the “New” Europe’, 42 American Journal of Comparative Law 25 (1994) Gerber, David, Law and Competition in Twentieth Century Europe: Protecting Prometheus, Clarendon Press, 1998, Gilbert, Richard, ‘Looking for Mr Schumpeter: Where Are We in the CompetitionInnovation Debate?’, in Adam Jaffe, Josh Lerner and Scott Stern, eds., Innovation Policy and the Economy, Vol. 6, MIT Press, 2006 Gilbert, Richard, ‘Ties that Bind: Policies to Promote (Good) Patent Pools’, 77 Antitrust Law Journal 1 (2010) Gilbert, Richard and Steven Sunshine, ‘Incorporating Dynamic Efficiency Concerns in Merger Analysis: The Use of Innovation Markets’, 63 Antitrust Law Journal 569 (1995) Gilliams, Hans, ‘Competition Law and Public Interest: Do We Need to Change the Law for the (Liberal) Professions’, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2004: the Relationship between Competition Law and (Liberal) Professions, Hart Publishing, 2005 Ginsburg, Douglas H. Ginsburg and Eric M. Fraser, ‘The Role of Economic Analysis in Competition Law’, in Ian McEwin, ed., Intellectual Property, Competition Law and Economics in Asia, Hart Publishing, 2011, chapter 3 Goddin, Gaëtane, ‘The Pfleiderer Judgment on Transparency; The National Sequel of the Access to Document Saga’, 3 Journal of European Competition Law and Practice 40 (2012) Govindarajan, Vijay and Chris Trimble, ‘Organizational DNA for Strategic Innovation’, 47 California Management Review 47 (2005) Grasso, Roberto and John Ratliff, ‘Unilateral conduct in the energy sector: An overview of EU and national case law’, 15 March 2012, e-Competitions, No. 44203

482  References Green, Jerry and Suzanne Scotchmer, ‘On the Division of Profits in Sequential Innovation’, 26 Rand Journal of Economics 20 (1995) Grosskettler, Hans, ‘On Designing an Institutional Infrastructure for Economies: The Freiburg Legacy after 50 Years’, 21 Journal of Economic Studies 4 (1994) Growitsch, Christian, Nicole Nulsch and Margarethe Rammerstorfer, ‘Preventing innovative co-operations: the legal exemption’s unintended side effect’, 33 European Journal of Law and Economics 1 (2012) H Hancher, Leigh and Wolf Sauter, eds., EU Competition and Internal Market Law in the Health Care Sector, Oxford University Press, 2012 Harden, Ian, ‘The Revision of Regulation 1049/2001 on Public Access to Documents’, 15 European Public Law 239 (2009) Henkel, Joachim, Markus Reitzig and Christopher Heath, ‘On sharks, trolls, and their patent prey. Unrealistic damage awards and firms’ strategies of “being infringed”’, 36 Research Policy 134 (2007) Hoekman, Bernard and Kamal Saggi, ‘Tariff bindings and bilateral cooperation on export cartels’, 83 Journal of Development Economics 141 (2007) Hovenkamp, Herbert, ‘Signposts of Anticompetitive Exclusion: Restraints on Innovation and Economies of Scale’, in Barry Hawk, eds, International Antitrust Law and Policy 2006: Fordham Competition Law, Juris Publishing, 2007, chapter 18 Howell, Bronwyn, ‘Politics and the pursuit of telecommunications sector efficiency in New Zealand’, 6 Journal of Competition Law and Economics 253 (2011) Hufbauer, Gary and Sherry Stephenson, ‘Services Trade: Past Liberalization and Future Challenges’, 10 Journal of International Economic Law 605 (2007) Hunt, Robert, ‘Patentability, Industry Structure and Innovation’, 52 Journal of Industrial Economics 401 (2004) I Idot, Laurence, ‘Les leçons du droit international privé’, Concurrences no. 2-2012, pp. 41 et seq. Italianer, Alexander, ‘Public and private enforcement of competition law’, speech, Brussels, 17 February 2012

References  

483  

J Jenny, Frederic, ‘International Cooperation on Competition: Myth, Reality and Perspective’, 48 The Antitrust Bulletin 973 (2003) Jenny, Frédéric, ‘Competition, Trade and Development Before and After Cancún’, in Barry Hawk, ed., International Antitrust Law & Policy: Fordham Corporate Law 2003, Juris Publishing, 2004, chapter 26 Jenny, Frédéric, ‘The Economic and Financial Crisis, Regulation and Competition’, 32 Law and Economics Review 449 (2009) Jenny, Frederic, ‘Export Cartels in Primary Products: The Potash Case in Perspective’, in Simon J. Evenett and Frédéric Jenny, eds., Trade, Competition, and The Pricing of Commodities, 2012, http://ssrn.com/abstract=2064686 Jenny, Frédéric and Pradeep Mehta, ‘Global problems & solutions’, The Financial Express, 2012, posted online 13 January 2012, http://www.financialexpress. com/news/global-problems-&-solutions/899105/0 Joerges, Christian, ‘A New Alliance of De-legalisation and Legal Formalism? Reflections on Responses to the Social Deficit of the European Integration Project’, 19 Law and Critique 235 (2008) K Kapp, Thomas, ‘Das Akteneinsichtsrecht kartellgeschädigter Unternehmen: Bonn locuta, causa finita?’, WuW, 05/2012, pp. 474 et seq. Katz, Michael Katz and Howard Shelanski, ‘Mergers and Innovation’ 74 Antitrust Law Journal 1 (2007) Kessides, Ioannis, Reforming Infrastructure; Privatization, Regulation, and Competition, World Bank, 2004 Kingston, Suzanne, Greening EU Competition Law and Policy, Cambridge University Press, 2012 Knable Gotts, Ilene, Scott Sher, and Michelle Lee, ‘Antitrust Merger Analysis in High-Technology Markets’, 4 European Competition Journal 463 (2008) Kobayashi, Bruce and Joshua Wright, ‘Federalism, Substantive Preemption, and Limits on Antitrust: An Application to Patent Holdup’, 5 Journal of Competition Law and Economics 469 (2009) Kjølbye, Lars and Luc Peeperkorn, ‘The New Technology Transfer Block Exemption Regulation and Guidelines’, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2005: The Interaction between Competition Law and Intellectual Property Law, Hart Publishing, 2007, pp. 161 et seq.

484  References Komninos, Assimakis, ‘Relationship between Public and Private Enforcement: quod Dei Deo, quod Caesaris Caesari’, in Philip Lowe and Mel Marquis, eds., European Competition Law Annual 2011: Integrating Public and Private Enforcement of Competition Law: Implications for Courts and Agencies, Hart Publishing, 2014 Körber, Torsten, Standard Essential Patents, FRAND Commitments, and Competition Law, Nomos, 2013 Kortmann, Jeroen and Christof Swaak, ‘The EC White Paper on Antitrust Damage Actions: Why the Member States are (Right to be) Less Than Enthusiastic’, 30 European Competition Law Review 340 (2009) L Larouche, Pierre, Competition Law and Telecommunications, Hart Publishing, 2000

Regulation

in

European

Larouche, Pierre, ‘Contrasting Legal Solutions and Comparability of the E.U. and U.S. Experiences’, TILEC Discussion Paper 28 (2006) Larouche, Pierre, ‘Europe and Investment Infrastructure with Emphasis on Electronic Communications’, TILEC Discussion Paper DP 2007/031 Larouche, Pierre, ‘The European Microsoft Case at The Crossroads of Competition Policy and Innovation’, 75 Antitrust Law Journal 601 (2008) Laurinen, Elina, ‘The Consumer detergents cartel’, (2011/2) Competition Policy Newsletter 3 Lemley, Mark, ‘Intellectual Property Rights and Standard Setting Organizations’, 90 California Law Review 1889 (2002) Lemley, Mark and Carl Shapiro, ‘Patent Holdup and Royalty Stacking’, 85 Texas Law Review 2163 (2007) Levenstein, Margaret and Valerie Suslow, ‘Private International Cartels and Their Effect on Developing Countries’, Background Paper for the World Bank’s World Development Report 2001, 9 January 2001, http://www-unix.oit. umass.edu/~maggiel /WDR2001.pdf Levenstein, Margaret and Valerie Suslow, ‘The Changing International Status of Export Cartel Exemptions’, 20 American University International Law Review 785 (2005) Levy, Nicholas, ‘The EU’s SIEC Test Five Years On: Has It Made a Difference?’, 6 European Competition Journal 211 (2010) Lowe, Philip, ‘Can EU competition policy create competition in the energy sector?’, The Beesley Lecture, London, 6 November 2008

References  

485  

Lowe, Philip and Mel Marquis, eds., European Competition Law Annual 2011: Integrating Public and Private Enforcement of Competition Law: Implications for Courts and Agencies, Hart Publishing, 2014 Lugard, Paul, ‘Technology Licensing: Evolving Antitrust Standards in the Smartphone and Other Sectors’, CPI Antitrust Chronicle 2012(2) Lugard, Paul and Martin Mollmann, ‘The European Commission’s Practice Under Article 9 Regulation 1/2003: A Commitment a Day Keeps the Court Away?’, CPI Antitrust Chronicle 2013(3) M Maes, Marc, ‘L’accès du Public aux documents des Institutions de la Communauté européenne : vers une révision du règlement (CE) 1049/2001’, (2007/2) Revue de Droit de l’Union Européenne 411 (2007) Maier-Rigaud, Frank, Federica Manca and Ulrich von Koppenfels, ‘Strategic underinvestment and gas network foreclosure – the ENI case’, (2011/1) Competition Policy Newsletter 18 Maks, Hans, ‘The “New” Horizontal Agreements Approach in the EU: An “Economic” Assessment’. 37 Intereconomics 28 (2002) Marino, Ignazio, ‘Autorità garante della concorrenza e del mercato e giustizia amministrativa’, in Diritto dell’economica, 3/1992, pp. 578 et seq. Marquis, Mel, ‘Abuse of Administrative Power to Restrict Competition in China: Four Reflections, Two Ideas and a Thought’, in Michael Faure and Xinzhu Zhang, eds., The Chinese Anti-Monopoly Law: New Developments and Empirical Evidence, Edward Elgar, 2013, pp. 73 et seq. Maton, Anthony et al., ‘The effectiveness of National Fora in Europe for the Practice of Antitrust Litigation’, 2 Journal of European Competition Law and Practice 489 (2011) Matutes, Carmen, Pierre Regibeau and Katharine Rockett, ‘Optimal patent design and the Diffusion of Innovations,’ 27 Rand Journal of Economics 60 (1996) McChesney, Fred, ‘Talkin’ Bout My Antitrust Generation: Competition For and in the Field of Competition Law’, 52 Emory Law Journal 1401 (2003) Mehta, Pradeep, Competition Regimes in the World: A Civil Society Report, CUTS International, 2006 Melamed, Douglas and David Lerch, ‘Uncertain Patents, Antitrust and Patent Pools’, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2005: The Interaction between Competition Law and Intellectual Property Law, Hart Publishing, 2007, pp. 273 et seq.

486  References Merlino, Pietro and Gianluca Faella, ‘L’obbligo di investire nello sviluppo di infrastrutture: note a margine di due storie parallele’, 14(1) Mercato concorrenza regole 117 (2012) Michel, Suzanne, ‘Bargaining For Rand Royalties In The Shadow Of Patent Remedies Law’, 77 Antitrust Law Journal 889 (2011) Mileva, Elitza and Nikolaus Siegfried, ‘Oil Market Structure, Network Effects and the Choice of Currency for Oil Invoicing’, European Central Bank Occasional Paper No 77 (2007) Miller, Roger and Marcel Côté, ‘The Faces of Innovation’ (2008), http://ssrn.com/ abstract=1106740 Monti, Giorgio, ‘Article 82 EC and New Economy Markets’ in Cosmo Graham and Fiona Smith, eds., Competition, Regulation and the New Economy, Hart Publishing, 2004, pp. 17 et seq. Monti, Giorgio, ‘Merger Defences’, in Giuliano Amato and Claus-Dieter Ehlermann, eds., EC Competition Law: A Critical Assessment, Hart Publishing, 2006, pp. 519 et seq. Monti, Giorgio, EC Competition Law, Cambridge University Press, 2007 Monti, Giorgio, ‘Managing the Intersection of Utilities Regulation and EC Competition Law’, 4 Competition Law Review 123 (2008) Moretti, Marta, ‘Valutazioni economiche complesse in materia antitrust e self restraint dei giudici dell’U.E.’, Diritto dell’Unione Europea, 2/2009, pp. 315 et seq. Motta, Massimo, Competition Policy: Theory and Practice, Cambridge University Press, 2004 N Nazzini, Renato, The Foundations of European Union Competition Law, Oxford University Press, 2012 Nitsche, Rainer and Lars Wiethaus, ‘Access regulation and investment in next generation networks – A ranking of regulatory regimes’, 29 International Journal of Industrial Organization 263 (2011) Noce, Alessandro, ‘Antitrust e regolazione nelle decisioni con impegni in materia di energia’, 13(2) Mercato concorrenza regole 333 (2011) O O’Donoghue, Robert and Jorge Padilla, The Law and Economics of Article 102, 2nd edition, Hart Publishing, 2013

References  

487  

O’Donoghue, Ted, ‘Patentability Requirement for Sequential Innovation’, 29 Rand Journal of Economics 654 (1998) Odudu, Okeoghene, The Boundaries of EC Competition Law: The Scope of Article 81, Oxford University Press, 2006 Osakwe, Chiedu, ‘Poverty Reduction and Development: The Interaction of Trade, Macroeconomic and Regulatory Policies’, 10th Joseph Mubiru Memorial Lecture Organized by the Bank of Uganda, 2001 P Parret, Laura, ‘Shouldn’t We Know What We Are Protecting? Yes We Should! A Plea for a Solid and Comprehensive Debate about the Objectives of EU Competition Law and Policy’, 6 European Competition Journal 339 (2010) Pera, Alberto and Giulia Codacci Pisanelli, ‘Decisioni con impegni e private enforcement del diritto antitrust’, 14(1) Mercato concorrenza regole 69 (2012) Petit, Nicolas, ‘The Proliferation of National Regulatory Authorities alongside Competition Authorities: A Source of Jurisdictional Confusion’, in Damien Geradin, Rodolphe Muñoz and Nicolas Petit, eds., Regulatory Authorities in the EC: A New Paradigm for European Governance, Edward Elgar, 2005, pp. 194 et seq. Posner, Richard, Antitrust Law, 2nd edn, Chicago University Press, 2001 Posner, Richard, ‘Antitrust in the New Economy’, 68 Antitrust Law Journal 925 (2001) Priest, George and Franco Romani, ‘L’antitrust negli Stati Uniti e in Europa. Analisi e psicoanalisi di una divergenza’, 4(1) Mercato concorrenza regole 151 (2002) R Régibeau, Pierre and Katharine Rockett, ‘Assessment of potential anticompetitive conduct in the field of intellectual property rights and assessment of the interplay between competition policy and IPR protection’ (November 2011), http://ec.europa.eu/competition/consultations/2012_technology_transfer/ study_ipr_en.pdf Rockett, Katharine, ‘Property Rights and Inventions’, in Bronwyn Hall and Nathan Rosenberg, eds., Handbook of the Economics of Innovation, Vol. I, North-Holland, 2010, Chapter 7 Röller, Lars-Hendrik, Johan Stennek and Frank Verboven, ‘Efficiency gains from mergers’, Discussion papers WZB, Wissenschaftszentrum Berlin für Sozialforschung, Forschungsschwerpunkt Marktprozeß und Unternehmensentwicklung, No. FS IV 00-09 (2000)

488  References Rousseva, Ekaterina and Mel Marquis, ‘Hell Freezes Over: A Climate Change for Assessing Exclusionary Conduct under Article 102 TFEU’, 1 Journal of European Competition Law and Practice 32 (2013) S Sabel, Charles and Jonathan Zeitlin, Experimentalist Governance in the European Union, Oxford University Press, 2010 Sadowska, Malgorzata, ‘Energy Liberalization in Antitrust Straightjacket: A Plant Too Far?’, 3 World Competition 449 (2011) Sauter, Wolf, ‘The impact of EU competition law on national healthcare systems’, 38 European Law Review 457 (2013) Scholz, Ulrich and Stephan Pups, ‘The Application of EC Competition Law in the Energy Sector’, 1 Journal of European Competition Law and Practice 37 (2010) Schumpeter, Joseph, Capitalism, Socialism and Democracy, Routledge, 1942 Schwartz, Mark, ‘The Not-So-New Antitrust Environment for Consolidation in the Defense Industry: The Martin Marietta-Lockheed Merger’, 1996 Columbia Business Law Review 329 (1996) Schweitzer, Heike, ‘Commitment Decisions under Article 9 of Regulation 1/2003: The Developing EC Practice and Case Law’, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2008: Antitrust Settlements, Hart Publishing, 2010, pp. 547 et seq. Schweitzer, Heike, ‘Competition Law and Public Policy: Reconsidering an Uneasy Relationship’, EUI Working Paper 2007/30 Scotchmer, Suzanne, Innovation and Incentives, MIT Press, 2004 Scott-Morton, Fiona, ‘Antitrust Enforcement in High Technology Industries: Protecting Innovation and Competition’, speech, New York, 7 December 2012, www.justice.gov/atr/public/speeches/290876.pdf Shakespeare, William, Twelfth Night (1601) Shapiro, Carl, ‘Competition and Innovation: Did Arrow Hit the Bull’s Eye?’, in Joshua Lerner and Scott Stern, eds., The Rate and Direction of Inventive Activity Revisited, University of Chicago Press, 2012 Shelanski, Howard, ‘Innovation, Information and Competition Policy for the Internet’ 161 University of Pennsylvania Law Review 1663 (2013) Sidak, Gregory, ‘The Meaning of Frand, Part I: Royalties’, 9 Journal of Competition Law and Economics 931 (2013) Sidak, Gregory and David Spulber, Protecting Competition from the Postal Monopoly, American Enterprise Institute Press, 1996

References  

489  

Singham, Shanker, ‘A General Theory of Trade and Competition: Trade Liberalization and Competitive Markets’, Cameron May, 2007 Siragusa, Mario, ‘Gli obblighi di non discriminazione nella regolazione settoriale e nella disciplina antitrust’, in Giulio Napolitano and Andrea Zoppini, eds., Annuario di diritto dell’energia 2012. Il regime giuridico delle infrastrutture di energia, Il Mulino, 2012, pp. 147 et seq. Sweeney, Brendan, ‘Export Cartels: Is There a Need for Global Rules?’, 10 Journal of International Economic Law 87 (2007) Spanjer, Aldo, ‘Do Article 22 exemptions adequately stimulate investment in European gas markets?’, (2008) Zeitschrift für Energiewirtschaft 46 Stern, Jon, ‘Developing Upstream Competition in the England and Wales Water Supply Industry: A New Approach’, CCRP Working Paper, March 2010 Stern, Jon and Jonathan Mirrlees-Black, ‘A Framework for Valuing Water in England and Wales from 2015 Onwards’, CCRP Working Paper, October 2011 Stewart, Richard, ‘The Importance of Law and Economics for European Environmental Law’, 2 Yearbook of European Environmental Law 1 (2002) Szydlo, Marek, ‘The promotion of investments in new markets in electronic communications and the role of national regulatory authorities after Commission v Germany’, 60 International and Comparative Law Quarterly 533 (2011) T Talus, Kim, EU Energy Law and Policy: A Critical Account, Oxford University Press, 2013 Tarasov, Aleksei, ‘The making of empires: Russia’s gas-exporting pipelines v Nabucco’, 4 Journal of World Energy Law and Business 77 (2011) Taylor, Robert and Diana Moss (2013), AAI Working Paper No. 13-05: The Fertilizer Oligopoly: The Case for Global Antitrust Enforcement, http:// www.antitrustinstitute.org/~antitrust/content/aai-working-paper-no-13-05fertilizer-oligopoly-case-global-antitrust-enforcement Teece, David, ‘Favouring Dynamic over Static Competition – Implications for Antitrust Analysis and Policy’, in Geoffrey Manne and Joshua Wright, eds., Competition Policy and Patent Law Under Uncertainty, Cambridge University Press, 2011 Teh, Robert, ‘Competition provisions in regional trade agreements’, in Antoni Estevadeordal, Kati Suominen and Robert Teh, eds., Regional Rules in the Global Trading System, Cambridge University Press, 2009, chapter 8

490  References Temple Lang, John, ‘Defining Legitimate Competition: Companies’ Duties to Supply Competitors and Access to Essential Facilities’, 18 Fordham International Law Journal 437 (1994) Temple Lang, John, ‘European competition policy and regulation: Differences, overlaps, and constraints’, in François Leveque and Howard Shelanski, eds., Antitrust and Regulation in the EU and U.S.: Legal and Economic Perspectives, Edward Elgar, 2008 Teravaninthorn, Supee and Gaël Raballand, Transport prices and costs in Africa: A review of the main international corridors, World Bank, 2009 U Ullrich, Hanns, ‘Patent Pools: Approaching a Patent Law Problem via Competition Poliy’, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2005: The Interaction between Competition Law and Intellectual Property Law, Hart Publishing, 2007, pp. 305 et seq. Ungerer, Herbert, ‘Use of EC Competition Rules in the Liberalisation of the European Union’s Telecommunications Sector’ (2001) V Van de Gronden, Johan and Wolf Sauter, ‘Taking the Temperature: EU Competition Law and Health Care’ 38 Legal Issues of Economic Integration 213 (2011) Van der Hout, Robin and Miriam Firmenich, ‘Access to documents containing confidential business information – The application of Regulation (EC) 1049/2001 in cartel cases and the need for reform’, Zeitschift für Europarechtliche Studien 647 (2011/4) Van der Vijver, Tjarda, ‘Third Party Access Exemption Policy in the EU Gas and Electricity Sectors: Finding the Right Balance between Competition and Investments’, in Martha Roggenkamp et al., eds., Energy Networks and the Law, Oxford University Press, 2012 Van Gerven, Gerwin, ‘The Application of Article 81 in the New Europe’, in Barry Hawk, ed., International Antitrust Law & Policy: Fordham Corporate Law 2003, Juris Publishing, 2004, pp. 415 et seq. Vandenborre, Ingrid, ‘The confidentiality of EU Commission cartel records in civil litigation: the ball is in the EU Court’, 32 European Competition Law Review 116 (2011) Varney, Christine, ‘Vigorous Antitrust Enforcement in This Challenging Era’, Remarks as Prepared for the United States Chamber of Commerce (12 May 2009), http://www.justice.gov/atr/public/speeches/245777.htm Vedder, Hans, Competition Law and Environmental Protection in Europe: Towards Sustainability?, Europa Law Publishing, 2003

References  

491  

Vedder, Hans, ‘Of Jurisdiction and Justification: Why Competition is Good for Non-Economic Goals, But May Need to be Restricted’, 6 Competition Law Review 51 (2009) Vejanovski, Cento, ‘Margin squeeze: An overview of EU and national case law’, e-Competitions n° 46, 2012 Vickers, John, ‘Competition Policy and Property Rights’ 120 Economic Journal 375 (2010) Völcker, Sven, case note, ‘Case C360/09, Pfleiderer AG v. Bundeskartellamt, Judgment of the Court of Justice (Grand Chamber) of 14 June 2011’, 49 Common Market Law Review 695 (2012) Von Hippel, Eric, The Sources of Innovation, Oxford University Press, 1988 Vogelaar, F.O.W., ‘Towards an Improved Integration of EC Environmental Policy and EC Competition Policy: An Interim Report’, in Barry Hawk, ed., International Antitrust & Policy: Fordham Corporate Law 1994, Juris Publishing, 1996, pp. 529 et seq. W Wainwright, Richard and André Bouquet, ‘State Intervention and Action in EC Competition Law’, in Barry Hawk, ed., International Law & Policy: Fordham Corporate Law Institute 2003, Juris Publishing, 2004 Wesseling, Rein, The Modernisation of EC Antitrust Law, Hart Publishing, 2000 Williamson, Oliver, ‘Economies as an Antitrust Defense: The Welfare Tradeoffs’, 58 American Economic Review 18 (1968) Wright, Joshua, ‘Antitrust, Multidimensional Competition and Innovation’, in Geoffrey Manne and Joshua Wright, eds., Competition Policy and Patent Law Under Uncertainty, Cambridge University Press, 2011. Wu, Tim, ‘Taking Innovation Seriously: Antitrust Enforcement if Innovation Mattered Most’, 78 Antitrust Law Journal 313 (2012) Y Young, Mike, ‘Towards a Generic Framework for the Abstraction and Utilisation of Water in England and Wales’, UCL Environment Institute Report, 2012 Z Zäch, Roger, ‘Freedom to compete and the more economic approach - limits imposed by law’, 40 IIC International Review of Intellectual Property and Competition Law 623 (2009)

492  References II. EU legislation and soft law 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 2299/89 on a code of conduct for computerised reservation systems, 1989 OJ L220/1, amended by Council Regulation 3089/93, 1993 OJ L278/1 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 Directive 2002/21/EC on a common regulatory framework for electronic communications networks and services (Framework Directive), 2002 OJ L108/33, modified by Directive 2009/140/EC amending Directives 2002/21/ EC on a common regulatory framework for electronic communications networks and services, 2002/19/EC on access to, and interconnection of, electronic communications networks and associated facilities, and 2002/20/ EC on the authorisation of electronic communications networks and services, 2009 OJ L337/37 Directive 2009/73 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC, 2009 OJ L211/94 Directive 2009/72/EC of the European Parliament and the Council of 13 July 2009, 2009 OJ L211/55, on common rules for the internal market in electricity Directive 2012/34/EC of the European Parliament and the Council of 21 November 2012 establishing a single European railway area, 2012 OJ L343/32 Regulation (EC) No 1049/2001 of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, 2001 OJ L145/43 Regulation 1/2003 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, 2003 OJ L1/1 Regulation 80/2009 of the European Parliament and the Council of 14 January 2009 on a Code of Conduct for computerised reservations systems and repealing Council Regulation 2299/89, 2009 OJ L35/47 Regulation 714/2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003, 2009 OJ L211/15 Regulation 1227/2011 on wholesale energy market integrity and transparency (REMIT), 2011 OJ L326/1

References  

493  

European soft law and consultation documents Commission, Energy 2020 – A strategy for competitive, sustainable and secure energy, COM(2010) 639 final of 10 November 2010 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 Commission Guidelines on the application of Article 81(3) of the EC Treaty, 2004 OJ C101/97 Commission Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, 2004 OJ C31/5 Commission Recommendation of 20 September 2010 on regulated access to Next Generation Access Networks (NGA Recommendation), 2010 OJ L251/35 Commission Recommendation on consistent non-discrimination obligations and costing methodologies to promote competition and enhance the broadband investment environment, C(2013) 5761 final Commission, Staff working document on Article 22 of Directive 2003/55/EC concerning common rules for the internal market in natural gas and Article 7 of Regulation (EC) No 1228/2003 on conditions for access to the network for cross-border exchanges in Electricity, SEC(2009)642 final (‘Exemption working document’) Commission, State Aid Modernisation, COM(2012)209 final III. Official and unofficial reports, newsletters, studies, papers, etc. A. National jurisdictions United Kingdom OFGEM, RIIO: A New Way to Regulate Energy Networks (October 2010) United States International Competition Policy Advisory Committee, Final Report to the Attorney General and Assistant Attorney General for Antitrust, http://www. justice.gov/atr/icpac/finalreport.html Antitrust Modernization Commission, Report and Recommendations, April 2007, http://govinfo.library.unt.edu/amc/report_recommendation/amc_final_report.pdf

494  References B. International organizations Organisation for Economic Cooperation and Development OECD Global Forum on Competition, Summary of Cartel Cases Described by Invitees (5 October 2001), CCNM/GF/COMP(2001)4 OECD, Competition, Patents and Innovation II (1 April 2010), DAF/COMP(2009) OECD, Background Note on Standard Setting (1 June 2010), DAF/COMP/ WP2(2010)4 OECD, State-Owned Enterprises and the Principle of Competitive Neutrality (20 September 2010), DAF/COMP(2009)37 OECD, Collusion and Corruption in Public Procurement (2010), http://www. oecd.org/competition/cartels/46235884.pdf OECD, Regulated Conduct Defence (2011), DAF/COMP(2011)3, http://www. oecd.org/daf/competition/mergers/48606639.pdf OECD, Fighting Bid Rigging in Public Procurement in Mexico (2013), http:// www.oecd.org/daf/competition/MexicoISSSTEBidRiggingENG.pdf OECD, Competition and Poverty Reduction (2013), http://www.oecd.org/daf/ competition/competition-and-poverty-reduction2013.pdf World Trade Organization World Trade Organization, ‘Special Study on Trade and Competition Policy’, in Annual Report for 1997 (Geneva, 1997), chapter IV World Trade Organization, WTO Dispute Settlement: One-Page Case Summaries (Geneva, 2006), http://www.wto.org/english/res_e/booksp_e/dispu_summary 06_e.pdf WTO Working Group on the Interaction between Trade and Competition Policy, Annual Report of the Working Group on the Interaction between Trade and Competition Policy to the General Council (Geneva, 1998), WT/WGTCP/2

Table of Cases

I. Judgments of the Courts of the European Union A Joined Cases C-204/00 P, C-205/00 P, C-211/00 P, C-213/00 P, C-217/00 P and C-219/00 P, Aalborg Portland A/S and Others v Commission [2004] ECR I-123

198

Case C-143/99 Adria-Wien Pipeline GmbH e Wietersdorfer & Peggauer Zementwerke GmbH v Finanzlandesdirektion für Kärnten [2001] ECR I-8365

128

Case T‑111/07 Agrofert Holding v Commission, judgment of 7 July 2010, 373 not reported Case C-389/96 Aher-Waggon GmbH v Germany [1998] ECR I-4473

118, 131

Case 66/86 Ahmed Saeed Flugreisen and Silver Line Reisebüro GmbH v Zentrale zur Bekämpfung unlauteren Wettbewerbs e.V. [1989] ECR 838

196

Case T-9/11 Air Canada v Commission (2011 OJ C72/25)

85, 367

Cases T-62/11, T-63/11 and T-67/11 Air France-KLM v Commission, Air France v Commission and Martinair Holland v Commission (2011 OJ C95/8-10)

85, 367

Case C-62/86 AKZO Chemie BV v Commission [1991] ECR I-3359 Case C-67/96 Albany International BV v Stichting Bedrijfspensioenfonds Textielindustrie [1999] ECR I-5751 Case C-280/00 Altmark Trans GmbH [2003] ECR I-7747

169, 218, 220, 231

124, 348, 454–55

224

Case 106/77 Amministrazione delle Finanze dello Stato v Simmenthal SpA 411 [1978] ECR 629

496  Table of Cases Case C-176/99 Arbed SA v Commission [2003] ECR I-10687

442

Case C-127/07 Société Arcelor Atlantique et Lorraine and Others v Premier ministre, Ministre de l’Écologie et du Développement durable and Ministre de l’Économie, des Finances et de l’Industrie [2008] ECR I-9895

344

Case C-35/99 Manuele Arduino and Compagnia Assicuratrice xxxiv, 449, 450 RAS SpA [2002] ECR I-1529 Case C-238/05 Asnef-Equifax and Administración del Estado [2006] ECR I-11125

70, 213

Case T-321/05 AstraZeneca AB and AstraZeneca plc v 280, 293, 301 Commission [2010] ECR II-2805, upheld on appeal: Case C-457/10 AstraZeneca v Commission [2012] ECR I-000 Case T-395/94 Atlantic Container Line AB v Commission [2002] ECR II-875

465

Case T-24/90 Automec v Commission [1992] ECR II-2223

290

B Case T-41/96 Bayer AG v Commission [2000] ECR II-3383, upheld on appeal: Joined Cases C-2/01 P and C-3/01 P Bundesverband der Arzneimittef-Importeure eV and Commission v Bayer AG [2004] ECR I-23

442, 463

Case 234/84 Belgium v Commission [1986] ECR 2263

230

Case 40/85 Belgium v Commission [1986] ECR 2321

230

Case C-70/93 BMW v ALD Auto-Leasing [1995] ECR I-3439

432

Case 30/87 Corinne Bodson v Pompes Funebres [1998] ECR 2479

459

Case C-415/93 Union royale belge des sociétés de football association 347, 454 ASBL v Jean-Marc Bosman and others and Union des associations européennesde football (UEFA) v Jean-Marc Bosman [1995] ECR I-4921 Case T‑111/00 British American Tobacco International (Investments) v Commission [2001] ECR II‑2997

370

Table of Cases  

497  

Case C-487/06 P British Aggregates Association v Commission 128, 129, 343 and United Kingdom [2008] ECR I-10505 Case T-210/02 RENV British Aggregates v Commission, judgment of 7 March 2012, not yet reported

128

Case T-219/99 British Airways plc v Commission [2003] ECR II-5917

168

Case C-95/04 P British Airways plc v Commission [2007] ECR I-2331

25, 66, 72, 125, 168, 258, 360

Case T-48/11 British Airways v Commission (2011 OJ C80/32)

85

Case C-185/91 Bundesanstalt für den Güterfernverkehr v Gebrüder Reiff GmbH & Co. KG [1993] ECR I-5801

347, 449

Case C‑536/11, Bundeswettbewerbsbehörde v Donau Chemie AG and Others, judgment of the Court of Justice of 6 June 2013, not yet reported

xxx, 385

C Case C-343/95 Diego Calì & Figli Srl v Servizi ecologici porto di Genova SpA (SEPG) [1997] ECR I-1547

120, 121

Case T-39/11 Cargolux Airlines v Commission (2011 OJ C80/26)

85

Case T-38/11 Cathay Pacific Airways v Commission (2011 OJ C72/32) 85, 367 Case T‑437/08 CDC Hydrogene Peroxide v Commission [2011] ECR II-825

xxx, 376, 377, 378

Case C-1/09 Centre d’exportation du livre français (CELF) and Ministre 346 de la Culture et de la Communication v Société internationale de diffusion et d’édition (SIDE) [2010] ECR I-2099 Case C-203/96 Chemische Afvalstoffen Dusseldorp BV and Others v 44, 126 Minister van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer [1998] ECR I-4075 Joined Cases C-94/04 and C-202/04 Cipolla and Macrino [2006] ECR I-11421

xxxiv, 450

498  Table of Cases Case T-79/12, Cisco Systems Inc., Messagenet SPA v Commission 274, 277 and Microsoft Corp., judgment of the General Court of 11 December 2013, not yet reported Case T-301/04, Clearstream Banking AG and Clearstream International SA v Commission [2009] ECR II-3155

299

Case C‑477/10 P Commission v Agrofert Holding [2012] ECR I-000 334, 365, 373, 374, 387 Case C-2/90 Commission v Belgium [1992] ECR I-4421

118, 131

334, 365, Case C‑404/10 P Commission v Éditions Odile Jacob [2012] ECR I-000 373, 374, 387 Case C-124/10 P Commission v Electricité de France (EDF) 334 [2012] ECR I-000 Case C-424/07 Commission v Germany [2009] ECR I-11431

23, 58, 59

Case 73/79 Commission v Italy [1980] ECR 1533

129

Case C-297/05 Commission v Netherlands [2007] ECR I-476 Case C-279/08 Commission v Netherlands [2011] ECR I-7671

118, 131 127, 129, 344

Case C-290/07 P Commission v Scott SA [2010] ECR I-7763 Case C‑139/07 P Commission v Technische Glaswerke Ilmenau [2010] ECR I‑5885

345, 346 371, 372, 378, 387

Case C-359/95 P Commission and France v Ladbroke Racing Ltd. [1995] ECR I-6265

196

Case-T-24/93 Compagnie maritime belge transports SA and Others v Commission [1996] ECR II-1201

301

Case T-86/95 Compagnie Générale Maritime and Others v Commission 70, 465 [2002] ECR II-1011 Case C-209/07 The Competition Authority v Beef Industry xxxiv, 444, 445 Development Society Ltd, Barry Brothers (Carrigmore) Meats [2008] ECR I-8637

Table of Cases   Case C-513/99 Concordia Bus Finland Oy Ab, formerly Stagecoach Finland Oy Ab v Helsingin kaupunki and HKL-Bussiliikenne [2002] ECR I-7213 Case C-198/01 Consorzio Industrie Fiammiferi (CIF) v Autorità Garante della Concorrenza e del Mercato [2003] ECR I-8055 Joined Cases C-101/07 and 110/07 Coop de France bétail et viande and Fédération nationale des syndicats d’exploitants agricoles (FNSEA) and Others v Commission [2008] ECR I-10193

499   131

14, 350 444

Joined Cases 56/64 and 58/64 Établissements Consten S.A.R.L. 68, 446, 463 and Grundig-Verkaufs-GmbH v Commission [1966] ECR 299 Case 6/64 Flaminio Costa v E.N.E.L. [1964] ECR 585 (Eng. Spec. edn)

411, 430

Case C-453/99 Courage Ltd v Bernard Crehan and Bernard 296, 333, 367, 385 Crehan v Courage Ltd and Others [2001] ECR I-6297 D Case T-151/01 Der Grüne Punkt - Duales System Deutschland GmbH [2007] ECR II-1607

290, 300

Case T-289/01 Der Grüne Punkt - Duales System Deutschland GmbH v Commission [2007] ECR II-1691

43, 44

Case C-385/07 P Der Grüne Punkt - Duales System Deutschland GmbH v Commission [2009] ECR I-6155

125

Case T-46/11 Deutsche Lufthansa and others v Commission (2011 OJ C80/31)

85, 367

Case T-271/03 Deutsche Telekom AG v Commission [2008] ECR II-477

193, 198

Case C-280/08 P Deutsche Telekom AG v Commission xxii, 71, 105, 168, [2010] ECR I-9555 170, 193, 195–96, 198, 199, 200, 217, 296, 341, 430, 434, 469

500  Table of Cases E Case T‑344/08 EnBW Energie Baden-Württemberg v Commission [2012] ECR II-000, set aside on appeal: Case C-365/12 P, Commission v EnBW Energie Baden-Württemberg, judgment of the Court of Justice of 27 February 2014, not yet reported

xxx, 142, 333, 377, 378, 380, 393

Case C-209/98 Entreprenørforeningens Affalds/Miljøsektion (FFAD) v Københavns Kommune (Sydhavnens Sten & Grus) [2000] ECR I-3743 Case C-199/11 European Union v Otis N.V. [2012] ECR I-000

44, 126

297

F Case C-475/99 Firma Ambulanz Glöckner v Landkreis Südwestpfalz [2001] ECR I-8089

120

Joined Cases T-217/03 and T-245/03 FNCBV and Others v 444, 445 Commission [2006] ECR II-4987, upheld on appeal: Joined Cases C-101/07 and 110/07 Coop de France bétail et viande and Fédération nationale des syndicats d’exploitants agricoles (FNSEA) and Others v Commission [2008] ECR I-10193 Case T-340/03 France Télécom SA v Commission [2007] ECR II-107 192, 193, 201, 301 Case C-202/07 P France Télécom SA v Commission [2009] 168, 170, 192, 193 ECR I-2369 Joined Cases T‑391/03 and T‑70/04 Franchet and Byk v Commission 370, 373 [2006] ECR II‑2023 G Case 13/77 SA G.B.-INNO-B.M. v Association des détaillants en tabac 296, 350 (ATAB) [1977] ECR 2115 Case C-156/98 Germany v Commission [2000] ECR I-6857

129

Case T‑168/01 Glaxo Wellcome v Commission, order of 5 August 2003, not reported

368

Table of Cases   Case T-168/01 GlaxoSmithKline Services v Commission [2006] ECR II-2969 Joined Cases C-501/06 P, C-513/06 P, C-515/06 P and C-519/06 P GlaxoSmithKline Services and Others v Commission and Others [2009] ECR I-9291

501  

70, 113, 446

68, 72, 242, 356

Case C-250/92 Gøttrup-Klim e.a. Grovvareforeninger v Dansk Landbrugs Grovvareselskab AmbA [1994] ECR I-5641

124

H Case T‑14/98 Hautala v Council [1999] ECR II‑2489

370

Case T-30/89 Hilti AG v Commission [1991] ECR II-1439 Case C-170/13 Huawei Technologies Co. Ltd v ZTE Corp., ZTE Deutschland GmbH, not yet decided

125, 458 xxviii, 283, 286

I Case C-418/01 IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG [2004] ECR I-5039

184, 296–98

Case T-66/01 Imperial Chemical Industries Ltd v Commission [2010] ECR II-2631 Case T-457/08 Intel v Commission, not yet decided

301

xxv, 20, 181,

Joined Cases C-327/03 and C-328/03 ISIS Multimedia and Firma O2 [2005] ECR I-8877 Case 41/83 Italy v Commission [1985] ECR 873

227 75, 227

J Case T-36/11 Japan Airlines v Commission (2011 OJ C80/25) Case T‑123/99 JT’s Corporation v Commission [2000] ECR II‑3269

85 370

K Case C-52/07 Kanal 5 Ltd e TV 4 AB v Föreningen Svenska 125 Tonsättares Internationella Musikbyrå (STIM) upa [2008] ECR I-9275

502  Table of Cases Case C-389/10 KME Germany AG, KME France SAS and KME Italy SpA v Commission [2011] ECR I-000 Case C-41/90 Klaus Höfner and Fritz Elser v Macrotron GmbH [1991] ECR I-1979

346 350, 355

Case T-28/11 Koninklijke Luchtvaart Maatschappij (2011 OJ C72/30)

85

Case C-52/09 Konkurrensverket v TeliaSonera Sverige AB 75, 168, 171, 194, [2011] ECR I-527 196, 199, 200–202, 214, 216,217, 222, 225, 295 Case C-109/03 KPN Telecom BV v Onafhankelijke Post en 57, 105, 216, 227 Telecommunicatie Autoriteit (OPTA) [2004] ECR I-11273 Case C-438/02 Krister Hanner [2005] ECR I-4551

348

L Case T-40/11 Lan Airlines and Lan Cargo v Commission (2011 OJ C80/27)

85, 367

Joined Cases T-268/08 and T-281/08 Land Burgenland (Autriche) and Republic of Austria v Commission [2012] ECR II-000, appeals dismissed: Joined Cases C-214/12 P, C-215/12 P and C-223/12 P, not yet reported Joined Cases C-468/06 to C-478/06 Sot. Lélos kai Sia EE v GlaxoSmithKline [2008] ECR I-7139

344

xxxv, 356, 363, 463, 465

M Joined Cases C-295/04 to C-298/04 Vincenzo Manfredi v Lloyd Adriatico Assicurazioni SpA [2006 ECR I-6619

xiii, 296

Case T-111/08 MasterCard, Inc and MasterCard Europe v Commission [2012] ECR II-000

70

Case C-382/12 MasterCard and Others v Commission, not yet decided

70

Case C-344/98 Masterfoods Ltd v HB Ice Cream Ltd. [2000] ECR I-11369 296 Case C-225/91 Matra SA v Commission [1993] ECR I-3203

402

Table of Cases   Case T-17/93 Matra Hachette SA v Commission [1994] ECR II-595

503  

50, 446

Case C-519/04 P David Meca-Medina and Igor Majcen v xxxiv, 347, 452, 453 Commission [2006] ECR I-6991 Case C-2/91 Wolf W. Meng [1993] ECR I-5751

347, 355, 448

Case C-179/90 Merci Conventionali Porto di Genova SpA v 350, 354, 355, 459 Siderurgica Gabriella SpA [1991] ECR I-5889 Case T-112/99 Métropole télévision (M6), Suez-Lyonnaise des eaux, France Télécom and Télévision française 1 SA (TF1) v Commission [2001] ECR II-2459

24, 62

Case T-201/04 Microsoft Corp. v Commission [2007] ECR II-3601 75, 168, 184, 242, 295, 299 Case T-167/08 Microsoft Corp. v Commission [2012] ECR II-000 286, 289, 290 Joined Cases C-544/03 and C-545/03 Mobistar and Belgacom Mobile [2005] ECR I-7723

227

Case C-12/08 Mono Car Styling SA, in liquidation v Dervis Odemis and Others [2009] ECR I-6653

297

N Joined Cases T-29/10 and T-33/10 Netherlands and ING Groep NV v Commission [2012] ECR I-000

344

Case T-233/04 Netherlands v Commission [2008] ECR II-591

118

Case T-380/08 Netherlands v Commission, judgment of the General Court 393 of 13 September 2013, not yet reported Case 322/81 NV Nederlandsche Banden Industrie Michelin v Commission [1983] ECR 3461

213, 360

O Case T-328/03 O2 (Germany) GmbH & Co. OHG v Commission [2006] ECR II-1231

62

Case T‑237/05 Éditions Odile Jacob v Commission [2010] ECR II‑2245

373

504  Table of Cases Case C-245/91 Ohra Schadeverzekeringen NV [1993] ECR I-5851 Case T-486/11 Orange Polska v Commission, not yet decided

347, 448 195

Case C-7/97 Oscar Bronner GmbH v Mediaprint Zeitungs-und xxii, 199, 202 Zeitschriftenverlag GmbH [1998] ECR I-7791 P Joined Cases C-180/98 to C-184/98 Pavel Pavlov and Others v Stichting Pensioenfonds Medische Specialisten [2000] ECR I-6451

450

Case C‑360/09 Pfleiderer AG v Bundeskartellamt [2011] xxx, 332, 334, 348, ECR I-5161 353, 356, 360, 378–86, 390, 393, 394 Case T-193/02 Laurent Piau v Commission [2005] ECR II-209, 347, 453, 454 appeal dismissed, Case C-171/05 [2006] ECR I-37* Case C-204/97 Portugal v Commission [2001] ECR I-3175

129

Case C-209/10 Post Danmark A/S v Konkurrencerådet xxiv, , 20, 21, 25, 66, [2012] ECR I-000 72, 166–69, 174–76, 184, 216, 217, 218, 220, 221, 223,224, 226, 234, 301 Case C-23/14 Post Danmark A/S v Konkurrencerådet (Post Danmark II), xxv not yet decided Case E-15/10 Posten Norge AS v EFTA Surveillance Authority [2012] EFTA Ct. Rep. 246

352

Case C-169/08 Presidente del Consiglio dei Ministri v Regione Sardegna 406 [2009] ECR I-821 Case C-379/98 PreussenElektra AG v Schhleswag AG, in the presence 118, 127 of Windpark Reußenköge III GmbH and Land Schleswig-Holstein [2001] ECR I-2099 Case C-375/09 Prezes Urzędu Ochrony Konkurencji i Konsumentów v Tele2 Polska sp. z o.o., devenue Netia SA [2011] ECR I-3055

67

Case T-111/96 ITT Promedia v Commission [1998] ECR II-2937 294, 295, 296

Table of Cases  

505  

Case T-119/09 Protégé International v Commission [2012] ECR II-000 294, 295 Case C-68/12 Protimonopolný úrad Slovenskej republiky v Slovenská sporitel’ňa a.s., judgment of the ECJ of 7 February 2013

48

Q Case C-157/96 The Queen v Ministry of Agriculture, Fisheries and Food, Commissioners of Customs & Excise, ex parte National Farmers’ Union [1998] ECR I-2211

402

R Joined Cases C-241/91 P and C-242/91 P Radio Telefís 184, 289, 296, 297, 298 Éireann (RTE) and Independent Television Publications Ltd (ITP) v Commission (Magill) [1995] ECR I-743 S Case C-70/10 Scarlet Extended SA v Société belge des auteurs, compositeurs et éditeurs SCRL (SABAM) [2011] ECR I-11959 Case T-398/07 Spain v Commission (Telefónica II) [2012] ECR II-527

295, 297

196, 197, 199, 200, 205

Joined Cases 40-48, 50, 54-56, 111, 113, 114/73 Suiker Unie v Commission 231 [1975] ECR 1663 Joined Cases C‑514/07 P, C‑528/07 P and C‑532/07 P Sweden v API [2010] ECR I‑8533

371

Case T-229/04 Sweden v Commission [2007] ECR II-2437

118

Joined Cases C‑39/05 P and C‑52/05 P Sweden and Turco v Council 370, 372 [2008] ECR I‑4723 Case C-53/03 Synetairismos Farmakopoion Aitolias & Akarnanias (Syfait) and Others v GlaxoSmithKline plc. and GlaxoSmithKline AEVE [2005] ECR I-4609

298, 464

506  Table of Cases T Case T‑237/02 Technische Glaswerke Ilmenau v Commission [2006] ECR II‑5131, set aside on appeal: Case C‑139/07 P

371

Case T-336/07 Telefónica, SA and Telefónica de España, SA v xxi, 194, 196, Commission [2012] ECR II-000 197, 198, 199, 201, 205, 217 Case C-295/12 P Telefónica SA and Telefónica de España SAU v xxii, 194 Commission, judgment of the ECJ of 10 July 2014, not yet reported Case T-486/11 Telekomunikacja Polska v Commission, not yet decided 40, 205 Case T-83/91 Tetra Pak International SA v Commission (Tetra Pak II) [1993] ECR II-755

125

Case T-158/99 Thermenhotel Stoiser Franz Gesellschaft mbH & Co. KG and Others v Commission [2004] ECR II-1

129

Case T-155/06 Tomra Systems SA and Others v Commission [2010] ECR II-4361

301

U Case 27/76 United Brands Company and United Brands Continentaal BV v Commission [1978] ECR 207

214

Case C-120/97 Upjohn Ltd v The Licensing Authority established by the Medicines Act 1998 and Others [1999] ECR I-223

402

V Case 267/86 Pascal Van Eycke v Aspa SA [1988] ECR 4769 Case 45/85 Verband der Sachversicherer e.V. v Commission [1987] ECJ 405 Case T-2/03 Verein für Konsumenteninformation v Commission [2005] ECR II‑1121

126, 196, 447, 448 430

370, 373, 375, 376, 377

Table of Cases  

507  

W Case C-309/99 J. C. J. Wouters, J. W. Savelbergh and xxxiv, 124, 125, 451, Price Waterhouse Belastingadviseurs BV v Algemene 452, 453 Raad van de Nederlandse Orde van Advocaten [2002] ECR I-1577 II. Decisions of the European Commission Antitrust Commission Decision of 20 July 1999 in Case IV/26.888 – 1998 Football 231 World Cup, 2000 OJ L5/55 Commission Decision of 21 December 1988 – Ansac, 1990 OJ L152/54

467

Commission Decision of 16 October 2003 in Case COMP/35473 – 43, 44, 124 ARA 2004 OJ L75/59, upheld in Case T-419/03 Altstoff Recycling Austria AG v Commission [2011] ECR II-975 Commission Decision of 14 January 1992 – Assurpol, 1992 OJ L37/16 Commission Decision of 15 June 2005 in Case COMP/A.37.507 – AstraZeneca

466

270, 280

Commission Decision of 5 May 1988 – Bayer/BP Chemicals, 1988 OJ L150/35

44

Commission Decision of 19 July 1984 – BPCL/ICI, 1984 OJ L212/1

442

Commission Decision of 24 January 1999 in xxv, 41, 43, 44, 48347, 466, 467 Case IV/36.718 – CECED, 2000 OJ L187/47 Commission Decision of 21 December 1988 – Decca Navigator System, 1989 OJ L43/27 Commission Decision of 20 March 2001 in Case COMP/35.141 – Deutsche Post AG, 2001 OJ L25/27

461

xxv, 167, 169, 174, 218, 219, 220, 221, 231

Commission Decision of 21 May 2003 in Case COMP/37.451 – Deutsche Telekom AG, 2003 OJ L263/9, upheld on appeal: Case C-280/08 P Deutsche Telekom AG v Commission [2010] ECR I-9555

193, 231

508  Table of Cases Commission Decision of 17 September 2001 – DSD, 2001 OJ L319/1 44, 124, 431, 466 EACEM, 1998 OJ C12/2 (Article 19(3) Notice)

466

Commission Decision of 15 June 2001 in Case/COMP/34.950 – Eco-Emballages, 2001 OJ L233/37

43, 124

Commission Decision of 29 September 2010 in Case COMP/39.315 – xxiii, 173, ENI, 2010 OJ C352/8 180, 185, 210, 213 Commission Decision of 4 December 1986 – ENI/Montedison, 1987 OJ L5/13

442

Commission Decision of 4 May 2010 Case COMP/39.317 – E.ON gas foreclosure

211, 213

Commission Decision of 14 January 1998 in Case IV/34.80 – Flughafen Frankfurt/Main AG, 1998 OJ L72/30

216

Commission Decision of 28 March 2012 in Case COMP/39462 – Freight Forwarding

85

Commission Decision of 2 April 2003 – French Beef, 2003 OJ L209/12 Commission Decision of 3 December 2009 in Case COMP/39.316 – GDF gas foreclosure Commission Decision of 26 November 2008 in Case COMP/39.388 – German electricity wholesale market, 2008 OJ C36/8

431, 444 210, 211, 213, 228 xxii, 173 212

Commission Decision of 27 August 2003 in Case COMP/37.685 – GVG/FS, 2004 OJ L11/17 Commission Decision of 13 May 2009 in Case COMP/37.990 – Intel, 2009 OJ C227/13

231 20, 217, 272–74

Commission Decision of 12 December 1990 – KSB/Goulds/Lowara/ITT, 1991 OJ L19/25

466

Commission Decision of 24 March 2004 in Case 217, 270, 272, COMP/37.792 –Microsoft, 2007 OJ L32/23, upheld 273, 274, 286, 297, 298

Table of Cases  

509  

(on the substantive aspects): Case T-201/04, Microsoft Corp. v Commission [2007] ECR II-3601

Commission Decision of 27 February 2008 in Case COMP/37.392 – Microsoft, 2009 OJ C166/20

286

Commission Decision of 30 March 1984 – Nuovo CEGAM, 1984 OJ L99/24 432 Commission Decision of 14 December 1979 – Pioneer Hi-Fi Equipment, 462 1980 OJ L60/21 Commission Decision of 21 December 1993 – Port of Rødby, OJ 1994 L55/52

231

Commission Decision of 9 December 2009 in Case 248, 272, 281, 290, 292, 296 COMP/38.636 – Rambus, 2010 OJ C30/17 Commission Decision of 20 December 2012 in Case COMP/39.654 – Reuters Instrument Codes (RICs)

289

Commission Decision of 15 November 2011 in Case COMP/39.592 – Standard & Poor’s

289

Commission Decision of 29 April 1994 – Stichting Baksteen, 1994 OJ L131/15 Commission Decision of 4 July 1984 – Synthetic Fibres, 1984 OJ L207/17

xxxiv, 443 xxxiv, 431, 442

Commission Decision of 22 June 2011 in Case COMP/39.525 – Telekomunikacja Polska, on appeal: Case T-486/11 Telekomunikacja Polska v Commission, not yet decided

170, 194–95, 198, 205

Commission Decision of 5 December 2012 in Case COMP/39.437 – TV and computer monitors Commission Decision of 13 December 1999 in Case IV/24.780 – Virgin/British Airways, 2000 OJ L30/1

85

Commission Decision of 4 July 2007 in Case COMP/38.784 – Wanadoo España v Telefónica, 2008 OJ C83/6 Commission Decision of 16 July 2003 in Case COMP/38.233 – Wanadoo Interactive

278

194

192, 205

510  Table of Cases Exemptions Arnoldstein/Tarvisio (AT/IT) – SG D(2010)16980 and Ares (2011) 42548 BritNED (UK/NL) – CAB D(2007) 1258

63

24, 62, 63

LNG Eemshaven (NL) – C (2009) 4006

64

Gazelle (CZ/DE) – C (2011) 3424

65

LNG Livorno (IT) – C (2009) 10172

64

Nabucco - HU – C (2009) 3034

65

National Grid Grain LNG (UK) – C(2013) 3443

64

LNG Porto Empedocle (IT) – C (2012) 3123 and LNG Eemshaven (NL) – C (2009) 4006

64

Trans Adriatic Pipeline – C(2013) 2949

65

Mergers Case COMP/M.3280 Air France/KLM 2004 OJ C60/5

85

Case COMP/M.5747 BA/Iberia Commission Decision of 14 July 2010

85

Case COMP/M. 6106 - Caterpillar/MWM 277 Case COMP/M.5669 - Cisco/Tandberg

277

Case COMP/M.5181 Delta Airlines/Northwest Airlines, 2008 OJ C281/3 Case COMP/M.5224 – EDF/British Energy Case COMP/M.6381 - Google/Motorola Mobility

85 211

277, 278, 281, 288, 293, 295, 299, 301

Case COMP/M.6447 IAG/bmi 85 Case COMP/M.5364 Iberia/Vueling/Clickair 2009 OJ C72/23

85

Case COMP/M.5984 - Intel/McAfee 277

Table of Cases  

511  

Case COMP/M.5141 KLM/Martinair 2009 OJ C51/ 4

85

Case COMP/M.5440 Lufthansa/Austrian Airlines 2010 OJ C16/11

85

Case COMP/M.5403 Lufthansa/bmi 2009 OJ C158/1

85

Case COMP/M.3940 Lufthansa/Eurowings 2006 OJ C18/22

85

Case COMP/M.5335 Lufthansa/SN Airholding (Brussels Airlines), 2009 OJ C295/11

85

Case COMP/M.3770 Lufthansa/Swiss 2005 OJ C204/3

85

Case COMP/M.4187- Metso/Ake Kvaerner 287 Case COMP/M.7047 - Microsoft/Nokia

278

Case COMP/M.6281 - Microsoft/Skype 277 Case COMP/M.5727 - Microsoft/Yahoo! Search Business 277 Case COMP/M.4942 - Nokia/Navteq

277, 278

Case COMP/M.5830 Olympic/Aegean 85 Case COMP/M.2537 - Philips/Marconi Medical Systems 277 Case COMP/M.4439 Ryanair/Aer Lingus 2008 OJ C47/9

85

Case COMP/M.2672 SAS/Spanair, 2002 OJ C93/7

85

Case COMP/M.6214 - Seagate Technology/The HDD Business of Samsung Electronics

277

Case COMP/M.5675 - Syngenta/Monsanto 277 Case COMP/M.3916 - T-Mobile, Tele.ring 278 Case COMP/M.4854 - TomTom/Tele Atlas Case COMP/M.4523 Travelport/Worldspan Case COMP/M.4600 TUI/First Choice 2007 OJ C137/6

69, 71, 278 86, 89 85

512  Table of Cases Case COMP/M.5658 - Unilever/Sarah Lee 277 Case COMP/M.5889 United Airlines/Continental Airlines 85 III. US cases (courts and the agencies) A.L.A. Schechter Poultry Corp. v United States, 295 U.S. 495 (1935)

420

Albrecht v Herald Co., 390 U.S. 145 (1968)

428

Andrx Pharms., Inc. v Biovail Corp. Int’l, 256 F.3d 799 (D.C. Cir. 2001)

426

E. Bement & Sons v National Harrow Co., 186 U.S. 70 (1902)

426

Cal. Retail Liquor Dealers Ass’n v Midcal Aluminum, Inc., 445 U.S. 97 (1980) In re Cardizem CD Antitrust Litig., 332 F.3d 896 (6th Cir. 2003) Chicago Board of Trade v United States, 246 U.S. 231 (1918)

428, 430 426 49

In re Ciprofloxacin Hydrochloride Antitrust Litig., 544 F.3d 1323 (Fed Cir. 2008)

426

Connell Constr. Co. v Plumbers, 421 U.S. 616 (1975)

422

Continental T.V., Inc. v GTE Sylvania, Inc., 433 U.S. 36 (1977)

359

Credit Suisse Securities (USA) v Billing, 551 U.S. 264 (2007) xxxiii, 197, 339, 340, 417, 424–29, 433, 435 In Re Compact Disc Minimum Advertised Price Antitrust Litigation, 216 FRD 197; 2003 US Dist LEXIS 11257 (Maine District Court, 2003)

262

Dr. Miles Med. Co. v John D. Park & Sons Co., 220 U.S. 373 (1911)

428

Federal Baseball Club of Baltimore, Inc. v National League of Professional Baseball Clubs, 259 U.S. 200 (1922)

358, 433

Federal Trade Commission v Watson Pharmaceuticals, Inc., 677 F.3d 1298 (11th Cir. 2012), reversed: Federal Trade Commission v Actavis, 570 U.S. __ (2013)

349, 426

Table of Cases  

513  

Flood v Kuhn, 407 U.S. 258 (1972)

433

FTC v Alliant Techsystems Inc., 808 F. Supp. 9 (D.D.C. 1992)

419

FTC v PPG Industries, Inc., 798 F.2d 1500 (D.C. Cir. 1986)

423

FTC v Weyerhauser, 665 F.2d 1072 (D.C. Cir. 1981)

423

Gordon v New York Stock Exchange, Inc., 422 U.S. 659 (1975)

424

In re Innovatio IP Ventures, LLC Patent Litig., 213 WL 5593609 (N.D. Ill. 2013)

292

International Business Machines Corp v United States, 298 U.S. 131 (1936) 262 In re K-Dur Antitrust Litigation, 686 F.3d 197 (3d Cir. 2012) Leegin Creative Leather Products, Inc. v PSKS, Inc., 551 U.S. 877 (2007)

426 242, 428

Microsoft Corp. v Motorola Inc., 2013 U.S. Dist. LEXIS 60233 (W.D. Wash. 2013)

292

Pacific Bell Telephone Co. v LinkLine, 555 U.S. 438 (2009) 201, 202, 203, 430 Posadas v National City Bank of New York, 296 U.S. 497 (1936)

422

Professional Real Estate Investors, Inc. v Columbia Pictures Industries, Inc., 508 U.S. 49 (1993)

422

Rambus Inc. v FTC, 522 F.3d 456 (D.C. Cir. 2008)

429

Silver v New York Stock Exchange, 373 U.S. 341 (1963) State Oil Co. v Khan, 522 U.S. 3 (1997)

424, 425 428

Superior Court Trial Lawyers Ass’n v FTC 856 F.2d 226 (D.C. Cir. 1988), 421 rev’d, 493 U.S. 411 (1990) In re Tamoxifen Citrate Antitrust Litig., 466 F.3d 187 (2d Cir. 2006)

426

United States v AMR Corp., 335 F. 3d 1109 (10th Cir. 2003)

183

514  Table of Cases United States v AT&T, 552 F.Supp. 131 (D.D.C. 1982) aff’d sub nom. Maryland v United States, 460 U.S. 1001 (1983)

171

United States v Butler, 297 U.S. 1 (1936)

420

United States v First City Nat. Bank of Houston, 386 U.S. 361 (1967)

423

United States v Line Material Co., 333 U.S. 287 (1948)

426

United States v Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001)

431, 435

U.S. Philips Corp. v International Trade Commission, 424 F.3d 1179 (Fed. Cir. 2005)

240

Verizon Communications Inc. v Law Offices of Curtis xxxiii, 173, 179, 186, v Trinko, 540 U.S. 398 (2004) 196, 207, 339, 340, 427, 428, 429, 433, 435, 470, 472 Re Vitamins Antitrust Litigation, Report of Special Master (99-197) (TFH), 2002 U.S. Dist. LEXIS 26490 January 23, 2002 D.D.C.

366

Re Vitamins Antitrust Litigation (99-197) (TFH), 2002 U.S. Dist. LEXIS 25815, December 18, 2002 D.D.C.

366

IV. Other national jurisdictions Canada Canada (Commissioner of Competition) v Superior Propane Inc, [2003] FCJ No 151 Canada (Competition Act, Director of Investigation and Research) v Tele-Direct (Publications) Inc, [1997] CCTD No 8, 73 CPR (3d) 1

264 262, 266

Canada (Competition Act, Director of Investigation and Research) v 263, 266 Warner Music Canada Ltd, [1997] CCTD No 53, 78 CPR (3d) 321 France Cour de Cassation, judgment of 19 January 2010, Semavem, no. 08-19.761

382

Cour d’Appel de Paris, judgment of 20 December 2012, 2011/05667, Vedettes Inter-Iles Vendéennes/Régie Départementale des Passages d’Eau de la Vendée

226

Table of Cases  

515  

Germany BGH, judgment of 13 July 2004, KZR 40/02, GRUR 2004, 966, 967 – Standard-Spundfass

293

Case KZR 39/06, [2009] 180 BGHZ 312 – Orange-Book-Standard, (translated in 41 IIC International Review of Intellectual Property and Competition Law 369 (2010))

291

Ireland Panda Waste Services v Dublin City Council and Others [2009] IEHC 589 (on appeal to the Irish Supreme Court)

43, 121

Italy AGCM decision no. 11421 of 21 November 2001, A329, Blugas-Snam, Bulletin no. 47/2002

228

AGCM decision of 30 November 2011, TNT Post Italia/Poste italiane, 224–25 Bulletin no. 48/2011; Tar Lazio, judgment of 25 June 2012, no. 5769 AGCM decision no. 15174 of 15 February 2006, A358, ENI-Trans xxv 228, 230 Tunisian Pipeline (ENI-TTPC), Bulletin no. 5/2006; TAR Lazio, 30 March 2007, no. 2798; Consiglio di Stato, 20 December 2010, no. 9306 Cons. Stato, VI, 23 April 2002, n° 2199

403

Cons. Stato, VI, 2 March 2004, n° 926

403

Cons. Stato, VI, 10 March 2006, n° 1271

403

Cons. Stato, VI, 8 February 2007, n° 515

403

Cons. Stato, VI, 12 February 2007, n° 550

403

Constitutional Court, decision n° 14 of 18 December 2003

399

TAR Lazio, Roma, I, 13 March 2006, n° 1898

403

TAR Lazio, Roma, I, 30 March 2007, n° 2798

228 403

TAR Lazio, Roma, I, 29 December 2007, n° 14157

403

516  Table of Cases TAR Lazio, Roma, I, 24 August 2010, n° 31278

403

TAR Lazio, Roma, III ter, 11 February 2011, n° 1336, upheld on appeal: Cons. Stato, 23 May 2011, n° 3106

413

The Netherlands Dutch Competition Authority, no. 6207/476, Sandd v PostNL 225 United Kingdom Colgate v Bacheler, Cro Eliz 872, 78 Eng Rep 1097 (1596)

438, 439

Days Medical Aids Ltd v Pihsiang Machinery Manufacturing Co [2004] EWHC 44 (Comm)

354

Homer v Ashford (1825) 3 Bing. 322

438, 439

National Grid Plc v ABB Ltd & Others [2012] EWHC 869 (Ch)

333, 382, 383, 384

Nordenfelt v Maxim Nordenfelt Guns & Ammunition Company [1894] AC 535

438